« 2009年5月31日 - 2009年6月6日 | トップページ | 2009年6月14日 - 2009年6月20日 »

2009年6月13日 (土)

[Claims] 電気機械スイッチのためのトーションバネおよびトーションバネを内蔵したカンチレバー型RFマイクロ電気機械スイッチ

【特許請求の範囲】
【請求項1】
電気機械スイッチのためのトーションバネであって、前記電気機械スイッチは自由端および入力線と出力線を有する電機子を含んでおり、前記トーションバネは、
前記スイッチの前記電機子の前記自由端から延びる少なくとも一つの歯部を含む一連の歯部と、
前記歯部によって回転的に懸架され、導電伝送路を含み、少なくとも一部は電気接触に露出される末端部と、
を含み、
前記導電伝送路の長さは、前記電気機械スイッチが閉鎖位置に移動させられた時に、前記末端部が前記歯部を介して回転して前記導電伝送路と前記スイッチの前記入力線と出力線との間に正角接触を形成することによって前記導電伝送路の前記露出部分が前記電気機械スイッチの前記入力線と出力線との間に回路を形成するように選択されることを特徴とするトーションバネ。
【請求項2】
導電伝送路の電気接触に露出される部分は複数の凹部の形状で提供され、該凹部は入力部と出力部との間に形成される接触にそれぞれ対応し、それによって前記凹部は末端部の回転特性と共に導電伝送路と入力線と出力線との間の正角接触を提供し、その間に回路を形成することを特徴とする請求項1記載のトーションバネ。
【請求項3】
歯部は窒化ケイ素、タイプIII~V半導体および二酸化ケイ素で成る群から選択される素材で構成されることを特徴とする請求項2記載のトーションバネ。
【請求項4】
導電伝送路はチタニウム接着層、金導体層およびその間の抗拡散浸透層によって形成されることを特徴とする請求項3記載のトーションバネ。
【請求項5】
導電伝送路はチタニウム接着層、金導体層およびその間の抗拡散浸透層によって形成されることを特徴とする請求項1記載のトーションバネ。
【請求項6】
導電伝送路の電気接触に露出される部分は複数の凹部の形状で提供され、該凹部は入力部と出力部との間の接触にそれぞれ対応し、それによって前記凹部は末端部の回転特性と連携して前記導電伝送路と前記入力部と出力部との間で正角接触を提供し、その間で回路を形成することを特徴とする請求項5記載のトーションバネ。
【請求項7】
歯部は窒化ケイ素、タイプIII~V半導体および二酸化ケイ素で成る群から選択される素材で構成されることを特徴とする請求項1記載のトーションバネ。
【請求項8】
導電伝送路はチタニウム接着層および金導体層およびその間の拡散浸透層によって形成される素材で構成されることを特徴とする請求項7記載のトーションバネ。
【請求項9】
一連の歯部は複数の歯部を含むことを特徴とする請求項1記載のトーションバネ。
【請求項10】
一連の歯部は2つの歯部を含むことを特徴とする請求項1記載のトーションバネ。
【請求項11】
電気機械スイッチであって、
基板と、
前記基板上部の上の入力線と、
前記基板上部の上にあり、前記入力線と分離された出力線と、
前記基板上部の上にあり、前記入力線および前記出力線の近くに配置されているが分離されている基板電極と、
電機子と、
を含む電気機械スイッチであって、
前記電機子は、
前記基板と機械的に接続された第一端部および少なくとも一つの歯部を含む一連の歯部およびその間に回転式に懸架された末端部を有する第二端部を含んでおり、該末端部は前記入力線および前記出力線の上に位置する導電伝送路を含んでおり、該導電伝送路の少なくとも一部は前記入力線と前記出力線との間の正角接触のために露出されている第一梁型構造体層と、
前記基板電極の真上に位置し、前記電機子上に懸架される電機子電極と、
電機子電極と基板電極との間に配置され、その間の短絡を防ぐ末端部と、
を含んでおり、
それによってスイッチが閉鎖位置に作動された場合に前記末端部が自由に回転して前記導電伝送路の前記露出部分と前記入力線と出力線との間の正角接触を確実にし、その間に回路を形成して通電を可能にすることを特徴とする電気機械スイッチ。
【請求項12】
導電伝送路の電気接触に露出される部分は複数の凹部で提供され、該凹部は入力部と出力部との接触にそれぞれ対応し、それによって前記凹部は末端部の回転特性と連携して前記導電伝送路と前記入力部と出力部との間で正角接触を提供し、その間で回路を形成することを特徴とする請求項11記載の電気機械スイッチ。
【請求項13】
歯部は窒化ケイ素、タイプIII~V半導体および二酸化ケイ素で成る群から選択される素材で構成されることを特徴とする請求項12記載の電気機械スイッチ。
【請求項14】
導電伝送路はチタニウム接着層、金導体層およびその間の抗拡散浸透層によって形成されることを特徴とする請求項13記載の電気機械スイッチ。
【請求項15】
導電伝送路はチタニウム接着層、金導体層およびその間の抗拡散浸透層によって形成されることを特徴とする請求項11記載の電気機械スイッチ。
【請求項16】
導電伝送路の電気接触に露出される部分は複数の凹部で提供され、該凹部は入力部と出力部との接触にそれぞれ対応し、それによって前記凹部は末端部の回転特性と連携して前記導電伝送路と前記入力部と出力部との間で正角接触を提供し、その間で回路を形成することを特徴とする請求項15記載の電気機械スイッチ。
【請求項17】
歯部は窒化ケイ素、タイプIII~V半導体および二酸化ケイ素で成る群から選択される素材で構成されることを特徴とする請求項11記載の電気機械スイッチ。
【請求項18】
導電伝送路はチタニウム接着層、金導体層およびその間の抗拡散浸透層によって形成されることを特徴とする請求項17記載の電気機械スイッチ。
【請求項19】
第一梁型構造体層は電機子電極の上方に配置されることを特徴とする請求項11記載の電気機械スイッチ。
【請求項20】
絶縁層は第二梁型構造体層として電機子電極の下に形成され、第一梁型構造体層および前記第二梁型構造体層を構成する素材はそれらの機械的および熱的特性がスイッチ起動時における湾曲の望む量を提供するように選択されることを特徴とする請求項19記載の電気機械スイッチ。
【請求項21】
導電伝送路の電気接触に露出される部分は複数の凹部で提供され、該凹部は入力部と出力部との接触にそれぞれ対応し、それによって前記凹部は末端部の回転特性と連携して前記導電伝送路と前記入力部と出力部との間で正角接触を提供し、その間で回路を形成することを特徴とする請求項20記載の電気機械スイッチ。
【請求項22】
歯部は第一あるいは第二梁型構造体層のどちらかと同じ素材で提供され、前記梁型構造体層は窒化ケイ素、タイプIII~V半導体および二酸化ケイ素で成る群から選択される素材で形成されることを特徴とする請求項21記載の電気機械スイッチ。
【請求項23】
導電伝送路はチタニウム接着層、金導体層およびその間の抗拡散浸透層によって形成されることを特徴とする請求項22記載の電気機械スイッチ。
【請求項24】
導電伝送路はチタニウム接着層、金導体層およびその間の抗拡散浸透層によって形成されることを特徴とする請求項23記載の電気機械スイッチ。
【請求項25】
導電伝送路の電気接触に露出される部分は複数の凹部で提供され、該凹部は入力部と出力部との接触にそれぞれ対応し、それによって前記凹部は末端部の回転特性と連携して前記導電伝送路と前記入力部と出力部との間で正角接触を提供し、その間で回路を形成することを特徴とする請求項24記載の電気機械スイッチ。
【請求項26】
歯部は窒化ケイ素、タイプIII~V半導体および二酸化ケイ素で成る群から選択される素材で形成されることを特徴とする請求項25記載の電気機械スイッチ。
【請求項27】
導電伝送路はチタニウム接着層、金導体層およびその間の抗拡散浸透層によって形成されることを特徴とする請求項26記載の電気機械スイッチ。
【請求項28】
導電伝送路の電気接触に露出される部分は複数の凹部で提供され、該凹部は入力部と出力部との接触にそれぞれ対応し、それによって前記凹部は末端部の回転特性と連携して前記導電伝送路と前記入力部と出力部との間で正角接触を提供し、その間で回路を形成することを特徴とする請求項19記載の電気機械スイッチ。
【請求項29】
歯部は第一あるいは第二梁型構造体層のどちらかと同じ素材で形成され、それぞれは窒化ケイ素、タイプIII~V半導体および二酸化ケイ素で成る群から選択される素材で形成されることを特徴とする請求項28記載の電気機械スイッチ。
【請求項30】
導電伝送路はチタニウム接着層、金導体層およびその間の抗拡散浸透層によって形成されることを特徴とする請求項29記載の電気機械スイッチ。
【請求項31】
導電伝送路はチタニウム接着層、金導体層およびその間の抗拡散浸透層によって形成されることを特徴とする請求項19記載の電気機械スイッチ。
【請求項32】
導電伝送路の電気接触に露出される部分は複数の凹部で提供され、該凹部は入力部と出力部との接触にそれぞれ対応し、それによって前記凹部は末端部の回転特性と連携して前記導電伝送路と前記入力部と出力部との間で正角接触を提供し、その間で回路を形成することを特徴とする請求項31記載の電気機械スイッチ。
【請求項33】
歯部は窒化ケイ素、タイプIII~V半導体および二酸化ケイ素で成る群から選択される素材で形成されることを特徴とする請求項19記載の電気機械スイッチ。
【請求項34】
導電伝送路はチタニウム接着層、金導体層およびその間の抗拡散浸透層によって形成されることを特徴とする請求項33記載の電気機械スイッチ。
【請求項35】
絶縁層は基板電極上の層として形成されることを特徴とする請求項19記載の電気機械スイッチ。
【請求項36】
導電伝送路の電気接触に露出される部分は複数の凹部で提供され、該凹部は入力部と出力部との接触にそれぞれ対応し、それによって前記凹部は末端部の回転特性と連携して前記導電伝送路と前記入力部と出力部との間で正角接触を提供し、その間で回路を形成することを特徴とする請求項35記載の電気機械スイッチ。
【請求項37】
歯部は第一梁型構造体層と同じ素材で構成され、該第一梁型構造体層は窒化ケイ素、タイプIII~V半導体および二酸化ケイ素で成る群から選択される素材で形成されることを特徴とする請求項36記載の電気機械スイッチ。
【請求項38】
導電伝送路はチタニウム接着層、金導体層およびその間の抗拡散浸透層によって形成されることを特徴とする請求項37記載の電気機械スイッチ。
【請求項39】
導電伝送路はチタニウム接着層、金導体層およびその間の抗拡散浸透層によって形成されることを特徴とする請求項35記載の電気機械スイッチ。
【請求項40】
導電伝送路の電気接触に露出される部分は複数の凹部で提供され、該凹部は入力部と出力部との接触にそれぞれ対応し、それによって前記凹部は末端部の回転特性と連携して前記導電伝送路と前記入力部と出力部との間で正角接触を提供し、その間で回路を形成することを特徴とする請求項39記載の電気機械スイッチ。
【請求項41】
歯部は窒化ケイ素、タイプIII~V半導体および二酸化ケイ素で成る群から選択される素材で形成されることを特徴とする請求項35記載の電気機械スイッチ。
【請求項42】
導電伝送路はチタニウム接着層、金導体層およびその間の抗拡散浸透層によって形成されることを特徴とする請求項41記載の電気機械スイッチ。
【請求項43】
一連の歯部は複数の歯部を含むことを特徴とする請求項11記載の電気機械スイッチ。
【請求項44】
一連の歯部は2つの歯部を含むことを特徴とする請求項11記載の電気機械スイッチ。
【請求項45】
トーションバネを有する電気機械スイッチのための電機子であって、
前記電気機械スイッチの基板に機械的に接続するための第一端部と、少なくともひとつの歯部を有する一連の歯部と、前記歯部間に回転式に懸架された末端部とを有しており、該末端部は前記入力線および前記出力線の上に位置する導電伝送路を含んでおり、該導電伝送路の少なくとも一部は前記入力線と前記出力線との間の正角接触のために露出されている第一梁型構造体層と、
前記電気機械スイッチの基板電極の真上に位置する電機子電極と、
前記電機子電極と前記基板電極との間に位置し、前記電機子が前記電気機械スイッチに取り付けられ起動される時にその間で短絡するのを防止する絶縁層と、
を含み、
ゆえに前記電機子が前記電気機械スイッチに取り付けられて“閉鎖位置”に移動されると、前記末端部は自由に回転して前記導電線の前記露出部分と前記入力線と出力線との間で正角接触を提供して回路を形成し、通電を提供することを特徴とする電機子。
【請求項46】
絶縁層は電機子電極の下の第二梁型構造体層として形成され、第一梁型構造体層および第二梁型構造体層はそれらの機械的および熱的特性がスイッチ起動時に望む量の湾曲を提供するように選択される素材で形成されることを特徴とする請求項45記載の電機子。
【請求項47】
導電伝送路の電気接触に露出される部分は複数の凹部で提供され、該凹部は入力部と出力部との接触にそれぞれ対応し、それによって前記凹部は末端部の回転特性と連携して前記導電伝送路と前記入力部と出力部との間で正角接触を提供し、その間で回路を形成することを特徴とする請求項46記載の電機子。
【請求項48】
歯部は窒化ケイ素、タイプIII~V半導体および二酸化ケイ素で成る群から選択される素材で形成されることを特徴とする請求項47記載の電機子。
【請求項49】
導電伝送路はチタニウム接着層および金導体層およびその間の抗拡散浸透層で形成されることを特徴とする請求項48記載の電機子。
【請求項50】
導電伝送路はチタニウム接着層および金導体層およびその間の抗拡散浸透層で形成されることを特徴とする請求項46記載の電機子。
【請求項51】
導電伝送路の電気接触に露出される部分は複数の凹部で提供され、該凹部は入力部と出力部との接触にそれぞれ対応し、それによって前記凹部は末端部の回転特性と連携して前記導電伝送路と前記入力部と出力部との間で正角接触を提供し、その間で回路を形成することを特徴とする請求項50記載の電機子。
【請求項52】
歯部は窒化ケイ素、タイプIII~V半導体および二酸化ケイ素で成る群から選択される素材で形成されることを特徴とする請求項46記載の電機子。
【請求項53】
導電伝送路はチタニウム接着層および金導体層およびその間の抗拡散浸透層で形成されることを特徴とする請求項52記載の電機子。
【請求項54】
導電伝送路の電気接触に露出される部分は複数の凹部で提供され、該凹部は入力部と出力部との接触にそれぞれ対応し、それによって前記凹部は末端部の回転特性と連携して前記導電伝送路と前記入力部と出力部との間で正角接触を提供し、その間で回路を形成することを特徴とする請求項45記載の電機子。
【請求項55】
歯部は第一梁型構造体層と同じ素材で形成され、梁型構造体層はそれぞれ窒化ケイ素、タイプIII~V半導体および二酸化ケイ素で成る群から選択される素材で構成されることを特徴とする請求項54記載の電機子。
【請求項56】
導電伝送路はチタニウム接着層および金導体層およびその間の抗拡散浸透層で形成されることを特徴とする請求項55記載の電機子。
【請求項57】
導電伝送路はチタニウム接着層および金導体層およびその間の抗拡散浸透層で形成されることを特徴とする請求項45記載の電機子。
【請求項58】
導電伝送路の電気接触に露出される部分は複数の凹部で提供され、該凹部は入力部と出力部との接触にそれぞれ対応し、それによって前記凹部は末端部の回転特性と連携して前記導電伝送路と前記入力部と出力部との間で正角接触を提供し、その間で回路を形成することを特徴とする請求項57記載の電機子。
【請求項59】
歯部は窒化ケイ素、タイプIII~V半導体および二酸化ケイ素で成る群から選択される素材で形成されることを特徴とする請求項45記載の電機子。
【請求項60】
導電伝送路はチタニウム接着層および金導体層およびその間の抗拡散浸透層で形成されることを特徴とする請求項59記載の電機子。
【請求項61】
絶縁層は基材電極上の層として提供されることを特徴とする請求項45記載の電機子。
【請求項62】
導電伝送路の電気接触に露出される部分は複数の凹部で提供され、該凹部は入力部と出力部との接触にそれぞれ対応し、それによって前記凹部は末端部の回転特性と連携して前記導電伝送路と前記入力部と出力部との間で正角接触を提供し、その間で回路を形成することを特徴とする請求項61記載の電機子。
【請求項63】
歯部は第一梁型構造体層と同じ素材で形成され、梁型構造体層はそれぞれ窒化ケイ素、タイプIII~V半導体および二酸化ケイ素で成る群から選択される素材で形成されることを特徴とする請求項62記載の電機子。
【請求項64】
導電伝送路はチタニウム接着層および金導体層およびその間の抗拡散浸透層で形成されることを特徴とする請求項63記載の電機子。
【請求項65】
導電伝送路はチタニウム接着層および金導体層およびその間の抗拡散浸透層で形成されることを特徴とする請求項61記載の電機子。
【請求項66】
導電伝送路の電気接触に露出される部分は複数の凹部で提供され、該凹部は入力部と出力部との接触にそれぞれ対応し、それによって前記凹部は末端部の回転特性と連携して前記導電伝送路と前記入力部と出力部との間で正角接触を提供し、その間で回路を形成することを特徴とする請求項65記載の電機子。
【請求項67】
歯部は窒化ケイ素、タイプIII~V半導体および二酸化ケイ素で成る群から選択される素材で形成されることを特徴とする請求項67記載の電機子。
【請求項68】
導電伝送路はチタニウム接着層および金導体層およびその間の抗拡散浸透層で形成されることを特徴とする請求項67記載の電機子。
【請求項69】
一連の歯部は複数の歯部を含むことを特徴とする請求項45記載の電機子。
【請求項70】
一連の歯部は2つの歯部を含むことを特徴とする請求項45記載の電機子。

1. A torsion spring for an electro-mechanical switch, where the electro-mechanical switch includes an armature having a free end and an input line and an output line, the torsion spring comprising: a set of tines including at least one tine extending from the free end of the armature of the switch; a terminus portion rotatably suspended from the tines, the terminus portion including a conducting transmission line, at least a portion of which is exposed for electrical contact, the conducting transmission line having a length selected such that the exposed portion of the transmission line forms a circuit between the input and output lines of the electro-mechanical switch when the electro-mechanical switch is urged into a closed position, with the terminus portion rotating via the tines to form a conformal connection between the exposed portion of the conducting transmission line and the input and output lines of the switch.

2. A torsion spring for an electro-mechanical switch as set forth in claim 1, wherein the portion of the conducting transmission line exposed for electrical contact is in the form of a plurality of dimples, with a dimple corresponding to the contact to be made between the input and the output, respectively; whereby the dimples combined with the rotatable nature of the terminus portion provide a conformal contact between the conducting transmission line and the input and the output to form a circuit therebetween.

3. A torsion spring for an electro-mechanical switch as set forth in claim 2, wherein the tines are constructed of a material selected from a group consisting of silicon nitride, Type III-V semiconductor materials, and silicon dioxide.

4. A torsion spring for an electro-mechanical switch as set forth in claim 3, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween and an anti- diffusion layer therebetween. 5. A torsion spring for an electro-mechanical switch as set forth in claim 1, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

6. A torsion spring for an electro-mechanical switch as set forth in claim 5, wherein the portion of the conducting transmission line exposed for electrical contact is in the form of a plurality of dimples, with a dimple corresponding to the contact to be made between the input and the output, respectively; whereby the dimples combined with the rotatable nature of the terminus portion provide a conformal contact between the conducting transmission line and the input and the output to form a circuit therebetween.

7. A torsion spring for an electro-mechanical switch as set forth in claim 1, wherein the tines are constructed of a material selected from a group consisting of silicon nitride, Type III-V semiconductor materials, and silicon dioxide.

8. A torsion spring for an electro-mechanical switch as set forth in claim 7, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

9. A torsion spring for an electro-mechanical switch as set forth in claim 1, wherein the set of tines includes a plurality of tines.

10. A torsion spring for an electro-mechanical switch as set forth in claim 1, wherein the set of tines includes two tines.

11. A micro-electro-mechanical switch comprising: a substrate; an input line on top of the substrate; an output line on top of the substrate and separated from the input line; a substrate electrode on top of the substrate, located near but separated from the input line and the output line; and an armature comprising: a first beam structural layer having a first end mechanically connected with the substrate and a second end including a set of tines including at least one tine with a terminus portion rotatably suspended therefrom, with the terminus portion including a conducting transmission line positioned over the input and output lines, with the at least a portion of the conducting transmission line exposed for conformal contact with the input and output lines; and an armature electrode positioned directly above the substrate electrode and suspended on the armature; an insulating layer positioned between the armature electrode and the substrate electrode to prevent short-circuiting therebetween; whereby when the switch is actuated into a "closed"position, the terminus portionf is free to rotate to ensure a conformal contact between the exposed portion of the conducting transmission line and the input and output lines in order to form a circuit therebetween to permit the flow of electricity.

12. A micro-electro-mechanical switch as set forth in claim 11, wherein the portion of the conducting transmission line exposed for electrical contact is in the form of a plurality of dimples, with a dimple corresponding to the contact to be made between the input and the output, respectively; whereby the dimples combined with the rotatable nature of the terminus portion provide a conformal contact between the conducting transmission line and the input and the output to form a circuit therebetween.

13. A micro-electro-mechanical switch as set forth in claim 12, wherein the tines are constructed of a material selected from a group consisting of silicon nitride, Type III-V semiconductor materials, and silicon dioxide.

14. A micro-electro-mechanical switch as set forth in claim 13, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

15. A micro-electro-mechanical switch as set forth in claim 11, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

16. A micro-electro-mechanical switch as set forth in claim 15, wherein the portion of the conducting transmission line exposed for electrical contact is in the form of a plurality of dimples, with a dimple corresponding to the contact to be made between the input and the output, respectively; whereby the dimples combined with the rotatable nature of the terminus portion provide a conformal contact between the conducting transmission line and the input and the output to form a circuit therebetween.

17. A micro-electro-mechanical switch as set forth in claim 11, wherein the tines are constructed of a material selected from a group consisting of silicon nitride, Type III-V semiconductor materials, and silicon dioxide.

18. A micro-electro-mechanical switch as set forth in claim 17, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

19. A micro-electro-mechanical switch as set forth in claim 11, wherein the first beam structural layer is positioned above the armature electrode.

20. A micro-electro-mechanical switch as set forth in claim 19, wherein the insulating layer is formed as a second beam structural layer under the armature electrode, with the first and the second beam structural layers formed of materials selected such that their mechanical and thermal properties provide a desired amount of bowing when the switch is activated.

21. A micro-electro-mechanical switch as set forth in claim 20, wherein the portion of the conducting transmission line exposed for electrical contact is in the form of a plurality of dimples, with a dimple corresponding to the contact to be made between the input and the output, respectively; whereby the dimples combined with the rotatable nature of the terminus portion provide a conformal contact between the conducting transmission line and the input and the output to form a circuit therebetween.

22. A micro-electro-mechanical switch as set forth in claim 21, wherein the tines are formed of the same material as either the first or the second beam structural layer and the beam structural layers are constructed of a material selected from a group consisting of silicon nitride, Type III-V semiconductor materials, and silicon dioxide.

23. A micro-electro-mechanical switch as set forth in claim 22, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

24. A micro-electro-mechanical switch as set forth in claim 23, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

25. A micro-electro-mechanical switch as set forth in claim 24, wherein the portion of the conducting transmission line exposed for electrical contact is in the form of a plurality of dimples, with a dimple corresponding to the contact to be made between the input and the output, respectively; whereby the dimples combined with the rotatable nature of the terminus portion provide a conformal contact between the conducting transmission line and the input and the output to form a circuit therebetween.

26. A micro-electro-mechanical switch as set forth in claim 25, wherein the tines are constructed of a material selected from a group consisting of silicon nitride, Type III-V semiconductor materials, and silicon dioxide.

27. A micro-electro-mechanical switch as set forth in claim 26, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

28. A micro-electro-mechanical switch as set forth in claim 19, wherein the portion of the conducting transmission line exposed for electrical contact is in the form of a plurality of dimples, with a dimple corresponding to the contact to be made between the input and the output, respectively; whereby the dimples combined with the rotatable nature of the terminus portion provide a conformal contact between the conducting transmission line and the input and the output to form a circuit therebetween.

29. A micro-electro-mechanical switch as set forth in claim 28, wherein the tines are formed of the same material as the first beam structural layer and the beam structural layers are each constructed of a material selected from a group consisting of silicon nitride, Type III-V semiconductor materials, and silicon dioxide.

30. A micro-electro-mechanical switch as set forth in claim 29, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

31. A micro-electro-mechanical switch as set forth in claim 19, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

32. A micro-electro-mechanical switch as set forth in claim 31, wherein the portion of the conducting transmission line exposed for electrical contact is in the form of a plurality of dimples, with a dimple corresponding to the contact to be made between the input and the output, respectively; whereby the dimples combined with the rotatable nature of the terminus portion provide a conformal contact between the conducting transmission line and the input and the output to form a circuit therebetween.

33. A micro-electro-mechanical switch as set forth in claim 19, wherein the tines are constructed of a material selected from a group consisting of silicon nitride, Type III-V semiconductor materials, and silicon dioxide.

34. A micro-electro-mechanical switch as set forth in claim 33, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

35. A micro-electro-mechanical switch as set forth in claim 19, wherein the insulating layer is formed as a layer on the substrate electrode.

36. A micro-electro-mechanical switch as set forth in claim 35, wherein the portion of the conducting transmission line exposed for electrical contact is in the form of a plurality of dimples, with a dimple corresponding to the contact to be made between the input and the output, respectively; whereby the dimples combined with the rotatable nature of the terminus portion provide a conformal contact between the conducting transmission line and the input and the output to form a circuit therebetween.

37. A micro-electro-mechanical switch as set forth in claim 36, wherein the tines are formed of the same material as the first beam structural layer and the beam structural layers are each constructed of a material selected from a group consisting of silicon nitride, Type III-V semiconductor materials, and silicon dioxide.

38. A micro-electro-mechanical switch as set forth in claim 37, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

39. A micro-electro-mechanical switch as set forth in claim 35, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

40. A micro-electro-mechanical switch as set forth in claim 39, wherein the portion of the conducting transmission line exposed for electrical contact is in the form of a plurality of dimples, with a dimple corresponding to the contact to be made between the input and the output, respectively; whereby the dimples combined with the rotatable nature of the terminus portion provide a conformal contact between the conducting transmission line and the input and the output to form a circuit therebetween.

41. A micro-electro-mechanical switch as set forth in claim 35, wherein the tines are constructed of a material selected from a group consisting of silicon nitride, Type III-V semiconductor materials, and silicon dioxide.

42. A micro-electro-mechanical switch as set forth in claim 41, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

43. A torsion spring for a micro-electro-mechanical switch as set forth in claim 11, wherein the set of tines includes a plurality of tines.

44. A torsion spring for a micro-electro-mechanical switch as set forth in claim 11, wherein the set of tines includes two tines.

45. An armature for a micro-electro-mechanical switch having a torsion spring, the armature comprising: a first beam structural layer having a first end for mechanically connecting with a substrate of a micro-electro-mechanical switch and a second end including a set of tines including at least one tine with a terminus portion rotatably suspended therefrom, with the terminus portion including a conducting transmission line configured to be positioned over the input and output lines of a micro-electro-mechanical switch, with the at least a portion of the conducting transmission line exposed for conformal contact with the input and output lines; and an armature electrode positioned directly above a substrate electrode of the micro-electro-mechanical switch and suspended on the armature; an insulating layer positioned between the armature electrode and the substrate electrode to prevent short-circuiting therebetween when the armature is assembled in a micro-electro-mechanical switch and actuated; whereby when the armature is assembled in a micro-electro-mechanical switch is actuated into a "closed"position, the terminus is free to rotate to ensure a conformal contact between the exposed portion of the conducting transmission line and the input and output lines in order to form a circuit therebetween to permit the flow of electricity.

46. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 45, wherein the insulating layer is formed as a second beam structural layer under the armature electrode, with the first and the second beam structural layers formed of materials selected such that their mechanical and thermal properties provide a desired amount of bowing when the switch is activated.

47. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 46, wherein the portion of the conducting transmission line exposed for electrical contact is in the form of a plurality of dimples, with a dimple corresponding to the contact to be made between the input and the output, respectively; whereby the dimples combined with the rotatable nature of the terminus portion provide a conformal contact between the conducting transmission line and the input and the output to form a circuit therebetween.

48. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 47, wherein the tines are constructed of a material selected from a group consisting of silicon nitride, Type III-V semiconductor materials, and silicon dioxide. 49. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 48, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

50. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 46, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

51. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 50, wherein the portion of the conducting transmission line exposed for electrical contact is in the form of a plurality of dimples, with a dimple corresponding to the contact to be made between the input and the output, respectively; whereby the dimples combined with the rotatable nature of the terminus portion provide a conformal contact between the conducting transmission line and the input and the output to form a circuit therebetween.

52. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 46, wherein the tines are constructed of a material selected from a group consisting of silicon nitride, Type III-V semiconductor materials, and silicon dioxide.

53. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 52, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

54. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 45, wherein the portion of the conducting transmission line exposed for electrical contact is in the form of a plurality of dimples, with a dimple corresponding to the contact to be made between the input and the output, respectively; whereby the dimples combined with the rotatable nature of the terminus portion provide a conformal contact between the conducting transmission line and the input and the output to form a circuit therebetween.

55. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 54, wherein the tines are formed of the same material as the first beam structural layer and the beam structural layers are each constructed of a material selected from a group consisting of silicon nitride, Type III-V semiconductor materials, and silicon dioxide.

56. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 55, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

57. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 45, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

58. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 57, wherein the portion of the conducting transmission line exposed for electrical contact is in the form of a plurality of dimples, with a dimple corresponding to the contact to be made between the input and the output, respectively; whereby the dimples combined with the rotatable nature of the terminus portion provide a conformal contact between the conducting transmission line and the input and the output to form a circuit therebetween.

59. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 45, wherein the tines are constructed of a material selected from a group consisting of silicon nitride, Type III-V semiconductor materials, and silicon dioxide. 60. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 59, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

61. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 45, wherein the insulating layer is formed as a layer on the substrate electrode.

62. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 61, wherein the portion of the conducting transmission line exposed for electrical contact is in the form of a plurality of dimples, with a dimple corresponding to the contact to be made between the input and the output, respectively; whereby the dimples combined with the rotatable nature of the terminus portion provide a conformal contact between the conducting transmission line and the input and the output to form a circuit therebetween.

63. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 62, wherein the tines are formed of the same material as the first beam structural layer and the beam structural layers are each constructed of a material selected from a group consisting of silicon nitride, Type III-V semiconductor materials, and silicon dioxide.

64. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 63, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

65. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 61, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

66. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 65, wherein the portion of the conducting transmission line exposed for electrical contact is in the form of a plurality of dimples, with a dimple corresponding to the contact to be made between the input and the output, respectively; whereby the dimples combined with the rotatable nature of the terminus portion provide a conformal contact between the conducting transmission line and the input and the output to form a circuit therebetween.

67. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 61, wherein the tines are constructed of a material selected from a group consisting of silicon nitride, Type III-V semiconductor materials, and silicon dioxide.

68. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 67, wherein the conducting transmission line is formed from a titanium adhesive layer and a gold conductor layer and an anti-diffusion layer therebetween.

69. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 45, wherein the set of tines includes a plurality of tines.

70. An armature for a micro-electro-mechanical switch having a torsion spring as set forth in claim 45, wherein the set of tines includes two tines.

「特表2004-535654およびWO2002073645より引用」

|
|

2009年6月12日 (金)

電気機械スイッチのためのトーションバネおよびトーションバネを内蔵したカンチレバー型RFマイクロ電気機械スイッチ

【発明の詳細な説明】
【優先権】
【0001】
本出願は、2001年3月12日に米国特許商標庁に出願された米国仮出願第60/275,171号の優先権を主張する。
【技術背景】
【0002】
(a)技術分野
本発明は、通常は電気機械スイッチに関し、より具体的にはマイクロ電気機械スイッチ(MEMS)に関し、最も具体的には高パワーRF MEMSに関する。
(b)関連技術の説明
通信関係の利用法では、スイッチはトランジスタまたはピンダイオード等の半導体素子を用いて構成されることが多い。しかしながら、マイクロ波周波数においてこれらの素子はいくつかの欠点を有する。ピンダイオードおよびトランジスタは、典型的には1dB以上の挿入損失を有し、これはスイッチ閉鎖時にスイッチを通る損失である。マイクロ波周波数で作動するトランジスタは、20dB未満のアイソレーション値を有する傾向にある。
【0003】
ゆえに、スイッチ開放時でさえ信号はスイッチを通って流れる。ピンダイオードおよびトランジスタは制限周波数応答を有し、典型的には20GHz未満の周波数のみに応答する。さらに、これらのスイッチの挿入損失および高アイソレーション値は、スイッチを通過する信号の周波数に応じて変化する。これらの特徴のため、半導体トランジスタおよびピンダイオードはマイクロ波を使用するスイッチの選択肢としては向かない。
【0004】
米国特許第5,121,089号(Larson)は、新規な部類のマイクロ波スイッチ、マイクロ電気機械(MEM)スイッチを開示している。MEMスイッチは、非常に低い挿入損失(45GHzで0.2dB未満)および開放時の高アイソレーション値(30dB以上)を有する。さらに、スイッチは半導体トランジスタおよびピンダイオードより大きい周波数応答およびバンド幅を有する。これらの特徴によって、MEMスイッチがマイクロ波回路内で従来の狭いバンド幅のPINダイオードおよびトランジスタに置き換わる可能性が提供される。
【0005】
LarsonのMEMスイッチは、電機子構造を使用している。金属電機子の1端部は出力線に固定され、電機子のもう1つの端部は入力線の上方に位置する。電機子はスイッチが開放位置にある時は入力線から電気的に絶縁される。電機子の下方の電極に電圧が加えられると、電機子は下方に引かれて入力線に接触する。これにより入力線と出力線との間に金属電機子を介して導電経路が提供される。
【0006】
しかしながら、前述の通常タイプのマイクロ電気機械スイッチは初期故障が多い。初期故障の原因は、電機子接点と基板接点との接触による衝撃から生じる損傷に関する。従来のMEMスイッチは一定角度で基板接触面に接する電機子接点を有するため、損傷は悪化される。斜め衝突によって全衝突エネルギーは比較的小さい領域に伝達されるため、単位面積ごとの衝撃増加、および実際の接触の小領域を通る電流密度の増加による抵抗加熱から生じる熱の両方によって最終的には初期故障が引き起こされる。
【発明の概要】
【0007】
本発明は、接触板の間に実質的に正角接触を提供するように構成されるトーションバネを提供することによって上記を含む問題点を解決する。その結果のMEMSは向上した耐久力を持ち、実質的な正角接触がより良い電気接触を提供するために、熱は減少し、MEMはより多くのパワーに対応できる。ゆえに、耐久力が向上した多目的MEMスイッチが提供される。
【0008】
本発明の1実施例は電機子搭載トーションバネを提供する。トーションバネは、電機子電極が電気機械的に基板電極に接触させられるとそれらの電極が相互間の実質的正角状態を提供して接触域を最大化し、摩耗およびオーム抵抗を減少させるような十分な柔軟性を提供するように構成される。
【0009】
本発明は、金属同士の正角接触を提供するためにRF線がトーションスプリングを搭載した新規なRFMEMスイッチを提供する。金属同士の正角接触は最大接触域および最小接触抵抗を確実にし、ゆえに重要な長期信頼性および良好な熱放散を提供して改良型高パワー対応をもたらす。
【0010】
本発明は、別実施例中で電気機械スイッチにトーションバネを提供する。トーションバネは、少なくとも一つの歯部を有する一連の歯部を含む。一連の歯部はスイッチの電機子の自由端から延び、歯部の間に回転式に懸架された末端部を含む。末端部は導電伝送路を含み、少なくともその一部は電気接触のために露出される。導電伝送路の長さは、マイクロ電気機械スイッチが閉鎖位置へ移動させられると伝送路の露出部分がマイクロ電気機械スイッチの入力部と出力部との間で回路を形成するように選択される。スイッチが閉鎖すると末端部は歯部を介して回転し、導電伝送路の露出部分とスイッチの入力部と出力部との間で正角接続を形成し、その間の通電を最適化する。
【0011】
トーションバネの別実施例では、電気接触のために露出される導電伝送路の部分は複数の凹部で提供される。それぞれの凹部は、入力部および出力部との接触にそれぞれ対応する。凹部は末端部の回転特性と共に導電伝送路と入力部と出力部との間の正角接触を提供し、その間に回路を形成する。
【0012】
好適には、トーションバネは窒化ケイ素、タイプIII~Vの半導体および二酸化ケイ素で成る群から選択される素材で構成される。好適には、導電伝送路はチタニウム接着層と金導体層とで形成される。
【0013】
一連のトーションバネは、好適には複数の歯部を含み、より好適には2つの歯部を含む。
【0014】
別実施例では、本発明は互いに分離した入力線、出力線および基板電極を上部に有する基板を含むマイクロ電気機械スイッチを提供する。スイッチは、基板に機械接続された第一端部および少なくとも一つの歯部を有する一連の歯部を含む第二端部を持つ第一梁型構造体層を有する電機子をさらに含む。末端部は歯部の間に懸架され、入力部および出力部の上方に位置する導電伝送路を含む。導電伝送路の少なくとも一部は、入力線および出力線との正角接触のために露出される。電機子は基板電極の直接の上方に位置し、電機子上に懸架された電機子電極をさらに含む。絶縁層は電機子電極と基板電極との間に提供され、その間の短絡を妨げる。そのため、スイッチが閉鎖位置に移動させられると末端部は自由に回転して導電伝送路の露出部分と入力線と出力線との間に正角接触を形成し、その間で通電する回路を形成する。
【0015】
スイッチの電機子は、好適にはトーションバネ実施例に関して前述されたのと同様に増強させて改良される。
【0016】
マイクロ電気機械の別実施例では、絶縁層は第二梁型構造体層として電機子電極の下部に形成される。第一および第二梁型構造体層の構成素材は、スイッチ作動時にその機械的および熱的特性が望む量の湾曲を提供するように選択される。
【0017】
本発明の別実施例は、トーションバネを有するマイクロ電気機械スイッチに1電機子を提供する。電機子は、マイクロ電気機械スイッチの基板に機械接続した第一端部および少なくとも一つの歯部を有する一連の歯部を含む第二端部を有する梁型構造体層を含む。末端部は歯部の間に懸架されており、マイクロ電気機械スイッチの入力線と出力線の上方に位置するように構成された導電伝送路を含む。導電伝送路の少なくとも一部は、入力線と出力線との正角接触のために露出される。電機子電極はマイクロ電気機械スイッチの基板電極の直接的上方に位置して電機子上に懸架されており、絶縁層は電機子電極および基板電極との間に位置し、電機子がマイクロ電気機械スイッチに取り付けられて閉鎖位置に移動させられる時にその間での短絡を防ぐ。電機子がマイクロ電気機械スイッチに取り付けられ、閉鎖位置に移動させられると、末端部は自由に回転して導電伝送路と入力線と出力線との間の正角接触を確実にし、その間で通電する回路を形成する。
【0018】
好適には、電機子はトーションバネ実施例に関して前述されたのと同様に増強させて改良される。
【0019】
電機子の別実施例では、第二梁型構造体層として電機子電極の下に形成され、第一および第二梁型構造体層の構成素材は、その機械的および熱的特性がスイッチ作動時に望む量の湾曲を提供するように選択される。
【図面の簡単な説明】
【0020】
「発明の概要」で記述された利点を含む特徴と利点は、以下で図面を利用してより明確に詳述される。
【0021】
図1は、導電伝送路が入力線および出力線の上方に位置する本発明のMEMスイッチ別実施例の上面図である。
【0022】
図2Aは、図1のMEMスイッチの側面図である。
【0023】
図2Bは、閉鎖位置にある図2AのMEMスイッチを示す。
【0024】
図3Aおよび図3Bは、電機子を含む層順序が変更された図1のMEMスイッチ別実施例の側面図である。
【0025】
図4A~図4Fは、本発明の製造工程の進行ステップ中にある図1のMEMスイッチの側面図である。
【0026】
図5は、MEMスイッチを50オーム供給ラインに適合させるためにスイッチが分路電気容量と共に構成された本発明のMEMスイッチを示す。
【0027】
本発明は、通常は電子およびマイクロ電気機械スイッチ(MEMS)に関し、より具体的には高パワーRF MEMSでの改良型入力/出力部接触を提供するトーションバネに関する。以下の記述は図面を利用して提供され、通常技術者が本発明を特定利用法に組み込むために製造および使用することを可能にする。様々な改良、別実施例中での様々な使用法および本明細書中に提供される一般的な本質は、当業者にとって明らかなように、広範囲の実施例に適応させることができる。ゆえに、本発明は記述される実施例には制限されず、本質と矛盾しない最大範囲と本書中で開示される新規な特性に従う。さらに、明記されない限りは本書中の図面は寸法を特定されずに図示されており、本発明の概念の性質を表している。
【0028】
本発明はトーションバネによって電気機械スイッチ中の電機子接点と電極との間の正角接触を確実にし、MEMSとの関連で特定使用法を有する。本発明が提供する接触の正角特性は利用可能な接触域を最大化し、熱放散は改良される。米国特許第6,046,659号に開示されているRFスイッチのごとき既存の単純な片持ち梁タイプのRF MEMSスイッチは、スイッチを切るとエッジ接触を形成するという不利点を持つスイッチの1例である。この種の古いスイッチの接触域は、通常は100平方ミクロン以上の全凹部(接触板)サイズの約10平方ミクロンである。電極と凹部エッジとの接触による衝撃は、過剰な摩耗と初期故障を生じる。この過剰な摩耗は、スイッチが故障以前に対応できる使用回数を制限する主要素である。本発明は、従来型RFスイッチに比べて、スイッチのサイクル寿命を増長させ、パワー対応能力を大幅に増加させる。
【0029】
スイッチは、米国特許第6,046,659号に開示された工程を含む既存の製造工程によって製造できる。既存のRFMEMスイッチの信頼性を研究したところ、スイッチサイクル時間を制限する主要因は起動時の凹部縁部の性質であることが示された。エッジ接触が電極と接触させるのは凹部面の10%未満のみである。このような接触域の制限は、2つの問題をもたらす。第一に、より小さい接触域によってスイッチシステムに衝撃に関するより大きい損傷が常に提供され、衝撃の集中地点からの分離を引き起こす。第二に、より小さい接触域によって接続は劣化し、それによってジュール(抵抗)加熱は増加し、初期故障あるいは動作劣化の要因ともなる。本発明のより大きい接触域は優れた接触、より良い熱放散およびパワー対応をもたらし、同時に接点での衝撃による損傷を軽減する。
【0030】
本構成の改良された信頼性およびパワー対応能力は、回転の自由と共に凹部接触を提供するように構成されたトーションバネの追加によって提供される。試験結果は本構造が衝撃による初期故障の可能性を軽減し、正角凹部接触を確実にし、ゆえに接触抵抗を最小化することを示している。さらに、非常に拡大された接触域によって熱放散が改良されるため、この種のスイッチのパワー対応能力が向上される。
【0031】
図1は、本発明によるマイクロ電気機械(MEM)スイッチ10の実施例の上面図である。この実施例では、導電伝送路28は第一および第二梁型構造体層26と27の間に位置し、入力線20または出力線18のどちらにも直接接続されていない。電機子バイアス電極30は、電機子16下部の大部分および梁型構造体層26と27の間の領域を覆っている。電機子16の一端部は、基板14に直接固定される。電機子バイアス電極30の上方に位置する電機子16の自由端は分岐して、歯部40の間に回転式に懸架されて入力線18と出力線20との間の隙間内に位置する末端部42と共に一対の歯部40を形成する。導電伝送路28は電機子16の末端部42上に提供されるため、スイッチ10が閉鎖した時に入力線20と出力線18との間の隙間を埋めることができる。基板バイアス電極22と電機子バイアス電極30との間に電圧が加えられると、電機子16は基板14の方へ湾曲する。これにより伝送路28は入力線20および出力線18の両方に電気接触し、スイッチを閉鎖する。この状態で、信号は入力線18から出力線20へ導電伝送路28を介して通過できる。
【0032】
図2Aおよび図2Bは、図1のMEMスイッチ10の側面図である。図2Aは開放状態にあるMEMスイッチを示し、図2Bは閉鎖状態にあるMEMスイッチを示す。梁型構造物体26は、固定アンカーバイアス32を介して基板14と接続される。懸架された電機子バイアス電極30は層26と27の間の構造物体内に位置し、バイアス線38と電機子バイアスパッド34との間の接点を介して電気的にアクセスされる。導電伝送路28は梁型構造体層26と27の自由端に位置し、(誘電性)梁型構造体層26と27によって懸架状電機子バイアス電極30から電気的に絶縁される。導電伝送路28には一対の凹部24が提供されるため、スイッチ10が閉鎖すると凹部24と導電伝送路28は連携して入力線18と出力線20との間に導電経路を形成し、通電が可能になる。凹部24と基板14との間の隙間は、第一構造体層26と基板14の間の距離よりも小さい。伝送路28および凹部24は、好適には歯部40によって懸架された電機子16の末端部42に封入される。歯部40は、凹部24と入力線18および出力線20との間に正角接触を確実に提供するために末端部42を回転させるように十分に柔軟な電機子16の一部を提供する。歯部40の厚みと形状の変動によって望ましいバネ特性が提供される。
【0033】
さらに、歯部40と末端部42との間の連結の提供は、通電を最適化するために凹部24と入力線18および出力線20との間の正角接触を確実にする望む回転度および回転方向を提供できるように変更できる。さらに、特定実施例および望む回転自由度に応じて、使用する歯部の数は1から特定実施例に実用的な数まで変更できる。歯部40は正角接触を提供するために湾曲した後に元来の形状を取り戻すように十分に堅固である必要がある。歯部が過度に堅固である場合、スイッチ10にかかる閉鎖強度に応じて脆性破壊が生じる可能性がある。懸架状接触凹部24は第一梁型構造体層26を通ってその下方に延び、入力線18および出力線20それぞれに金属接触域を提供する。基板バイアス電極22は、基板14表面上の懸架状電機子バイアス電極30の下方に位置し、空隙44によってそれから分離されている。懸架状電機子バイアス電極30と基板バイアス電極22との間に電圧が加えられると、静電引力が電機子バイアス電極30を付着した電機子16と共に基板バイアス電極22の方へ引きつける。
【0034】
動作中、電機子バイアス電極30を基板バイアス電極22の方へ近づかせるように電機子16が湾曲した点において凹部24は最初に入力線18に接触する。その結果、懸架状接触凹部24は入力線20および出力線18に接触するため、導電伝送路28は入力線18および出力線20との間の隙間を埋めて、MEMスイッチ10を図2Bが示す位置へ閉鎖する。このように、凹部24と入力線18との間の金属接触の力は、電機子電極30を基板電極22に引きよせる引力ではなく、主に歯部40の柔軟性と凹部24の構造に応じる。
【0035】
接触凹部を持たない金属スイッチは、温度、年齢およびMEMスイッチの使用量につれて変化する要因である電機子柔軟性およびバイアス強度に従う接触を有することは記述に値する。信頼性の改良に加えて、接触自体の品質は凹部を追加することで改良される。凹部は環境よりも製造に応じる特徴である制御可能なサイズや表面構成を有するためである。ゆえに、凹部24を持たないMEMスイッチは時間に応じて変化する接触特性を持つことが多く、そのためにスイッチをいくつかの回路実施例に使用することを困難あるいは不可能にする。
【0036】
図3Aと図3Bは図1、図2Aおよび図2BのMEMS10別実施例の側面図である。この場合は、懸架状電機子バイアス電極30は電機子16下部の上の露出層である。懸架状電機子バイアス電極30の露出特性により、スイッチが閉鎖位置にある場合に電機子バイアス電極30と基板バイアス電極22が電気的に短絡することを防ぐために、追加的絶縁層48が基板バイアス電極22上部の上に提供される。
【0037】
図4A~図4Fは、図1および図2のMEMスイッチ10の製造に使用できる製造工程を示している。工程は基板14から開始する。好適実施例では、基板にはGaAs(ガリウムヒ素)を使用している。しかしながら、Inp、セラミック、クオーツ、あるいはケイ素のごとき別素材を使用できる。これによって、回路素子およびMEMスイッチ10は金属スパッタリングおよびマスキングのごとき標準集積回路製造技術を用いて基板上で同時に製造できる。基板は主に回路技術および接続されるMEMスイッチに基づいて選択されるため、MEMスイッチと回路は同時に製造できる。たとえば、InPは低ノイズHEMTMMICS(高電子移動トランジスタモノリシックマイクロ波集積回路)に使用でき、GaAsは典型的にPHEMT(疑似形態HEMT)パワーMMICSに使用できる。MEMスイッチ10製造は柔軟性を持つため、様々な回路への使用が可能である。これにより、本MEMスイッチを使用して構成される回路のコストおよび複雑性が軽減される。
【0038】
図4Aは、電機子バイアスパッド34用に基板14上に金属層を蒸着する第一ステップ後に、入力線18、出力線20および基板バイアス電極22が完成したMEMスイッチ10の側面図である。金属層は、抵抗リフトオフ、抵抗提供(resistdefinition)あるいは金属エッチングのような標準集積回路製造技術を使用してリトグラフィー式に蒸着できる。好適実施例では、金属層の主な組成として金(Au)が使用される。Auは、その低抵抗率のためにRF利用法では好適である。Auを基板に確実に接着させるために、チタニウム(Ti)の薄層(250~500オングストローム)、続いて約1000オングストロームのプラチナ(Pt)層、最後にAuを蒸着する。Ptは拡散隔膜(抗拡散浸透層)の働きをしてAuがTiと混合するのを防止し、金属が基板との接着力を失うのを防ぐ。特定実施例の要求に応じて別の抗拡散浸透層を使用できる。III~V半導体基板の場合には、金ゲルマニウム(AuGe)共晶の薄層を最初に蒸着して、III~V MESFETあるいはHEMTへの標準オーム金属加工に類似した半導体とAuGeの合金化によりAuの接着を確実にする。
【0039】
次に図4Bでは、犠牲層46が金属層(好適にはAu)の上部の上に配置されてエッチングされるため、電機子16は犠牲層46の上方に形成することができる。犠牲層46は、典型的にはスパッタ蒸着あるいはPECVD(プラズマ助長化学気相蒸着)を使用して蒸着された2ミクロンのSiO2を含む。バイアホール52は犠牲層42内に彫刻されるため、金属電機子バイアスパッド34は固定アンカーバイアス42の形成を開始するために露出される。バイアホール52は、犠牲層46の標準抵抗リトグラフィーおよびエッチングで提供される。SiO2以外の別素材を犠牲層46に使用できる。犠牲層46の重要な特徴は、高エッチング率、良好な厚さ均一度、および基板14に既に蒸着されている金属の酸化物による正角塗装である。酸化物の厚さは、スイッチ10を閉鎖するのに必要な電圧およびスイッチ10開放時の電気的絶縁を決定するのに重要なスイッチ開口部(溝44)の厚さを部分的に決定する。図4Fで示すように、犠牲層46は電機子16を開放するために最終ステップで取り除かれる。
【0040】
SiO2を犠牲層46に使用する別利点は、SiO2が高温に耐えられることである。有機ポリイミドのごとき犠牲層の別タイプは、高温下ではかなり堅くなる。ゆえに、ポリイミド犠牲層は後で取り除くことが難しい。犠牲層46は、梁型構造体層として窒化ケイ素が蒸着された際に高温に露出される。より低いBOEエッチング率を窒化ケイ素に提供するように、窒化ケイ素の蒸着には高温蒸着が望まれるためである。低BOEエッチング率は、SiO2がエッチング処理で除かれると失われる第一梁型構造体層26の量を最小化する。
【0041】
図4Cは第一梁型構造体層26の製造を示す。第一および第二梁型構造体層26と27は電機子16の支持機構を提供し、好適には窒化ケイ素で構成されるが、タイプIII~V半導体物質および二酸化ケイ素のごとき窒化ケイ素とは別の素材を使用できる。窒化ケイ素は蒸着することができ、それによって梁型構造体層26と27に中性ストレスが発生するために好適である。中性ストレス製造はスイッチ起動時に起こり得る湾曲を減少させる。第一梁型構造体層26に使用される素材は犠牲層46より低いエッチング率を持つ必要があり、それによって電機子16を開放するために犠牲層46が取り除かれる際に第一梁型構造体層26はエッチングで除去されない。第一梁型構造体層26は、標準リトグラフィーおよびエッチング方法を用いて模様付けおよびエッチングできる。
【0042】
図3Aおよび図3Bのように、梁型構造体層26は電機子バイアス電極30の下方あるいは上方のどちらかに形成できる。梁型構造体層26が電機子バイアス電極30の片側のみに提供される場合、スイッチ作動時に梁型構造体層26内のストレスが電機子バイアス電極30内のストレスと異なる時に、電機子16内で湾曲が生じる。電機子16はどちらの物質がより高ストレスであるかに応じて上方あるいは下方へ湾曲する。湾曲はスイッチ起動に必要な電圧を変化させることができ、極度に湾曲した場合は起動電圧に関わらずスイッチを開放(下方に湾曲)あるいは閉鎖(上方に湾曲)から妨げることができる。ゆえに、電機子バイアス電極30層のストレスレベルを梁型構造体層26(適応できれば27も)と一致させることが好適である。しかしながら、歯部40および末端部42に提供される本発明のトーションバネ品質は、必要な場合にはある程度まで湾曲に反発できる。
【0043】
梁型構造体層はまた、図2に関して前述したように、電機子16の湾曲を最小化するために第一梁型構造体層26および第二梁型構造体層27として電機子バイアス電極30の上方および下方の両方に形成できる。電機子バイアス電極30の両側面に梁型構造体層26と27を形成することにより、第二梁型構造体層27は電機子バイアス電極30の上方に位置し、第一梁型構造体層26と同じように湾曲するため(電機子バイアス電極30の下方)、物質圧力が異なる際の影響は最小化される。電機子バイアス電極30は第一および第二梁型構造体層26と27に拘束され、構造体層26と27と共に湾曲するため、スイッチ内の湾曲は最小化される。2つの梁型構造体層が使用される場合、スイッチの湾曲の望む最小限レベルを提供するために同じ素材でそれらを構成することが好適である。しかしながら、異なる素材の組み合わせを使用して、歯部に湾曲に補正度を提供して所定レベルの湾曲を提供することも可能である。
【0044】
図4Dでは、凹状受領部54が第一梁型構造体層26および犠牲層46に彫刻されている。凹状受領部54は凹部24が後に蒸着される開口部である。凹状受領部54は第一梁型構造体層26の標準リトグラフィーおよびドライエッチング、次に犠牲層46の部分的エッチングを使用して形成される。開口部によって凹部は第一梁型構造体層26を通って突起できる。
【0045】
次に、懸架状電機子バイアス電極30、伝送路28および凹部24を形成する第二金属層が第一梁型構造体層26上に蒸着される。好適実施例では、金属層はTi薄層(200オングストローム)をスパッタ蒸着し、続いて1000オングストロームのAuを蒸着することにより成る。金属層はウエハー中で整合しなくてはならず、Auのメッキ面として作用する。メッキされるスイッチの領域を開くためにメッキは金属リトグラフィーを使用して実行される。ウエハーの端部上の膜金属を電気的に接触し、メッキ溶液に金属模様ウエハーを置くことによってAuは電気メッキされる。電気回路を完成させるために膜金属がメッキ溶液に露出される部分でのみメッキは発生し、ウエハー上に電気的絶縁抵抗が残る部分では発生しない。2ミクロンのAuがメッキされた後、残りはウエハーから取り除かれ、膜金属を取り除くために全表面はイオンエッチングされる。イオンエッチング中にAuの一部もメッキされたAuの上部から取り除かれるが、膜の厚さは1200オングストロームのみであるため損失は最小である。
【0046】
この過程の結果として、好適実施例では主にAuである第二金属層内に電機子バイアス電極30、導電伝送路28および凹部24が形成される。さらに、Auはバイアホール52を満たし、電機子バイアス電極30を電機子バイアスパッド34に接続する。Auはその低抵抗率のために第二金属層の好適な選択肢である。第二金属層の金属および梁型構造体層26と28への素材を選択する際、起動の際に電機子16が上方あるいは下方に湾曲しないように梁型構造体層26と28のストレスが電機子バイアス電極30および導電伝送路28のストレスと等しくなるような素材を選択することが重要である。これは構造体層への蒸着パラメータを注意深く決定することにより達成される。窒化ケイ素はその絶縁特性のみではなく、主にこれらの蒸着パラメータの制御性および結果として生じた膜のストレスレベルによって構造体層に選択された。また、電機子バイアス電極30を導電伝送路28より電気的に絶縁するために、その間の第二金属層の一部は取り除かれなければならない。第一梁型構造体層26および第二梁型構造体層27が共に使用される場合、この過程は第二梁型構造体層27を電機子バイアス電極30上に蒸着する以前に実行されなければならない。
【0047】
第二梁型構造体層27を蒸着後、リトグラフィーおよびエッチングでスイッチ製造を完成させる。最終的に、犠牲層46は電機子16を開放するために取り除かれる。さらに、歯部40および末端部42を形成する梁型構造体層26と27の部分は、望む厚さを提供するためにエッチングされる。
【0048】
犠牲層46がSiO2を含む場合、最終製造手順においてフッ化水素酸(HF)溶液を使用して典型的にはウェットエッチングで取り除かれる。除去過程を最適化するように除去液の滑らかで平らな流れを犠牲層46に提供するために、液体の流れる小孔部は特に電機子16の電機子バイアス電極30部が大領域を有する位置および電機子16の末端部42内に層を通って形成される。エッチングと洗浄処理は、犠牲層46が取り除かれる時に電機子16を基板14と接触させないために臨界点ドライヤー内で後工程で実行される。
【0049】
この処理中に接触が発生した場合は、装置固着およびスイッチ故障が起こり得る。スイッチを直接的ではなく、液相と気相の間に超臨界相を導入して液相(例:HF)環境から気相(例:空気)環境へ移動させることによって接触を防ぐ。サンプルはHF中でエッチングされ、希釈によってDI水で洗浄されるため、過程中にスイッチは液体から取り除かれない。DI水は同様にメタノールで置き換えられる。サンプルは臨界点ドライヤーに移動され、房は密閉される。房内でメタノールは高圧力液体CO2に置き換えられるため、サンプルを取り囲むのはCO2のみである。房は加熱されるため、CO2は超臨界相に変化する。続いて圧力が開放されるため、CO2は気相に変化する。サンプルが気体のみに囲まれれば、房から大気中に取り出してよい。犠牲層46が除去された後のMEMスイッチの側面図は図4Fに示されている。
【0050】
図5の示すMEMスイッチ10”は、50オーム直通線と接続されるように構成される。入力線18および出力線20は50オーム直通線の一部である。短絡入力線と出力線と共に構成される場合は必要ないが、インピーダンス正角回路網は入力線20および出力線18において反射を発生させずに5GHz以上の周波数でスイッチ10”を操作するために使用できる。スイッチ10”は、50オーム直通線と直列で配置された場合には誘導リアクタンスを単独で提供する。スイッチ10”のインピーダンスを50オーム直通線と一致させるために、装置に分路電気容量が加えられる。
【0051】
この構成の1実施例中では、分路電気容量は入力線20および出力線18の両方にマイクロストリップ放射状スタブ60を追加することで提供できる。スタブ60間の連続的静電結合を避けるために、スタブ60はMEMスイッチから一定角度で離される。このような連続的結合は、開放位置にあるMEMスイッチ10”の絶縁を弱める。結果の回路は低ミリメーター波範囲(約40GHz)までの周波数に対する低域フィルターとして機能する。電気容量値は、特定の通過域リップルを維持する一方で遮断周波数を最大化するように選択される。
【0052】
要約すれば、従来型電機子MEMsでは電機子16が基板バイアス電極22に引かれて入力線と出力線18と20の方へ湾曲すると、凹部24は電機子16の湾曲により生じる角偏向によって入力線と出力線18と20と正角接触できない。その結果、凹部24はその表面領域の約10%のみが入力線と出力線18と20と実際に接触を形成する質の悪い接触しか提供できない。この質の悪い接触によって、機械的損傷および小領域を通る電気流による熱が発生する。本発明は、末端部42と共に付着された複数の歯部40を含む電機子搭載のトーションバネを提供することによってこの問題を克服する。歯部40は、装置の起動時に末端部42を回転させて末端部42上の凹部24の入力線18と出力線20との正角接触を最大化させ、通電を可能にする。
【0053】
当業者にとって明らかなように、本明細書で開示された以外にも本発明はその他多くの構造が可能である。たとえば、導電伝送路層、バイアス電極、バイアスパッド、入力線および出力線に別素材を使用できる。梁型構造体層26と27および犠牲層46は、適切な対応素材であれば窒化ケイ素および二酸化ケイ素とは別の素材で製造できる。放射状スタブ実施例で開示したものとは別のインピーダンス正角回路網あるいは回路もまた使用できる。ゆえに、先行する詳述は例示のみのために提供されており、制限をするためではない。本発明の範囲はすべての均等物を含む以下の「特許請求の範囲」で制限される。

TORSION SPRING FOR ELECTRO-MECHANICAL SWITCHES AND A CANTILEVER-TYPE RF MICRO-ELECTROMECHANICAL SWITCH INCORPORATING THE TORSION SPRING

[02] PRIORITY CLAIM

[03] This application claims the benefit of priority to the following provisional application ; 60/275,171, titled Priority-Based Dynamic Slot Reservation for Airborne Supported Wireless Ad-hoc Networks filed with the United States Patent and Trademark Office on 3/12/2001.

[04] BACKGROUND OF THE INVENTION

[05] (a) Technical Field of the Invention

[06] The present invention generally relates to electro-mechanical switches, more particularly to micro-electromechanical switches (MEMS), and most particularly to high power RF MEMS.

[07] (b) Description of Related Art

[08] In communications applications, switches are often designed with semiconductor elements such as transistors or pin diodes. At microwave frequencies, however, these devices suffer from several shortcomings. Pin diodes and transistors typically have an insertion loss greater than 1 dB, which is the loss across the switch when the switch is closed. Transistors operating at microwave frequencies tend to have an isolation value less than 20 dB. This allows a signal to'bleed'across the switch even when the switch is open. Pin diodes and transistors have a limited frequency response and typically only respond to frequencies below 20 GHz. In addition, the insertion losses and high isolation value for these switches vary depending on the frequency of the signal passing through the switches. These characteristics make semiconductor transistors and pin diodes a poor choice for switches in microwave applications.

[09] U. S. Pat. No. 5,121,089, to Larson, disclosed a new class of microwave switch-- the micro-electro-mechanical (MEM) switch. The MEM switch has a very low insertion loss (less than 0.2 dB at 45 GHz) and a high isolation when open (greater than 30 dB). In addition, the switch has a large frequency response and a large bandwidth compared to semiconductor transistors and pin diodes. These characteristics give the MEM switch the potential to replace traditional narrow- bandwidth PIN diodes and transistor switches in microwave circuits.

[10] The Larson MEM switch utilizes an armature design. One end of a metal armature is affixed to an output line, and the other end of the armature rests above an input line. The armature is electrically isolated from the input line when the switch is in an open position. When a voltage is applied to an electrode below the armature, the armature is pulled downward and contacts the input line. This creates a conducting path between the input line and the output line through the metal armature.

[11] Micro-electromechanical switches of the general type described above are, however, prone to premature failure. The cause of the premature failure is linked to the damage resulting from the impact of the armature contact with the substrate contact. This damage is exacerbated by the fact that conventional MEM switches have armature contacts that impinge on the substrate contact surface at an angle. The angled impact results in all the impact energy being transferred to a relatively small area, thereby ultimately causing premature failure due to both increased impact per unit area and heat caused by resistive heating due to increased current density through the small area of actual contact.

[12] SUMMARY OF THE INVENTION

[13] The present invention solves this and other problems by providing a torsion spring which is configured to result in a substantially conformal contact between the contact plates. The resultant MEMS has increased durability, and because the substantially conformal contact results in a better electrical contact, there is less heating and the MEM switch can handle more power. Thus, a more durable and versatile MEM switch results.

[14] One embodiment of the present invention provides an armature mounted torsion spring, wherein the torsion spring is configured to provide sufficient flexibility such that when an armature electrode is electromechanically brought into contact with a substrate electrode the electrodes provide substantial conformity with one another and thus maximize contact area, reduce wear and reduce Ohmic resistance.

[15] This invention provides a new RF MEM switch in which the RF line is loaded with a torsion spring to achieve a conformal metal to metal contact. A conformal metal to metal contact assures a maximum contact area and lowest contact resistance, and, therefore, provides for critical long term reliability and good heat dissipation thus allowing for improved high-power handling.

[16] In another embodiment, the present invention provides a torsion spring for an electro-mechanical switch. The torsion spring comprises a set of tines with the set of tines having at least one tine. The set of tines extends from a free end of the armature of the switch, and includes a terminus portion rotatably suspended between the tines. The terminus portion includes a conducting transmission line, at least a portion of which is exposed for electrical contact. The conducting transmission line have a length selected such that the exposed portion of the transmission line forms a circuit between the input and output of the micro- electro-mechanical switch when the micro-electro-mechanical switch is urged into a closed position. When the switch closes, the terminus portion rotates via the tines to form a conformal connection between the exposed portion of the conducting transmission line and the input and output of the switch, thus optimizing the electrical flow therebetween.

[17] In another embodiment of the torsion spring, the portion of the conducting . transmission line exposed for electrical contact is in the form of a plurality of dimples. Each dimple corresponds to the contact to be made with the input and the output, respectively. The dimples combined with the rotatable nature of the terminus provide a conformal contact between the conducting transmission line and the input and the output to form a circuit therebetween.

[18] The torsion spring is preferably constructed of a material selected from a group consisting of silicon nitride, Type III-V semiconductor materials, and silicon dioxide. The conducting transmission line is preferably formed from a titanium adhesive layer and a gold conductor layer.

[19] The set of tines of the torsion spring preferably includes a plurality of tines, and more preferably includes two tines.

[20] In another embodiment, the present invention provides a micro-electro- mechanical switch comprising a substrate with an input line, an output line, and a substrate electrode formed on the top of the substrate, all separated from each other. The switch further includes an armature having a first beam structural layer with a first end mechanically connected with the substrate and a second end including a set of tines with at least one tine. A terminus portion is suspended between the tines, and includes a conducting transmission line positioned over the input and output lines. At least a portion of the conducting transmission line is exposed for conformal contact with the input and output lines. The armature further includes an armature electrode positioned directly above the substrate electrode and suspended on the armature. An insulating layer is positioned between the armature electrode and the substrate electrode to prevent short-circuiting therebetween. Thus, when the switch is actuated into a "closed"position, the terminus is free to rotate to ensure a conformal contact between the exposed portion of the conducting transmission line and the input and output lines in order to form a circuit therebetween to permit the flow of electricity.

[21] The armature of the switch is preferably modified with the same enhancements discussed relative to the torsion spring embodiment above.

[22] In another embodiment of the micro-electro-mechanical switch, the insulating layer is formed as a second beam structural layer under the armature electrode.

The first and the second beam structural layers are formed of materials selected such that their mechanical and thermal properties provide a desired amount of bowing when the switch is activated.

[23] Another embodiment of the present invention provides an armature for a micro- electro-mechanical switch having a torsion spring. The armature comprises a first beam structural layer having a first end for mechanically connecting with a substrate of a micro-electro-mechanical switch and a second end including a set of tines including at least one tine. A terminus portion is suspended between the tines, and includes a conducting transmission line configured to be positioned over the input and output lines of a micro-electro-mechanical switch. At least a portion of the conducting transmission line is exposed for conformal contact with the input and output lines. An armature electrode is directly above a substrate electrode of the micro-electro-mechanical switch and suspended on the armature, and an insulating layer is positioned between the armature electrode and the substrate electrode to prevent short-circuiting therebetween when the armature is assembled in a micro-electro-mechanical switch and actuated into a closed position. When the armature is assembled in a micro-electro-mechanical switch and is actuated into a"closed"position, the terminus is free to rotate to ensure a conformal contact between the exposed portion of the conducting transmission line and the input and output lines in order to form a circuit therebetween to permit the flow of electricity.

[24] The armature is preferably modified with the same enhancements discussed relative to the torsion spring embodiment above.

[25] In another embodiment of the armature, the insulating layer is formed as a second beam structural layer under the armature electrode, with the first and the second beam structural layers formed of materials selected such that their mechanical and thermal properties provide a desired amount of bowing when the switch is activated.

[26] BRIEF DESCRIPTION OF THE DRAWINGS

[27] The advantages set forth in the SUMMARY above, and other features and advantages will become more apparent from a detailed consideration of the invention when taken in conjunction with the drawings in which: FIG. 1 is a top overview of an alternative embodiment of a MEM switch embodying the present invention, wherein the conducting transmission line is positioned over both the input line and the output line; FIG. 2A shows a side elevation view of the MEM switch depicted in FIG. 1 ; FIG. 2B is the MEMS switch of FIG. 2A in a closed position; [28] FIGS. 3A and 3B are side elevation views of an alternative embodiment of the MEM switch depicted in FIG. 1 wherein the ordering of the layers comprising the armature is altered; FIGS. 4A-4F are side elevation views of the MEM switch of FIG. 1 during progressive steps of a fabrication process further embodying the present invention; and FIG. 5 is a MEM switch embodying the present invention, wherein the switch is configured with a shunt capacitance to match the MEM switch to a 50 ohm feed line.

[29] DESCRIPTION OF THE PREFERRED EMBODIMENT

[30] The present invention generally relates to electro and micro-electromechanical switches (MEMS), and more particularly to a torsion spring for providing improved input/output contact in high power RF MEMS. The following description, taken in conjunction with the referenced drawings, is presented to enable one of ordinary skill in the art to make and use the invention and to incorporate it in the context of particular applications. Various modifications, as well as a variety of uses in different applications, will be readily apparent to those skilled in the art, and the general principles defined herein, may be applied to a wide range of embodiments. Thus, the present invention is not intended to be limited to the embodiments presented, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein. Furthermore it should be noted that unless explicitly stated otherwise, the figures included herein are illustrated diagrammatically and without any specific scale, as they are provided as qualitative illustrations of the concept of the present invention.

[31] This invention teaches a torsion spring to assure a conformal contact between the armature contact and the electrodes in electro-mechanical switches, and has particular use with MEMS. The conforming nature of the contact provided by the present invention is intended to maximize the available contact area so that the contact resistance is minimized, and heat dissipation is improved. The existing simple cantilever beam type of RF MEMS switch such as the RF switch disclosed in U. S. patent 6,046,659 which is herein incorporated for all that it discloses, is an example of a switch having the disadvantages of making edge contact as the switch is snapped down. The contact area in this older switch is usually around 10 microns2 of the total dimple (contact plate) size of greater than 100 microns2. The impact of the contact of the electrode with the edge of the dimple can result in excessive wear and premature failure. This excessive wear is the primary limiting factor to the number of cycles that the switch will accommodate prior to failure. The present invention simultaneously increases the switch cycle-lifespan and substantially increases the switch's power handling capacity, as compared to conventional RF switches.

[32] The switch can be fabricated using existing fabrication processes including those disclosed in U. S. patent 6,046,659. Reliability studies on existing RF MEM switches indicate that a dominant factor limiting the switch cycling times is the nature of the edge contact of dimples upon actuation. Edge contact allowed less than 10% of the dimple surface to touch the electrode. This limitation on contact area results in a two-fold problem: First, a smaller contact area necessarily results in greater impact related damage on the switching system, resulting in part from a concentrated point of impact, and second, the smaller contact area results in an inferior connection and an associated increase in Joule (resistive) heating, which can also contribute to premature failure and inferior performance. The larger contact area provided by the present invention results in superior contact, better heat dissipation and power handling and simultaneously reduces the impact related damage at the point of contact.

[33] The improved reliability and power handling capability of the present design is achieved with the addition of a torsion spring, configured to provide the dimple contact with a freedom of rotation. Experimental results have shown that this structure reduces the likelihood of premature impact-related failure and also assures conformal dimple contact, thus minimizing contact resistance.

Additionally, this type of switch has an improved power-handling capability, as heat dissipation is improved by the much larger contact area.

[34] FIG. 1 depicts a top view of an embodiment of a micro-electro-mechanical (MEM) switch 10 according to the present invention. In this embodiment, a conducting transmission line 28 is between first and second beam structural layers 26 and 27 and is not connected directly to either the input line 20 or the output line 18. An armature bias electrode 30 covers a large portion of the underside of the armature 16 and the area between the beam structural layers 26 and 27. One end of the armature 16 is affixed directly to the substrate 14. The free end of the armature 16, beyond the armature bias electrode 30, is bifurcated, forming a pair of tines 40, with a terminus portion 42 rotatably supported on the tines 40 and resting across a gap between the input line 18 and the output line 20. The conducting transmission line 28 is positioned on the terminus portion 42 of the armature 16, so that it can bridge the gap between the input line 20 and the output line 18 when the switch 10 is closed. When a voltage is applied between the substrate bias electrode 22 and the armature bias electrode 30, the armature 16 will bend towards the substrate 14. This forces the conducting transmission line 28 into electrical contact with both the input line 18 and the output line 20, thereby closing the switch. A signal can then pass from the input line 18 to the output line 20 through the conducting transmission line 28.

[35] FIGS. 2A and 2B are side-elevational views of the MEMS switch 10 shown in FIG. 1. FIG. 2A depicts the MEMS switch in an"open"state and FIG. 2B depicts it in a"closed"state. The beam structural material 26 is connected to the substrate 14 through the fixed anchor vias 32. The suspended armature bias electrode 30 is nested within the structural material between layers 26 and 27, and is electrically accessed via contact between the bias line 38 and the armature bias pad 34. The conducting transmission line 28 is at the free end of the beam structural layers 26 and 27 and is electrically isolated from the suspended armature bias electrode 30 by the (dielectric) beam structural layers 26 and 27.

The conducting transmission line 28 is provided with a pair of dimples 24 so that when the switch 10 is closed, the combination of the dimples 24 and the transmission line 28 form a conductive pathway between the input line 18 and the output line 20, allowing electricity to flow therethrough. The gap between the dimples 24 and the substrate 14 is less than the distance between the first structural layer 26 and the substrate 14. The transmission line 28 and the dimples 24 are preferably encased within the terminus portion 42 of the armature 16, which is suspended by the tines 40. The tines 40 provide a portion of the armature 16 that is sufficiently flexible to permit the terminus portion 42 to rotate in order to ensure a conforming contact between the dimples 24 and the input line 18 and the output line 20. Variations in the thickness and shape of the tines 40 can be used to provide desirable"spring"properties. Additionally, the placement of the attachment between the tines 40 and the terminus portion 42 can be varied in order to accommodate a desired degree and direction of rotation to ensure a conformal contact between the dimples 24 and the input line 18 and the output line 20 in order to optimize the flow of electricity. Furthermore, depending on the particular embodiment and the degree of rotational freedom desired, the number of tines used may vary from one to as many as practical for a particular application. Note that the tines 40 must be sufficiently rigid to ensure that they resume their former shape after bending to provide the conforming contact. If they are formed too rigidly, depending on the closure force exerted on the switch 10, a brittle fracture may result. The suspended contact dimples 24 extend through and below the first beam structural layer 26 and define the areas of metal contact for the input and output lines 18 and 20, respectively. The substrate bias electrode 22 is below the suspended armature bias electrode 30 on the surface of the substrate 14, and is separated therefrom by an air gap 44. When a voltage is applied between the suspended armature bias electrode 30 and the substrate bias electrode 22, an electrostatic attractive force pulls the suspended armature bias electrode 30, along with the attached armature 16, toward the substrate bias electrode 22. In action, the dimple 24 first contacts the input line 18, at which point the armature 16 bends to allow the suspended armature bias electrode 30 to approach the substrate bias electrode 22. As a result, the suspended contact dimples 24 touch the input line 20 and the output line 18, so the conducting transmission line 28 bridges the gap between the input line 18 and the output line 20 thereby closing the MEM switch 10 to the position shown in FIG. 2B. Thus, the force of the metallic contact between the dimple 24 and the input line 18 is primarily dependent on the flexibility of the tines 40 and the geometry of the dimple 24, and not on the attractive forces of the armature electrode 30 to the substrate electrode 22.

[36] It is worth noting that metallic switches that do not have contact dimples have contacts that depend on armature flexibility and bias strength, factors which vary with the temperature, age, and the amount of use of the MEM switch. In addition to improving repeatability, the quality of the contact itself is improved by the addition of the dimple because the dimple has a controllable size and surface texture, characteristics that are dependent on the fabrication rather than on the environment. Thus, MEM switches without the dimple 24 are more likely to have time-varying contact characteristics, a feature that may make them difficult or impossible to use in some circuit implementations.

[37] FIGS. 3A and 3B are side-elevational views of an alternative embodiment of the MEMS 10 in FIG. 1, FIG. 2A, and FIG. 2B. In this case, the suspended armature bias electrode 30 is an exposed layer on the bottom of the armature 16.

Because of the exposed nature of the suspended armature bias electrode 30, an additional insulating layer 48 is provided on top of the substrate bias electrode 22 to keep the armature bias electrode 30 and the substrate bias electrode 22 from becoming electrically shorted when the switch is in the closed position.

[38] FIGS. 5A-5F illustrate a manufacturing process that may be employed for fabricating the MEM switch 10 of FIGS. 1 and 2. The process begins with a substrate 14. In a preferred embodiment, GaAs is used as the substrate. Other materials may be used, however, such as InP, ceramics, quartz, or silicon. This allows the circuit elements as well as the MEM switch 10 to be fabricated simultaneously on the same substrate using standard integrated circuit fabrication technology such as metal sputtering and masking. The substrate is chosen primarily based on the technology of the circuitry and the MEM switch is to be connected with so that the MEM switch and the circuit may be fabricated simultaneously. For example, InP can be used for low noise HEMT MMICS (high electron mobility transistor monolothic microwave integrated circuits) and GaAs is typically used for PHEMT (pseudomorphic HEMT) power MMICS. The flexibility in the fabrication of the MEM switch 10 allows it to be used in a variety of circuits. This reduces the cost and complexity of circuits designed using the present MEM switch.

[39] FIG. 4A shows a profile of the MEM switch 10 after the first step of depositing a metal layer onto the substrate 14 for the armature bias pad 34, the input and output lines 18 and 20, and the substrate bias electrode 22 is complete. The metal layer may be deposited lithographically using standard integrated circuit fabrication technology, such as resist lift-off or resist definition and metal etch.

In the preferred embodiment, gold (Au) is used as the primary composition of the metal layer. Au is preferred in RF applications because of its low resistivity.

In order to ensure the adhesion of the Au to the substrate, a thin layer (250-500 angstroms) of titanium (Ti) is deposited, followed by an approximately 1000 angstrom layer of platinum (Pt), and finally the Au. The Pt acts as a diffusion barrier (an anti-diffusion layer) to keep the Au from intermixing with the Ti and causing the metal to lose adhesion strength to the substrate. Other anti-diffusion layer materials may be used, depending on the needs of a particular embodiment.

In the case of a III-V semiconductor substrate, a thin layer of gold germanium (AuGe) eutectic metal may be deposited first to ensure adhesion of the Au by alloying the AuGe into the semiconductor similar to a standard ohmic metal process for any III-V MESFET or HEMT.

[40] Next, as shown in FIG. 4B, a sacrificial layer 46 is placed on top of the metal layer (preferably Au) and etched so that the armature 16 may be produced above the sacrificial layer 46. The sacrificial layer 46 is typically comprised of 2 microns of Si02 which may be sputter-deposited or deposited using PECVD (plasma enhanced chemical vapor deposition). A via hole 52 is etched in the sacrificial layer 46 so that the metal armature bias pad 34 is exposed in order to begin forming the fixed anchor vias 42. The via hole 52 definition may be performed using standard resist lithography and etching of the sacrificial layer 46. Other materials besides Si02 may be used as a sacrificial layer 46. The important characteristics of the sacrificial layer 46 are a high etch rate, good thickness uniformity, and conformal coating by the oxide of the metal already on the substrate 14. The thickness of the oxide partially determines the thickness of the switch opening (gap 44), which is critical in determining the voltage necessary to close the switch 10 as well as the electrical isolation of the switch 10 when it is open. The sacrificial layer 46 will be removed in the final step to release the armature 16, as shown in FIG. 5F.

[41] Another advantage of using Si02 as the sacrificial layer 46 is that Si02 can withstand high temperatures. Other types of sacrificial layers, such as organic polyimides, harden considerably if exposed to high temperatures. This makes the polyimide sacrificial layer difficult to later remove. The sacrificial layer 46 is exposed to high temperatures when the silicon nitride for the beam structural layer is deposited, as a high temperature deposition is desired when depositing the silicon nitride to give the silicon nitride a lower BOE etch rate. A low BOE etch rate minimizes the amount of the first beam structural layer 26 that is lost when the Si02 is etched away.

[42] FIG. 4C shows the fabrication of the first beam structural layer 26. The first and second beam structural layers 26 and 27 provide the supporting mechanism of the armature 16 and are preferably made out of silicon nitride, although other materials besides silicon nitride may be used, examples of which include Type III-V semiconductor materials, and silicon dioxide. Silicon nitride is preferred because it can be deposited so that neutral stress exists in the beam structural layers 26 and 27. Neutral stress fabrication reduces the bowing that may occur when the switch is actuated. The material used for the first beam structural layer 26 must have a low etch rate compared to the sacrificial layer 46 so that the first beam structural layer 26 is not etched away when the sacrificial layer 46 is removed to release the armature 16. The first beam structural layer 26 is patterned and etched using standard lithographic and etching processes.

[43] As depicted by the embodiment of FIG. 3A and 3B, a beam structural layer 26 may be formed either below or above the armature bias electrode 30. If a beam structural layer 26 is fabricated on only one side of the armature bias electrode 30, bowing will occur in the armature 16 when the switch is actuated if the stress in the beam structural layer 26 differs from the stress in the armature bias electrode 30. The armature 16 will bow either upward or downward, depending upon which material has the higher stress. Bowing can change the voltage required to activate the switch and, if the bowing is severe enough, can prevent the switch from either opening (bowed downward) or closing (bowed upward) regardless of the actuating voltage. Therefore, it is preferable to match the stress levels of the layer of the armature bias electrode 30 with beam structural layer 26 (and 27, if applicable). Note, however, that the torsion spring-quality of the present invention provided by the tines 40 and the terminus portion 42 can aid to some degree in countering bowing when necessary.

[44] The beam structural layer may also be formed both above and below the armature bias electrode 30, as previously discussed relative to FIG. 2, as a first beam structural layer 26 and a second beam structural layer 27 in order to minimize the bowing in the armature 16. By fabricating the beam structural layers 26 and 27 on both sides of the armature bias electrode 30, the effect of differing material stresses is minimized because the second beam structural layer 27 is above the armature bias electrode 30, and will flex in the same manner as the first beam structural layer 26 (below the armature bias electrode 30). The armature bias electrode 30 is constrained by the first and second beam structural layers 26 and 27, and will therefore flex with the structural layers 26 and 27 so that the bowing in the switch is minimized. When two beam structural layers are used, it is preferable that they be constructed of the same material in order to provide a desired minimum level of bowing of the switch. However, it is also feasible that combinations of different materials may be used to provide a customized level of bowing with the tines providing a degree of compensation for the bowing.

[45] In FIG. 4D, a dimple receptacle 54 is etched into the first beam structural layer 26 and the sacrificial layer 46. The dimple receptacle 54 is an opening where the dimple 24 will later be deposited. The dimple receptacle 54 is created using standard lithography and a dry etch of the first beam structural layer 26, followed by a partial etch of the sacrificial layer 46. The opening allows the dimple to protrude through the first beam structural layer 26.

[46] Next, a second metal layer forming the suspended armature bias electrode 30, the conducting line 28, and the dimples 24 is deposited onto the first beam structural layer 26. In the preferred embodiment, the metal layer is comprised of a sputter deposition of a thin film (200 angstroms) of Ti followed by a 1000 angstrom deposition of Au. The metal layer must be conformal across the wafer and acts as a plating plane for the Au. The plating is done by using metal lithography to open up the areas of the switch that are to be plated. The Au is electroplated by electrically contacting the membrane metal on the edge of the wafer and placing the metal patterned wafer in the plating solution. The plating occurs only where the membrane metal is exposed to the plating solution to complete the electrical circuit and not where the electrically insulating resist is left on the wafer. After 2 microns of Au is plated, the rest is stripped off of the wafer and the whole surface is ion milled to remove the membrane metal. Some Au will also be removed from the top of the plated Au during the ion milling, but the loss is minimal because the membrane is only 1200 angstroms thick.

[47] The result of this process is that the armature bias electrode 30, the conducting transmission line 28, and the dimple 24 are created in the second metal layer, primarily Au in the preferred embodiment. In addition, the Au fills the via hole 52 and connects the armature bias electrode 30 to the armature bias pad 34. Au is a preferred choice for second metal layer because of its low resistivity. When choosing the metal for the second layer and the material for the beam structural layers 26 and 28, it is important to select the materials such that the stress of the beam structural layers 26 and 28 match the stress of the armature bias electrode 30 and the conducting transmission line 28 so that the armature 16 will not bow upwards or downwards when actuating. This is done by carefully determining the deposition parameters for the structural layer. Silicon nitride was chosen for this structural layer not only for its insulating characteristics but in large part because of the controllability of these deposition parameters and the resultant stress levels of the film. Also note that in order to electrically isolate the armature bias electrode 30 from the conducting transmission line 28, a portion of the second metal layer therebetween must be removed. In cases where both a first beam structural layer 26 and a second beam structural layer 27 are used, this process must be performed before depositing the second beam structural layer 27 on the armature bias electrode 30.

[48] After deposition of the second beam structural layer 27, it is lithographically defined and etched to complete the switch fabrication. Finally, the sacrificial layer 46 is removed to release the armature 16. Furthermore, the portion of the beam structural layers 26 and 27 forming the tines 40 and the terminus portion 42 is etched in order to provide a desired thickness.

[49] If the sacrificial layer is comprised of Si02, then it will typically be wet etched away in the final fabrication sequence by using a hydrofluoric acid (HF) solution. It is noteworthy that small fluid-flow holes may be formed through the layers, particularly where the armature bias electrode 30 portion of the armature 16 has a large area and in the terminus portion 42 of the armature 16 in order to permit smooth and even flow of the removal liquids to the sacrificial layer 46 to optimize the removal process. The etch and rinses are performed with post- processing in a critical point dryer to ensure that the armature 16 does not come into contact with the substrate 14 when the sacrificial layer 46 is removed. If contact occurs during this process, device sticking and switch failure are likely.

Contact is prevented by transferring the switch from a liquid phase (e. g. HF) environment to a gaseous phase (e. g. air) environment not directly, but by introducing a supercritical phase in between the liquid and gaseous phases. The sample is etched in HF and rinsed with DI water by dilution, so that the switch is not removed from a liquid during the process. DI water is similarly replaced with methanol. The sample is transferred to the critical point dryer and the chamber is sealed. High pressure liquid CO2 replaces the methanol in the chamber, so that there is only CO2 surrounding the sample. The chamber is heated so that the CO2 changes into the supercritical phase. Pressure is then released so that the CO2 changes into the gaseous phase. Once the sample is surrounded only by gas, it may be removed from the chamber into room air. A side elevational view of the MEM switch after the sacrificial layer 46 has been removed is shown in FIG. 4F.

[50] FIG. 5 depicts a third MEM switch 10"where the MEM switch 10"is configured to be connected to a 50 ohm through line. The input line 18 and the output line 20 are segments of the 50 ohm through line. An impedance matching network, not needed when designed with short input and output lines, may be used to operate the switch 10"at frequencies greater than 5 GHz without generating reflections at the input line 20 or the output line 18. By itself, the switch 10"presents an inductive reactance when placed in series with the 50 ohm through line. In order to match the impedance of the switch 10"to the 50 ohm line, a shunt capacitance is added to the device. The shunt capacitance, in one embodiment of the design, can be realized by adding a microstrip radial stub 60 to both the input line 20 and the output line 18. The stubs 60 are angled away from the MEM switch to avoid series capacitive coupling between the stubs 60.

Such series coupling would degrade the isolation of the MEM switch 10"in the open position. The resulting circuit acts as a lowpass filter for frequencies up to the low millimeter-wave range (approximately 40 GHZ). The capacitance values are chosen to maximize the cutoff frequency while maintaining a specified passband ripple.

[51] In summary, in traditional armature MEMs, when the armature 16 flexes toward the input and output lines 18 and 20, drawn by a substrate bias electrode 22, the dimples 24 are unable to conformably contact the input and output lines 18 and 20 due to the angular deflection caused by bending of the armature 16. As a result, the dimples 24 have relatively poor contact with the input and output lines 18 and 20, with only about 10% of the surface area of the dimples 24 actually making contact. This poor contact results in mechanical wear as well as heading due to the flow of electricity through a small area. The present invention overcomes this problem by providing an armature-mounted torsion spring comprising a plurality of tines 40, attached with a terminus portion 42. The tines 40 permit rotation of the terminus portion 42 in order to maximize the conformal contact of dimples 24 on the terminus portion 42 with the an input line 18 and an output line 20 when the device is actuated in order to allow electricity to pass therethrough.

[52] As can be surmised by one skilled in the art, there are many more configurations of the present invention that may be used beyond those presented herein. For example, other metals can be used to form the conducting transmission line layer, the bias electrodes and pads, and the input and output lines. The beam structural layers 26 and 27 and the sacrificial layer 46 may be fabricated with materials other than silicon nitride and silicon dioxide as long as suitable counterpart materials are selected. Other impedance-matching networks or circuitry can also be substituted for the radial stub embodiment shown. It is therefore intended that the foregoing detailed description be regarded as illustrative rather than limiting and that it be understood that it is the following claims, including all equivalents, are intended to define the scope of this invention.

「特表2004-535654およびWO2002073645より引用」

|
|

[Claims] CMOS適合性基板上にマイクロ電気機械スイッチを作製する方法

【特許請求の範囲】
【請求項1】
マイクロ電気機械(MEM)スイッチを作製する方法であって、
a)複数の導電性相互接続線がその中に形成された第1の誘電層を基板上に堆積させるステップと、
b)前記複数の導電性相互接続線の少なくとも1本と接触する導電性バイアをそれを通して形成する第2の誘電層を堆積させるステップと、
c)前記第2の誘電層に彫り込まれた空洞を形成するステップと、
d)前記空洞を犠牲材料で充填し、前記犠牲材料を平坦化するステップと、
e)第3の誘電層を堆積させ、前記導電性バイアと接触させた状態で導電性はりを形成するステップとを含む方法。
【請求項2】
f)第4の誘電層を堆積させ、前記第1の空洞に合った第2の空洞をパターン形成するステップと、
g)前記第2の空洞を犠牲材料で充填し、前記犠牲材料を平坦化するステップと、
h)前記第2の空洞を覆う第5の層を堆積させるステップと、
i)前記犠牲材料の上に複数の孔をパターン形成し、エッチングするステップと、
j)前記導電性はりの少なくとも一端が固定され、前記導電性はりの残りの部分が空気で取り囲まれるように、前記犠牲材料を選択的に除去するステップとをさらに含む、請求項1に記載の方法。
【請求項3】
前記第2の空洞を密閉して、前記導電性はりの露出部分を保護し、前記第5の誘電層のリリースバイアを閉じる、第6の誘電層を付加するステップをさらに含む、請求項2に記載の方法。
【請求項4】
前記保護層がTaまたはTaNで構成される、請求項3に記載の方法。
【請求項5】
前記ステップe)を、前記第2の誘電層から犠牲材料を選択的に除去するステップで置き換え、前記選択的な除去が、前記第1の空洞の形状に合わせて行われる、請求項1に記載の方法。
【請求項6】
前記空洞が、前記導電性はりの周囲から誘電材料を選択的に除去することによって形成される、請求項1に記載の方法。
【請求項7】
前記ステップe)を、
e1)第3の誘電層をパターン形成して、前記導電性はりを前記第3の誘電層から選択的に解放するステップと、
e2)前記第2の誘電層内の前記犠牲材料を除去するステップとで置き換えた、請求項1に記載の方法。
【請求項8】
前記導電性相互接続線が銅製であり、誘電体内に埋め込まれた、請求項1に記載の方法。
【請求項9】
前記相互接続銅線の全体または一部を、前記MEMスイッチが閉じた状態にあるときに容量性空気ギャップが形成される程度まで凹ませた、請求項8に記載の方法。
【請求項10】
スティクション効果を最小限に抑えるために、前記第1の誘電層の上面よりも前記導電性線を凹ませた、請求項1に記載の方法。
【請求項11】
前記凹ませた導電性線をカプセル化するステップをさらに含む、請求項10に記載の方法。
【請求項12】
前記導電性はりがその端部の1つまたは2つで固定される、請求項2に記載の方法。
【請求項13】
前記MEMスイッチが、様々な構成で配列されたその他の複数の金属/誘電体/金属スイッチと結合される、請求項1に記載の方法。
【請求項14】
前記第1の空洞の露出領域内に形成された前記導電性線が、前記導電性はりを引き下げるアクチュエータ電極として作用し、1つまたは複数の電気信号経路を提供する、請求項1に記載の方法。
【請求項15】
下側電極として作用する前記複数の導電性相互接続線と前記導電性はりとの間に形成される隙間のサイズによって決まる厚さまで、前記第2の誘電層を堆積させる、請求項1に記載の方法。
【請求項16】
基板上に堆積させた第1の誘電層に埋め込まれた導電性線であって、前記第1の誘電層の上面より凹んだ導電性線と、
第2の誘電層上に位置する第1の空洞と、
前記第1の空洞の上に位置し、その少なくとも一端が第3の誘電層によって固定された導電性はりと、
前記第3の誘電層の上に重ねられた第4の誘電層に彫り込まれた、前記第1の空洞の外周と一致する外周を有する第2の空洞と、
前記導電性はりを保護するために前記第4の層の上に重ねられた第5の誘電層とを含むマイクロ電気機械(MEM)スイッチ。
【請求項17】
前記導電性はりが、その端部の1つまたは2つで固定された、請求項16に記載のマイクロ電気機械(MEM)スイッチ。

1. A method of fabricating a micro-electromechanical (MEM) switch comprising the steps of : a) depositing a first dielectric layer on a substrate, said first dielectric layer having a plurality of conductive interconnect lines formed therein ; b) depositing a second dielectric layer through which conductive vias are formed, said vias contacting at least one of said plurality of conductive interconnect lines ; c) forming a cavity that is carved out from said second dielectric layer ; d) filling said cavity with sacrificial material and planarizing said sacrificial material ; and e) depositing a third dielectric layer and forming a conductive beam, having said conductive vias contact said conductive beam.

2. The method as recited in claim 1, further comprising the steps of : f) depositing a fourth dielectric layer and patterning a second cavity conforming to said first cavity ; g) filling said second cavity with sacrificial material and planarizing said sacrificial material h) depositing a fifth layer to cover said second cavity ; i) patterning and etching a plurality of holes over said sacrificial material ; and j) selectively removing said sacrificial material such that said conductive beam is anchored at at least one end, and leaving the remainder of said conductive beam surrounded by air.

3. The method as recited in claim 2, further comprising the step of adding a sixth dielectric layer to seal said second cavity to protect exposed portions of said conductive beam and to close off released vias in said fifth dielectric layer.

4. The method as recited in claim 3, wherein said protective layer is made of Ta or TaN.

5. The method as recited in claim 1, wherein said step e) is replaced by the step of selectively removing the sacrificial material from said second dielectric layer, said selective removal conforming to the shape of said first cavity.

6. The method as recited in claim 1, wherein said cavity is formed by selectively removing dielectric material from about said conductive beam.

7. The method as recited in claim 1, wherein said step e) is replaced by the step of el) patterning the third dielectric layer to selectively free said conductive beam from said third dielectric layer ; and e2) removing said sacrificial material within said second dielectric layer.

8. The method as recited in claim 1, wherein said conductive interconnect lines are made of copper, and wherein said conductive interconnect lines are inlaid in a dielectric.

9. The method as recited in claim 8, wherein all or portions of said copper interconnect lines are recessed to a degree sufficient to provide a capacitive air gap when said MEM switch is in a closed state 10. The method as recited in claim 1, wherein said conductive lines are recessed with respect to a top surface of said first dielectric layer to minimize stiction effects.

11. The method as recited in claim 10, further comprising the step of encapsulating said r recessed conductive lines.

12. The method as recited in claim 2, wherein said conductive beam is anchored at one or at two of its ends.

13. The method as recited in claim 1, wherein said MEM switch is coupled to a plurality of other metal-dielectric-metal switches that are arranged in a variety of configurations.

14. The method as recited in claim 1, wherein said conductive lines formed in an exposed area of said first cavity act as actuator electrodes for pulling down said conductive beam and provide one or more electrical signal paths.

15. The method as recited in claim 1, wherein said second dielectric layer is deposited to a thickness that is determined by the size of the gap to be formed between said plurality of conductive interconnect lines acting as a lower electrode and said conductive beam.

16. A micro-electromechanical (MEM) switch comprising : conductive lines embedded in a first dielectric layer deposited on a substrate, said conductive lines being recessed with respect to the top surface of said first dielectric layer ; a first cavity positioned on a second dielectric layer ; a conductive beam positioned over said first cavity and anchored at at least one end of said conductive beam by a third dielectric layer ; a second cavity carved out of a fourth dielectric layer superimposed on top of said third dielectric layer, said second cavity having a perimeter matching the perimeter of said first cavity ; a fifth dielectric layer superimposed on said fourth layer to provide protection for said conductive beam.

17. The MEM switch recited in claim 16, wherein said conductive beam is anchored at one or two of its ends.

「特表2005-514728およびWO2003054938より引用」

|
|

CMOS適合性基板上にマイクロ電気機械スイッチを作製する方法

【発明の詳細な説明】
【技術分野】
【0001】
本発明は一般に、マイクロ電気機械スイッチ(MEMS)の作製に関し、特に、現況技術の半導体作製プロセスに統合することができるMEMSの製造に関する。
【背景技術】
【0002】
切換動作は、数多くの電気的、機械的および電気機械的な応用分野の基礎の1つである。ここ数年の間に、MEMスイッチが大きな関心を集めるようになり、MEMS技術を用いた製品が、生体医用システム、航空宇宙システムおよび通信システムで広く用いられている。
【0003】
従来のMEMSでは、通常、例えばマクミラン(McMillan)他の米国特許第6160230号、フェング(Feng)他の米国特許第6143997号、カレー(Carley)他の米国特許第5970315号、タム(Tham)他の米国特許第5880921号などに記載されるように、カンチレバー・スイッチ、メンブレン・スイッチおよび同調可能コンデンサ構造を利用する。MEMSデバイスは、マイクロ電気機械技術を用いて製造され、電気的信号、機械的信号または光学的信号の流れを制御するために使用される。しかし、このようなデバイスは、その構造および本質的な材料の特性により従来の半導体加工とは別のラインで製造しなければならないので、多くの問題がある。これは通常、異なる材料および処理が相互に適合性がなく、そのために標準的な半導体作製処理に統合できないことによる。
【0004】
金など、MEMSの製造で通常使用される材料を使用すると、デバイスを直接オンチップ・アプリケーションに統合する上で明らかな問題が生じる。文献に広く見られるポリシリコンを用いても、温度サイクル、ならびに実際の半導体デバイスを作製する前工程(FEOL)のツールと相互接続金属を加工する後工程(BEOL)との間によく見られる分離による問題が生じる。通常は、能動デバイスの金属汚染を防止するために、これら2つのセットが一方から他方へのプロセスのクロスオーバを有することはできない。したがって、後工程でポリシリコンの堆積が起こる可能性は低い。
【0005】
ほとんどの既存のプロセスには、標準的な金属被覆を使用することによって、金属を保護するためのカプセル化が行われないという深刻な欠点がある。さらに、複数の基板を使用することが多く、それに伴う固有の欠点もある。
【0006】
その他の既存の技術は、構造の上部において切換機能を提供するだけであるので、本発明で以下に述べるようにあらゆるレベルで統合を実現できる可能性が低い。
【0007】
したがって、従来のBEOLまたは相互接続レベルと共に、またはその付加モジュールとしてデバイスを製造することができるように完全に統合できる処理と結びついた既成のBEOL材料を使用して、MEMSデバイスを提供することができるプロセスが必要とされている。
【0008】
本発明がよりよく理解されるように、変形可能なはり1の両端が誘電体4内に固定されたMEMスイッチの断面図である図1を参照して、従来のMEMスイッチについて述べる。最下レベルは、導電性エレメント2、2aおよび3を含む誘電体材料5からなり、これらのエレメントは、後にデバイスの様々な電気部分を接続または形成するために使用される。参照番号2および2aで示す導体は、はりを変形させる動作電位を提供するために使用される。信号を伝える導体3は、動作中にはりに接続される。
【0009】
図2は、図1と同じ従来技術のMEMデバイスの平面図である。通常の実施態様では、変形可能なはり1は、例えばSiO2などの誘電体の上にポリシリコンを重ねて形成され、その周囲の材料をエッチングで除去して、隆起した構造、すなわち、その前に形成した、またはそれ自体はポリシリコン製である導体の上に架かるはりが残るようにする。その後、このデバイスに、ポリシリコンに付着して導電性エレメント1、2、2aおよび3を形成する、通常は金の無電解めっきを施す。このスイッチは、はりと電極2および2aとの間に電位差を発生させることにより動作する。この電圧により、はり1を引き寄せて電極3と接触させてスイッチを閉じる静電引力が生じる。
【0010】
これらは全て、通常は、従来の半導体デバイスと比較して大きなトポグラフィを有する隆起構造であることに留意されたい。これにより、これらのデバイスは、半導体チップ作製プロセスに統合することが実質的に不可能となる。これらのデバイスは、通常は、表面微細加工技術を用いて作製される。表面微細加工技術は、フォトレジストの上またはシリコンなどの基板の上にデバイスを構築し、その後デバイスの下の基板の一部分を基板の裏側から除去することを含むので、やはり標準的な半導体処理と統合することができなくなる。
【0011】
図3は、はり1の一端のみが誘電体4内に固定された、別のタイプの従来のMEMスイッチを示す断面図である。その他全ての部分は図1で述べたのと同様に動作する。図3に示す対応するデバイスの上面図である図4についても同じことが言える。この後者の場合、スイッチは、はり1と制御電極2の間に電圧を印加することによって動作する。これにより、はりは下方に引かれて信号電極3と接触する。電圧が降下すると、はり1は元の位置に戻る。
【0012】
通常は、はりと制御電極の間の隙間によって、はりを引き下げるのに必要な電圧は実質的に決定される。ほとんどの文献には、1マイクロメートルから数マイクロメートルの範囲の隙間を有するデバイスが記載されている。これらの隙間は大きく、したがって、必要な電圧も大部分の民生用アプリケーションで望ましい電圧より高くなる。報告されている活動化電圧は、約30ボルトから75ボルトの範囲である。これは、通常3から5ボルトで動作する携帯電話などのアプリケーションにとっては高すぎる。本発明の構造は、200オングストロームから数千オングストロームの範囲の隙間で動作し、活動化電圧が5ボルト以下のスイッチをもたらす。
【0013】
前述の例示的なスイッチ構成は、当技術分野で既知の多くの可能な構造の一部に過ぎない。MEMスイッチは、複数のはりを様々な組合せで配線した形で構成することもできることに留意されたい。
【0014】
スティクションは、MEMSデバイスにおける主要な関心事である。スティクションは、接触している2つ以上の表面が離れる際に何らかの損傷をデバイスに与えることとして定義される。衝突がこの現象の主原因である。本発明では、少なくとも1実施形態で、図20を参照して詳細に述べるように、スイッチが閉じたときに空気ギャップ200を形成することによってこの問題に対処する。表面張力も、もう1つのスティクションの主な原因と考えられている。これが、本発明で、可動部分の解放およびその後の処理にドライ・エッチングおよびドライ処理を利用する理由である。
【特許文献1】米国特許第6160230号
【特許文献2】米国特許第6143997号
【特許文献3】米国特許第5970315号
【特許文献4】米国特許第5880921号
【非特許文献1】ジョシ(A.Joshi)およびニマガッダ(R.Nimmagadda)、「酸素プラズマ中のダイヤモンド膜とグラファイトのエロージョン(Erosionof diamond films and graphite in oxygen plasma)」、材料研究学会(Materials ResearchSociety)、ジャーナル・オブ・マテリアルズ・リサーチ(Journal of Materials Research)、Vol.6、No.7、1484ページ、1996年
【発明の開示】
【発明が解決しようとする課題】
【0015】
したがって、本発明の目的は、CMOSウェハ、バイポーラ・ウェハまたはBiCMOSウェハ内に完全に一体化されるMEMスイッチおよびその他の同様の構造を構築することである。
【0016】
別の目的は、改変したダマシン・プロセスを用いてMEMスイッチおよびその他の同様の構造を製造することである。
【0017】
別の目的は、保護用のバリヤ材料内にカプセル化した銅を利用するMEMスイッチおよびその他の同様の構造を構築することである。
【0018】
別の目的は、カプセル化をBEOLの銅と、このプロセスに適した温度で確実に統合できるようにすることである。
【課題を解決するための手段】
【0019】
本発明では、適合性のあるプロセスおよび材料を用いて従来の半導体相互接続レベルと一体化されたMEMSスイッチを作製する方法を提供することによって、上記その他の目的に対処する。
【0020】
本明細書に記載の本発明は、接点切換に使用される様々な構成または金属/誘電体/金属スイッチあるいはその両方を製造するのに適した容量性スイッチを作製する方法を提供する。
【0021】
好ましい実施形態では、このプロセスは、誘電体内に埋め込まれた金属導体で構成された銅ダマシン(copper damascene)相互接続層から始まる。銅相互接続の全体または一部分を、スイッチが閉じたときに容量性空気ギャップが形成され、さらに望ましい場合にはTa/TaN製の保護層のための空間が確保される程度に凹ませる。スイッチによって特定される領域内に画定される金属構造は、アクチュエータ電極として機能し、可動導電性はりを引き下げ、切換信号が流れる1つまたは複数の経路を提供する。このような空気ギャップの利点は、誘電体(空気)が、信頼性および電圧ドリフトの問題を引き起こす電荷の蓄積または捕捉を行わないことである。
【0022】
本発明は、二酸化ケイ素や窒化ケイ素など、任意数の誘電体材料を用いて実施することができ、これらの材料は全て、下側電極と容量性スイッチの変形可能なはりとの間に配置することができるので有利である。
【0023】
ギャップを形成するために電極を凹ませる代わりに、電極の周囲または電極の上、あるいはその両方に誘電体を付加するだけでもよい。その後、第2の誘電層を、下側電極と可動はりの間に形成するギャップの所望の厚さまで堆積させる。第2の誘電層を貫通するようにバイアを形成し、前述の金属相互接続層と、可動はりを含む次の金属層との間の接続をもたらす。次いで、バイアを含む層をパターン形成およびエッチングして、下側活動化電極および信号経路を含む空洞領域を形成する。次いで、この空洞を犠牲リリース材料(release material)で埋め戻す。この犠牲リリース材料は、好ましくはSiLKであり、γ-ブチロラクトン、Bステージ高分子、およびメシチレンからなる高分子樹脂の形態の半導体誘電体である。次いで、このリリース材料を、誘電体の上部と面一になるように好ましくは化学機械研磨(CMP)でほぼ平坦にする。空洞領域を画定するために使用したのと同じマスクを反対極性のフォトレジストを用いて再度イメージングし、その後上面からリリース材料をエッチングで除去して、空洞が形成された誘電体とほぼ面一にすることによって、リリース材料を平坦化することもできる。これにより、はり層をその上に構築できる平坦な表面が形成される。
【0024】
可動はりを含む層は、標準的なCMOSデバイスの作製に通常使用されるその他の相互接続層と同様である。はり構造は、それより先に画定された空洞領域を部分的にまたは完全に横切って延び、スイッチの動作中に所望の電気接続を提供するようになっていることが好ましい。可動はりに孔を形成することができることにも留意されたい。これらの孔は、最初は、はりを含む層を形成するのと同じ誘電体で充填されている。その後、はりの中に存在するこれらの誘電体領域をエッチングで除去し、はりの下にある犠牲材料へのより大きなアクセスを提供して、リリース・プロセスの助けとなるようにする。次に、空洞領域のパターンまたはこのパターンの一部を画定し、はりの孔を通る領域も含めてはりの周囲の誘電体材料を除去し、その下にあるリリース材料へのアクセスを提供する。この時点で、リリース材料を除去することができる。これで、空洞領域の上に架かるスイッチは、完全に機能できる状態となる。
【発明を実施するための最良の形態】
【0025】
本発明の前述その他の目的、態様および利点は、以下の本発明の好ましい実施形態の詳細な説明を添付の図面を参照して読めば、よりよく理解できるであろう。
【0026】
この主要な実施形態は、CMOS、バイポーラ、BiCMOSまたはその他の一般的な半導体チップの標準的なプロセスに完全に組み込むことができる方法でMEMSデバイスを構築することを教示するものである。
【0027】
図5は、最初の構造、すなわち銅ダマシン相互接続層を示す図である。この層の下で、FETやNPNトランジスタ、HBT(ヘテロ接合バイポーラ・トランジスタ)、抵抗器などのデバイスを、シリコン基板、あるいは好ましくはSiO2製の低損失基板上に配置することができる。これは、SiO2製であることが好ましい誘電層20に埋め込んだ銅線30、40および50からなる。参照番号30で示す領域は、誘電層100に埋め込んだバイア110の受け金属(landing metal)である。参照番号40で示す金属はアクチュエータまたは制御電極となり、金属50は信号電極である。この金属の厚さは、通常は3000から5000オングストローム程度であるが、この範囲に限定されるわけではない。誘電層150内に形成されるはり160は、両端を固定した状態で示してある。はりの中心は、空洞120の上で垂直方向に動くようになっている(図20参照)。
【0028】
さらに図5を参照すると、図示の導線30、40および50の上面60は、スティクションができるだけ小さくなるように凹部になっている。これはウェット・エッチングで行うことが好ましいが、その他の手段でもよい。凹みの程度は所望の最終的な構造によって決まる。好ましい実施形態では、酢酸および過酸化水素の希釈溶液からなるウェット・エッチング液を一般に使用する。銅は、約800オングストロームの深さまでエッチングすることが好ましい。
【0029】
図6は、図5に示すMEMSデバイスの上面図である。空洞120の上に掛けられたはり160は、複数の孔175を有するものとして示してある。これらの孔については、プロセスの説明(図9~19)の際により詳細に述べる。これらの孔は、空洞120を構築するときに犠牲材料の除去を容易にするのに役立つ。
【0030】
図7は、スイッチの一端のみが固定された、別のタイプのMEMSデバイスの断面図である。図5の場合と同様に、導線30、40および50は誘電層20内に形成されるが、バイア110は誘電層100内に形成される。はり160は一端が浮いた状態なので、空洞120の上で垂直方向に自由に動くことができる。
【0031】
図8は、図7に示すMEMSデバイスに対応する上面図である。この場合も、可動はり160には孔175が形成されている。
【0032】
次に、図5~6および図7~8に示すMEMSスイッチの構築に使用されるプロセスについて詳細に述べる。
【0033】
図9を参照すると、誘電層20内に形成された導線30、40および50が示してある。配線の上面は、誘電層20の上面とほぼ面一である。
【0034】
次に図10を参照すると、配線の上面60は誘電層20の上面より凹んでいる。金属を凹ませる1つの方法は、所望の深さを実現するのに必要な時間だけ当該金属に対して有効なウェット・エッチングを用いるものである。例えば、好ましい実施形態では、水、酢酸および過酸化水素(それぞれ3リットル、15ml、9ml)からなる溶液を2.5分間用いて、600から800オングストロームの深さを実現する。
【0035】
次に図11を参照すると、バリヤ材料70の層を堆積させて、銅をその後の処理から保護するためにカプセル化してある。好ましい実施形態では、厚さ100オングストロームのTa層を堆積させ、その後、厚さ400オングストロームのTaNを堆積させる。別法として、バリヤ層は、誘電体で構成する、または任意数のその他の適当な金属バリヤで構成することもできる。
【0036】
図12は、誘電体20の最上面80からバリヤ70を除去した結果を示している。これを実施するには、化学機械研磨(CMP)が最適である。これで、バリヤ材料75は凹部領域のみに残る。CMPプロセスは誘電体20の一部も腐食させることがあるので、スイッチが閉じたときに適当な隙間200を維持したい場合には、誘電体の腐食を最小限に抑えるように注意しなければならない。CMPプロセスが完了すると、誘電体上部150からバリヤ75までの深さは約200オングストロームとなる。これは、容量性スイッチの最終的な隙間を形成するためにカスタマイズすることができる。
【0037】
導体をカプセル化する別の方法は、導体を画定したのと同じマスクを用いて、ただし反対の極性のフォトレジストを用いてブランケット・バリヤ70にパターン形成し、バリヤ層を最上面からエッチングすることである。
【0038】
図5から図8のいずれかを参照して、この好ましい実施形態の変形形態では、凹部または堆積部あるいはその両方を改変して、平坦なまたは隆起した金属接点を実現し、金属/金属接点スイッチまたは金属/絶縁体/金属スイッチを作製することもできる。金属接点スイッチの場合には、1つまたは複数の下側電極、特に下側電極50を、誘電体20とほぼ面一になるレベル、または誘電体20の表面より若干隆起したレベルまでカプセル化することができる。この構成では、スイッチが閉じたときに、信号電極50とはり160の間に直接的な金属/金属接触が形成される。同様に、誘電率の高いものであることが好ましい誘電体材料で下側電極をカプセル化して、金属/絶縁体/金属スイッチを構築することができる。これにより、接触領域の減少という利点のない好ましい空気ギャップ・スイッチと同様の容量性スイッチが得られる。
【0039】
次に図13を参照すると、もう1つの誘電層100を付加した状態が示してあり、この誘電層100内に、標準的な加工技術を用いてバイアが作製されている。誘電層100の厚さは、下側電極と後に形成する上部はりとの間の必要な間隔によって決まる。好ましい実施形態では、誘電体100の厚さは約1000オングストロームである。バイア110は、標準的なダマシン加工によってパターン形成し、エッチングし、充填し、平坦化したものである。
【0040】
図14は、誘電体100にエッチングで形成した空洞120を示す。この空洞は、従来のフォトリソグラフィおよび反応性イオンエッチング(RIE)を用いて作製する。下側電極90の元の構造が確実に保存されるように、すなわち、下側誘電体の表面80と凹部金属75の間の関係が変わらないように注意しなければならない。金属/金属または金属/絶縁体/金属タイプのMEMSデバイスを構築するときには、この関係は重要ではない。
【0041】
図15は、SiLKやDLCなどの犠牲材料130を付加した後の構造を示す図である。(SiLKは、DowChemical社製の半導体誘電体である。)この材料は、Porous SiLKの名称で知られる製品の様々な調合物を含む。これは、γ-ブチロラクトン、Bステージ高分子、およびメシチレンからなる高分子樹脂である。優先的に用いられる別の材料は、前述のDLCやダイヤモンド状炭素である。ダイヤモンド状炭素は、コーティングを含む炭素原子の一部がダイヤモンドと同様の結合をし、多くの点でダイヤモンドに似ている無定形炭素である。同様の方法で揮発させることができるその他の有機材料について述べる。ダイヤモンド状炭素は、エネルギー・ボンバードメント状態で炭素を堆積させると生成される。高温と高圧が瞬間的に局在することにより、炭素原子の一部がダイヤモンドのように結合することになる。これらの条件は、プラズマ化学蒸着(PACVD)で得られる。堆積は、アセチレンなどの炭素含有ガスを導入して高エネルギー炭素イオンを生じることによって行う。これらの材料は、曝露される酸化性材料がない場合には、酸素プラズマ曝露によって除去することができる。有機材料の除去中に曝露される酸化性材料がある場合には、H2/CO2/CO/N2型プラズマ除去を使用する。これらのガス混合物は、反応性イオンエッチング処理の当業者には分かるであろう。最後に、誘電体100の上面とほぼ一致する程度までこの構造を平坦化する。
【0042】
図16に示す本発明の別の実施形態では、誘電層100全体を、SiLKなどのリリース材料100Aで構成する。これにより、図27および図28に示すように排気する領域を後から画定することができるので、底部空洞120(図4参照)を形成するステップをなくすことができる。空洞の縁部は、リリース材料を横方向に除去することによって画定される。この材料は、その材料を分解する酸素プラズマまたは水素プラズマに曝露することによってリリースされる。さらなる詳細は、材料研究学会(Materials Research Society)のジャーナル・オブ・マテリアルズ・リサーチ(Journal ofMaterials Research)Vol.6、No.7、1484ページ、1996年、に発表された、ジョシ(A.Joshi)およびニマガッダ(R.Nimmagadda)による「酸素プラズマ中のダイヤモンド膜とグラファイトのエロージョン(Erosionof diamond films and graphite in oxygen plasma)」と題する論文に見ることができる。
【0043】
図17は、はり160のためのレベルの形成を示す図である。この目的のために、この構造をよりよく理解するために、図5から図8を再度参照すると役立つことがある。はりの高さで、誘電材料150を再度堆積させる。金属はり160は、誘電体150の中に作製することが好ましい。このはりは、図5~図6に示すように両端を固定することも、図7~図8に示すように一端を固定することもできる。前述のように、このはりには、犠牲材料の除去を実質的に容易にするのに役立つ孔175が開けられている。図11~図12に参照番号75で示し前述したように、このはりも銅を保護するためにカプセル化することができる。
【0044】
図18は、はりの周囲の誘電体、およびはりを貫通し空洞の上にある誘電体をエッチングで除去する、フォトリソグラフィ・ステップおよびRIEステップを示す図である。これにより、はり160の可動部分が上側誘電体150から解放され、また、リリース材料100Aへのアクセスが提供される。
【0045】
図19は、リリース材料100Aを除去した後の構造130を示す図である。これは、酸素プラズマを用いて存在する有機材料を分解し、それらを排出することによって行うことが好ましい。このステップの後で、従来の方法を用いて、あるいは図23~図26を参照して本明細書の以下で述べるように誘電体カバーのリリース孔(release holes)を埋めることによって、デバイス全体のカプセル化を行うことができる。
【0046】
図20は、完成したデバイスの閉じた状態を示す図である。両端が固定されたはり160の中心を下向きに屈曲させて示してある。図面は一定の比率で描いたものではない。したがって、最終的なデバイスがよりよくイメージされ、その働きがよりよく理解されるように、このデバイスの様々な部分の真の縮尺について述べる。例えば、可動はりの厚さが10000オングストロームであるのに対して、はりが動く距離を規定する空洞はわずか200から1000オングストロームにすることができる。したがって、はりの変形はそれほど大きくない。
【0047】
図21から図24は、図17以降に挿入できる、MEMSデバイスをカプセル化する方法を述べるのに有用な代替方法を示す図である。図21を参照すると、さらに別の誘電層210が追加され、はり220の上にもう1つの空洞を形成している(図22参照)。上側空洞220をエッチングするときに、はりを通る空洞領域およびはりの周囲の空洞領域から誘電体を除去する。
【0048】
次に、図23において、エッチングした領域を、前述のようにリリース材料230で、好ましくはSiLKまたはDLCで充填し、これを平坦化する。個々でも、その下に配置したものと同じ犠牲層を使用し、このリリース・プロセスで全ての層のリリース材料が除去されるようにする。
【0049】
図24は、さらに別の誘電層240を追加し、その中にバイア250をパターン形成してエッチングした状態を示す図である。これらのバイアにより、除去すべきリリース材料へのアクセスが提供される。
【0050】
図25は、前述のようにバイア250を介してリリース材料を除去したデバイスを示す図である。
【0051】
図26は、小さなバイア250を埋める別の誘電体260を堆積させる最終的なカプセル化ステップを示す図である。
【0052】
図27および図28は、図16で参照したように、空洞/バイア誘電層をリリース材料100Aのみから形成する代替方法を示す図である。図27は、図22を参照して前述したように、はりが既に形成され、上側空洞領域220がパターン形成およびエッチングされた状態を示す。
【0053】
次に、プラズマ・プロセスを用いたリリース・プロセスを図28に示す。このリリース・プロセスは等方性であり、したがって横方向の除去速度を有する。横方向速度は、使用するリリース材料およびその材料を除去するプロセスによって決まる。リリース材料は、開いた領域の真下だけでなく、各境界から距離250だけ横方向に離れた位置でも除去される。
【0054】
好ましい実施形態に関して本発明について述べたが、前述の説明に照らせば、多くの代替形態、修正形態および変形形態が当業者には明らかであることを理解されたい。したがって、添付の特許請求の範囲の趣旨および範囲内となるこのような全ての代替形態、修正形態および変形形態を含むものとする。本明細書に記載した、または添付の図面に図示した全ての事項は、例示的かつ非限定的なものとして解釈されたい。
【産業上の利用可能性】
【0055】
通信分野、特に携帯電話などに使用される。

METHOD OF FABRICATING MICRO-ELECTROMECHANICAL SWITCHES ON CMOS COMPATIBLE SUBSTRATES

Technical Field

This invention generally relates to the fabrication of micro-electromechanical switches (MEMS), and more particularly, to the manufacture of MEMS which can be integrated into current state of the art semiconductor fabrication processes.

Background Art

Switching operations are a fundamental part of many electrical, mechanical and electromechanical applications. MEM switches have drawn considerable interest over the last few years. Products using MEMS technology are widespread in biomedical, aerospace, and communications systems.

Conventional MEMS typically utilize cantilever switches, membrane switches, and tunable capacitor structures as described, e. g., in U. S. Patents No. 6, 160, 230 to McMillan et al., No. 6, 143, 997 to Feng et al., No. 5, 970, 315 to Carley et al., and No. 5, 880, 921 to Tham et al.. MEMS devices are manufactured using micro-electromechanical techniques and are used to control electrical, mechanical or optical signal flows. Such devices, however, present many problems because their structure and innate material properties require them to be manufactured in lines that are separate from conventional semiconductor processing. This is usually due to the different materials and processes which are not compatible and, therefore, which cannot be integrated in standard semiconductor fabrication processes.

The use of materials typically used in the manufacture of MEMS, such as gold, pose obvious integration problems for integrating devices directly to on-chip applications. Even the use of polysilicon, which is widely found in the literature, poses problems due to the temperature cycles and the usual segregation of front-end of the line (FEOL) tools where the actual semiconductor devices are fabricated and the back-end of the line (BEOL) where interconnect metals are processed. Typically, the two sets are not allowed to have process crossovers from one to the other in order to prevent metallic contamination of the active devices. It is therefore unlikely to see polysilicon deposition in the back-end of the line.

Most existing processes suffer from a serious drawback in that by using standard metalization, no encapsulation is provided to protect the metal. Moreover, more than one substrate is used, oftentimes bonded together, with corresponding inherent disadvantages.

Other existing techniques only provide switching capabilities at the top of the structure, making it unlikely that integration can be achieved at all levels, as will be described hereinafter in the present invention.

Accordingly, there is a need for a process that is capable of providing MEMS devices using established BEOL materials coupled to processing that can be fully integrated so that these devices can be manufactured either in conjunction with or as an add-on module to the conventional BEOL or interconnect levels.

In order to gain a better understanding of the present invention, a conventional MEM switch will now be described with reference to Fig. 1, which shows a cross-section view of a MEM switch having both ends of a deformable beam 1 anchored in dielectric 4. The lowest level consists of a dielectric material 5 containing conductive elements 2, 2a, and 3 which will be used subsequently to connect or form the various electrical portions of the device. The conductors referenced by numerals 2 and 2a are used to provide an operating potential that causes the beam to deform. Conductor 3, which conducts a signal, is in turn connected to the beam when it is in operation.

Fig 2 shows a planar view of the same prior art MEM device of Fig. 1. In a typical implementation, deformable beam 1 is formed by polysilicon over dielectric 4, e. g., Silo2, and the surrounding material is etched away leaving a raised structure, i. e., the beam suspended above the conductors that were previously formed or which, themselves, are made of polysilicon. Then the device is subjected to electroless plating, usually of gold, that adheres to the polysilicon forming the conductive elements 1, 2, 2a and 3. The switch is operated by providing a potential difference between the beam and electrodes 2 and 2a. This voltage generates an electrostatic attraction which pulls beam 1 in contact with electrode 3, thus closing the switch.

One should note that these are all typically raised structures having a large topography when compared to conventional semiconductor devices. This in itself makes them virtually impossible to integrate into the semiconductor chip fabrication process. These devices are typically made using surface micro-machining techniques which include building on photoresist or building on a substrate, such as silicon, and then removing a portion of the substrate under the device from the backside of the substrate, again precluding integration with standard semiconductor processing.

Figure 3 illustrates a cross-section view of another version of a conventional MEM switch, wherein only one end of the beam 1 is anchored within the dielectric 4. All the other parts perform as described in Fig. 1. The same applies with regard to Fig. 4, illustrating a top-down view of the corresponding device illustrated in Fig. 3. In the latter case, the switch is operated by applying a voltage between beam 1 and control electrode 2. This causes the beam to be pulled down into contact with the signal electrode 3. When the voltage is dropped, beam 1 returns to its original position.

Typically, the gap between the beam and the control electrode substantially determines the voltage required to pull down the beam. Most literature describes devices having gaps ranging from 1 to several micrometers. These gaps are large and the voltage required is therefore higher than would be desired for most consumer applications. Reported activation voltages range from around 30 to 75 volts. This is far too high for applications like cell phones which typically operate between 3 to 5 volts. The structure of the present invention operates with gaps ranging from 200 angstroms to several thousand angstroms, producing switches having an activation voltage below 5 volts.

The aforementioned illustrative switch configurations are only some of many possible structures which are known in the art. It is worth noting that MEM switches may also be configured in an arrangement of multiple beams wired in a variety of combinations.

Stiction is of primary concern in MEMS devices. Stiction is defined as two or more surface making contact that will not release without causing some damage to the device.

Impingement is a major cause of this phenomena. The present invention addresses this problem in at least one embodiment by providing an air gap 200 when the switch is closed, as will be shown in detail with reference to Fig. 19A. Surface tension is also believed to be another major cause of striction. That explains why the present invention utilizes dry etches and processes for the release of the moving parts and subsequent processing.

Accordingly, it is an object of the invention to build MEM switches and other similar structures which are fully integrated within CMOS, bipolar or BiCMOS wafers.

It is another object to manufacture MEM switches and other similar structures with a modified damascene process.

It is a further object to build MEM switches and other similar structures utilizing copper encapsulated in a barrier material to protect the metal.

It is yet another object to ensure that the encapsulation can be integrated into BEOL copper at a temperature compatible with such a process.

Disclosure of Invention These and other objects are addressed by the present invention by providing a method of fabricating MEMS switches integrated with conventional semiconductor interconnect levels, using compatible processes and materials.

The invention described herein provides a method of fabricating a capacitive switch adaptable to produce various configurations used for contact switching and/or metal-dielectric-metal switches.

In a preferred embodiment, the process starts with a copper damascene interconnect layer made of metal conductors inlaid within a dielectric. All or portions of the copper interconnects are recessed to a degree sufficient to provide a capacitive air gap when the switch is closed and, additionally, to provide a space for a protective layer of Ta/TaN, if so desired. The metal structures that are defined within the area specified by the switch act as actuator electrodes to pull down a moveable conductive beam providing one or more paths for the switching signals to traverse. The advantage of such air gap is that the dielectric (air) is not subject to charge storage or trapping that causes reliability and voltage drift problems.

The present invention can be embodied using any number of dielectric materials, such as silicon dioxide or silicon nitride, all of which may advantageously be placed between the lower electrodes and the deformable beam of the capacitive switch.

It is worth noting that instead of recessing the electrodes to provide a gap, one may just add dielectric around and/or on the electrode. A second dielectric layer is then deposited to the desired thickness of the gap formed between the lower electrodes and the moveable beam. Vias are the formed through the second dielectric layer to provide connections between the aforementioned metal interconnect layer and the next metal layer which includes the movable beam. The layer containing the vias is then patterned and etched to provide a cavity area which contains the lower activation electrodes as well as the signal paths. The cavity is then back-filled with sacrificial release material, preferably SiLK, which is a semiconductor dielectric in the form of a polymer resin consisting of gamma-butyrolactone, B-staged polymer and mesitylene. This release material is then made substantially planar, preferably, by way of chemical-mechanical polish (CMP) to align it with the top of the dielectric. It is also possible to planarize the release material by re-imaging the same mask that was used to define the cavity area using the opposite polarity photoresist and then etching the release material from the upper surface to make it substantially co-planar with the dielectric in which the cavity was formed. This provides a planar surface upon which the beam layer can be constructed.

The layer containing the movable beam is similar to other interconnect layers typically used for the fabrication of standard CMOS devices. The beam structure preferably extends, partially or completely, across the cavity area previously defined, such that when the switch is in operation, it provides the desired electrical connections. It should also be noted that holes may be incorporated into the moveable beam. These holes are initially filled with the same dielectric that forms the layer containing the beam. These areas of dielectric within the beam are then etched out to provide greater access to the sacrificial material under the beam and, thus, aiding the release process. Next, the pattern of the cavity area or some subset of the pattern is defined and the dielectric material surrounding the beam is removed, including areas through the holes in the beam, providing access to the underlying release material. At this point the released material can be removed. The switch, suspended over the cavity area is now fully functional.

Brief Description of the Drawings The foregoing and other objects, aspects and advantages of the invention will be better understood from the following detailed description of a preferred embodiment of the invention with reference to the accompanying drawings, in which : Figs. 1 and 2 respectively show a cross-sectional view and a top-down view seen along the length of the beam of a prior art MEM switch, wherein both ends of the beam are anchored ; Figs. 3 and 4 respectively show a cross-sectional and a top-down view of another type of a prior art MEM switch wherein only one end of the beam is anchored ; Figs. 5 and 6 are, respectively, a cross-sectional view and top-down view of a MEM switch fabricated in accordance with the present invention, with both ends of the beam anchored ; Figs. 7 and 8 show respectively cross-sectional and top-down views of a MEM switch fabricated in accordance with the present invention, wherein only one end of the beam is anchored ; Figs. 9-19 illustrate the series of steps required to fabricate the MEMS device, in accordance to a preferred embodiment of the invention ; Fig. 19A shows the completed MEM switch, according to the invention, in its closed position.

Figs. 20-25 show an alternate process flow providing a fully encapsulated version of the MEM switch of the preferred embodiment.

Figs. 26 and 27 illustrate an alternative process of making the cavity and vias dielectric layer entirely out of the release material.

Best Mode for Carrying Out the Invention The primary embodiment teaches building a MEMS device in a way that may be fully integrated into a standard CMOS, bipolar, BiCMOS or other common semiconductor chip process.

Fig. 5 illustrates the initial structure, i. e., a copper damascene interconnect layer.

Underlying this, may be placed devices such as FETs, NPN transistors or HBTs (Hetero-junction Bipolar Transistors), resistors, etc., on a silicon substrate or, alternatively, on a low-loss substrate, preferably made of SiO2. It consists of copper wiring 30, 40 and 50, embedded within dielectric layer 20, preferably made of SiO2. The area referenced by numeral 30 represents the landing metal for vias 110 embedded in dielectric layer 100. The metal referenced by numeral 40 forms the actuator or control electrodes, and metal 50 represents the signal electrode. The thickness of the metal is typically on the order of 3000 to 5000 Angstroms, but it is not limited to this range. The beam 160, formed within dielectric layer 150, is shown anchored at both ends. The center of the beam is designed to move vertically over cavity 120 (see Fig. 19A).

Still referring to Fig. 5, the top surface 60 of the copper wiring 30, 40 and 50 is shown to be recessed to minimize stiction, preferably by way of a wet etch, although other means are suitable. The extent of the recess depends upon the final structure desired. In the preferred embodiment, a wet etch consisting of a dilute solution of acetic acid and hydrogen peroxide is generally used. The copper is preferably etched to a depth of about 800 Angstroms.

Fig. 6 shows a top-down view of the MEMS device illustrated in Fig. 5. Beam 160, suspended over the cavity 120, is shown having a plurality of holes 175 which will be described in more detail during the process description (Figs. 9-19). These holes are useful to facilitate the removal of the sacrificial material when constructing the cavity 120.

Fig. 7 shows a cross-section view of another type of MEMS device, wherein only one end of the switch is anchored. As in Fig. 5, copper wiring 30, 40 and 50 arc formed in dielectric layer 20, whereas vias 110 are formed in dielectric layer 100. Beam 160 is unattached at one end, allowing it to freely move in a vertical motion over cavity 120.

Fig. 8 is a corresponding top-down view of the MES device described in Fig. 7, again with moveable beam 160 being provided with holes 175.

The process used to construct the MEM switches illustrated in Figs 5-6 and 7-8 will now be described in detail.

Referring to Fig. 9, there is shown the copper wiring 30, 40, and 50 formed in dielectric layer 20, wherein the top surface of the wiring is substantially planar with respect to the top surface of dielectric layer 20.

Next, and with reference to Fig. 10, the top surface 60 of the wiring is shown recessed with respect to the top layer of dielectric layer 20. One method for recessing the metal is by way of a wet etch which is efficacious for that metal for the time required to achieve the desired depth. For instance, in the preferred embodiment, a solution consisting of water, acetic acid and hydrogen peroxide (3 liters, 15 ml and 9 ml, respectively) applied for 2. 5 minutes achieves a depth of 600 to 800 Angstroms.

Referring now to Fig. 11, a layer of barrier material 70 is deposited to encapsulate the copper to protect it for the subsequent processing. The preferred embodiment incorporates a layer 100 Angstroms thick of Ta followed by 400 Angstroms of TaN. Alternatively, the barrier material can also be made of dielectric or of any number of other suitable metallic barriers.

Fig. 12 shows the result of removing the barrier 70 from the uppermost surface of dielectric 80. This is best accomplished by a chemical-mechanical polish (CMP). The barrier material 75 is now left only in the recessed areas. Since the CMP process may erode some of the dielectric 20, care must be taken to minimize erosion of the dielectric if one is to maintain a suitable gap 200 when the switch is closed. When the CMP process is completed, the depth from the upper dielectric 80 down to the barrier 75 will be about 200 Angstroms. This can be customized in order to form the final gap for the capacitive switch.

Another method of encapsulating the conductors is to pattern the blanket barrier 70 with the same mask that defined the conductors, but with an opposite polarity photoresist, and etching the barrier layer from the uppermost surface.

In a variation to the preferred embodiment, and referring to any of Figs. 5 through 8, one may modify the recess and/or the depositions to achieve a planar or even a raised metal contact to fabricate a metal-to-metal contact switch or a metal-insulator-metal switch. In the case of a metal contact switch, one may encapsulate the lower electrode (s), especially 50, to a level that is substantially planar with dielectric 20 or even slightly raised above that surface. In this configuration, when the switch is closed, a direct metal-to-metal contact is established between signal electrode 50 and beam 160. Similarly, the lower electrodes can be encapsulated with dielectric material, preferably, one having a high dielectric constant, in order to construct a metal-insulator-metal switch. This generates a capacitive switch similar to the preferred air gap switch without the advantages of the reduced contact area.

Referring now to Fig. 13, the addition of another dielectric layer 100 is shown within which vias are fabricated using standard processing techniques. The thickness of dielectric layer 100 depends on the required spacing between the lower electrodes and the upper beam to be formed later. In the preferred embodiment, dielectric 100 is about 1000 Angstroms thick. Vias 110 are patterned, etched, filled and planarized according to standard damascene processing.

Fig. 14 shows the formation of cavity 120 etched in dielectric 100. This cavity is fabricated using conventional photolithography and reactive-ion-etching (RIE). Care must be taken to ensure that the original structure of the lower electrodes 90 is preserved, i. e., ensuring that the relationship between the lower dielectric surface 80 and the recessed metal 75 is not altered. This relationship is not critical when constructing the metal-to-metal or metal-insulator-metal versions of the MEMS device.

Fig. 15 illustrates the structure after applying a sacrificial material 130, such as SiLK or DLC. [Note : SiLK. is a semiconductor dielectric manufactured by Dow Chemical Corp. This material includes various formulations of the product, known also under the name of Porous SiLK. It is a polymer resin comprised of gamma-butyrolactone, B-staged polymer and mesitylene. Another material which is preferentially used is the aforementioned DLC, a diamond like carbon, which is an amorphous carbon containing a coating wherein a proportion of the carbon atoms is bonded in a similar manner to diamond and which resembles in many ways to diamond. Other organic materials that are capable of being volatized in a similar manner will be described. Diamond-like carbon is produced when carbon is deposited under energetic bombardment. The instantaneous localized high temperature and pressure induce a proportion of the carbon atoms to bond as diamond. These conditions are obtained during plasma assisted chemical vapor deposition (PACVD). The deposition is done with carbon containing gas, such as acetylene, which is introduced to provide the energetic carbon ions.

These materials can be removed by way of oxygen plasma exposure if there is no oxidizable material that is exposed. If there is an oxidizable material that will be exposed during the removal of the organic material, then an HJCO2/CO/N2 type plasma removal is to be used.

These gas mixtures will be recognizable to practitioners skilled in the art of reactive-ion-etch processes. The structure is finally planarized to an extent where it is substantially aligned with respect to the upper surface of dielectric 100.

In another embodiment of the invention, shown in Fig. 16, the entire dielectric layer 100 is formed of release material 100A, such as SiLK. This allows eliminating the steps of forming the bottom cavity 120 (see Fig 14) since the area evacuated can be defined later, as will be shown in Figs. 26 and 27. The edges of the cavity are defined by the lateral removal of the release material. The material is released by exposing it to an oxygen or hydrogen plasma which decomposes the material. Additional details may be found in the paper"Erosion of diamond films and graphite in oxygen plasma"by A. Joshi and R. Nimmagadda, published in the Journal of Materials Research., Vol. 6, No. 7, p. 1484, 1996, by the Materials Research Society.

Fig 17 illustrates the formation of the level reserved for beam 160. To this end, it may be useful to refer back to Figs. 5-8 for better understanding the structure. At the beam level, dielectric material 150 is again deposited. The metal beam 160 is preferably fabricated within dielectric 150. The beam may be fixed at both ends, as shown in Figs. 5-6, or at one end as illustrated with reference to Figs. 7-8. As previously mentioned, the beam may be provided with holes 175 which are useful to subsequently facilitate the removal of the sacrificial material. The beam may also be encapsulated as described earlier with reference to Figs.

11-12, referenced by numeral 75, in order to protect the copper.

Fig. 18 illustrates the photolithography and RIE steps to etch out the dielectric around and through the beam and over the cavity. This frees the moveable section of the beam 160 from the upper dielectric 150 and provides access to release material 130.

Fig. 19 illustrates the structure after removal of the release material 130. This is preferably done using an oxygen plasma to break down the organic materials that are present, allowing them to be evacuated. Following this step, encapsulation of the entire device may be performed using conventional methods or, alternatively, by pinching offthe release holes in a dielectric covering, as will be described hereinafter with reference to Figs. 22-25.

Fig. 19A depicts the completed device in its closed state. Beam 160, anchored at both ends, is shown with its center bent downwards. The drawings are not to scale. Accordingly, in order to better visualize the final device and better understand its working, the true scale of the various parts of the device will be described. For example, the cavity that defines the distance the beam is to move, may only be 200 to 1000 Angstroms for a moveable beam 10, 000 Angstroms thick. Thus, the deformation of the beam is fairly minor.

Figs. 20 through 23 shows an alternate method to be inserted from Fig. 17 onward and which is useful to describe how to encapsulate the M : EMS device. Referring to Fig. 20, yet another dielectric layer 210 is now added to form another cavity above the beam 220 (see Fig. 21). When etching the upper cavity 220, the dielectric is removed from the cavity area through and surrounding the beam.

Next, in Fig 22 the etched areas are filled with more release material 230 as previously described, preferably, with SiLK or DLC, and it is planarized. The same sacrificial material that was placed below is now used again, such that the release process removes all the layers of release material.

Fig 23 illustrates the addition of yet another dielectric layer 240 within which vias 250 are patterned and etched. These vias provide access to the release material to be removed.

Fig. 24 shows the device with the release material removed through vias 250, as described earlier.

Fig. 25 shows the final encapsulation step of depositing another dielectric 260 which pinches offthe small vias 250, thus sealing the structure.

Figures 26 and 27 illustrate an alternate method of making the cavity/vias dielectric layer entirely out of release material 100A, as referenced in Fig. 16. Fig. 26 shows the beam already formed and the upper cavity area 220 patterned and etched, as described earlier with reference to Fig. 21.

Next, the release process using the plasma process is shown in Fig. 27. The release process is isotropic and, thus, has a lateral removal rate. The lateral rate depends on the release material used and the process by which it is removed. The release material is removed directly below the open areas as well as at some lateral distance 250 from each boundary.

While the invention has been described in conjunction with a preferred embodiment, it is to be understood that many alternatives, modifications and variations will be apparent to those skilled in the art in light of the aforementioned description. Accordingly, it is intended to embrace all such alternatives, modifications and variations which fall within the spirit and scope of the appended claims. All matters set forth herein or shown in the accompanying drawings are to be interpreted in an illustrative and non-limiting sense.

Industrial Applicability Used in field of communications, and more particularly in cell phones and the like.

「特表2005-514728およびWO2003054938より引用」

|
|

[Claims] 光学回路製造方法及び装置

【特許請求の範囲】
【請求項1】
半導体基板と、2つ又はそれ以上の光学部品とを備え、1つ又はそれ以上の中空コア光導波路が前記半導体基板に形成されて、前記2つ又はそれ以上の光学部品が光学的にリンクされるようになったことを特徴とする光通信用光回路装置。
【請求項2】
前記半導体基板が1つ又はそれ以上の位置合わせスロットを含み、各々の位置合わせスロットが、光学部品を位置合わせ状態で受け取るようにされた請求項1に記載の装置。
【請求項3】
前述の2つ又はそれ以上の光学部品のうちの1つ又はそれ以上が、前記半導体基板材料から形成された前記請求項のいずれか1項に記載の装置。
【請求項4】
前記半導体基板がシリコンからなる前記請求項のいずれか1項に記載の装置。
【請求項5】
前記半導体基板が、絶縁体上シリコン(SOI)ウエハからなる前記請求項のいずれか1項に記載の装置。
【請求項6】
前記半導体基板が前記光通信用光回路装置の基体部分を形成し、蓋部分が付加的に形成されて、前記1つ又はそれ以上の中空コア光導波路が形成されるようになった前記請求項のいずれか1項に記載の装置。
【請求項7】
1つ又はそれ以上の光学部品が前記蓋部分に取り付けられた請求項5に記載の装置。
【請求項8】
前記蓋部分が半導体材料からなる請求項6ないし請求項7のいずれかに記載の装置。
【請求項9】
前記蓋部分の前記半導体材料がシリコンである請求項8に記載の装置。
【請求項10】
1つ又はそれ以上の光学部品が、前記蓋部分の前記半導体材料に形成された請求項8ないし請求項9のいずれか1項に記載の装置。
【請求項11】
前記1つ又はそれ以上の中空コア光導波路の少なくとも幾らかの内側表面が反射コーティングを支持する前記請求項のいずれか1項に記載の装置。
【請求項12】
前記反射コーティングが1つ又はそれ以上の材料層を含み、作動波長域内での導波路コアの屈折率より低い有効屈折率を有する表面を与えるようになった請求項11に記載の装置。
【請求項13】
前記反射コーティングが、金、銀、又は銅のいずれか1つで構成された少なくとも一層を含む請求項12に記載の装置。
【請求項14】
前記反射コーティングが、少なくとも一層の誘電材料を含む請求項12又は請求項13のいずれか一項に記載の装置。
【請求項15】
前記反射コーティングが、少なくとも一層の炭化ケイ素を含む請求項12に記載の装置。
【請求項16】
1つ又はそれ以上の中空コア光導波路のうちの少なくとも1つが、基本モードの伝播を支持する前記請求項のいずれか1項に記載の装置。
【請求項17】
1つ又はそれ以上の中空コア光導波路のうちの少なくとも1つが、多モードの伝播を支持する請求項1ないし請求項15のいずれかに記載の装置。
【請求項18】
前記多モード領域が、イメージ再形成が生じる長さである請求項17に記載の装置。
【請求項19】
1つ又はそれ以上の中空コア光導波路のうちの少なくとも1つが、ほぼ長方形の内側断面を有する前記請求項のいずれか1項に記載の装置。
【請求項20】
1つ又はそれ以上の中空コア光導波路のうちの少なくとも1つが、ほぼ正方形の内側断面を有する請求項19に記載の装置。
【請求項21】
前記長方形の中空コア光導波路が、第1の導波路壁に対して平行な第1の断面寸法と、前記第1の断面寸法に対して垂直の第2の断面寸法とを有し、前記第1の断面寸法が、少なくとも10%だけ前記第2の断面寸法より大きい請求項19に記載の装置。
【請求項22】
前記少なくとも1つの長方形の内側断面の中空コア光導波路を定める表面の屈折率が、ほぼ等しい請求項19から請求項21までのいずれか1項に記載の装置。
【請求項23】
前記長方形の内側断面の中空コア光導波路を形成する向かい合う表面が、ほぼ等しい有効屈折率を有し、該前記長方形の内側断面の中空コア光導波路を形成する表面のうちの隣接する表面が、異なる有効屈折率を有する請求項19から請求項21までのいずれか1項に記載の装置。
【請求項24】
前記長方形の内側断面の中空コア光導波路の一対の向かい合う表面が、高屈折率のコーティングをもつ請求項23に記載の装置。
【請求項25】
0.1μmから20μmまでの波長帯における放射線で作動する前記請求項のいずれか1項に記載の装置。
【請求項26】
3μmから5μmまでの波長域における放射線で作動する前記請求項のいずれか1項に記載の装置。
【請求項27】
8μmから12μmまでの波長域における放射線で作動する請求項1から請求項25までのいずれか1項に記載の装置。
【請求項28】
1.4μmから1.6μmまでの波長域における放射線で作動する請求項1から請求項25までのいずれか1項に記載の装置。
【請求項29】
前記半導体基板が、光学繊維ケーブルを受け取るようにされ、かつ、前記光学繊維ケーブルを前記半導体基板の前記1つ又はそれ以上の中空コア光導波路のうちの1つと光学的に結合するようにされた少なくとも1つの位置合わせスロットを備えた前記請求項のいずれか1項に記載の装置。
【請求項30】
モード適合手段が、付加的に、前記位置合わせスロット付近に設けられて、光学繊維のモードと、異なるコア直径の中空コア光導波路の類似したモードとの間の結合を可能にするようになった請求項29に記載の装置。
【請求項31】
前記モード適合手段が、GRIN又は球面レンズのいずれか1つである請求項30に記載の装置。
【請求項32】
前記位置合わせスロットが中空コア光導波路を受け取るように配置された請求項29から請求項31までのいずれか1項に記載の装置。
【請求項33】
前記位置合わせスロットが、レンズ形状の光学繊維を受け取るようにされた請求項29に記載の装置。
【請求項34】
前記2つ又はそれ以上の光学部品のうちの少なくとも1つが、マイクロ電子機械(MEMS)装置を含むようになった前記請求項のいずれか1項に記載の装置。
【請求項35】
前記2つ又はそれ以上の光学部品のうちの少なくとも1つが、光を、前記半導体基板の平面の外に向けるように傾斜された鏡面を含むようになった前記請求項のいずれか1項に記載の装置。
【請求項36】
前述の請求項のいずれか1項に記載の装置であって、さらに少なくとも1つのマイクロ波部品を含むようになった装置。
【請求項37】
前記半導体基板が、付加的に、中空コアマイクロ波導波路を含むようになった前述の請求項のいずれか1項に記載の装置。
【請求項38】
1つ又はそれ以上の中空チャネルが形成された半導体基板を含む光通信用光回路の基体部分であって、前記基体部分が、適切な蓋部分と組み合わされたときに、少なくとも1つの中空コア光導波路が形成されるように配置されることを特徴とする基体部分。
【請求項39】
少なくとも1つのスロットが、前記基体部分の前記半導体基板に形成されて、光学部品を位置合わせ状態で受け取るようになった請求項38に記載の光通信用光回路の基体部分。
【請求項40】
1つ又はそれ以上の中空導波路チャネルと、光学部品を位置合わせ状態で受け取る少なくとも1つのスロットとが形成された半導体基板を備えることを特徴とする光通信用光回路の基体部分。
【請求項41】
請求項38から請求項40までのいずれか1項に記載された基体部分を取り出す段階と、これに蓋を取り付ける段階とを含む光通信用光回路を製造する方法。
【請求項42】
使用において、中空コア光導波路として好適な1つ又はそれ以上の中空チャネルを半導体基板に微細加工する段階を含む光通信用光回路を製造する方法。
【請求項43】
光学部品の受動的な位置合わせのために、スロットを前記半導体基板に製造する付加的な段階を含む請求項42に記載される方法。
【請求項44】
前記中空チャネルの内側表面を、作動波長域内での前記導波路コアの屈折率より低い屈折率を有する材料層により被覆する付加的な段階を含む請求項42ないし請求項43のいずれかに記載の方法。
【請求項45】
(a)少なくとも1つの中空コア光導波路と、光学部品を受け取る少なくとも1つのスロットとが形成された半導体基板を取り出す段階と、
(b)光学部品を前記少なくとも1つのスロットに導入する段階と、
からなり、前記光学部品を少なくとも1つのスロットに導入する段階が、さらに前記光学部品を位置合わせするように作用することを特徴とする光通信用光回路を形成する方法。
【請求項46】
変形可能な材料の層にパターンを形成するのに適したマスターであって、前記マスターが、前記変形可能な材料において、少なくとも1つの中空導波路チャネルと、光学部品を位置合わせ状態で受け取るように配置された少なくとも1つの位置合わせスロットとを形成するように適切にパターン形成された半導体材料を備えることを特徴とするマスター。
【請求項47】
(a)請求項43に記載されたマスターを用いて、変形可能な材料の層に、恒久的にパターンを形成する段階と、
(b)少なくとも1つの光学部品を、前記変形可能な材料に形成された少なくとも1つの位置合わせスロットの中に導入する段階と、
からなる光通信用光回路を形成する方法。

1. A photonic light circuit device comprising a semiconductor substrate and two or more optical components wherein one or more hollow core optical waveguides are formed in the semiconductor substrate to optically link said two or more optical components.

2. A device as claimed in claim 1 wherein the semiconductor substrate comprises one or more alignment slots, each alignment slot being adapted to receive in alignment an optical component.

3. A device as claimed in any preceding claim wherein one or more of said two or more optical components are formed from the material of the semiconductor substrate.

4. A device as claimed in any preceding claim wherein the semiconductor substrate comprises silicon.

5. A device as claimed in any preceding claim wherein the semiconductor substrate comprises a silicon on insulator (SOI) wafer.

6. A device as claimed in any preceding claim wherein the semiconductor substrate forms a base portion of the photonic light circuit device and a lid portion is additionally provided to form said one or more hollow core optical waveguides.

7. A device as claimed in claim 5 wherein one or more optical components are attached to the lid portion.

8. A device as claimed in any of claims 6-7 wherein the lid portion comprises semiconductor material. 9. A device as claimed in claim 8 wherein the semiconductor material of the lid portion is silicon.

10. A device as claimed in any one of claims 8-9 wherein one or more optical components are formed in the semiconductor material of the lid portion.

11. A device according to any preceding claim wherein at least some of the internal surfaces of said one or more hollow core optical waveguides carry a reflective coating.

12. A device as claimed in claim 11 wherein the reflective coating comprises one or more layers of material to provide a surface having an effective refractive index lower than that of the waveguide core within the operating wavelength band.

13. A device as claimed in claim 12 wherein the reflective coating comprises at least one layer of any one of gold, silver or copper 14. A device as claimed in any one of claims 12 or 13 wherein the reflective coating comprises at least one layer of dielectric material.

15. A device as claimed in claim 12 wherein the reflective coating comprises at least one layer of Silicon Carbide.

16. A device as claimed in any preceding claim wherein at least one of the one or more hollow core optical waveguides support fundamental mode propagation.

17. A device as claimed in any of claims 1-15 wherein at least one of the one or more hollow core optical waveguides support multi-mode propagation.

18. A device according to claim 17 wherein the multi-mode region is of a length such that re-imaging occurs. 19. A device as claimed in any preceding claim wherein at least one of the one or more hollow core optical waveguides has a substantially rectangular internal cross- section.

20. A device as claimed in claim 19 wherein at least one of the one or more hollow core optical waveguides has a substantially square internal cross-section.

21. A device as claimed in 19 in which the rectangular hollow core optical waveguide has a first cross-sectional dimension parallel to a first waveguide wall and a second cross-sectional dimension orthogonal to said first cross-sectional dimension wherein the first cross-section dimension is at least 10% greater than the second cross- sectional dimension.

22. A device as claimed in any on of claims 19 to 21 wherein the refractive indices of the surfaces defining the at least one rectangular internal cross-section hollow core optical waveguide are substantially equal.

23. A device as claimed in any one of claims 19 to 21 wherein opposite surfaces forming the rectangular internal cross-section hollow core optical waveguide have substantially equal effective refractive indices and adjacent surfaces forming the rectangular internal cross-section hollow core optical waveguide have different effective refractive indices.

24. A device as claimed in claim 23 wherein a pair of opposed surfaces of the rectangular internal cross-section hollow core optical waveguide carry a high refractive index coating.

25. A device as claimed in any preceding claim for operation with radiation within the wavelength range of 0. 1 tm to 20um.

26. A device as claimed in any preceding claim for operation with radiation within the wavelength bands of 3Rm to 5, um.

27. A device as claimed in any one of claims 1 to 25 for operation with radiation within the wavelength bands of 8Fm to 12Fm.

28. A device as claimed in any one of claims 1 to 25 for operation with radiation within the wavelength bands of 1. 4Rm to 1. 6um.

29. A device according to any preceding claim wherein the semiconductor substrate comprises at least one alignment slot arranged to receive an optical fibre cable and to optically couple said optical fibre cable with one of said one or more hollow core optical waveguide of the semiconductor substrate.

30. A device according to claim 29 wherein a mode matching means is additionally provided in the vicinity of the alignment slot to allow coupling between the modes of an optical fibre and the analogous modes of a hollow core optical waveguide of a different core diameter.

31. A device according to claim 30 wherein the mode matching means is any one of a GRIN or ball lens.

32. A device according to any one of claims 29 to 31 wherein the alignment slot is arranged to receive a hollow core optical fibre.

33. A device according to claim 29 wherein the alignment slot is arranged to received a lensed optical fibre.

34. A device according to any preceding claim wherein at least one of said two or more optical components comprises a micro-electro-mechanical (MEMs) device.

35. A device according to any preceding claim wherein at least one of said two or more optical components comprises a mirrored surface that is angled to direct light out of the plane of the semiconductor substrate.

36. A device as claimed in any one of the preceding claims and further comprising at least one micro-wave component.

37. A device as claimed in any one of the preceding claims wherein the semiconductor substrate additionally comprises a hollow core microwave waveguide.

38. A base portion for a photonic light circuit comprising a semiconductor substrate having one or more hollow channels formed therein, wherein said base portion is arranged such that when combined with an appropriate lid portion at least one hollow core optical waveguide is formed.

39. A base portion for a photonic light circuit as claimed in claim 38 wherein at least one slot is formed in the semiconductor substrate of the base portion to receive in alignment an optical component.

40. A base portion for a photonic light circuit comprising a semiconductor substrates in which one or more hollow waveguide channels and at least one slot to receive in alignment an optical component are formed.

41. A method of fabricating a photonic light circuit comprising the steps of taking a base portion as claimed in any one of claims 38 to 40 and attaching a lid thereto.

42. A method of fabricating a photonic light circuit device comprising the step of micro-fabricating one or more hollow channels in a semiconductor substrate that are suitable, in use, as hollow core optical waveguides.

43. A method as claimed in claim 42 and comprising the additional step of fabricating slots in the semiconductor substrate for the passive alignment of optical components therein. 44. A method as claimed in any of claims 42-43 and comprising the additional step of coating the internal surfaces of the hollow channel (s) with a layer of material having a refractive index lower than that of the waveguide core within the operating wavelength band.

45. A method of forming a photonic light circuit comprising the steps of; (a) taking a semiconductor substrate in which at least one hollow core optical waveguide and at least one slot to receive an optical component are formed, and (b) introducing an optical component into the at least one slot, whereby the step of introducing the optical component into the at least one slot also acts so as to align said optical component.

46. A master suitable for forming a pattern in a layer of deformable material wherein the master comprises semiconductor material appropriately patterned to form in said deformable material, at least one hollow waveguide channel and at least one alignment slot wherein said at least one alignment slot is arranged to receive in alignment an optical component.

47. A method of forming a photonic light circuit comprising the steps of; (a) using a master as claimed in claim 43 to permanently form a pattern in a layer of deformable material ; and (b) introducing at least one optical component into the at least one alignment slot formed in the deformable material.

「特表2005-516253およびWO2003065091より引用」

|
|

光学回路製造方法及び装置

【発明の詳細な説明】
【技術分野】
【0001】
本発明は、一般に、統合された光学装置に関し、より具体的には、改良された光通信用光回路(PLC)装置に関する。
【背景技術】
【0002】
光通信用回路モジュールは、多くの光通信、センサ、及び器械装置の一体部分を形成する。このような光通信用回路装置においては、多数の光学部品が適当な位置に剛に保持されており、典型的には光学繊維の長さの導波路が、要求に応じて、部品を光学的に接続するのに用いられる。光学部品及び相互接続繊維は、好適な基板上の適当な位置に保持される。
【0003】
シリコン光学ベンチ(SiOB)は、光通信用回路のための組立技術の一例である。名称が示唆するように、SiOBは、シリコン又は同様の半導体材料から形成された光学ベンチである。溝及びスロットが微細加工工程を用いてシリコン材料にエッチングされて、種々の光学部品を保持する。高精度の微細加工工程は、光学部品及び光学繊維が、種々のスロット及び溝において、互いに対して精密に位置合わせされることを可能にする。このことは、部品のいわゆる「受動的な位置合わせ」であり、光学回路の種々の部品を、互いに対して位置合わせすることを積極的に確実にする必要性を減少させる。光はさらに、レンズなどのような自由空間光学を用いて、種々の光学部品間に向けられることがある。
【0004】
さらに、種々の材料層を堆積して、SiOBと一体的な導波路を形成できることが周知であり、例えば、US4902086及びEP0856755を参照されたい。典型的には、シリカのような基部層がシリコン基板上に形成される。高屈折率をもつドープ処理されたシリカ層、すなわちコア層が、次に、低屈折率の基部層の上部に堆積される。コア層は、適切な導波路を形成するようにパターン形成される。任意的に、低屈折率材料の上方被覆層を、さらにパターン加工されたコア層上に堆積する。換言すると、導波路は、別々の光学繊維として製造されるのではなく、シリコン基板上に直接形成される。
【0005】
SiOBに基づくものを含む公知の光通信用回路装置の欠点は、効率的な光学接続を確実にするために各光学部品が関連する導波路と位置合わせされなければならない高精度である。さらに、光学繊維と光学部品との正確な物理的位置合わせを確実にするために、各シリカ導波路の端部からの望ましくない反射を最小にすることも必要である。このことは、光学部品に対する導波路の屈折率整合、又は、ゲル又は反射防止コーティングの使用を必要にする。レンズは、さらに、部品間の光の自由空間結合を可能にすることが要求される。これらの必要条件は、光通信用回路製造の複雑さを増加させ、したがって、費用を増加させることになる。
【発明の開示】
【0006】
本発明の目的は、少なくとも幾つかの上述の欠点を緩和することである。
【0007】
本発明の第1の態様においては、光通信用光回路装置は、半導体基板と、2つ又はそれ以上の光学部品とを備え、前述の2つ又はそれ以上の光学部品を光学的にリンクさせる1つ又はそれ以上の中空コア光導波路が該半導体基板に形成される。
【0008】
本発明は、部品間の相互接続のために光学繊維を与えること、又は、中実コア導波路を形成する材料層を堆積することに対する必要を取り除くため、従来技術の回路装置に比べて有利である。本発明は、従来技術の装置より製造が容易であり、したがって、低価格である光通信用光回路(PLC)を与える。
【0009】
中空光導波路により部品をリンクさせることにおけるさらに別の利点は、光学部品を相互接続するのに中実コア(典型的には、シリカ又はシリコン)導波路を用いる従来技術の光通信用回路に対して、回路が取り扱うことができる光学的パワーが増加することである。さらに、屈折率整合ゲル又はエポキシ、或いは反射防止コーティングは、導波路の面に必要とされない。
【0010】
中空導波路は、光をPLCの部品間に導くように形成される。光学部品は、例えば、ビーム分割器/再結合器、エタロン構造、レンズ、波長板、変調器、レーザ、光検知器、又は作動された光学部品のような光信号を生成するか、検知するか、又はこれに作用するあらゆる装置である。光学部品という用語は、さらに、中空導波路に形成されるか又は該中空導波路から形成される表面格子特性などのような光学的構造を含むものとする。中空コア導波路は、平面、すなわち以下に説明するように二次元的に導波するものとすることができる。光学部品は、さらに、例えば、光をPLCに対して、又は該PLCから外に結合するのに用いられる光学繊維ケーブルのような光学繊維とすることができる。
【0011】
半導体基板は、微細加工技術を用いて、高精度でエッチングすることができる。基板は、例えば、SiGe又は絶縁体上シリコン(SOI)又はガラス上シリコンのような多層ウエハを含むという利点を有する。当業者は、微細加工技術は、典型的には、パターンを定める平板印刷技術段階と、これに続いて、このパターンを基板材料上であるか該基板材料における1つ又はそれ以上の層に変換する段階とを含むことが分かるであろう。平板印刷技術段階は、写真平板技術、x線、又はeビーム平板印刷技術を含むことができる。エッチング段階は、イオンビームミリング、化学的エッチング、ドライ・プラズマ・エッチング、又は深奥乾式エッチング(深奥シリコンエッチングともいわれる)を用いて実行することができる。この種類の微細加工技術は、さらに、スパッタリング、CVD、及び電気めっきのような種々の層堆積技術と両立性がある。
【0012】
半導体基板は、光学部品を位置合わせ状態で受け取るようにされた1つ又はそれ以上の位置合わせスロットを含むことが有利である。位置合わせスロットは、光学部品を受け入れるのに必要とされる形状に形成され、したがって、中空コア光導波路より深い/浅い、及び/又は、幅広い/狭いことになる。
【0013】
位置合わせスロットは、したがって、これらが受け取る光学部品を位置合わせするのに十分な精度をもって製造されることになる。光学部品をこのような位置合わせスロットに置くことは、該光学部品を本質的に位置合わせすることになり、部品の位置合わせ又は調整段階は、必要とされない。電子回路などの製造に用いられる種類の通常のピック・アンド・プレイス技術を用いて、光学部品を位置合わせスロットに置いでもよい。
【0014】
或いは、ピック・アンド・プレイス技術は、必要な位置合わせを与えることができる。例えば、部品が置かれ、次いで位置合わせ状態のまま残るように固定された(例えば、のりづけされた)とき、この部品を正確に位置合わせすることができる。
【0015】
位置合わせスロットと、(特に)光学部品は、特定の大きさの許容差で製造される。光学部品と関連する中空コア光導波路との間の結合効率は、該中空コア導波路に対する該光学部品の位置合わせにおける角度方向誤差が減少すると、増加する。しかし、中空コア導波路の横断面寸法を減少させることは、角度方向位置合わせ許容範囲を増加させるものとなるが、コア寸法の減少及び横方向位置合わせ許容誤差の増大(厳格化)のために光導波路における損失が僅かに増加するという犠牲を伴う。
したがって、特定の光部品により達成される位置合わせ許容誤差の知識(例えば、光学部品の製造許容差の知識による)は、中空コア導波路の寸法を、高い結合効率が確実になるように選択することを可能にする。
【0016】
位置合わせスロットは、さらに、中実コアコア光学繊維を適当な位置でクランプし、したがって、光学的入力/出力がPLCに対してなされるのを可能にするように形成することができる。階段状の光学繊維位置合わせスロットをさらに形成して、バッファ層及び被覆の両方を保持することができる。例えば、位置合わせスロットにおける光学繊維の被覆をクランプすることにより達成された、中空コア光学繊維のコアとPLC上の中空コア導波路との位置合わせは、空心対空心の接続は、如何なる望ましくない反射もないために、特に有利である。
【0017】
光学繊維のコアとPLCの中空コア導波路との間に有効な結合を与えるために、該中空コア導波路の断面は、該光学繊維のコアの断面に適したものであるべきである。中実コアコア繊維の場合には、被覆の中への漏れは、該繊維により支持されるモード幅が、実際にはコア直径より大きいことを意味し、例えば、典型的には、10μmの中実コア単一モードガラス繊維は、約14μm直径の合計フィールド幅を有する。モード幅が、中空コア導波路のそれとは異なる場合には、レンズ(例えば、ボール又はGRINロッドなど)を用いて、光が、PLCの中空コア導波路のそれとは異なる大きさのコアをもつ繊維に/から結合されるように光学的フィールドを拡げるか又は減少させることができる。中実コア繊維の繊維端部は、反射防止性のものとすることができる。
【0018】
2つ又はそれ以上の光学部品のうちの1つ又はそれ以上は、半導体基板材料から形成されており、すなわち、モノリシック部品を形成できることが好都合である。
【0019】
或いは、PLCを構成し、かつ半導体基板に形成された中空コア導波路を介して相互接続される幾つか又はすべての光学部品を、上述のように半導体基板に取り付けることができ、換言すると、混成装置を形成することができる。
【0020】
前述の2つ又はそれ以上の光学部品の少なくとも1つは、マイクロ電子機械(MEMS)装置を含むという利点を有する。MEMS部品は、混成であってもモノリシックであってもよい。MEMSは、マイクロマシン加工要素、マイクロシステム技術、マイクロロボティクス、及びマイクロ工学を含むものとされる。MEMS光学部品の例は、位置合わせ要素、ポップダウン・フレスネルレンズ、ジャイロスコープ、可動ミラー、調整可能なファブリ・ペロキャビティ、適応光学要素、スイッチ、可変光学減衰器、フィルタなどを含む。
【0021】
半導体基板は、光通信用光回路装置の基体部分を形成し、前述の中空コア光導波路を形成するために、蓋部分が付加的に形成されることが好都合である。
【0022】
1つ又はそれ以上の光学部品が蓋部分に取り付けられることが有利である。光学部品は、単独で蓋に取り付けてもよいし、単独で基体部分に取り付けてもよいし、又は、該蓋及び該基部の両方に取り付けてもよい。
【0023】
蓋部分は、シリコンのような半導体材料から形成することができ、1つ又はそれ以上の光学部品をその上に形成できるという利点を有する。或いは、蓋部分はガラスから形成することができる。例えば、蓋が基板と同じ半導体材料から形成されることにより、該蓋は、該基板と同じ熱膨張特性を有するべきであることが好ましい。
【0024】
蓋に取り付けられた部品の場合においては、基体部分は、中空コア導波路構造を形成するように、及び蓋部分から形成された、又は該蓋部分に取り付けられた光学部品のための凹部を与えるようにエッチングされる。蓋部分を基体部分に取り付けることは、光学部品を該基体部分の光導波路と位置合わせすることを可能にする。当業者は、精密な位置合わせ嵌合部品又はウエハ又はチップ位置合わせ工具のような種々の技術を、蓋及び基部の正確な位置合わせを確実にするために与えることができることが分かるであろう。或いは、光学部品の幾つか又はすべてを、直接、基体部分に形成された位置合わせスロットに取り付けることができる。このことは、蓋部分及び基体部分を精密に位置合わせすることなく、該蓋部分を該基体部分に取り付けることを可能にする。
【0025】
蓋部分は、反射コーティングを支持することが好都合である。反射コーティングは、要求に応じて、蓋部分のすべて、又は該蓋部分の選択された部分だけを覆うことができる。反射コーティングは、作動波長域内での導波路コアの屈折率より低い屈折率を有する、例えば、金、銀、又は銅のような材料層を含むという利点を有する。或いは、1つ又はそれ以上の誘電材料層であるか又は炭化ケイ素の層を形成することができる。
【0026】
当業者は、蓋部分及び基体部分を互いにどのように接合できるかが分かるであろう。例えば、導電又は非導電エポキシのような中間層を用いることができる。或いは、金属層が低屈折率層として用いられる場合には、金属-半導体共晶接合を形成することができる。ガラスフリット技術を用いて、蓋を半導体基体部分に接合するか、又は、蓋部分がガラスから形成されている場合には、陰極技術を用いることができる。
【0027】
半導体基板は、シリコンからなることが有利である。これは、例えば、ウエハ形態(例えば、Si、絶縁体上シリコン、又はガラス上シリコン)、のような種々の形態であるか、又はSi基板上のエピタキシャル層(例えば、SiGe又はGaAs)として形成することができる。SOIを用いることが有利である。
【0028】
1つ又はそれ以上の中空コア光導波路を形成する第1の内側表面の光学的特性は、その中空コア光導波路を形成する第2の内側表面の光学的特性とは異なることが好都合である。このことは、以下に図6に関してより詳細に説明されるように、特定の偏光の光をより効率的に導く中空導波路を形成することを可能にする。
【0029】
前述の1つ又はそれ以上の中空コア光導波路の少なくとも幾つかの内側表面は、反射コーティングを支持することが有利である。
【0030】
反射コーティングは、作動波長域内での導波路コアの屈折率より低い屈折率を有する材料層を含むという利点を有する。
【0031】
中空コア導波路より低い屈折率を有する材料層は、PLC導波路内の光の全内反射(TIR)を与え、したがって、光損失の量を減少させる。
【0032】
中空コア光導波路構造が生成されたときには、中空コアは、空気で充填される傾向がある。したがって、ここでは、コアの屈折率は、気圧及び温度における空気の屈折率であると仮定される
イメージ ID=000003

しかしながら、これは、本発明の範囲を制限するものであると決して見なされるべきではない。中空コアは、あらゆる流体(例えば、液体であるか又は窒素のような不活性ガス)を含むことができ、又は真空であってもよい。中空コアという用語は、単純に、如何なる固体材料もないコアのことを意味する。さらに、全内反射(TIR)という用語は、ここでは、減衰された全内反射(ATIR)を含むものとする。
【0033】
中空コア光導波路の内側表面上に支持される反射材料は、金、銀、又は銅のような金属であることが好都合である。金属は、金属の物理的特性により管理される波長帯において、好適な低い屈折率を示し、1998年、ロンドン、アカデミックプレスのE.D.Palilkによる「光学的定数の手引き」のような標準的な教科書は、種々の材料の波長に依存する屈折率についての正確なデータを与える。具体的には、金は、約500nmから2.2μmまでの範囲内の波長において、空気の屈折率より低い屈折率を有し、これは、1400nmから1600nmまでの重要な電気通信帯域内の波長を含む。銅は、560nmから2200nmまでの波長帯において、1より低い屈折率を示し、銀は、320nmから2480nmまでの波長帯において、同様の屈折率特性を有する。
【0034】
当業者に周知の種々の技術を用いて、金属層を堆積することができる。これらの技術は、スパタリング、蒸着、化学気相反応法(CVD)、及び(電気又は無電解)めっきを含む。CVD及びめっき技術は、顕著な方向に依存する厚さの変動なしで、金属層を堆積することを可能にする。回転サンプル及び/又はソースを用いるスパッタリングは、さらに、カバーさえも与える。めっき技術は、バッチ(すなわち、多基板平行)処理を担うことを可能にするため、特に有利である。
【0035】
当業者は、金属層を堆積する前に、接着層及び/又は隔壁層を、中空導波路上に堆積できることが分かるであろう。例えば、金を堆積する前に、クロム又はチタンの層を接着層として設けることができる。さらに、金を堆積する前に、プラチナのような隔壁層を接着層の上に堆積することができる。或いは、組み合わされた接着及び拡散層(窒化チタン、チタンタングステン合金、又は絶縁層のような)を用いることができる。
【0036】
反射コーティングを、すべて絶縁性の又は金属誘電体のスタックにより、中空導波路(あらゆる蓋部分を含む)の内側表面上に形成できることが好都合である。当業者は、誘電体層の光学的厚さは、コーティングの反射特性を定める干渉効果を与えることが分かるであろう。誘電材料は、CVDであるか、スパッタリングであるか、又は反応性スパッタリングにより堆積することができる。或いは、誘電体層は、堆積された金属層との化学反応により形成することができる。例えば、銀の層をハロゲン化物と化学反応させて、ハロゲン化銀の薄い表面層を生成することができる。
【0037】
換言すると、反射コーティングは、すべて絶縁性であるか又は金属誘電体のスタックにより形成することができる。当業者は、誘電体層の光学的厚さは、要求される干渉効果を与え、したがって、該コーティングの反射特性を定めることが分かるであろう。コーティングの反射特性は、さらに、中空コア導波路が形成される材料の特性に、ある程度依存することがある。したがって、下にある半導体基板はさらに、基部層を形成することができ、あらゆるこのような多層の誘電体スタックの一部になる。
【0038】
さらに、中空コア導波路の内側表面に支持される材料層は、炭化ケイ素であることが好都合である。
【0039】
上述のように、あらゆる要求される波長において、有効な作動を与えるように、低屈折率材料の層を選択することができる。炭化ケイ素は、10.6μmにおいて0.06の屈折率を有し、このような材料を、このような波長で作動する装置に含むのに特に適したものにする。
【0040】
1つ又はそれ以上の中空コア光導波路のうちの少なくとも1つのは、ほぼ長方形の(ここでは正方形を含むものとする)断面を有することが有利である。正方形の、又はほとんど正方形の断面の中空コア導波路は、損失が、実質的に、偏光非依存性のものであり、光の偏光状態が分からないか又は変化する場合に好ましい導波路をもたらす。
【0041】
長方形の中空コア導波路は、第1の導波路壁に対して平行な第1の断面寸法と、前述の第1の断面寸法に対して垂直な第2の断面寸法とを有し、該第1の断面寸法は、該第2の断面寸法より少なくとも5%であるか、10%であるか、15%であるか、25%であるか、又は50%だけ大きいことが好ましい。図7Dに関して述べられるように、このような導波路は、公知の偏光の直線偏光に好ましい。
【0042】
少なくとも1つの長方形の内側断面の中空コア光導波路を定める表面の屈折率は、ほぼ等しいことが有利である。このことは、導波路における偏光依存の損失を減少させることができる。
【0043】
長方形の内側断面の中空コア導波路を形成する向かい合う表面は、ほぼ等しい有効屈折率を有し、該長方形の内側断面の中空コア導波路を形成する隣接する表面は、異なる有効屈折率を有することが好ましい。以下の図7Aから図7Cまでを参照して述べられるように、対向する導波路壁の対の屈折率を調整することは、公知の直線偏光を導くときに、透過損失を減少させることを可能にする。
【0044】
長方形の内側断面の中空コア光導波路の一対の向かい合う表面は、高屈折率のコーティングを支持することが有利である。このことは、以下に述べられるように、s偏光された光が反射される場合に好ましい高屈折率を与える。
【0045】
基板の半導体材料を、さらに、ドープ処理して、中空コア導波路の損失を減少させるようにその光学的特性を修正することができる。
【0046】
1つ又はそれ以上の中空コア導波路のうちの少なくとも1つは、基本モードの伝播を支持することが好都合である。さらに、1つ又はそれ以上の中空コア導波路のうちの少なくとも1つは、多モードの伝播を支持するという利点を有する。多モード領域は、以下により詳細に説明されるように、イメージ再形成が生じる長さであることが好ましい。
【0047】
当業者は、中空コア導波路の形状及び寸法は、関連する光学的導波特性に影響を与えることが分かるであろう。例えば、先細になった中空導波路を用いて、ビームの拡がり又は圧縮機能を与えることができる。微細加工技術を用いて、中空コア導波路を製造することができる高解像度は、要求に応じて、PLC作動を最適化するように誘導特性を調整することを可能にする。しかしながら、当業者は、中空コア光導波路の形状は、用いられる微細加工工程の種類によって、ある程度決定されることが分かるであろう。例えば、v字形態の溝を[100]シリコンに容易に湿式エッチングすることができ、長方形の導波路を湿式エッチングにより[110]シリコンに容易に形成することができる。しかしながら、深奥反応性イオンエッチング(DRIE)は、製造において最大の容易さを与える。
【0048】
この装置は、0.1μmから20μmまで、0.8μmから1.6μmまでの波長帯内、より好ましくは1.4μmから1.6μmまでの範囲における放射線による作動のために形成されることが有利である。金、銀、及び銅のコーティングは、したがって、これらの金属を、電気通信波長域における作動のために(すなわち、約1.55μmに集中した波長における使用)、PLC装置に含むのに特に適したものにする。この装置は、3ないし5μm又は8ないし12μmの赤外線帯域で作動できるという利点を有する。
【0049】
半導体基板は、光学繊維ケーブルを受け取るようにされ、かつ前述の光学繊維ケーブルを前述の半導体基板の1つ又はそれ以上の中空コア光導波路のうちの1つと光学的に結合するように配置された少なくとも1つの位置合わせスロットを含むことが好都合である。
【0050】
さらに、モード適合手段を位置合わせスロット付近に設けて、光学繊維のモードと、異なるコア直径の中空コア光導波路の類似したモードとの間の結合を可能にするという利点を有する。例えば、基本モードの光学繊維の場合においては、モード適合手段は、この繊維の基本モードと中空コア導波路の基本モードとを結合する。多モードの伝播の場合においては、光学繊維のモードスペクトルは、中空コア導波路のモードスペクトルに適合させられる。モード適合手段は、GRINロッド、球面レンズ、通常のレンズ、又はフレスネルレンズを含むという利点を有する。
【0051】
或いは、位置合わせスロットは、レンズ形状にされた光学繊維を受け取るように配置することができる。
【0052】
位置合わせスロットは、中空コアの光学繊維を受け取るように配置されることが好ましい。光学繊維は、多モードであってもよいし、単一モードであってもよい。
【0053】
前述の2つ又はそれ以上の光学部品のうちの少なくとも1つは、光を、半導体基板の平面の外に向けるように傾斜された鏡面を含むことが有利である。鏡面は、モノリシック構成(例えば、図15において述べられるような傾斜した半導体表面)であってもよいし、ハイブリッド構成であってもよい。換言すると、PLCは、光を基板表面の平面に導波することに限定されていない。光は、基板の平面の外に向けることができる。例えば、スタックされた又は3次元のPLCを、本発明により製造することができる。
【0054】
PLCは、さらに、少なくとも1つのマイクロ波部品及び/又は中空コアマイクロ波導波路を含むことができるという利便性を有する。換言すると、光学的/マイクロ波混成回路を形成することができる。
【0055】
本発明の第2の態様においては、光通信用光回路のための基体部分は、1つ又はそれ以上の中空チャネルが形成された半導体基板を含み、前述の基体部分は、適切な蓋部分と組み合わせされた場合には、少なくとも1つの中空コア光導波路が形成されるように配置される。
【0056】
少なくとも1つのスロットを基体部分の半導体基板に形成して、光学部品を位置合わせ状態で受け取ることが好都合である。
【0057】
本発明の第3の態様においては、光通信用光回路のために基体部分は、1つ又はそれ以上の中空導波路チャネルと、光学部品を位置合わせ状態で受け取る少なくとも1つのスロットとが形成された半導体基板を含む。
【0058】
本発明の第4の態様においては、光通信用光回路を製造する方法は、本発明の第2又は第2の態様により、基体部分を取り出す段階と、これに蓋を取り付ける段階とを含む。
【0059】
本発明の第5の態様においては、光通信用光回路装置を製造する方法は、使用において、中空コア導波路として作用するのに適した1つ又はそれ以上の中空チャネルを半導体基板に微細加工する段階を含む。
【0060】
光学部品の適切な受動的な位置合わせのために、スロットを半導体基板に製造する付加的な段階を実行することが好都合である。スロットは、微細加工技術を用いて、又はレーザ加工のような精密工学により製造することができる。
【0061】
この方法は、中空チャネルの内側表面を、作動波長域内での導波路コアの屈折率より低い屈折率を有する材料層によって被覆する付加的な段階を含むという利点を有する。
【0062】
本発明の第6の態様においては、光通信用光回路を製造する方法は、(a)少なくとも1つの中空コア導波路と、光学部品を受け取る少なくとも1つのスロットが形成された半導体基板を取り出す段階と、(b)光学部品を、該少なくとも1つのスロットの中に導入する段階とを含み、光学部品を少なくとも1つのスロットの中に導入する段階は、さらに、前述の光学部品を位置合わせするように作用する。
【0063】
本発明の第7の態様においては、変形可能な材料層にパターンを形成するのに適したマスターが設けられ、該マスターは、前述の変形可能材料において、少なくとも1つの中空導波路チャネル及び少なくとも1つの位置合わせスロットを形成するように適切にパターン加工された半導体材料を含み、前述の少なくとも1つの位置合わせスロットは、光学部品を位置合わせ状態で受け取るように構成されている。
【0064】
或いは、マスターは、サブマスターの製造を可能にする半導体材料に形成することができる。サブマスターは、次いで、変形可能な材料に要求されるパターンを形成して、PLCを定めるように用いることができる。マスター又はサブマスターは、さらに、要求されるパターンを固定可能な層に形成する金型として用いることができる。
【0065】
本発明の第8の態様においては、光通信用光回路を形成する方法は、(a)本発明の第7の態様によるマスターを用いて、変形可能材料層に、恒久的にパターンを形成する段階と、(b)少なくとも1つの光学部品を、変形可能な材料に形成された少なくとも1つのスロットの中に導入する段階とを含む。
【0066】
光通信用光回路は、このようにして、1つ又はそれ以上の中空コア光導波路が形成された半導体基板を含むものとして説明される。
【発明を実施するための最良の形態】
【0067】
ここで、本発明を、例示に過ぎないものとして、添付の図面を参照して説明する。
図1を参照すると、典型的な従来技術のシリコン光学ベンチ装置が示されている。
図1Aは、微細加工により作られた中空チャネル4と、一対のはんだ接続器6とを有するシリコン光学ベンチ2を示す。シリコン光学ベンチ2は、レーザ8及びシリカ光学繊維ケーブル10を保持するように構成されている。
図1Bは、シリカ光学繊維10及びレーザダイオード8がシリコン光学ベンチ2上に取り付けられた状態を示す。中空チャネル4は、十分に高い精密さで形成されるため、レーザ8からの光学的出力は、シリカ光学繊維10の端部と精密に位置合わせされる。はんだ接続器6は、電気接続を与え、レーザダイオード8を基板に取り付ける。
【0068】
各々のシリカ導波路の端部からの好ましくない反射を最小にするために、反射防止コーティング(図示せず)が与えられる。或いは、シリカ導波路は、(例えば、屈折率整合ゲルを用いて)屈折率整合されて、直接、光学部品の各々と接続されることができる。反射防止コーティング及び/又は屈折率整合についての要求は、装置全体の費用に加えられることになり、製造をより複雑にし、時間がかかるものにする。
【0069】
単純さのために、単一の光学繊維ケーブル(すなわち、シリカ光学繊維ケーブル10)及び光学部品(すなわち、レーザ8)が図1に示されるが、当業者であれば、複雑な多部品光通信用回路を、同じ原理を用いて製造できることを認識するであろう。多数の光学部品をシリコン光学ベンチ上に配置することができ、種々の長さのシリカ光学繊維導波路を用いて、光学リンクを部品間に確立することができる。光学部品は、例えば、光学モジュレータ、ビーム分割器、ビーム再結合器、検知器などを含むことができる。
【0070】
図2を参照すると、シリコン光学ベンチの一部として用いられる従来技術の統合された光導波路が示されている。
低屈折率シリカ層20が、シリコン光学ベンチ基板22上に堆積される。ドープ処理されたシリカの高屈折率層がシリカ層20上に形成され、高屈折率導波路コア24は該ドープ処理されたシリカの高屈折率層の一部をエッチングによって除去することにより形成される。
【0071】
高屈折率導波路コア24は、光導波路として働き、被覆と比較すると高い屈折率のコアは、全内反射により導光を与える。このことは、シリコン光学ベンチと一体であり、単に該シリコン光学ベンチと接続された状態で保持されているだけではない光導波路を与える。中実コアコアの一体的な光導波路は、したがって、シリコン光学ベンチ上の溝に取り付けられた光学繊維に対する公知の代替技術である。しかしながら、統合された光導波路の使用は、導波路を光学部品と屈折率整合させる必要、すなわち反射防止コーティングを与える必要を減らすものではない。シリコン基板上に付加的な材料の層を堆積することは、さらに、光通信用回路を製造するにおいて複雑さを増加させる。
【0072】
図3を参照すると、本発明の装置の一部を形成する中空コア導波路光通信用光回路(PLC)40が示されており、図3AはPLCの斜視図を示し、図3Bは、図3Aで「A」と印が付けられた破線に沿ったPLCの断面を示す。
【0073】
中空コア導波路PLC40は、シリコン基部42及びシリコン蓋を備える。レーザ8がシリコン基部42に取り付けられて位置合わせされる。レーザ8により射出される光は、シリコン基部及びシリコン蓋44により形成される単一モードの中空コア導波路46に結合される。換言すると、中空コア導波路はシリコンに直接形成され、ここからPLC基部及び蓋が製造される。単純さのために、レーザ8に対する電気接続は示されていないが、当業者は、例えば、電気接続は、基部42においてダイオードの隔離を用いるトラック埋め込みのような様々な方法で作ることができることが分かるであろう。
【0074】
図3の中空コア導波路46は、長方形の断面を有することが分かる。ほぼ等しい深さ及び幅を有する長方形導波路の使用は(ここでは、長方形という用語は、正方形を含むものとする)、偏光に依存する損失を減少させ、多くの電気通信用途において利点をもたらすことができる。
【0075】
長方形の導波路が示されるが、導波路の断面は、要求に応じて成形することができる。例えば、円形の又は放物線状の断面の導波路であるか、又は、V字形態の導波路を、適切なエッチング技術を用いてシリコン基部に形成することができる。中空導波路構造は、さらに、シリコン蓋44に形成することができる。しかしながら、このことは、基部及び蓋の両方をパターン加工することを必要とし、さらに、該蓋及び該基部を精密に位置合わせしなければならないことを意味する。中空コア導波路の寸法は、要求に応じて、基本モード又は多モードの伝播を支持するように選択することができ、以下により詳細に述べられる。
【0076】
図3を参照して説明される例においては、当業者に周知の種類の微細加工技術を用いて、非常に高精度にまでエッチングすることができるために、PLCを形成するのにシリコンが用いられる。しかしながら、当業者であれば、さらに、本発明のPLCを形成するのに、どんな微細加工により作られた半導体材料を用いてもよいことが分かるであろう。
【0077】
レーザ8は、シリコン基部42に接合された別々の部品であり、換言すると、これは混成構成である。当業者は、さらに、レーザ8を蓋に接合すること、又はシリコン自体に光学部品を製造できることがわかるであろう。レーザ8だけが、図3を参照して説明されているが、多数の光学部品をシリコン基部又は蓋の上に配置又は形成し、及び/又は位置合わせすることができる。蓋に形成された位置合わせスロットは、さらに、光学部品を位置合わせ状態で受け取るように用いることができる。この技術は、したがって、複雑な多部品PLCの製造を可能にする。光学部品は、例えば、光変調器、ビーム分割器、ビーム再結合器、検知器、格子、ミラー、GRIN(屈折率勾配)レンズなどを含むことができる。本発明のPLCに形成することができる光学部品の種類の幾つかの例は、以下でより詳細に説明される。
【0078】
中空コア導波路46を通る光透過を最大にするために、金層48が中空コア導波路46の内面に与えられる。金層をシリコン基部及び蓋上に堆積することは、例えば、スパッタリング又はめっき加工のような適切な金属堆積技術を用いて容易に達成することができる。
【0079】
蓋は、当業者に周知の種々の方法により基部に接合することができる。中空光導波路の一部を形成しないシリコンの領域は、蓋部分及び/又は基体部分上に露出したまま残されることになり、蓋及び基部は、金-シリコン共晶接合により接合されることになる。さらに、銀充填エポキシ、はんだ、又はポリマー接着剤を用いて、蓋と基部とを接合することができる。蓋は、要求に応じて、基部の一部だけを覆うことができる。
【0080】
金層48の存在は、電気通信波長域内の波長(すなわち、約1.55μmの波長)をもつ光のATIRを中空コア装置内に与える。これらの電気通信波長においては、金は、n<1の必要屈折率特性を有する。
【0081】
金層48が説明されたが、当業者は、導波路が作動される波長において、空気(又はキャビティ内に含まれるあらゆるもの)より少ない屈折率を有するどんな材料も中空コア導波路を定める表面上に堆積できることが分かるであろう。異なる材料の屈折率は、1998年、ロンドン、アカデミックプレス、E.D.Palikによる「光学定数についての手引き」のような種々の刊行物に見出すことができる。金属は、典型的には、所定の波長帯、すなわち、金属の物理特性に依存する特定の波長帯において、空気より少ない屈折率を有する。
【0082】
金層48はATIRを与えるが、中空コア導波路46を付加的な低屈折率材料層により被覆することは必要不可欠なことではないことに注目されたい。シリコンの屈折率は、0.5μmと300μmとの間の波長においては約3.4であり、したがって、被覆されていないシリコンから形成された中空コア(すなわち、空気により充填された)導波路は、このような波長帯内では光のATIRを与えない。しかしながら、被覆されていないシリコンは、依然として、フレスネル反射によって導光を与える。導光のためにフレスネル反射を用いる中空コア導波路は、TIRを与える導波路より多い光損失をもたらすが、特定の状況においては、この増加したレベルの光損失は、許容できる。
【0083】
反射コーティングが与えられる場合には、基板は、半導体以外の微細加工により作られた材料から形成することができる。例えば、プラスチックの導波路装置は、高温エンボス加工又は射出成形を含む技術により製造することができる。この技術は、マスターを形成することを含む。マスターは、深奥乾式エッチングを用いて、シリコンのような半導体材料に形成することができる。或いは、マスターは、LIGA又はUVLIGA技術を用いて、層の電着により形成することができる。マスターが形成されると、中空コア導波路は、スタンピング(すなわち、プレシング)、又はホットスタンピングによりプラスチック基板に形成することができる。さらに、プラスチック基板に中空コア導波路を形成するために用いることができるサブマスターを形成するのに好適なマスターを製造することができる。中空プラスチック導波路は、このようにして形成され、反射コーティングにより被覆されることになる。反射コーティングを支持するプラスチックの中空コア導波路は、さらに、プラスチック又はポリマーにより形成できる。例えば、中空コア導波路は、石版印刷法を用いて、「スピン-オン」ポリマーコーティング(例えば、Microchem.社から入手可能なSU8)上に形成することができる。
【0084】
単純なPLCが図3を参照して説明されたが、当業者は、本発明が、複雑なPLCにも同じように適用可能であることが分かるであろう。例えば、多数の光学部品をPLC上に取り付け、PLC基板から形成された中空コア導波路を介してリンクさせることができる。このようなPLCは、光信号処理の基本、及び/又は、光信号ルーティング及び分析システムを形成することができる。このようなPLCの幾つかの例が以下に与えられる。
【0085】
図4Aを参照すると、シリコン基板62に形成された中空コア導波路構造60a、60b、及び60cが平面図で示されている。傾斜表面(例えば、表面64)は、光が90°で導かれるように形成される。
【0086】
反射における位相摂動を最小にするために、傾斜表面64は、理想的には、λ/10、又はそれ以上、より好ましくは、λ/20より多く平らな表面仕上げを必要とする。したがって、1.5μmの波長を用いる場合には、150nmより多く平らな表面仕上げが必要とされる。このレベルの精度は、微細加工技術を用いて容易に到達することができ、典型的には、30-50nmの解像度をもたらすことができる。
【0087】
傾斜表面64は、このようにして、中空コア導波路の部分が互いに対してあらゆる角度に配向されることを可能にするミラーを与える。このような鋭角を通して光繊維ケーブルを湾曲するのは可能なことではない。公知のSiOB技術を用いて、同様の回路が製造される場合には、光繊維部分間で光を結合する別々の(良好に位置合わせされた)ミラーをもつ2つの光繊維部分を形成することが必要になる。本発明は、したがって、従来技術のSiOB装置より複雑でコンパクトな回路設計を与えることができる。
【0088】
モニリシックミラーが図4Aに示されるが、混成構成もまた、同じ光学機能を与えることができることを認識されたい。例えば、研磨ミラーを位置合わせ状態で受け取る位置合わせスロットを製造することができる。混成構成は、例えば、偏光非依存性多層コーティングを支持できるといった、最小の偏光依存性を有するように設計することができる高品質の光学ミラーの使用を可能にするため、有益である。
【0089】
図4Aを参照して説明された導波路構造は、すべてほぼ直線であり、適切に置かれたミラーにより接続されている。しかしながら、中空コア導波路構造は、湾曲していてもよい。例えば、図4Bを参照すると、シリコン基板62に形成された湾曲導波路66が示されている。当業者であれば、達成可能な最大曲率は、ガイド厚に依存することを認識するであろう。
【0090】
図5Aを参照すると、図4Aを参照して説明された種類の中空コア導波路構造60を通る断面が示されている。中空コア導波路構造60は、シリコン基板62に形成されており、要求される中空コア導波路を与える上述された方法により、該基板62に取り付けることができるシリコン蓋部分68も形成されている。
【0091】
図5Bに示されるように、中空コア69を形成する壁の各々の内面は、付加的に、例えば銅、金、又は銀のような材料の層70で被覆して、TIRにより、1.55μmの放射線の反射率を高めることができる。
【0092】
公知の偏光の直線偏光を導く場合には、異なる内面が異なる光学的特性を有する中空コア導波路を形成して、該導波路に関連する光損失をさらに減少させることができる。
【0093】
図6は、86°の角度で、空気から表面上に入射する光に対するフレスネル反射係数を、s偏光(Rs)及びp偏光(Rp)に対する表面の屈折率(n)及び吸収(k)の関数として示す。図6から、フレスネル反射係数は、光の偏光に強く依存することが分かる。したがって、長方形の中空コア導波路により導かれる光の偏光状態が分かっている場合には、該導波路を形成する一対の向かい合う表面は、p偏光に対する反射率を最適化するために低い屈折率を有するように構成される一方で、第2の対の向かい合う表面は、s偏光に対する反射率を最大にするためにより高い屈折率を有するように構成されることになる。
【0094】
異なる内面が異なる光学的特性を有する導波路を形成するのに用いることができる多数の技術が図7を参照して説明される。
【0095】
図7Aは、絶縁体上シリコン(SOI)製造技術を用いて製造された、SOIウエハ80に形成された中空導波路を示す。ウエハ80は、基板84上に支持されるSiO2材料の絶縁層86を含み、その上にシリコン層86を有する。シリコンの層86は、絶縁層82まで下方にエッチングされて、要求されるチャネル88を形成する。SiO2から形成される蓋部分90もまた設けられる。
【0096】
このようにして、中空コア導波路は、シリコンで構成される第1の表面92及び第2の表面94を有するように形成され、第3の表面96及び第4の表面98は二酸化ケイ素で構成される。シリコンの屈折率は約3.5であり、二酸化ケイ素は、約1.5の屈折率を有する。したがって、導波路における光損失は、該導波路に沿ってy軸方向に伝播する光がz軸方向に偏光されたときに減少され、すなわち、第1の表面92及び第2の表面94においてRs反射があり、第3の表面96及び第4の表面98からRp反射があるときに、減少される。
【0097】
図7Bを参照すると、シリコン基板102に形成され、蓋部分103を有する中空導波路100が示されている。上方壁104(すなわち、蓋部分103により定められる壁)及び下方壁106は第1の材料により被覆され、側壁108及び側壁110は、第2の材料により被覆される。導波路に沿って、y軸方向に伝播するz軸方向に偏光される光の光損失を最小にするために、第1の材料及び第2の材料は、それぞれ、低い屈折率及び高い屈折率を有するように選択される。
【0098】
図7Bは、導波路のすべての4つの壁にコーティングが適用された状態を示すが、単一の壁だけであるか、又は、一対の向かいあう壁だけを要求に応じて被覆することができる。換言すると、1つ又はそれ以上の壁は被覆されないまま残ることができ、したがって、基板を形成するのに用いられた半導体材料の光学的特性を有することができる。
【0099】
さらに、物理的構造をシリコンに形成して、要求に応じて、所定の光の偏光に対する反射率を高めることができる。図7Cは、エタロン側壁構造122を備えたシリコン基板に、どのように中空コア導波路を形成することができるかを示す。この場合においては、エタロン側壁構造は、反射率を高める。中空部分が空気で充填されているエタロン構造が示されるが、別の材料(例えば、液体又はガス)を空気の代わりに用いて、反射率を高めることができる。
【0100】
中空コア導波路に関連する光損失はまた、導波路コアの形状を制御することによりさらに減少させることができる。例えば、導波路コアが広ければ広いほど、関連する光損失は低くなる。図7Dは、シリコン層130に形成され、シリコン蓋部分134を有する長方形断面の導波路132を示す。導波路132の中空コアは、深さ「b」より少ない幅「a」を有する。z軸方向に偏光され、導波路132に沿って伝播される光は、このようにして、深さ「a」の導波路を通って伝播される場合より低い損失を経験する。
【0101】
例えば、光が一平面内にのみ導かれる構造をPLCに形成することができ、例えば、垂直軸に沿って自由伝播空間があるが、導波は水平軸上にのみ生じるように配置することができることに、さらに注目されたい。この場合においては、導波路は平坦な導波路と呼ばれ、すなわち、これらは、1つの平面にのみ導かれる。平坦な導波路を用いることができ、ここでは、一次元のビームの拡がりを要求するが、導波によって、二次元のビーム幅は制限される。導波が水平面においてのみ要求される場合には、蓋部分は、必要ではない。光が蓋と導波路の床との間に導かれるが、横方向平面には導かれない場合には、逆の状況も可能である。
【0102】
図8を参照すると、ビーム分割器及び偏光フィルタを、どのように中空導波路PLCに形成できるかが示されている。
図8Aは、シリコン基板160に形成された中空導波路構造から製造されたビーム分割器を示す。ビーム分割器は、入力中空コア導波路162、第1出力中空コア導波路164、及び第2出力中空コア導波路166を備える。入力中空コア導波路162を通って伝播する光は、薄いシリコン壁168から第1出力中空コア導波路164の中に部分的に反射され、さらに、部分的に、第2出力中空コア導波路166の中に透過されて結合される。
【0103】
入力中空コア導波路162と第1出力中空コア導波路164との間の角度(θ)は、薄いシリコン壁168上への光の入射角を定める。図8Cに示されるように、シリコンの反射特性は、入射する光の入射角と偏光状態との両方に依存する。入力中空コア導波路162から第1及び第2の出力中空コア導波路164及び166の中に結合されたパワーの相対的な比率は、このようにして、特定の角度(θ)で装置を製造することにより選択することができる。
【0104】
さらに、図8Bに示されるように、偏光分割器は、角度θをブルースター角と等しくなるように構成することにより製造することができる。この場合においては、θ=74°の角度は、z軸方向に偏光された光をもたらし、第1出力中空コア導波路164を介して装置から導かれ、x軸方向に偏光された光は、第2の出力中空コア導波路166を介して装置から導かれる。
【0105】
薄いシリコン壁168の代わりに、エタロンフィルタを、図8A及び図8Bを参照して説明された種類の装置に形成することができる。このことは、異なる波長の光に対して異なる反射特性を有する光学装置を与え、したがって、この装置はさらにスペクトルフィルタとしても作動することができる。
【0106】
モノリシックビーム分割器及びブルースタープレートが説明されたが、当業者は、さらに、混成構成を用いて、同様の光学機能を実施できることを理解するであろう。位置合わせスロットを基板に形成して、必要な光学部品を受け取ることができる。
【0107】
図9参照すると、光合焦機能を与えるのに、さらに、どのようにシリコン材料の基板を形成できるかを見ることができる。
【0108】
図9Aは、シリコンレンズ構造192と中空コア導波路194とが形成されたシリコン基板190を示す。レンズ構造192は、中空コア導波路194に沿って導かれた光196が点198に合焦されることを可能にするレンズとして働く。このようなレンズは、例えば、光を検出器要素に対して合焦するのに用いることができる。
【0109】
図9Bに示されるように、さらに、成形されたシリコンリフレクタ200をシリコン基板202に形成して、光学的に中空コア導波路をリンクさせることができる。リフレクタ200は、第1中空コア導波路204から第2中空コア導波路206に、或る角度(この場合は90°)で光を導き、さらに、光208を合焦する機能を果たす。ここでも、このような要素は、PLCにおいて様々な異なる方法で用いることができ、これは反射防止コーティングを必要としないために、実現するのは比較的単純である。
【0110】
ここに述べられた種類のPLCは完全な光学回路を含むことができるが、さらに、典型的には光学繊維を介して、PLCの中に又は該PLCの外に光を結合することが必要になることがある。
【0111】
図10は、シリコン基板220に形成され、光を第1入力光学繊維222から受け取るように配置されたPLCを示す。入力光学繊維222は、中空コアを有し、そこからの光は、入力球面レンズ226を用いて、入力中空コア導波路224に結合される。入力中空コア導波路224に沿って伝播する光は、エタロン構造228に向けられる。エタロン構造228は、そのスペクトル特性によって、光を、第1出力中空コア導波路230又は第2出力中空コア導波路232の中に、スペクトル的にフィルタ処理する。第1出力中空コア導波路230を通って伝播する光は、球面レンズ236を介して、第1出力光学繊維234に結合され、第2出力中空コア導波路232を通って伝播する光は、球面レンズ240を介して、第2出力光学繊維238に結合される。ここでも、示されるモノリシック要素の代わりに、混成エタロンフィルタを用いることができる。
【0112】
図10には球面レンズが示されるが、代替的に、GRINロッドレンズのような他のレンズを用いることができる。エタロン構造228は、さらに、必要に応じて、ビーム分割器又はブルースタープレートと置き換えることができる。光学繊維は、要求に応じて、単一モードであっても多モードであってもよい。
【0113】
SOI技術は、光学繊維が結合されるPLCを形成するのに特に適している。これは、典型的なSOIウエハが、製造工程中に、非常に正確に定められた厚さを有するシリコン層を含むためである。SOIウエハのシリコン層における中空導波路構造の製造においては、シリカ絶縁層は、エッチング工程に関する限りは、垂直「停止部」として働く。したがって、SOI技術は、μ以下のチャネル深さ精度を与えることができる。
【0114】
SOIのエッチング精度を、エッチング深さの約数パーセントの精度を有する純粋なシリコンにおけるチャネルエッチングと対比されたい。純粋なシリコンウエハにおいてチャネルをエッチングしてファイバを取り出すことは(125.0μmの被覆直径まで剥離される)、エッチングチャネルの深さに3.0から4.0μmの不正確さを生成することになる。ファイバコアは、典型的には、10.0μmの直径に過ぎないため、或る繊維を/から他の部品から/に(例えば、半導体レーザから)結合する場合に、この垂直方向の位置ずれの大きさは、有害なものとなる。したがって、SOIベースの製造ルートは、位置合わせ及び導波路の断面精度に対する利点を有し、偏光に依存する損失を減少させる。
【0115】
上述のように、本発明のPLCは、単一モードであるか又は多モードの伝播を可能にする中空コア導波路を含むことができる。特定の状況においては、さらに、例えば、効率的に光を異なる光学部品の中に又はこれの外に結合するために、中空コア導波路の寸法を変更することが必要になることがある。
【0116】
図11Aを参照すると、シリコン基板262に形成された中空コア導波路構造260が示されている。幅広い(125μm)直径の出力導波路264が、先細になった導波路部分268によって、より狭い(62.5μm)直径の入力導波路266に光学的にリンクされている。先細になった部分の長さは、1.875mmである。
【0117】
図11Bは、入力導波路266における基本モードの入力ビームによりもたらされる出力導波路264を伝播する光の電界強度を示す。図11Cに示されるように、出力導波路264における出力光は、主として基本モードで伝播する。換言すると、先細になった導波路は、ビームの大きさを拡げることを可能にし、出力ビームのパワーの大部分が基本モードに結合されるのを確実にする。
【0118】
本発明のPLCは、さらに、基板に形成された中空コア多モード干渉(MMI)装置を含むことができる。ビーム分割及びビーム再結合MMI装置の例は、US5410625に与えられている。US5410625の基本的なMMI装置に対する変更及び改良もまた周知である。例えば、US5379354は、入力ガイド位置の変動をどのように用いて、入力放射線を、異なる強度を有する出力ビームに分割する多方向ビーム分割器を得ることができるかを説明する。レーザキャビティを形成するのにMMI装置を用いることが、さらに、US5675603に示されている。種々の組み合わせのMMI分割器及び再結合装置が、さらに、光学的ルーティング能力を与えるように用いられており、例えば、US5428698を参照されたい。すべての上述の場合において、MMI装置は、シリコンその他の適切な半導体材料のいずれかにおける中空コア導波路として製造して、PLCの一体部分を形成することができる。
【0119】
MMI装置は、基板に形成された多モード領域から製造することができ、これに入力及び出力の単一モード光学繊維ケーブルが結合されている。この方法により、分割ビームが入力ビームのイメージであるビーム分割/結合を得ることができる。
【0120】
具体的には、長方形であるか又は正方形断面の中空多モード導波路を設計して、導波路の長さが、その幅と適切な関係を有するように設計することにより、対称的な、反対称的な又は非対照的な光学的フィールドのイメージ再形成を与えることができる。例えば、正方形断面の導波路における対称的なフィールドにおいては、イメージ再形成長さは、Lがガイド長さであり、wがその幅であり、λが放射線の波長としたとき、伝播放射線の波長における波長幅の自乗、すなわちL=w2/λにより与えられる。対称的なフィールドのイメージ再形成は、この長さにおいて、及びこの長さの倍数で生じ、すなわち、nを整数としたとき、n.w2/λで生じる。
【0121】
50.0μm幅の中空導波路及び1.55μmの放射線の場合においては、イメージ再形成長さは、502/1.55=1613μm=1.613mmにより与えられる。対称的なフィールドは、この長さでイメージ再形成され、さらに、この長さでの整数の倍数でイメージ再形成され、これはすなわち、3.23mm、4.84mmなどである。例えば、単一モードの光学繊維からのTEM00ガウスの入力ビームは、1.613mmの距離でイメージ再形成することができる。イメージ再形成点においては、あらゆる要求される光学部品を位置させることができる。この方法により、イメージ再形成現象は、光を一連の部品間で導く付加的な方法を与える。
【0122】
或いは、非対称的な光学的フィールドの場合においては、イメージ再形成は、対称的な光学的フィールドに要求される長さの8倍で生じ、すなわち、50.0幅の中空導波路に対して12.09mm(8×1.613mm)で生じる。非対称的なフィールドの鏡像はさらに、この長さの半分、すなわち6.05mmで形成される。
【0123】
導波路の水平方向幅及び垂直方向幅が実質的に異なる長方形導波路の場合においては、2つの幅に関連するイメージ再形成長さ自体が異なる。しかしながら、長方形の中空コア導波路の幅の間の関係を、イメージ再形成が各幅に対して同一の長さで生成されるように構成することにより、あらゆるフィールドをイメージ再形成できるようになる。
【0124】
例えば、対称的なフィールドは、幅w1及びw2の軸に関連するイメージ再形成長さを同一に構成することにより、L1=n1.w12/λ、及びL2=n2.w22/λ、中空の長方形導波路にイメージ再形成することができる。このことは、前述のようにn1及びn2が整数であるものとして、w2=(n1/n2)1/2.w1により達成することができる。
【0125】
本発明のPLCに含むのに適した別の種類のMMI装置は、係属中のPCT特許出願番号GB2002/004560に説明され、図12に示される波長デ・マルチプレクサである。
【0126】
デ・マルチプレクサ300は、シリコンSOI基板302に形成され、入力基本モード導波路304、中央多モード領域306、及び4つの出力導波路308aないし308d(総合的に308といわれる)を含む。導波路の寸法及び位置は、(GB2002/004560に説明されるように)入力基本モード導波路304から多モード領域306に入る4つの波長成分が分離されて、出力導波路308を介して別々に出力するように選択される。
【0127】
さらに、光が隣接する導波路に近接結合されるPLC装置を形成することが可能である。図13を参照すると、第1の中空コア導波路340、第2の中空コア導波路342、及び第3の中空コア導波路344がシリコン基板346に形成され、蓋部分347を有するように示されている。シリコン壁348及び350の厚さ「c」は、光が隣接する導波路に透過するのに十分なだけ薄いものである。この種類の近接結合器部品は、ビーム分割部品を光路の中に挿入することなく、例えば、少ない百分率の伝播ビームを取り出すようにビーム分割器として用いることができる。
【0128】
図14を参照すると、PLCは、中空コアのシリコン導波路及び中実コアの(例えば「リッジ」)シリコン導波路の両方を含むことができる。このことは、中実コア及び中空コア技術の両方において光学的機能を実現することを可能にする。
【0129】
図14Aは、SOI基板上に形成された中空コア導波路400及び中実コア導波路402の間のブルースターインターフェースを示す。図14Bは、中空コア導波路400のB-Bに沿った断面を示し、図14Cは、中実コア導波路402のA-Aに沿った断面を示す。中空コア導波路400は、中実コア導波路402の傾斜したインターフェース404で終端する。導波路400及び402は、インターフェースがブルースター角にあるように配置される。このことは、中空コア導波路と中実コア導波路との間に効率的な結合を与える。
【0130】
図15を参照すると、中空コア導波路450及び傾斜表面452が示されている。表面452は、基板の平面に対しておおよそ45°だけ傾斜されており、光454は該基板の平面外で結合される。図15に示される構成は、基板の平面に対して異なる垂直平面に配置された他の回路又は装置の中に又は該回路又は装置から光を結合するのに用いることができる。この方法により、3次元のスタックされたPLC(例えば、3次元の光スイッチ)を生成することができる。
【0131】
このようなミラーをモノリシックに製造する1つの方法は、約8ないし9°だけオフセットされた[100]シリコン材料における精密な軸外しカットである。このような傾斜表面を製造する幾多の代替的な方法が当業者に明らかである。混成ミラー構成を、さらに用いることができる。
【0132】
図16を参照すると、本発明のPLCに組み込むのに適した中空コア光導波路のガイド長さに依存する光学的透過特性を示す実験データが示されている。
【0133】
曲線500は、50μmの幅及び深さの正方形の内部コアを有するシリコン基板に形成された中空コア導波路の予測光学的透過を示し、点502aないし点502cは、該中空コア導波路の測定光学的透過を示す。曲線504は、銅被覆がその内面の各々に適用された同様の導波路の予測光学的特性を示し、点506aないし点506cは、該導波路の測定光学的特性を示す。
【0134】
曲線508は、125μmの幅及び深さの正方形の内部コアを有するシリコン基板に形成された中空コア導波路の予測光学的透過を示し、点510aないし点510cは、該中空コア導波路の測定光学的透過を示す。曲線512は、銅被覆がその内面の各々に適用された同様の導波路の予測光学的特性を示す。図16に示されるすべての場合において、1.55μmの波長を有する放射線が用いられた。
【0135】
したがって、導波路の寸法を増加させることは、光学的損失を減少させ、反射コーティング(この場合においては銅)の包含は、損失をさらに減少させることがわかる。しかしながら、角度方向位置合わせ許容差は減少する。
【0136】
図17を参照すると、部品の角度方向位置合わせ効果が示されている。
【0137】
図17Aは、第1の中空コア導波路602、第2の中空コア導波路604、及び第3の中空コア導波路606が形成されたシリコン基板600を示す。ビーム分割要素608が位置合わせスロット610に配置される。この要素608は、該要素及びスロット製造許容差により求められる角度方向位置ずれ
イメージ ID=000004

を有することが分かる。
【0138】
図17Bは、種々のモードの中空コア導波路に対するのパワー結合効率を角度方向位置ずれ
イメージ ID=000005

の関数として示す。曲線620は、基本モードに結合されたパワーを示すが、曲線622は、より高いオーダーのモードに結合された光学的パワーを示す。
【0139】
図18を参照すると、横方向位置合わせの効果が示される。
【0140】
図18Aは、第2の中空コア導波路652から
イメージ ID=000006

だけ横方向に変位された第1の中空コア導波路650を示す。横方向変位の関数としてのパワー結合係数が図18Bに示され、ここでは、曲線654は基本モードに結合されたパワーを示し、曲線656は、より高いオーダーのモードに結合された光学的パワーを示す。
【0141】
上述により、中空導波路により相互接続された部品の統合されたシステムを通る基本モードの伝播は、導波路寸法及び位置合わせ許容差が適切に選択されているのであれば得ることができる。このことは、基本導波路モードにおけるパワー量は、どれだけの光が単一モードの繊維に/から結合されるかを決定するため、光を単一モードの光学繊維に/から結合する部品のシステムにおいては特に重要である。導波路において、効率の高い基本モードの伝播を確実にすることは、単一モード繊維の基本モードに対する良好な結合、及び全体的に低い挿入損失を確実にする。
【0142】
換言すると、導波路幅と、有効な基本モードが達成されることを確実にするために必要とされる(導波路及び部品の両方の)角度及び横方向の位置合わせ許容差との間には、折り合いがある。減衰係数は導波路幅と逆に関係しているため、より低い減衰係数は、ガイドの断面(幅)を十分に大きくすることで得ることができる。導波路幅をより大きくすることは、さらに、横方向の位置合わせ許容差を緩和するが、角度方向位置合わせ許容差はきつくなることが分かる。
【0143】
図19を参照すると、スロットに置かれた部品の正確な位置合わせを確実にする技術が示されている。
【0144】
シリコン基板700は、光学部品702を保持するために形成されたスロットを有する。多数のばねクリップ704(マイクログリッパとも呼ばれる)が、周知の微細加工技術によりシリコンに形成される。これらのクリップ704は、変位されると、横方向の力を与えるようにされている。この方法により、部品は、スロット内にしっかりと位置合わせされた状態で保持される。
【0145】
図19は、クリップが光学部品を取り囲むように示されているが、該部品をスロットの側壁のような基準表面に対して押し付けることも可能である。さらに、シリコンウエハにおける酸化物の犠牲層を除去することにより製造される、ばねその他のMEMS特徴は、或る量のアンダーカットをもたらすことになることを当業者は理解するであろう。このアンダーカットは、位置合わせスロットと関連している場合には何の効果もなく、さらに、モード分布が典型的には円形であるか又は楕円形である長方形の中空コア導波路における伝播とほとんど変わらない。
【0146】
本発明のPLCは、幾多の異なる光学回路を実施するのに用いることができる。これらの幾つかの例は、干渉計、(例えば、マイケルソン、又はマッハツェンダ)、分光計、ライダー、及びMIM装置の光学的読み出し(センサ又はアクチュエータのような)を含む。電気通信回路(ルータ、マルチプレクサ、デ・マルチプレクサなど)もまた実施することができる。光学部品が上述されたが、代替的に又は付加的に、マイクロ波成分及びマイクロ波放射線を導く中空コア導波路を含むことができない理由は述べられていない。したがって、光学-マイクロ波を単一の回路で統合することは、可能である。

Optical circuit fabrication method and device

The present invention relates generally to integrated optics devices, and more particularly to improved photonic light circuit (PLC) devices.

Photonic circuit modules form an integral part of many optical communication, sensor and instrumentation devices. In such photonic circuit devices a number of optical components are rigidly held in place and waveguides, typically lengths of optical fibre, are used to optically connect the components as required. The optical components and interconnecting fibres are held in place on a suitable substrate.

Silicon optical benches (SiOBs) are one example of an assembly technology for photonic circuits. As the name suggests, SiOBs are optical benches formed from silicon or a similar semiconductor material. Grooves and slots are etched in the silicon material, using micro-fabrication processes, to hold the various optical components. The high accuracy of the micro-fabrication process allows the optical components and optical fibres to be precisely aligned relative to one another in the various slots and grooves. This provides so called"passive alignment"of the components and reduces the need to actively ensure the various components of the optical circuit are aligned with one another. Light may also be directed between the various optical components using free space optics such as lenses etc.

It is also known, for example see US4902086 and EP0856755, that it is possible to deposit various layers of material to form waveguides that are integral with the SiOB. Typically a base layer, such as silica, is formed on the silicon substrate. A layer of doped silica with a high refractive index, i. e. the core layer, is then deposited on top of the low refractive index base layer. The core layer is patterned to form appropriate waveguides. Optionally, an upper cladding layer of low refractive index material is also deposited on the patterned core layer. In other words, waveguides are formed directly on the silicon substrate rather than being fabricated as separate optical fibres.

A disadvantage of known photonic circuit devices, including those based on SiOBs, is the high degree of accuracy with which each optical component has to be aligned with the associated waveguides to ensure an efficient optical connection. In addition to ensuring accurate physical alignment of the optical fibres and optical components, it is also necessary to minimise unwanted reflections from the end of each silica waveguide. This requires refractive index matching of the waveguides to the optical components, or the use of a gel or antireflection coating. Lenses may also be required to facilitate the free space coupling of light between components.

These requirements increase the complexity, and hence cost, of photonic circuit fabrication.

It is an object of the present invention to mitigate at least some of the disadvantages described above.

According to a first aspect of the invention, a photonic light circuit device comprises a semiconductor substrate and two or more optical components wherein one or more hollow core optical waveguides are formed in the semiconductor substrate to optically link said two or more optical components.

The present invention is advantageous over prior art photonic circuit devices as it removes the requirement to provide optical fibres for interconnects between components or to deposit layers of material to form solid core waveguides. This invention provides a photonic light circuit (PLC) that is easier to fabricate, and hence lower cost, than prior art devices.

A further advantage of linking the components with hollow optical waveguides is the increased optical power the circuit can handle over prior art photonic circuits that use solid core (typically Silica or silicon) waveguides to interconnect the optical components. Furthermore, index matching gels or epoxies, or antireflection coatings are not required on the faces of the waveguides.

The hollow waveguides are formed so as to guide light between optical components of the PLC. The optical components are any devices that will create, detect or act on an optical signal; for example beam splitters/recombiners, etalon structures, lenses, waveplates, modulators, lasers, photo-detectors, or actuated optical components.

The term optical component should also be taken to include optical structures, such as surface grating profiles etc, that are formed in or from the hollow waveguides.

The hollow core waveguides may be planar or two dimensionally guiding as described below. An optical component may also be an optical fibre cable; for example an optical fibre cable that is used to couple light in to, or out from, the PLC.

Semiconductor substrates can be etched to a high degree of accuracy using micro- fabrication techniques. The substrate may advantageously comprise a multiple layer wafer; for example SiGe or silicon-on-insulator (SOI) or silicon-on-glass. A person skilled in the art would recognise that micro-fabrication techniques typically involve a lithography step to define a pattern, followed by an etch step to transform the pattern in to one or more layers on, or in, the substrate material. The lithography step may comprise photolithography, x-ray or e-beam lithography. The etch step may be performed using ion beam milling, a chemical etch, a dry plasma etch or a deep dry etch (also termed deep silicon etch). Micro-fabrication techniques of this type are also compatible with various layer deposition techniques such as sputtering, CVD and electro-plating.

Advantageously, the semiconductor substrate comprises one or more alignment slots, each alignment slot being adapted to receive in alignment an optical component. The alignment slots are formed to the shape required to accept the optical components and may thus be deeper/shallower and/or wider/narrower than the hollow core optical waveguides.

The alignment slots can thus be fabricated with sufficient accuracy to align the optical component they receive. Placing an optical component in such an alignment slot inherently aligns the optical component and a component alignment or adjustment step is not required. Conventional pick and place techniques of the type used in the manufacture of electronic circuits and the like could be used to place the optical components in the alignment slots.

Alternatively, pick and place techniques may provide the necessary alignment. For example, a component could be accurately aligned when placed and then fixed (e. g. glued) to remain in alignment.

The alignment slots and (especially) the optical components are manufactured with a certain size tolerance. The coupling efficiency between a optical component and an associated hollow core optical waveguide will reduce as the angular error of alignment of the optical component with respect to the hollow core waveguide increases. However, reduction of the cross-sectional dimensions of the hollow core waveguide will increase the acceptable angular alignment tolerance, albeit at the expense of slightly increased losses in the optical waveguide due to the reduced core dimensions and increased (tighter) lateral alignment tolerances. Therefore, knowledge of the alignment tolerances that will be achieved with a certain optical component (e. g. from knowledge of the manufacturing tolerances of the optical component) will permit the dimensions of the hollow core waveguide to be selected to ensure a high coupling efficiency.

The alignment slots may also be formed so as to clamp a solid core optical fibre in place thereby allowing optical inputs/outputs to be made to the PLC. Stepped optical fibre alignment slots may also be provided to hold both the buffer layer and the cladding. The alignment of the core of a hollow core optical fibre with a hollow core waveguide on the PLC, achieved for example by clamping the optical fibre cladding in a alignment slot, would be especially advantageous as the air core to air core connection would be free from any unwanted reflections.

To provide efficient coupling between the core of an optical fibre and a hollow core waveguide of the PLC, the cross-section of the hollow core waveguide should be appropriate for the cross-section of the optical fibre core. In the case of solid core fibres, leakage into the cladding means that the width of the mode carried by the fibre is actually greater than the core diameter; for example typically the 10pm solid core of a single mode glass fibre has a total field width of around 14Am diameter. If the mode width is different to that of the hollow core waveguide, lenses (e. g. ball or GRIN rod etc) can be used to expand or reduce the optical field to enable light to be coupled to/from fibres with a different size core to that of the hollow core waveguide of the PLC. Fibre ends of solid core fibres may be anti-reflection.

Conveniently, one or more of the two or more optical components are formed from the material of the semiconductor substrate; i. e. monolithic components may be formed.

Alternatively, some or all of the optical components that make up the PLC, and which are interconnected via the hollow core waveguides formed in the semiconductor substrate, may be attached to the semiconductor substrate as described above; in other words, a hybrid device may be formed.

At least one of said two or more optical components may advantageously comprise a micro-electro-mechanical (MEMS) device. The MEMS component may be hybrid or monolithic. MEMS is taken to include micro-machined elements, micro-systems technology, micro-robotics and micro-engineering. Examples of MEMS optical components include alignment elements, pop-down Fresnel lenses, gyroscopes, moveable mirrors, tuneable Fabry-Perot cavities, adaptive optics elements, switches, variable optical attenuators, filters etc.

Conveniently, the semiconductor substrate forms a base portion of the photonic light-cir. cuit-device-and-a-lid-por-tion-is-additionally-provided-in-order-to-for-m-said-- hollow core optical waveguides.

Advantageously, one or more optical components are attached to the lid portion.

Optical components may be mounted on the lid alone, on the base portion alone, or on both the lid and the base.

The lid portion may be formed from semiconductor material, such as silicon, and advantageously one or more optical components may be formed thereon.

Alternatively, the lid portion may be formed from glass. Preferably, the lid should have the same thermal expansion properties as the substrate; for example, by the lid being formed from the same semiconductor material as the substrate.

In the case of lid mounted components, the base portion is etched to form the hollow waveguide structures and to provide recesses for optical components that are formed from, or attached to, the lid portion. Mounting the lid portion on the base portion allows the optical components to be brought into alignment with the optical waveguides of the base portion. A person skilled in the art would recognise that various techniques, such as precision alignment mating parts or wafer or chip alignment tools, may be provided to ensure accurate alignment of the lid and base.

Alternatively, some or all of the optical components may be directly mounted in alignment slots formed in the base portion. This enables the lid portion to be mounted on the base portion without a requirement to precisely align the lid and base portions.

Conveniently, the lid portion carries a reflective coating. The reflective coating may cover all, or just selected parts, of the lid portion as required. Advantageously, the reflective coating may comprises a layer of material having a refractive index lower than that of the waveguide core within the operating wavelength band; for example, -gold, sil-ve} or-copper. A-lternat-i-vely, one-or-more-l-a-yers-of-dielee-tric-material-ora- layer of Silicon Carbide may be provided.

A person skilled in the art would recognise how the lid portion and base portion cold be bonded together. For example, an intermediate layer such as conductive or non-conductive epoxy could be used. Alternatively, and in the case of a metal layer being used as a low refractive index layer, a metal-semiconductor eutectic bond could be formed. Glass frit techniques could be employed to bond the lid to the semiconductor base portion or, if the lid portion is formed from glass, anodic techniques could be used.

Advantageously, the semiconductor substrate comprises silicon. This may be provided in a variety of forms, for example in wafer form (e. g. Si, silicon-on- insulator or silicon-on-glass) or as a epitaxial layer (e. g. SiGe or GaAs) on a Si substrate. Advantageously, SOI is used.

Conveniently, the optical properties of a first internal surface forming one or more of the hollow core optical waveguides are different to the optical properties of a second internal surface forming that hollow core optical waveguide. This enables hollow waveguides to be formed that more efficiently guide light of a certain polarisation as described in more detail with reference to figure 6 below.

Advantageously, at least some of the internal surfaces of said one or more hollow core optical waveguides. carry a reflective coating.

The reflective coating may advantageously comprise a layer of material having a refractive index lower than that of the waveguide core within the operating wavelength band.

The-layer-of-material-having-a-reefr-ac-tive-inde-x-lower-than-the-hollow-w-aveguide- core provides total internal reflection- (TIR) of light within the PLC waveguides, thereby decreasing the amount of optical loss.

It should be noted that when hollow core optical waveguide structures are produced, the hollow core is likely to fill with air. Herein the refractive index of the core is thus assumed to be that of air at atmospheric pressure and temperature (i. e. n = 1).

However, this should be seen in no way as limiting the scope of this invention. The hollow core may contain any fluid (for example a liquid or an inert gas such as nitrogen) or be a vacuum. The term hollow core simply means a core which is absent any solid material. Also, the term total internal reflection (TIR) shall be taken herein to include attenuated total internal reflection (ATIR).

Conveniently, the reflective material carried on the internal surface of the hollow core optical waveguides is a metal such as gold, silver or copper. Metals will exhibit a suitably low refractive index over a wavelength range that is governed by the physical properties of the metal; standard text books such as"the handbook of optical constants"by E. D. Palik, Academic Press, London, 1998, provide accurate data on the wavelength dependent refractive indices of various materials. In particular, gold has a refractive index less than that of air at wavelengths within the range of around 500nm to 2. 2gm ; this encompasses wavelengths within the important telecommunications band of 1400nm to 1600nm. Copper exhibits a refractive index less than unity over the wavelength range of 560nm to 2200nm, whilst silver has similar refractive index properties over a wavelength range of 320nm to 2480nm.

A layer of metal may be deposited using a variety of techniques known to those skilled in the art. These techniques include sputtering, evaporation, chemical vapour deposition (CVD) and (electro or electro-less) plating. CVD and plating techniques allow the metal layers to be deposited without significant direction dependent <BR> <BR> - thickness-variations-Sputtering-using-a-rotating-sample-and/or-source-would-also- provide even coverage. Plating techniques are especially advantageous as they permit batch (i. e. multi-substrate parallel) processing to be undertaken.

A skilled person would recognise that adhesion layers and/or barrier diffusion layers could be deposited on the hollow waveguide prior to depositing the layer of metal.

For example, a layer of chrome or titanium could be provided as an adhesion layer prior to the deposition of gold. A diffusion barrier layer, such as platinum, may also be deposited on the adhesion layer prior to gold deposition. Alternatively, a combined adhesion and diffusion layer (such as titanium nitride, titanium tungsten alloy or an insulating layer) could be used.

Conveniently, the reflective coating may be provided on the internal surfaces of the hollow waveguides (including any lid portion) by an all-dielectric, or a metal- dielectric, stack. A person skilled in the art would recognise that the optical thickness of the dielectric layer (s) provides an interference effect that will determine the reflective properties of the coating. The dielectric material may be deposited by CVD or sputtering or reactive sputtering. Alternatively, a dielectric layer could be formed by chemical reaction with a deposited metal layer. For example, a layer of silver could be chemically reacted with a halide to produce a thin surface layer of silver halide.

In other words the reflective coating may be provided by an all-dielectric, or a metal-dielectric, stack. A person skilled in the art would recognise that the optical thickness of the dielectric layer (s) gives the required interference effects and thus determines the reflective properties of the coating. The reflective properties of the coating may also be dependent, to some extent, on the properties of the material in which the hollow core waveguides are formed. Hence, the underlying semiconductor substrate may also form a base layer, and be a part of, any such multiple layer dielectric stack.

Furthermore, the layer of material carried on the internal surface of the hollow core waveguides is conveniently Silicon Carbide.

As described above, the additional layer of low refractive index material can be selected to provide efficient operation at any required wavelength. Silcon Carbide has a refractive index of 0.06 at 10. 6m, making such material particularly suited for inclusion in devices operating at such a wavelength.

Advantageously, at least one of the one or more hollow core optical waveguides have a substantially rectangular (which herein shall include square) cross-section.

A square, or almost square, cross-section hollow core waveguide provides a waveguide in which the losses are substantially polarisation independent and is preferred when the polarisation state of the light is unknown or varying.

Preferably, the rectangular hollow core optical waveguide has a first cross-sectional dimension parallel to a first waveguide wall and a second cross-sectional dimension orthogonal to said first cross-sectional dimension wherein the first cross-section dimension is at least 5% or 10% or 15% or 25% or 50% greater than the second cross-sectional dimension. As described with reference to figure 7d below, such a waveguide is preferred for linearly polarised light of known polarisation.

Advantageously, the refractive indices of the surfaces defining the at least one rectangular internal cross-section hollow core optical waveguide are substantially equal. This can reduce polarisation dependent losses in the waveguide.

Preferably, opposite surfaces forming the rectangular internal cross-section hollow core optical waveguide have substantially equal effective refractive indices and adjacent surfaces forming the rectangular internal cross-section hollow core optical waveguide have different effective refractive indices. As described with reference to figure-s-7a-to-7G--blovc--tailoring--the-refr-active--inclices---of--opposing-pair-s--of- waveguide walls enables transmission losses to be reduced when guiding light of a known linear polarisation.

Advantageously, a pair of opposed surfaces of the rectangular internal cross-section hollow core optical waveguide carry a high refractive index coating. This provides the high refractive index preferred when s-polarised light is to be reflected as described below.

The semiconductor material of the substrate may also be doped to modify its optical properties to reduce hollow core waveguide losses.

Conveniently, at least one of the one or more hollow core optical waveguides support fundamental mode propagation. Also, at least one of the one or more hollow core optical waveguides may advantageously support multi-mode propagation.

Preferably, the multi-mode region is of a length such that re-imaging occurs as described in more detail below.

A person skilled in the art would recognise that the shape and dimensions of the hollow waveguide will affect the associated optical guiding properties. For example, tapered hollow waveguides could be used to provide a beam expansion or compression function. The high resolution with which hollow core waveguides can be fabricated using micro-fabrication techniques allows the guiding properties to be tailored as required to optimise PLC operation. A person skilled in the art would however recognise that the shape of the hollow core optical waveguides may be dictated to some extent by the type of micro-fabrication process used. For example, v-grooves can readily be wet etched in [100] silicon whilst rectangular waveguides can be easily provided in [110] silicon by wet etching. However, deep reactive ion etching (DRIE) provides the greatest ease of manufacture. <BR> <BR> <P>Advantageously,-the-deviee-i-s-pr-ovided-for--oper-ation-with-r-adi-ation--within-the wavelength ranges of 0. 1, um to 20um, 0. 8, m to 1. 6m or more preferably in the range of 1. 4Fm to 1. 6, um. The optical properties of gold, silver and copper coating therefore make these metals particularly suited to inclusion in PLC devices for operation in the telecommunications wavelength band (i. e. for use with wavelengths centred around 1. 55, um). Advantageously, the device may operate in the thermal infra-red bands of 3-5um or 8-12, um..

Conveniently, the semiconductor substrate comprises at least one alignment slot arranged to receive an optical fibre cable and to optically couple said optical fibre cable with one of said one or more hollow core optical waveguide of the semiconductor substrate.

Furthermore, a mode matching means may be advantageously provided in the vicinity of the alignment slot to allow coupling between the modes of an optical fibre and the analogous modes of a hollow core optical waveguide of a different core diameter. For example, in the case of a fundamental mode optical fibre the mode matching means couples the fundamental mode of the fibre and the fundamental mode of the hollow core waveguide. In the case of multi-mode propagation, the mode spectrum of the optical fibre is matched to the mode spectrum of the hollow core waveguide. The mode matching means may advantageously comprise a GRIN rod, a ball lens, a conventional lens or a Fresnel lens.

Alternatively, the alignment slot may be arranged to received a lensed optical fibre.

Preferably, the alignment slot is arranged to receive a hollow core optical fibre. The optical fibre may be multi-mode or single mode.

Advantageously, at least one of said two or more optical components comprises a -mirr-ored-s-ur-faee-that is-angled--to-dir-ect-light-out-of-the--plane-of-the-semiconduetor- substrate. The mirrored surface may be a monolithic (e. g. an angled semiconductor surface as described in figure 15) or hybrid arrangement. In other words, the PLC is not restricted to guiding light in the plane of the substrate surface. Light may be directed out of the plane of the substrate. For example, stacked or three dimensional PLCs could be fabricated in accordance with this invention.

Conveniently, the PLC may further comprise at least one micro-wave component and/or a hollow core microwave waveguide. In other words, an optical/microwave hybrid circuit may be provided.

According to a second aspect of the invention, a base portion for a photonic light circuit comprises a semiconductor substrate having one or more hollow channels formed therein, wherein said base portion is arranged such that when combined with an appropriate lid portion at least one hollow core optical waveguide is formed.

Conveniently, at least one slot is formed in the semiconductor substrate of the base portion to receive in alignment an optical component.

According to a third aspect of the invention, a base portion for a photonic light circuit comprises a semiconductor substrate in which one or more hollow waveguide channels and at least one slot to receive in alignment an optical component are formed.

According to a fourth aspect of the invention, a method of fabricating a photonic light circuit comprising the steps of taking a base portion according to the second or third aspects of the invention and attaching a lid thereto.

According to a fifth aspect of the invention a method of fabricating a photonic light circuit device comprises the step of micro-fabricating one or more hollow channels in a semiconductor substrate that are suitable, in use, for acting as hollow core waveguides.

Conveniently, the additional step of fabricating slots in the semiconductor substrate for the appropriate passive alignment of optical components therein is performed.

The slots may be fabricated using micro-fabrication techniques, or by precision engineering techniques such as laser machining.

Advantageously, the method comprises the additional step of coating the internal surfaces of the hollow channel (s) with a layer of material having a refractive index lower than that of the waveguide core within the operating wavelength band.

According to a sixth aspect of the invention, a method of forming a photonic light circuit comprising the steps of (a) taking a semiconductor substrate in which at least one hollow core optical waveguide and at least one slot to receive an optical component are formed, and (b) introducing an optical component into the at least one slot, whereby the step of introducing the optical component into the at least one slot also acts so as to align said optical component.

According to a seventh aspect of the invention, a master suitable for forming a pattern in a layer of deformable material is provided wherein the master comprises semiconductor material appropriately patterned to form in said deformable material at least one hollow waveguide channel and at least one alignment slot wherein said at least one alignment slot is arranged to receive in alignment an optical component.

Alternatively, a master could be formed in semiconductor material that allows production of a sub-master. The sub-master may then be used to form the required pattern in a deformable material to define a PLC. A master or sub-master may also be used as a mould to form the required pattern in a fixable layer.

According to an eighth aspect of the invention a method of forming a photonic light circuit comprising the steps of; (a) using a master according to the seventh aspect of the invention to permanently form a pattern in a layer of deformable material and (b) introducing at least one optical. component into the at least one alignment slot formed in the deformable material.

A photonic light circuit device is thus described that comprises a semiconductor substrate wherein one or more hollow core optical waveguides are formed in the semiconductor substrate.

The invention will now be described, by way of example only, with reference to the accompanying drawings in which; Figure 1 shows a typical prior art SiOB comprising a plurality of optical components; Figure 2 shows a integrated solid core waveguide as used in certain prior art SiOB devices; Figure 3 shows a portion of a PLC according to the present invention; Figure 4 shows a number of hollow core waveguides according to the present invention; Figure 5 gives cross-sectional views of various hollow core waveguides; Figure 6 shows the Fresnel reflectance coefficient of a copper coated surface for s- polarised and p-polarised light; Figure 7 provides a cross sectional view of four additional hollow core waveguides; Figure 8a shows a hollow core beamsplitter, Figure 8b shows a Brewster plate and Figure 8c shows the reflectivity of silicon as a function of the angle of incidence of s-polarised and p-polarised light; Figure 9 illustrates a monolithic lens (figure 9a) and focussing mirror (figure 9b) fabricated in a Silicon substrate; Figure 10 illustrates a PLC in which light is coupled into and out of optical fibre cables; Figure 11 shows a tapered waveguide formed in a silicon substrate; Figure 12 shows a hollow core wavelength de-multiplexer formed in a silicon substrate; Figure 13 shows a hollow core proximity coupler; Figure 14 show a PLC having both hollow core and solid core waveguides; Figure 15 shows a PLC having a mirrored surface angled to couple light out of the plane of the substrate; Figure 16 show the optical loss of hollow core waveguides uses in PLCs of the present invention; Figure 17 illustrates the effect of angular misalignment ; Figure 18 illustrates the effect of lateral misalignment; and Figure 19 shows a means of holding a component in alignment in an alignment slot.

Referring to figure 1, typical prior art silicon optical bench apparatus is shown.

Figure la shows a silicon optical bench 2 having a micro-fabricated hollow channel 4 and a pair of solder connectors 6. The silicon optical bench 2 is configured to hold a laser 8 and a silica optical fibre cable 10.

Figure 1b shows the silica optical fibre 10 and the laser diode 8 mounted on the silicon optical bench 2. The hollow channel 4 is formed with high enough precision so that the optical output from the laser 8 is precisely aligned with the end of the silica optical fibre 10. The solder connectors 6 provide an electrical connection and attach the laser diode 8 to the substrate.

To minimise unwanted reflections from the end of each silica waveguide an antireflection coating (not shown) is provided. Alternatively, the silica waveguides can be refractive index matched (e. g. using an index matching gel) and connected directly to each of the optical components. The requirement for anti-reflection coatings and/or index matching adds to the cost of the overall device, and makes fabrication more complex and time consuming.

Although, for simplicity, a single optical fibre cable (i. e. silica optical fibre cable 10) and a optical component (i. e. the laser 8) are shown in figure 1, a person skilled in the art would recognise that complex multi-component photonic circuits can be fabricated using the same principle. Many optical components can be located on the silicon optical bench, and optical links can be established between the components using various lengths of silica optical fibre waveguides. The optical components may include, for example, optical modulators, beam splitters, beam recombiners, detectors etc.

Referring to figure 2, a prior art integrated optical waveguide for use as part of a silicon optical bench is shown.

A low refractive index silica layer 20 is deposited on the silicon optical bench substrate 22. A high refractive index layer of doped silica is formed on the silica layer 20, and a high refractive index waveguide core 24 is formed by etching away portions of the high index layer of doped silica. A capping layer 26 of low refractive index silica covers the high refractive index waveguide core 24.

The high refractive index waveguide core 24 acts as an optical waveguide, and the high refractive index of the core compared to the cladding provides light guiding by total internal reflection. This provides a optical waveguide that is integral with, and not merely held in connection with, the silicon optical bench. Solid core integral optical waveguides are thus a known alternative to optical fibres mounted in grooves on a silicon optical bench. However the use of integrated optical waveguide does not lessen the requirement to index match the waveguides to the optical components, or provide anti-reflection coatings. Depositing additional layers of material on the silicon substrate also increases the fabrication complexity of the photonic circuit.

Referring to figure 3, a hollow core waveguide photonic light circuit (PLC) 40 that forms part of a device of the present invention is shown; figure 3a giving a perspective view of the PLC and figure 3b showing a cross-section of the PLC along the dashed line marked"A"in figure 3 a.

The hollow core waveguide PLC 40 comprises a silicon base 42 and a silicon lid 44.

A laser 8 is attached to and aligned in the silicon base 42. Light emitted by the laser 8 is coupled in to the single mode hollow core waveguide 46 that is formed by the silicon base and the silicon lid 44. In other words, hollow core waveguides are formed directly in the silicon from which the PLC base and lid are fabricated. For simplicity, electrical connections to the laser 8 are not shown as a person skilled in the art would recognise the various ways in electrical connections could be made; for example, track implantation using diode isolation in the base 42.

The hollow core waveguide 46 of figure 3 can be seen to have a rectangular cross section. The use of rectangular waveguides (herein the term rectangular shall include square) having a substantially equal depth and width reduces polarisation dependent losses which can provide advantageous in many telecommunication applications.

Although rectangular waveguides are shown, the waveguide cross-section could be shaped as required. For example, circular or parabolic cross-section or V-shaped waveguides could be formed in the silicon base using appropriate etching techniques. Hollow waveguide structures could also be formed in the silicon lid 44.

However, this requires both the base and the lid to be patterned and also means the lid and base have to be precisely aligned. The dimensions of the hollow core waveguide can be selected to support fundamental mode or multi-mode propagation as required and are described in more detail below.

In the example described with reference to figure 3, Silicon is used to form the PLC as it can be etched to a very high degree of accuracy using micro-fabrication techniques of the type known to those skilled in the art. However, a person skilled in the art would also recognise that any micro-fabricated semiconductor material could be employed to form a PLC of the present invention.

The laser 8 is a separate component that is bonded to the silicon base 42; in other words it is a hybrid arrangement. A person skilled in the art would also recognise that it would be possible to bond the laser 8 to a lid, or to fabricate optical components in the silicon itself. Although only a laser 8 is described with reference to figure 3, many optical components could be located or formed and/or aligned on the silicon base or lid. Alignment slots formed in the lid may also be used to receive in alignment optical components. This technique thus allows complex multi- component PLCs to be fabricated. The optical components may include, for example, optical modulators, beam splitters, beam recombiners, detectors, gratings, mirrors, GRIN (graded refractive index) lenses etc. Examples of some of the types of optical components that could be formed in a PLC of the present invention are described in more detailed below.

To maximise optical transmission through the hollow core waveguide 46, a layer of gold 48 is provided on the internal surface of the hollow core waveguides 46. The deposition of a layer of gold onto the silicon base and lid can be readily achieved, for example using appropriate metal deposition techniques such as sputtering or plating.

The lid may be bonded to the base in a variety of ways known to those skilled in the art. Areas of silicon that do not form part of the hollow optical waveguides may be left exposed on the lid portion and/or the base portion, and the lid and base may be bonded via a gold-silicon eutectic bond. Silver loaded epoxy, solder or polymer adhesive may also be used to bond the lid and base. The lid may only cover a part of the base as required.

The presence of the layer of gold 48 provides ATIR within the hollow core device for light with a wavelength within the telecommunications wavelength band (i. e. for wavelengths around 1. 55, um). At these telecommunication wavelengths, gold has the required refractive index properties of n < 1.

Although a gold layer, 48 is described, a person skilled in the art would recognise that any material having a refractive index less than air (or whatever is contained within the cavity) at the wavelengths at which the waveguide is to be operated could be deposited on the surfaces defining the hollow core waveguide. The refractive indices of different materials can be found in various publications, such as"the handbook of optical constants"by E. D. Palik, Academic Press, London, 1998.

Metals typically posses a refractive index less than air over a given wavelength range; the particular wavelength range depending on the physical properties of the metal.

It should be noted that although the layer of gold 48 provides ATIR, coating the hollow core waveguide 46 with an additional layer of low refractive index material is not essential. The refractive index of silicon is around 3.4 at wavelengths between 0. 5Am and 30ors, and hence hollow core (i. e. air filled) waveguides formed from uncoated silicon will not provide ATIR of light within such a wavelength range.

However, uncoated silicon will still provide light guiding by way of Fresnel reflection. Hollow core waveguides that use Fresnel reflections to guide light will introduce more optical loss than waveguides that provide TIR, but in certain situations this increased level of optical loss is acceptable.

If a reflective coating is provided, the substrate may be formed from a micro- fabricated material other than a semiconductor. For example, plastic waveguide devices may be fabricated by techniques including hot embossing or injection moulding. The technique involves forming a master. The master may be formed in semiconductor material, such as silicon, using a deep dry etch. Alternatively, the master may be formed by electro deposition of layers using the LIGA or UV LIGA technique. Once the master is formed, the hollow core waveguides may be formed in a plastic substrate by stamping (i. e. pressing) or hot stamping. A master may also be fabricated which is suitable for forming a sub-master that can be used to form the hollow core waveguides in the plastic substrate. Hollow plastic waveguides can thus be formed and coated with a reflective coating. The plastic hollow core waveguides that carry the reflective coating may also be formed from plastic or a polymer. For example, the hollow core waveguides could be formed using a lithographic process on a"spin-on"polymer coating (e. g. SU8 available from Microchem. Corporation) Although a simple PLC is described with reference to figure 3, a person skilled in the art would recognise that the present invention is equally applicable to complex PLCs. For example, multiple optical components could be mounted on the PLC and linked via hollow core waveguide formed from the PLC substrate. Such PLCs could form the basis of optical signal processing, and/or optical signal routing and analysis system. Some examples of such PLCs are given below.

Referring to figure 4a, hollow core waveguide structures 60a, 60b and 60c formed in a silicon substrate 62 are shown in plan view. Angled surfaces (e. g. surface 64) are provided to guide the light through 90°.

To minimise phase perturbations on reflection, the angled surfaces 64 ideally require a surface finish that is flat to better than A/10 or more preferably to better than X/20. If using a wavelength of 1. 5jim, a surface finish flat to better than 150nm is thus required. This level of accuracy is readily attainable using micro-fabrication techniques which can typically provide a resolution of 30-50nm.

The angled surfaces 64 thus provide mirrors that allow sections of hollow waveguide to be orientated at any angle to one another. It would not be possible to bend an optical fibre cable through such an acute angle. If a similar circuit were to be fabricated using known SiOB techniques, it would be necessary to provide two sections of optical fibre with a separate (well aligned) mirror to couple light between the optical fibre sections. The present invention can thus provide more complex and compact circuit layouts than prior art SiOB devices.

Although monolithic mirrors are shown in figure 4a, it should be recognised that a hybrid arrangement could also provide the same optical function. For example, alignment slots could be fabricated to receive in alignment polished mirrors. The hybid arrangement is useful as it allows the use of high optical quality mirrors that can be designed to have a minimal polarisation dependence; for example they may carry a polarisation independent multiple layer coating.

The waveguide structures described with reference to figure 4a are all substantially straight and connected by appropriately placed mirrors. However, the hollow waveguide structures could also be curved. For example, and with reference to figure 4b, a curved waveguide 66 formed in a silicon substrate 62 is shown. A skilled person would recognise that the maximum curvature attainable would depend on the guide thickness.

Referring to figure 5a, a cross section through a hollow core waveguide structure 60 of the type described with reference to figure 4a is shown. The hollow core waveguide structure 60 is formed in the silicon substrate 62,, and a silicon lid portion 68 is also provided that can be attached to the substrate 62 in the manner described above to provide the required hollow core waveguide.

As shown in figure 5b, the internal surfaces of each of the walls forming the hollow. core 69 may additionally be coated with a layer of material 70, for example copper, gold or silver to enhance the reflectivity of the 1. 55 ! 1m radiation via TIR.

If guiding linearly polarised light of known polarisation, hollow core waveguides in which different internal surfaces have different optical properties can be provided to further decrease the optical losses associated with the waveguide.

Figure 6 shows the Fresnel Reflection coefficient for light incident on a surface from air at an angle of 86° as a function of the refractive index (n) and absorption (k) of that surface for s-polarised (Rs) and p-polarised (Rp) light. It can be seen from figure 6 that the Fresnel reflection coefficient is strongly dependent on the polarisation of the light. Therefore, if the polarisation state of the light that is to be guided by a rectangular hollow core waveguide is known a pair of opposed surfaces forming the waveguide could be configured to have a low refractive index to optimise reflectivity for p-polarised light whilst the second pair of opposed surfaces could be arranged to have a much higher refractive index to maximise reflectivity for s-polarised light.

A number of techniques are described with reference to figure 7 that can be used to form waveguides in which different internal surfaces have different optical properties.

Figure 7a illustrates a hollow waveguide formed in an SOI wafer 80 fabricated using silicon on insulator (SOI) fabrication techniques. The wafer 80 comprises an insulating layer 82 of Si02 material carried on a substrate 84 and having a layer of silicon 86 located thereon. The layer of silicon 86 is etched down to the insulating layer 82 to form the required channel 88. A lid portion 90 formed from Si02 material is also provided.

A hollow core waveguide is thus formed having a first surface 92 and a second surface 94 that consist of silicon, whilst a third surface 96 and a fourth surface 98 are silicon dioxide. The refractive index of silicon is around 3.5, whilst silicon dioxide has a refractive index of around 1.5. Hence, optical losses in the waveguide are reduced when light propagating in the y-direction along the waveguide is polarised in the z-direction; i. e. there is Rs reflection at the first surface 92 and second surface 94 and Rp reflection from the third surface 96 and fourth surface 98.

Referring to figure 7b, a hollow waveguide 100 formed in a silicon substrate 102 and having a lid portion 103 is shown. The upper wall 104 (i. e. the wall defined by the lid portion 103) and the lower wall 106 are coated with a first material, whilst the side-wall 108 and side-wall 110 are coated with a second material. The first material and second material are selected to have low and high refractive indices respectively in order to minimise optical losses of light polarised in the z-direction that propagates along the waveguide in the y-direction.

Although figure 7b shows coatings applied to all four walls of the waveguide, it would be appreciated that only a single wall, or a pair of opposed walls, could be coated as required. In other words, one or more of the walls could remain uncoated and thus have the optical properties of the semiconductor material used to form the substrate.

Furthermore, physical structures can be formed in silicon to enhance reflectivity for a given polarisation of light as required. Figure 7c shows how a hollow core waveguide can be formed in a silicon substrate that comprises etalon side-wall structures 122. In this case, the etalon side-wall structures will enhance reflectivity.

Although an etalon structure is shown in which the hollow portions are filled with air, another material (e. g. a liquid or gas) could be used instead of air to enhance reflectivity. The optical losses associated with hollow core waveguides can also be reduced further by controlling the shape of the waveguide core. For example, the wider the waveguide core, the lower the associated optical losses. Figure 7d shows a rectangular cross-section waveguide 132 formed in a silicon layer 130 and having a silicon lid portion 134. The hollow core of the waveguide 132 has a width"a"that is less than its depth"b". Light polarised in the z-direction and propagating along the waveguide 132 will thus experience lower losses than if it were to propagate through a waveguide of depth"a".

It should further be noted that structures can. be formed in a PLC in which light is guided in only one plane; for example, it could be arranged for there to be free- space propagation along a vertical axis but waveguiding on the horizontal axis. In this case the waveguides are referred to as planar waveguides; i. e. they only guide in one plane. Planar waveguides may be employed where beam expansion in one dimension is required whilst constraining the beam width by guiding in a second dimension. If guiding is required in only the horizontal plane, a lid portion is not required. The converse situation is also possible where light is guided between the lid and the floor of the waveguide but not in the lateral plane.

Referring to figure 8, it is demonstrated how beam splitters and polarisation filters may be formed in a hollow waveguide PLC.

Figure 8a shows a beam splitter fabricated from hollow waveguide structures formed in a silicon substrate 160. The beam splitter comprises an input hollow core waveguide 162, a first output hollow core waveguide 164 and a second output hollow core waveguide 166. Light propagating through the input hollow core waveguide 162 is partially reflected from a thin silicon wall 168 into the first output hollow core waveguide 164 and also partially transmitted and coupled into the second output hollow core waveguide 166.

The angle (0) between the input hollow core waveguide 162 and the first output hollow core waveguide 164 determines the angle of incidence of light on the thin silicon wall 168. As shown in figure 8c, the reflectance properties of silicon depends on both the angle of incidence, and the polarisation state, of the incident light. The relative proportion of the power coupled from the input hollow core waveguide 162 into the first and second output hollow core waveguides 164 and 166 can thus be selected by fabricating the device with a certain angle (0).

Furthermore, as shown in figure 8b, a polarisation splitter can be fabricated by arranging for the angle 0 to equal the Brewster angle. In this case, an angle of H = 74° will result in light polarised in the z-direction being routed from the device via the first output hollow core waveguide 164 whilst light polarised in the x-direction will be routed from the device via the second output hollow core waveguide 166.

An etalon filter may be formed in a device of the type described with reference to figures 8a and 8b instead of the thin silicon wall 168. This would provide an optical element that would have different reflective properties for different wavelengths of light, and hence the device could also operate as a spectral filter.

Although a monolithic beam splitter and Brewster plate are described, the skilled person would also appreciate that a similar optical function could be implemented using a hybrid arrangements. Alignment slots could be formed in the substrate to receive the necessary optical components.

Referring to figure 9, it can be seen how the silicon material of a substrate can also be formed to provide a light focussing function.

Figure 9a shows a silicon substrate 190 in which a silicon lens structure 192 and hollow core waveguide 194 have been formed. The lens structure 192 will act as a lens to enable light 196 guided along the hollow core waveguide 194 to be focussed to a point 198. Such lenses may be used, for example, to focus light to a detector element.

As shown in figure 9b, a shaped silicon reflector 200 can also be formed in a silicon substrate 202 to optically link hollow core waveguides. The reflector 200 performs the function of routing light through a certain angle (in this case 90°) from a first hollow core waveguide 204 to a second hollow core waveguide 206 whilst also focussing light 208. Again, such an element may be used in a variety of different ways in PLCs and would be relatively simple to realise as it does not require an anti-reflection coating.

Although a PLC of the type described herein may comprise a complete optical circuit, it may also be necessary to couple light into or out of a PLC, typically via optical fibres.

Figure 10 shows a PLC formed in a silicon substrate 220 and arranged to receive light from a first input optical fibre 222. The input optical fibre 222 has a hollow core, and light therefrom is coupled into the input hollow core waveguide 224 using an input ball lens 226. Light propagating along the input hollow core waveguide 224 is directed to an etalon structure 228. The etalon structure 228 spectrally filters light into the first output hollow core waveguide 230 or the second output hollow core waveguide 232 depending on its spectral characteristics. Light propagating through the first output hollow core waveguide 230 is coupled in to a first output optical fibre 234 via ball lens 236, and light propagating through the second output hollow core waveguide 232 is coupled in to a second output optical fibre 238 via ball lens 240. Again, a hybrid etalon filter could be used instead of the monolithic element shown.

Although ball lenses are shown in figure 10, other lenses such as GRIN rod lenses may alternatively be used. The etalon structure 228 may also be replaced with beam splitter or a Brewster plate as necessary. The optical fibre may be single or multiple mode as required.

SOI technology is particularly suited to forming PLCs to which optical fibre are coupled. This is because typical SOI wafers comprise a silicon layer that has a thickness which is very accurately defined during the manufacturing process. In the fabrication of hollow waveguide structures in the silicon layer of a SOI wafer, the silica insulating layer acts as a vertical"stop"as far as the etching process is concerned. SOI techniques can thus provide sub-um channel depth accuracy.

The SOI etching accuracy should be contrasted to channel etching in pure silicon which has an accuracy around a few percent of the etch depth. Etching a channel in a pure silicon wafer to take a fibre (stripped to its cladding diameter of 125. 0Fm) would produce a 3. 0, um to 4. 0, um inaccuracy in the depth of the etch channel. As the core of the fibre is typically only 10. 0, um in diameter a vertical misalignment of this magnitude when coupling to/from a fibre from/to some other component (e. g. from a semiconductor laser) could prove detrimental. Therefore, an SOI based fabrication route would have advantages for alignment and waveguide cross-section accuracy which would reduce polarisation dependent losses.

As described above, a PLC of the present invention could comprise hollow core waveguides that allow single or multiple mode propagation. In certain circumstances it may also be necessary to alter the dimensions of the hollow core waveguide; e. g. to efficiently couple light into or out of different optical components.

Referring to figure I I a, a hollow core waveguide structure 260 formed in a silicon substrate 262 is shown. A wide (125pm) diameter output waveguide 264 is optically linked to a narrower (62. 5pm) diameter input waveguide 266 via a tapered waveguide portion 268. The length of the tapered portion is 1.875mm.

Figure 11b shows the intensity field of light propagating in the output waveguide 264 that results from a fundamental mode input beam in the input waveguide 266.

As shown in figure 11c, the output light in the output waveguide 264 is propagating predominantly in the fundamental mode. In other words, the tapered waveguide allows expansion of the beam size whilst ensuring the majority of the output beam power is coupled into the fundamental mode.

PLCs of the present invention may also comprise hollow core multi-mode interference (MMI) devices formed in the substrate. An example of a beam splitting and beam recombining MMI device is given in US5410625. Variations and improvements to the basic MMI devices of US5410625 are also known. For example, US5379354 describes how variation of input guide location can be used to obtain a multi-way beam splitter that provides division of the input radiation into outputs beams having differing intensities. Use of MMI devices to form laser cavities has also been demonstrated in US5675603. Various combinations of MMI splitter and recombiner devices have also been used to provide an optical routing capability; for example, see US5428698. In all the above cases, the MMI device could be fabricated as hollow core waveguides in silicon, or any other appropriate semi-conducting material, and form an integral part of the PLC.

The MMI device may be fabricated from a multi-mode region formed in the substrate to which input and output single mode optical fibre cables are coupled. In this manner, beam splitting/combining can be obtained in which the split beams are images of the input beam.

In particular, rectangular or square cross-section hollow multi-mode waveguides can be designed to provide re-imaging of symmetric, anti-symmetric or asymmetric optical fields by designing the length of the waveguide to have an appropriate relationship to its width. For example, for a symmetric field in a square sectioned waveguide the re-imaging length is given by the square of the waveguide width over the wavelength of the propagating radiation, i. e. L = w2/X, where L, is the guide length, w, is its width, and, X is the wavelength of the radiation. Re-imaging of the symmetric field occurs at this length and multiples of this length, i. e. at n. w2/X, where, n, is an integer number.

For the case of a 50. 0um wide hollow waveguide and 1. 55, um radiation, the re- imaging length is given by 502/1. 55 = 1613Rm = 1.613 mm. The symmetric field would be re-imaged at this length and also at integer multiples of this length, i. e.

3.23 mm, 4. 84 mm etc. For example, a TEMoo gaussian input beam from a single mode optical fibre could be re-imaged at distances of 1.613 mm. At the re-imaging points any required optical components could be situated. In this manner the re- imaging phenomena provides an additional way of guiding light between a series of components.

Alternatively, for the case of an asymmetric optical field, re-imaging occurs at eight times the length required for symmetric field re-imaging, i. e. at 12.09 mm (8 x 1.613 mm) for a 50. 0, um wide hollow waveguide. A mirror image of the asymmetric field is also formed at half this length i. e. at 6.05 mm.

In the case of a rectangular waveguide where the horizontal and vertical widths of the waveguide are substantially different the re-imaging lengths associated with the two widths are themselves different. However, by arranging that the relationship between the widths of the rectangular hollow waveguide is such that re-imaging is produced at identical lengths for each width, any field can be re-imaged.

For example, a symmetric field can be re-imaged in a hollow rectangular waveguide by arranging that the re-imaging lengths, Li = ni. Wi, and, L2 = n2. w22/S, associated with axes of width wl and w2, are identical. This can be achieved by making w2 = (nl/n2) l/2. wl, here, as previously, ni and n2 are integer numbers.

Another type of M : MI device suitable for inclusion in a PLC of the present invention is the wavelength de-multiplexer described in co-pending PCT patent application number GB2002/004560 and shown in figure 12.

The demultiplexer 300 is formed in a silicon SOI substrate 302 and comprises an input fundamental mode waveguide 304, a central multi-mode region 306 and four output waveguides 308a-308d (collectively referred to as 308). The dimensions and positions of the waveguides are selected (as described in GB2002/004560) such that the four wavelengths components entering the multi-mode region 306 from the input fundamental mode waveguide 304 are separated and separately output via the output waveguides 308.

It is also possible to form PLC devices in which light is proximity coupled into adjacent waveguide. Referring to figure 13, a first hollow core waveguide 340, a second hollow core waveguide 342 and a third hollow core waveguide 344 formed in a silicon substrate 346 and having a lid portion 347 are shown. The thickness"c" of the silicon walls 348 and 350 is sufficiently thin to enable light to be transmitted to adjacent waveguides. A proximity coupler component of this type may be used as beam splitter ; for example to tap off a small percentage of a propagating beam without having to insert a beam-splitting component into the optical path.

Referring to figure 14, the PLC may comprise both hollow and solid core (e. g.

"ridge") silicon waveguides. This enables the realisation of optical functions in both solid and hollow core technologies.

Figure 14a shows a Brewster interface between a hollow core waveguide 400 and a solid core waveguide 402 both of which are formed on an SOI substrate. Figure 14b shows a cross-section along B-B of the hollow core waveguide 400, and figure 14c shows a cross-section along A-A of the solid core waveguide 402. The hollow core waveguide 400 terminates at the angled interface 404 of the solid core waveguide 402. The waveguide 400 and 402 are arranged such that the interface is at the Brewster angle. This provide efficient coupling between the hollow and solid core waveguides.

Referring to figure 15, a hollow core waveguide 450 and an angled surface 452 are shown. The surface 452 is angled at approximately 45° to the plane of the substrate such that light 454 is coupled out of the plane of the substrate. The arrangement shown in figure 15 may be used to couple light into or from other circuits or devices located in a different vertical plane to the plane of the substrate. In this manner a three dimensional stacked PLC (e. g. a three dimensional optical switch) could be produced.

One way of monolithically fabricating such a mirror is a precision off-axis cut in [100] silicon material that is offset at an angle of about 8-9°. Numerous alternative ways to manufacture such an angled surface would be apparent to a person skilled in the art. Hybrid mirror arrangements could also be used.

Referring to figure 16, experimental data showing the guide length dependent optical transmission characteristics of hollow core optical waveguides suitable for incorporation in a PLC of the present invention are shown.

Curve 500 shows the predicted, and points 502a-502c the measured, optical transmission of a hollow core waveguide formed in a silicon substrate having a square internal core of 50, um width and depth. Curve 504 shows the predicted, and points 506a-506c the measured, optical properties of a similar waveguide in which a copper coating has been applied to each of its internal surfaces.

Curve 508 shows the predicted, and points 510a-510c the measured, optical transmission of a hollow core waveguide formed in a silicon substrate having a square internal core of 125 u, m width and depth. Curve 512 shows the predicted optical properties of a similar waveguide in which a copper coating has been applied to each of its internal surfaces. In all cases shown in figure 16 radiation having a wavelength of 1. 55} im was used.

It can thus be seen that increasing the dimensions of a waveguide reduces optical losses and the inclusion of a reflective coating (in this case copper) reduces losses even further. However, allowable angular alignment tolerances are reduced.

Referring to figure 17, the effect of angular alignment of components is shown.

Figure 17a shows a silicon substrate 600 in which a first hollow waveguide 602, second hollow waveguide 604 and a third hollow waveguide 606 are formed. A beam splitting element 608 is located in alignment slot 610. It can be seen that the element 608 has an angular misalignment (30) determined by the element and slot manufacturing tolerances.

Figure 17b shows the power coupling efficiency into the various modes of a hollow core waveguide as a function of angular misalignment (axa). Curve 620 shows the power coupled into the fundamental mode, whilst curves 622 show the optical power coupled into the higher order modes.

Referring to figure 18, the effect of lateral alignment is demonstrated.

Figure 18a shows a first hollow core wave guide 650 laterally displaced from a second hollow core waveguide 652 by al. The power coupling coefficient as a function of lateral displacement is shown in figure 18b where curve 654 shows the power coupled into the fundamental mode, whilst curves 656 show the optical power coupled into higher order modes.

It can be seen from the above that fundamental mode propagation through an integrated system of components interconnected by hollow waveguides can be attained if the waveguide dimensions and alignment tolerances are appropriately selected. This is especially important in a system of components which couples light to/from single mode optical fibres because the amount of power in the fundamental waveguide mode dictates how much light is coupled to/from the single mode fibre.

Ensuring high efficiency fundamental mode propagation in the waveguides ensures good coupling to the fundamental mode of a single mode fibre and an overall low insertion loss.

In other words, there is a trade off between the width of the waveguide and the angular and lateral alignment tolerances that are required (of both waveguides and components) in order to ensure that efficient fundamental mode propagation is achieved. Lower attenuation coefficients can be obtained by making the guide cross-section (width) large enough because the attenuation coefficient is inversely related to waveguide width. Making the waveguide width larger also eases lateral alignment tolerances, but it can be seen to tighten angular alignment tolerances.

Referring to figure 19, a technique for ensuring accurate alignment of components placed in a slot is shown.

A silicon substrate 700 has a slot formed therein to hold an optical component 702.

A number of spring clips 704 (also termed micro-grippers) are formed in the silicon by known micro-fabrication techniques. These clips 704 are such that when displaced they provide a lateral force. In this manner, the component is held firmly in alignment in the slot.

Although figure 19 shows clips surrounding the optical component, it is also possible to press the component against a reference surface such as the side-wall of the slot. It would also be appreciated by the skilled person that springs or other MEMS features fabricated by the removal of a sacrificial layer of oxide in a silicon wafer would result in a certain amount of undercut. This undercut would have no effect if associated with alignment slots, and would also make little difference to propagation in a rectangular hollow waveguide where the mode distribution is typically circular or elliptical.

A PLC of the present invention could be used to implement numerous different optical circuits. A few examples of these include interferometers (e. g. Michelson or Mach-Zender), spectrometers, lidar and optical readout of MEM devices (such as sensors or actuators). Telecoms circuits (routers, multiplexers, demultiplexers etc) could also be implemented. Although optical components are described above, there is no reason why the PLC could not alternatively or additionally comprise microwave components and hollow core waveguides to guide the microwave radiation. Opto-microwave integration in a single circuit would thus be possible.

「特表2005-516253およびWO2003065091より引用」

|
|

[Claims] 活性表面増強ラマン分光(SERS)基板としての金属被覆ナノ結晶性シリコン

【特許請求の範囲】
【請求項1】
a)金属被覆多孔質基板を作成する段階と、
b)前記基板を少なくとも1個の被分析試料試を含む試料にさらす段階と、
c)少なくとも1個の被分析試料を検出および/または同定のためにレーザー励起および分光を用いる段階と
を備える、方法。
【請求項2】
前記基板が多孔質半導体基板である、請求項1に記載の方法。
【請求項3】
前記基板がナノ結晶性シリコン、単結晶シリコン、多結晶シリコン、非晶質シリコンおよびレーザーでアニールされたシリコンからなる群から選択される、請求項2に記載の方法。
【請求項4】
金属ナノ粒子が前記金属被覆多孔質基板に添加される、請求項1に記載の方法。
【請求項5】
前記金属被覆が銀、金、白金、銅および/またはアルミニウムを含む、請求項1に記載の方法。
【請求項6】
前記分光がラマン分光である、請求項1に記載の方法。
【請求項7】
前記ラマン分光が表面増強ラマン分光(SRRS)、表面増強共鳴ラマン分光(SERRS)、ハイパーラマンおよび/または反ストークスラマン分光(CARS)である、請求項6に記載の方法。
【請求項8】
前記被分析試料がアミノ酸、ペプチド、ポリペプチド、タンパク質、糖タンパク質、リポタンパク質、ヌクレオシド、ヌクレオシド、オリゴヌクレオチド、核酸、糖、炭水化物、オリゴ糖、多糖、脂肪酸、脂質、ホルモン、代謝物、サイトカイン、ケモカイン、受容体、神経伝達物質、抗原、抗体、基質、代謝物、コファクター、阻害剤、薬剤、医薬品、栄養物、プリオン、毒素、毒、爆発性物質、殺虫剤、化学兵器、生物兵器、バクテリア、ウイルス、放射性同位元素、ビタミン、複素環式芳香族化合物、発癌性物質、突然変異原、麻酔剤、アンフェタミン、バルビツール酸塩、幻覚発現物質、廃棄物および/または汚染物を含む群から選ばれる、請求項1に記載の方法。
【請求項9】
前記被分析試料がヌクレオシド、ヌクレオチド、オリゴヌクレオチド、核酸、アミノ酸、ペプチド、ポリペプチドまたはタンパク質である、請求項8に記載の方法。
【請求項10】
前記少なくとも1個の被分析試料が少なくとも1個のラマン標識でラベルされている、請求項1に記載の方法。
【請求項11】
各被分析試料が識別可能なラマン標識でラベルされている、請求項10に記載の方法。
【請求項12】
少なくとも1個の捕捉分子が金属被覆多孔質シリコン基板に付着している、請求項1に記載の方法。
【請求項13】
前記捕捉分子がオリゴヌクレオチド、核酸、抗体、抗体断片、抗原、エピトープ、レシチン、タンパク質、ポリペプチド、ペプチド、受容体タンパク質、リガンド、ホルモン、ビタミン、代謝物、基質、阻害剤、コファクター、医薬品、アプタマー、サイトカインおよび神経伝達物質を含む群から選ばれる、請求項12に記載の方法。
【請求項14】
a)金属被覆ナノ結晶性多孔質シリコン基板と、
b)レーザーと
c)ラマン検出器と
を備える、装置。
【請求項15】
前記多孔質シリコンを犠牲層として使用する、請求項14に記載の装置。
【請求項16】
前記犠牲層が金属で置き換えられる、請求項15に記載の装置。
【請求項17】
金属ナノ粒子を更に有する、請求項14に記載の装置。
【請求項18】
前記金属が銀、金、白金、銅および/またはアルミニウムを有する、請求項14に記載の装置。
【請求項19】
前記ラマン検出器と動作可能に結合するコンピュータを更に備える、請求項14に記載の装置。
【請求項20】
ラマン検出器と動作可能に結合するフロースルーセルを更に備え、前記フロースルーセル内部の前記金属被覆ナノ結晶性多孔質シリコン基板を通って流体が流れる、請求項14に記載の装置。
【請求項21】
前記金属被覆多孔質シリコン基板がマイクロ電気機械システム(MEMS)中に組み込まれる、請求項14に記載の装置。
【請求項22】
前記金属被覆多孔質シリコン基板が集積チップの一部として構成される、請求項21に記載の装置。
【請求項23】
前記金属被覆多孔質シリコン基板がウェーハから取り出され、MEMSに組み込まれる、請求項21に記載の装置。
【請求項24】
金属被覆ナノ結晶性多孔質シリコンの層を備えるウェーハ。
【請求項25】
前記金属被覆が銀、金、白金、銅および/またはアルミニウムを含む、請求項24に記載のウェーハ。
【請求項26】
前記多孔質シリコンがシリコン、酸化シリコン、二酸化シリコンおよび/または窒化シリコンを含む、請求項24に記載のウェーハ。
【請求項27】
前記多孔質シリコンが金属で被覆される前に二酸化シリコンによって酸化される、請求項24に記載のウェーハ。
【請求項28】
a)表面積の大きい金属被覆材料を調製する段階と、
b)材料を少なくとも1個の被分析試料を含む試料にさらす段階と、
c)レーザー励起と分光を用いて少なくとも1個の被分析試料を検出および/または同定する段階と
を備える、方法。
【請求項29】
前記表面積の大きい材料が多孔質シリコンである、請求項28に記載の方法。
【請求項30】
前記分光がラマン分光である、請求項28に記載の方法。

What is claimed is: 1. A method comprising: a) providing a metal-coated, porous substrate; b) exposing the substrate to a sample comprising one or more analytes; c) using laser excitation and spectroscopy to detect and/or identify one or more analytes.

2. The method of claim 1, wherein the substrate is a porous semiconductor substrate.

3. The method of claim 2, wherein the substrate is selected from the group consisting of nanocrystalline silicon, single crystal silicon, polycrystalline silicon, amorphous silicon and laser annealed silicon.

4. The method of claim 1, wherein metal nanoparticles are added to the metal-coated porous substrate.

5. The method of claim 1, wherein the metal coating comprises silver, gold, platinum, copper and/or aluminum.

6. The method of claim 1, wherein the spectroscopy is Raman spectroscopy.

7. The method of claim 6, wherein the Raman spectroscopy is surface enhanced Raman spectroscopy (SERS), surface enhanced resonance Raman spectroscopy (SERRS) hyper-Raman and/or coherent anti-Stokes Raman spectroscopy (CARS).

8. The method of claim 1, wherein the analyte is selected from the group consisting of an amino acid, peptide, polypeptide, protein, glycoprotein, lipoprotein, nucleoside, nucleotide, oligonucleotide, nucleic acid, sugar, carbohydrate, oligosaccharide, polysaccharide, fatty acid, lipid, hormone, metabolite, cytokine, chemokine, receptor, neurotransmitter, antigen, allergen, antibody, substrate, metabolite, cofactor, inhibitor, drug, pharmaceutical, nutrient, prion, toxin, poison, explosive, pesticide, chemical warfare agent, biohazardous agent, bacteria, virus, radioisotope, vitamin, heterocyclic aromatic compound, carcinogen, mutagen, narcotic, amphetamine, barbiturate, hallucinogen, waste product and contaminant.

9. The method of claim 8, wherein the analyte is a nucleoside, nucleotide, oligonucleotide, nucleic acid, amino acid, peptide, polypeptide or protein.

10. The method of claim 1, wherein the one or more analytes are labeled with one or more Raman labels.

11. The m ethod o f claim 1 0, wherein e ach analyte i s 1 abeled w ith a distinguishable Raman label.

12. The method of claim 1, wherein one or more capture molecules is attached to the metal-coated porous silicon substrate.

13. The method of claim 12, wherein the capture molecule is selected from the group consisting of oligonucleotides, nucleic acids, antibodies, antibody fragments, antigens, epitopes, lectins, proteins, polypeptides, peptides, receptor proteins, ligands, hormones, vitamins, metabolites, substrates, inhibitors, cofactors, pharmaceuticals, aptamers, cytokines and neurotransmitters.

14. An apparatus comprising: a) a metal-coated, nanocrystalline porous silicon substrate ; b) a laser; and c) a Raman detector.

15. The apparatus of claim 14, wherein the porous silicon is used as a sacrificial layer.

16. The apparatus of claim 15, wherein the sacrificial layer is replaced by a metal.

17. The apparatus of claim 14, further comprising metal nanoparticles.

18. The apparatus of claim 14, wherein the metal comprises silver, gold, platinum, copper and/or aluminum.

19. The apparatus of claim 14, further comprising a computer operably coupled to the Raman detector.

20. The apparatus of claim 14, further comprising a flow through cell operably coupled to the Raman detector, wherein flow passes through the metal-coated, nanocrystalline porous silicon substrate inside the flow through cell.

21. The apparatus of claim 14, wherein the metal-coated porous silicon substrate is incorporated into a micro-electro-mechanical system (MEMS).

22. The apparatus of claim 21, wherein the metal-coated porous silicon substrate is constructed as part of an integrated chip.

23. The apparatus of claim 21, wherein the metal-coated porous silicon substrate is removed from a wafer and incorporated into a MEMS.

24. A wafer comprising a layer of metal-coated nanocrystalline porous silicon.

25. The wafer of claim 24, wherein the metal coating comprises silver, gold, platinum, copper and/or aluminum.

26. The wafer of claim 24, wherein the porous silicon comprises silicon, silicon oxide, silicon dioxide and/or silicon nitride.

27. The wafer of claim 24, wherein the porous silicon is oxidized to silicon dioxide before it is coated with metal.

28. A method comprising: a) providing a metal-coated, high surface area material; b) exposing the material to a sample comprising one or more analytes; c) using laser excitation and spectroscopy to detect and/or identify one or more analytes.

29. The method of claim 28, wherein the high surface area material is porous silicon.

30. The method of claim 28, wherein the spectroscopy is Raman spectroscopy.

「特表2005-524857およびWO2003106943より引用」

|
|

活性表面増強ラマン分光(SERS)基板としての金属被覆ナノ結晶性シリコン

【発明の詳細な説明】
【技術分野】
【0001】
本発明の方法および装置300は、ラマン分光による分子の検出およびキャラクタリゼーションの分野に関する。具体的には、本明細書発明の方法および装置300は、表面増強ラマン分光(SERS)、表面増強共鳴ラマン分光(SRRRS)、ハイパーラマンおよび/またはコヒーレント反ストークスラマン分光(CARS)用の基板240、340としての金属被覆多孔質シリコンに関する。
【背景技術】
【0002】
生物試料およびその他の試料由来の1個の分子を感度良く正確に検出および/または同定することは、医療診断学、病理学、毒性学、環境サンプリング、化学分析、科学捜査その他の多数の分野で広く潜在的に求められてきたが達成しにくい目標と考えられてきた。この目標を達成するため、ラマン分光および/または表面プラズモン共鳴を使用することが試みられている。光が実際の媒体を通過する場合、光はある割合でその最初の方向を変えるが、これはラマン散乱と呼ばれる現象である。光の吸収とより高いエネルギー状態へ電子が励起されるため、散乱光の一部は最初の励起光と周波数が異なり、異なった波長で光を放出する。ラマン発光スペクトルの波長は、試料中の光吸収分子の化学組成と構造に特徴的であるが、光散乱強度は試料中の分子の濃度に依存する。
【発明の開示】
【発明が解決しようとする課題】
【0003】
励起光線と試料中の各分子との間で生じるラマン相互作用の確率は極めて低いために感度が低く、ラマン分析の用途は限られる。銀の粗い表面近くの分子は、6から7倍のオーダーも大きい増強ラマン散乱を示すことが観察されている。この表面増強ラマン分光(SERS)効果は、プラズモン共鳴現象と関連しているが、金属中の伝導電子の総体的な結合により金属のナノ粒子が入射電磁波に応答して顕著な光共鳴を示す。要するに、金、銀、銅その他の金属は微小アンテナとして作用し、電磁波の局所効果を増強する事が出来る。この様な粒子の近傍に位置する分子は、ラマン分光分析に対しはるかに高い感度を示す。
【0004】
SERSを分子の検出と分析に利用する事、具体的には基板の表面を金属ナノ粒子で被覆、または粗い金属フィルムを作成し、試料を金属被覆表面に載せる事が試みられている。しかしながら、平坦な表面に堆積できる金属粒子の数は限られ、SERSおよびこの様な表面を利用するラマン技術の増強因子は比較的低くなる。金属粒子の密度がより高いSERS活性基板の製造法、およびこの様な基板を備える装置が必要である。
【発明を実施するための最良の形態】
【0005】
以下の図面は本明細書の一部を形成し、本発明の開示された実施形態の特徴の一部をさらに説明するために記載される。本明細書に提示された具体的な実施形態の詳細な説明と組み合わせたこれらの本発明の実施形態のいくつかを参照することにより、本発明の実施形態をより良く理解されると考えられる。
【0006】
開示された方法および装置300は、表面増強ラマン分光(SERS)、表面増強共鳴ラマン分光(SERRS)および/または反ストロークラマン分光(CARS)検出による被分析試料の検出および/または同定に有用である。既存の技術と比較して、開示された方法と装置300は、金属粒子密度が増加しSERS増強場がより深いSERS活性基板240、340を提供し、より効果的なラマン検出および/または同定分析を可能にする。
【0007】
様々な分析のSERS検出の従来の方法は、例えば基板および/または支持体上に被覆した凝集銀ナノ粒子等のコロイド状金属粒子を使用してきた(例えば米国特許第5,306,403号;6,149,868号;6,174,677号;6,376,177号)。この様な構成はSERS検出の感度を106~108倍に増加させることもあるが、本明細書に開示される様なヌクレオチド等の小さな被分析試料等の1個の分子を検出する事は出来ない。ラマン検出の感度増大は明らかにコロイド状粒子凝集物内で均一でなく、“ホットスポット”の存在に依存する。この様な“ホットスポット”の物理的な構造、感度増大を生じるナノ粒子からの距離の範囲、および感度増大が可能な凝集ナノ粒子と被分析試料間の空間的関係は明らかにされていない。さらに、凝集ナノ粒子は溶液中で本質的に不安定であり、単一分子の被分析試料検出の再現性に悪影響を及ぼす。本発明の方法および装置300はSERS検出のための安定なマイクロ環境を提供し、その中でラマン活性金属基板の物理的配座および密度を正確に制御でき、溶液中の被分析試料の再現性があり、感度が高く正確な検出を可能にする。
【0008】
以下の詳細な説明には、本発明の開示された実施形態をより完全に理解するための様々な具体例が含まれる。しかしながら、本発明の実施形態はこれらの実施例がなくても実行可能であることは、当業者に自明であると思われる。他の実例では、公知の装置、方法、手順および個々の要素を本発明中で詳細に説明しない。
【0009】
定義
本明細書に使用する単数形の名詞は、一つまたは一つ以上の品目を意味する。
【0010】
本明細書では、“約”はその値の10パーセント以内を意味する。例えば“約100”は90~110の値を意味する。
【0011】
本明細書では、“被分析試料”と言う用語は検出および/または同定の対象としての任意の原子、薬剤、分子、化合物、組成物または凝集物を意味する。被分析試料の非限定例にはアミノ酸、ペプチド、ポリペプチド、タンパク質、糖タンパク質、リポタンパク質、ヌクレオシド、ヌクレオチド、オリゴヌクレオチド、核酸、糖、炭水化物、オリゴ糖、多糖、脂肪酸、脂質、ホルモン、代謝物、サイトカイン、ケモカイン、受容体、神経伝達物質、抗原、アレルゲン、抗体、基質、代謝物、コファクター、阻害剤、薬物、医薬品、栄養物質、プリオン、トキシン、毒素、爆発性物質、殺虫剤、化学兵器、生物兵器、放射性同位体、ビタミン、複素環式芳香族化合物、発癌性物質、突然変異原、麻酔剤、アンフェタミン、バルビツール酸塩、幻覚発現物質、廃棄物および/または汚染物が含まれる。本発明のある実施形態では、以下に開示する様に少なくとも1個の被分析試料が少なくとも1個のラマン標識物でラベルされる。
【0012】
本明細書で用いられる“捕獲”分子とは、少なくとも1個の標的被分析試料と結合する任意の分子を意味する。“捕獲”分子の非限定例には抗体、抗体断片、遺伝子組み替え抗体、単一鎖抗体、受容体タンパク質、結合タンパク質、酵素、阻害タンパク質、レシチン、細胞付着タンパク質、オリゴヌクレオチド、ポリヌクレオチド、核酸およびアプタマーが含まれる。
【0013】
本明細書で用いる“ナノ結晶性シリコン”という用語は、ナノメートルスケール、具体的には1~100ナノメートルのサイズの範囲のシリコン結晶を含むシリコンを意味する。“多孔質シリコン”110、210はエッチングまたは他の処理により多孔質構造となったシリコンを意味する。
【0014】
本明細書で用いる“動作可能に結合する”とは、少なくとも2個の装置300および/またはシステムの機能的相互作用を意味する。例えば、コンピュータ160、395が検出器380によりラマン信号に関するデータを取得、処理、記憶および/または伝達し得る場合、ラマン検出器380はコンピュータ160、395と“動作可能に結合”し得る。
【0015】
ナノ結晶性多孔質シリコン
ナノ結晶性シリコン
本発明のある実施形態は少なくとも1層のナノ結晶性シリコンを有する装置300に関する。様々なナノ結晶性シリコン製造法が公知である(例えばペトロワ-コッホ(Petrova-Koch)ら、“迅速熱酸化多孔質シリコン-優れた燐光性Si(Rapid-thermal-oxidized poroussilicon-the superior photoluminescent Si)”、アプライドフィジクス、レターズ(Appl.Phys.Lett.)、61:943、1992;エデルベルグ(Ederberg)ら、“プラズマ増強化学蒸着で製造されたナノ結晶性シリコンフィルムからの可視燐光(Visible luminescence from nanocrystallinesilicon films produced by plasma enhanced vapor deposition”、アプライドフィジクス、レターズ(Appl.Phys.Lett.)、68:1415-1417、1996;シェーンフェルド(Schoenfield)ら、“ナノ結晶性シリコンにおけるSi量子ドットの作成(Foematon of Si quantum dots in nanocrystallinesilicon)”、修飾された半導体構造に関する第7回国際会議プロシーディングス(Proc.7th Int.Conf.on Modulated SemiconductorStructures)、マドリッド(Madrid)、p605-608、1995;ザオ(Zao)ら、“ナノ結晶性Si:Si量子ドットで製造した材料(Nanocrystalline Si:a material constructed by Si quantumdots)”、低次元構造および装置に関する第1回会議(1st Con.on Low Dimensional Structure andDevices)、シンガポール(Singapore)、p467-471、1995;ルーツェン(Lutzen)ら、“非晶質シリコンのアニーリングで製造した超薄膜ナノ結晶性シリコンフィルムの構造的特徴(Structural characteristics of ultrathinmamocrystalline silicon films formed by annealing amorphoussilicon)”、ジャーナル オフ バキューム サイエンステクノロジー(J.Vac.Sci.Technology)、B16:282-05、1998;米国特許第5,770,022号;第5,994,164号;第6,268,041号;第6,294,442号;第6,300,193号)。本明細書に開示される方法と装置300はナノ結晶性シリコンの製造法に限られず、任意の方法も使用し得ると理解される。
【0016】
ナノ結晶性シリコンの非限定的製造法の例にはシリコンリッチな酸化物中へのシリコン(Si)注入およびアニーリング;金属核形成触媒による固相結晶化;化学蒸着;PECVD(プラズマ増強化学蒸着法);ガス気化;気相熱分解;気相光熱分解;電気化学エッチング;シランおよびポリシランのプラズマ蒸着;高圧液相酸化還元反応;非晶質シリコン層の迅速アニーリング;LPCVD(低圧気相蒸着)による非晶質シリコン層の蒸着後のRTA(迅速化学アニーリング)サイクル;シリコン陽極を用いるプラズマ電気アークとシリコンの焼結(米国特許第5,770,022号;第5,994,164号;第6,268,041号;第6,294,442号;第6,300,193号)が含まれる。プロセスによっては、サイズが1~100nm以上のいずれかのSi結晶をチップ上の薄層、独立層および/または凝集結晶として形成する。本発明のある実施形態では、基板層110、210に付着したナノ結晶性シリコンを含む薄層を使用し得る。
【0017】
本発明の様々な実施形態において、ナノ結晶性シリコンを開示された方法および装置300で使用し得ることが予想される。しかしながら、実施形態は出発原料の組成に限定されず、本発明の他の実施形態では他の原料も使用し得ることが予想され、唯一の要請はその材料が、図2に示される様なラマン感受性金属で被覆し得る多孔質基板110、210を形成し得るものでなければならないことである。
【0018】
本発明のある実施形態では、シリコン結晶および/または多孔質シリコンにおける細孔サイズをある限界内で、例えば金属被覆多孔質シリコン240、340のプラズモン共鳴周波数を最適化する様に選ばなければならない(例えば米国特許第6,334,272号参照)。プラズモン共鳴周波数は多孔質シリコン240、340を被覆する金属層の厚さでも調整し得る(米国特許第6,344,272号)。ナノスケールシリコン結晶のサイズを制御する技術は公知である(例えば米国特許第5,944,164号および第6,294,442号)。
【0019】
多孔質シリコン
本発明のある実施形態は、ラマン活性金属被覆基板240、340を使用する装置300および方法に関する。様々な実施形態で、基板はナノ結晶性多孔質シリコン110、210を含む。上記の議論の如く、基板110、210は純シリコンに限定されず、窒化シリコン、ゲルマニウムおよび/またはチップ製造業者に公知の他の物質も包含し得る。金属核化触媒および/またはドーパント等の他の微量物質も存在し得る。唯一の要請は、基板物質が図2に例示する様にラマン感受性金属で被覆し得る多孔質基板110、210を形成し得るものでなければならないことである。多孔質シリコンは783m2/cm3もの大きな表面積を有し、表面増強ラマン分光技術のための極めて大きい表面を提供する。
【0020】
多孔質シリコン110、210は1950年代の末に、希弗化水素酸水溶液中でのシリコンの電解研磨で発見された。公知の様に、多孔質シリコン110、210は、電気化学セル120中での希弗化水素酸(HF)150中でのシリコン基板110、210のエッチングによって製造し得る。場合によっては、低電流密度でシリコンを最初にHF150でエッチングしてもよい。最初の細孔形成後、シリコンを電気化学セル120から取り出し、電気化学セル120中で形成した細孔を広げるため濃HF150中でエッチングする。シリコンがドープされているかどうか、ドーパントのタイプおよびドーピングの程度に依存して、シリコン基板110、210の組成も細孔サイズに影響する。シリコン細孔サイズに対するドーピングの効果は公知である。大きな生物分子の検出および/または同定を含む本発明の実施形態では、約2nm~100または200nmの細孔サイズが選ばれる。多孔質シリコン中の細孔の配向も、本発明の具体的な実施形態では選ばれる。例えば、エッチングした(1、0、0)結晶構造は結晶に垂直に配向した細孔を有するが、(1、1、1)または(1、1、0)結晶構造は結晶軸と対角線に配向した細孔を有する。細孔配向に対する結晶構造の効果は公知である。結晶組成と多孔質も、ラマン信号を増大しバックグラウンドの雑音を低減するために多孔質シリコンの光学的性質を変える様に制御し得る。多孔質シリコンの光学的性質は公知である(例えばクリス(Culliss)ら、ジャーナルオフ アプライド フィジクス(J.Appl.Phys.)、82:909-965;コリンズ(Collims)ら、フィジクストゥデイ(Physics Today)、50:24-31、1977)。
【0021】
多孔質シリコン基板製造用の方法および装置100の非限定例を図1に示す。シリコンウェーハ110が、不活性材料、例えばテフロン(登録商標)を備える電気化学セル120の内部に置かれる。ウェーハ110を定電流電源130の陽極と接続し、電気化学セルの陰極110とする。定電流電源130の陰極は白金陰極140等の陰極に接続される。電気化学セル120にHFの希エタノール溶液150を満たす。または、HFをペンタンまたはヘキサン等の公知のアルコールおよび/または界面活性剤に溶解してもよい。本発明のある実施形態では、コンピュータ160、395を定電通電源130に動作可能に結合し、電流、電圧および/または電気化学エッチング時間を制御する。電気化学セル120中でHF電解質150にさらされたシリコンウェーハ110はエッチングされて多孔質シリコン基板110、120を生成する。公知の様に、多孔質シリコン層210の厚さとシリコン210の多孔質度を陽極酸化の時間および/または電流密度、および電解質溶液中のHF150の濃度を調節して制御し得る(例えば米国特許第6,358,815号)。
【0022】
本発明の様々な実施形態において、シリコンウェーハ110の一部をポリメチルメタクリレート等の公知のレジスト化合物で被覆してHF150によるエッチングから保護し得る。シリコンウェーハ110の選ばれた部分をHF150にさらするための、フォトリソグラフィー等のリソグラフ法は公知である。ラマン分光で用いられる多孔質Siチャンバーのサイズと形を制御するために、選択的エッチングは有用である。本発明のある実施形態では、直径約1μm(マイクロメートル)の多孔質シリコンチャンバーが用いられる。本発明の他の実施形態では、幅約1μmの多孔質シリコン溝またはチャネルが用いられる。多孔質シリコンチャンバーのサイズに制限はなく、任意のサイズまたは形の多孔質シリコンチャンバーを使用し得ると考えられる。例えば、1μmサイズのチャンバーをサイズが1μmである励起レーザーで使用し得る。
【0023】
上記の方法の例は多孔質シリコン基板110、210の製造を制限するものではなく、公知の任意の方法を使用し得ると考えられる。多孔質シリコン基板110、210の非限定製造法の例にはシリコンウェーハ110の陽極エッチング、電気めっき、およびシリコン/酸素含有材料の蒸着後の制御アニーリングが含まれる(例えばカンハム(Canham)、“ウェーハの電気化学および化学溶解によるシリコン量子ワイヤアレイ製造(Silicon quantumwire array fabrication by electrochemical and chemical dissolution ofwafers)”、Appl.Phys.Lett.、57:1046、1990;米国特許第5,561,304号;第6,153,489号;第6,171,945号;第6,322,895号;第6,358,613号;第6,358,815号;第6,359,276号)。本発明の様々な実施形態において、多孔質シリコン層210をバルクシリコン、石英、ガラスおよび/またはプラスチック等の少なくとも1つの支持層に付着し得る。ある実施形態では、窒化シリコン等のエッチング停止層をエッチングの深さを制御するために使用し得る。多孔質シリコン層を、チップ製造の公知の方法を用いて半導体チップ中に含ませる事も出来る。本発明のある実施形態では、金属被覆多孔質シリコン240、340チャンバーを組み込みチップの一部として設計し、様々なチャネル、マイクロチャネル、ナノチャネル、マイクロ流動チャネル、反応チャンバー等に接続し得る。また別な実施形態では、金属被覆多孔質シリコン基板240、340チャンバーをシリコンウェーハから切り出し、チップおよび/または他の装置に組み込む事が出来る。
【0024】
また別な実施形態では、被覆前後に多孔質シリコン基板110、210の更なる変更が行えると考えられる。例えば、多孔質シリコン基板110、210をエッチング後、多孔質シリコン基板を公知の方法を用いて酸化シリコンおよび/または二酸化シリコンに酸化し得る。例えば酸化を多孔質シリコン基板110、210の機械的強度と安定性を増加させるために使用し得る。または、金属被覆シリコン基板240、340を更にエッチングして、中空で残されるか、若しくは別なラマン活性金属等の他の材料で満たされた金属外皮を残してシリコン材料を除去する事が出来る。
【0025】
多孔質シリコンの金属被覆
多孔質シリコン基板110、210を金、銀、白金、銅またはアルミニウム等のラマン活性金属で、公知の任意の方法で被覆し得る。この方法の非限定例には電気めっき、陰極電気移動、金属の蒸発とスパッタリング、メッキを触媒する種結晶の使用(即ち、金めっきのための銅/ニッケル種結晶の使用)、イオン注入、拡散、またはシリコン基板110、210、240、340上に金属薄膜をメッキするための任意の他の方法(例えばロペス(Lopez)およびフォーチェット(Fauchet)、“多孔質シリコン1次元フォトンバンドギャップ構造からのエルビウム放射(Erbium emission fromporous silicon one dimensional photonic band gap structure)”、アプライドフィジクスレターズ(Appl.Phys.Lett.)、77:3740-6、2000;米国特許第5,561,304号;第6,171,945号;第6,359,276号参照)がある。金属被覆のその他の非限定例には無電解めっきがある(ゴール(Gole)ら、“直接電気接触のための無電解溶液から多孔質シリコンのパターン金属化(Patterned metallization of porous silicon fromelectroless solution for direct electrical contact)”、ジャーナル オフ エレクトロケミカルソサイエティ(J.Electrochem.Soc.)、147:3785、2000)。金属層の組成および/または厚さを、金属被覆多孔質シリコン240、340のプラズモン共鳴周波数を最適化する様に制御する事が出来る。
【0026】
本発明の別な実施形態では、被分析試料検出に用いられるラマン活性基板240、340は金属被覆ナノ結晶性多孔質シリコン基板240、340の異なったタイプの基板上に被覆された銀または金ナノ粒子等の固定化金属コロイド、および/または金属被覆ナノ結晶性多孔質シリコン基板240、340上に被覆された固定化金属コロイドを備える。後者の組成ではラマン活性金属の密度が極めて高いと思われ、溶液中の被分析試料が基板中に入るチャネルは相対的に小さい。このことは大きなタンパク質または核酸等の大きな被分析試料にとって不都合であるが、1個のヌクレオチドまたはアミノ酸等の小さな被分析試料の感度と検出が向上すると考えられる。
【0027】
ナノ粒子
本発明のある実施形態では、金または銀ナノ粒子等のラマン活性金属粒子を、ラマン信号をさらに増強するため金属被覆多孔質シリコン基板240、340に添加する事が出来る。本発明の様々な実施形態では、直径1nm~2μmのナノ粒子を使用し得る。本発明のまた別な実施形態では、直径2nm~1μm、5nm~500nm、10nm~200nm、20nm~100nm、30nm~80nm、40nm~70nm、または50nm~60nmのナノ粒子が検討される。本発明のある実施形態では、平均直径10~50nm、50~100nmまたは約100nmのナノ粒子が考慮される。ナノ粒子のサイズは金属被覆多孔質シリコン240、340中の細孔の直径に依存し、ナノ粒子が最高の内部にはまり込む様に選ばれる。ナノ粒子の形はほぼ球形であるが、任意の形または不規則な形のナノ粒子を使用してもよい。ナノ粒子の製造法は公知である(例えば米国特許第6,054,495号;第6,127,120号;第6,149,868号;リー(Lee)およびマイセル(Meisel)、ジャーナル オフ フィジカルケミストリー(J.Phys.Chem.)、86:3391-3395、1982)。ナノ粒子はまた、ナノプリズムの形でも製造し得る(ジン(Jin)ら、“銀ナノスフェアのナノプリズムへの光誘発転換(Photoinduced conversion of silver nanospherestonanoprism)”、サイエンス(Science)、294:1901、2001)。市販品からナノ粒子を入手する事も出来る(例えばナノプローブ社(Nanoprobes、Inc.、)、ヤファンク(Yaphank)、NY;ポリサイエンス社(Polyscience、INC.)、ワーリントン(Warrington、PA)。
【0028】
本発明のある実施形態では、ナノ粒子はナノ粒子の不規則な凝集体である(コロイド状ナノ粒子)。本発明の他の実施形態では、ナノ粒子を架橋して二量体、三量体、四量体その他の凝集体等のナノ粒子の粒子状凝集体を製造し得る。本発明の別な実施形態では、サイズの異なる凝集体の不均一混合物を使用し得るが、また別な実施形態ではナノ粒子凝集体の均一な集合を使用し得る。本発明のある実施形態では、蔗糖勾配溶液中の超遠心等の公知の技術により特定の数のナノ粒子を含む凝集体(二量体、三量体等)の割合を増加させるか、そのような凝集体が精製される。
【0029】
ナノ粒子を架橋する方法は公知である(例えばフェルドハイム(Feldheim)、“分子架橋を用いる金属ナノ粒子アレーの集合体(Assemblyof Metal Particle Array using molecular bridges)”、電気化学協会インターフェース(TheElectrochemical SocietyInterface)、フォール(Fall)、2001、p22-25参照)。末端チオールまたはスルフヒドリル基を有するリンカー化合物と金ナノ粒子殿反応は公知である(フェルドハイム(Feldheim)、2001)。本発明のある実施形態では、1個のリンカー化合物が両端のチオール基で誘導体化される。金ナノ粒子と反応すると、リンカーはリンカーの長さで隔てられたナノ粒子二量体を形成する。本発明の他の実施形態では、3個、4個またはそれ以上のチオール基を有するリンカーを複数のナノ粒子に同時に付着させるために使用される(フェルドハイム(Feldheim)、2001)。リンカー化合物に対し過剰のナノ粒子を使用すると、重複架橋を形成しナノ粒子が沈殿することを防止できる。銀ナノ粒子の凝集体を公知の標準合成法で生成し得る。
【0030】
本発明の具体的な実施形態では、金または銀ナノ粒子をアミノシラン、3-グリシドキシプロピルトリメトシキシラン(GOP)またはアミノプロピルトリメトシキシラン(APTS)等のシラン誘導体で被覆し得る。シラン末端の反応性基をナノ粒子の架橋凝集体を生成するために使用する。リンカー化合物は殆ど任意の長さであると考えられ、その長さは0.05、0.1、0.2、0.5、0.75、1、2、3、4、5、6、7、8、9、10、11、12、13、14、15、16、17、18、19、20、21、22、23、24、25、27、30、35、40、45、50、55、60、65、70、90、90~100nmまたはそれ以上の範囲である。本発明のある実施形態では長さが不均一なリンカーを使用し得る。
【0031】
本発明の他の実施形態では、ナノ粒子をリンカー化合物と結合する前に様々な反応基を含む様に修飾し得る。ナノプローブ社(Nanoprobes、Inc.、ヤファンク(Yaphank)、NY)製のナノゴールド(登録商標)ナノ粒子等の修飾ナノ粒子は市販されている。ナノゴールド(登録商標)ナノ粒子はナノ粒子1個当たり1個または複数のマレイミド、アミンまたは他の基が結合したナノゴールド(登録商標)ナノ粒子が入手可能である。ナノゴールド(登録商標)ナノ粒子はまた、正または負に荷電した形で入手でき、電磁場中でナノ粒子を取り扱いやすくなる。この様な修飾ナノ粒子を様々な公知のリンカー化合物と結合させ、二量体、三量体その他のナノ粒子凝集体を調整し得る。
【0032】
溶液中で沈殿しないナノ粒子の小さな凝集体を形成し得るかぎり、リンカー化合物のタイプは制限されない。本発明のある実施形態では、リンカー基はフェニルアセチレンポリマーを有する(フェルドハイム(Feldheim)、2001)。また、リンカー基はポリテトラフルオロエチレン、ポリビニルピロリドン、ポリスチレン、ポリプロピレン、ポリアクリルアミド、ポリエチレンその他の公知のポリマーでもよい。使用するリンカー化合物はポリマーに限られず、シラン、アルケン、シラン誘導体またはアルカン誘導体等の他のタイプの分子を含み得る。本発明の具体的な実施形態では、アルカンまたはシラン等の比較的簡単な化学構造のリンカー化合物が、被分析試料から放射されるラマン信号と干渉することを避けるために使用される。
【0033】
マイクロ電気機械システム(MEMS)
本発明のある実施形態では、ラマン活性金属被覆多孔質シリコン基板230、340を大きな装置300および/またはシステムに組み込む事が出来る。ある実施形態では、基板230、340をマイクロ電気機械(MEMS)に組み込む事が出来る。MEMSは機械要素、センサー、アクチュエータおよび電子回路を備える複合システムである。これらの部品は全て、シリコン系または等価の基板でなる共通のチップ上に、公知のマイクロ加工技術により製造し得る(例えばボルドマン(Voldman)ら、アニューアル レビュー オフ バイオメディカルエンジニアリング(Ann.Rev.Biomed.Eng.)、1:401-425、1999)。MEMSのセンサー部品は機械的、熱的、生物学的、化学的、光学的および/または電磁現象を測定するために使用し得る。電子回路はセンサーからの情報を処理し、ポンプ、バルブ、ヒーター、クーラー、フィルター等のアクチュエータ部品を制御し、MEMSの機能を制御し得る。
【0034】
MEMSの電子部品を集積回路(IC)プロセス(例えばCMOS、バイポーラーまたはBICMOSプロセス)を用いて製造する事が出来る。電子部品をコンピューターチップ製造業者に公知のフォトリゾグラフィーおよび/またはエッチング法を用いてパターン化する事が出来る。電気機械部品を、シリコンウェーハの一部を取り除くか、新しい構造層を付け加えて機械的および/または電気機械的部品を生成する互換性のある“マイクロ加工”プロセスを用いて製造し得る。
【0035】
MEMS製造の基本技術には材料の薄膜の蒸着、フォトリゾグラフィーイメージングまたは他の公知のリゾグラフィー法によるフィルム上へのパターンマスク作成、およびフィルムの選択的エッチングが含まれる。薄膜は数ナノメートルから100マイクロメートルの範囲の厚さである。使用する蒸着技術には化学蒸着(CVD)、電気蒸着、エピタキシーおよび熱酸化等の化学処理法、および物理蒸着(PVD)およびキャスティング等の物理的処理法がある。ナノ電気機械システムの製造法は本発明のある実施形態で使用し得る(例えばクレイグヘッド(Craighead)、サイエンス(Science)、290:1532-36、2000参照)。
【0036】
本発明のある実施形態では、金属被覆多孔質シリコン基板240、340をマイクロ流動チャネル、ナノチャネルおよび/またはマイクロチャネル等の様々な液体で満たされたコンパートメントに接続する。装置300のこれらの部品を例えば半導体チップおよび/またはマイクロキャピラリーまたはマイクロ流動チップ等の形の単一ユニットに形成し得る。または、金属被覆多孔質シリコン基板240、340をシリコンウェーハから取り外し、装置300の他の部品に取り付けてもよい。シリコン、二酸化シリコン、窒化シリコン、ポリジメチルシロキサン(PDMS)、ポリメチルメタクリレート(PMMA)、プラスチック、ガラス、石英等の、この様なチップで用いる公知の任意の材料を開示された装置300に使用し得る。
【0037】
チップのバッチ式製造技術はコンピューターチップ製造業者および/またはマイクロキャピラリーチップ製造業者の分野で公知である。この様なチップはフォトリゾグラフィーおよびエッチング、レーザー切削、射出成形、キャスティング、分子ビームエピタキシー、デイップペンナノリゾグラフィー、化学蒸着(CVD)、電子ビームまたは焦点イオンビーム技術または印刷技術等の公知の方法で製造し得る。非限定例にはプラスチックまたはガラス等の流動性で工学的に透明な材料による通常の成形体;二酸化シリコンのフォトリゾグラフィーおよび乾式エッチング;ポリメチルメタクリレートレジストを用いるエレクトロンビームリソグラフィーによる二酸化シリコン基板上のアルミニウムマスク形成および反応性イオンエッチングが含まれる。ナノ電気化学システムの製造法は本発明のある実施形態で使用される(例えばクレイグヘッド(Craighead)、サイエンス(Science)、290:1531-36、2000参照)。マイクロ加工チップの様々な形は、例えばカリパーテクノロジース社(Caliper Technologies Inc.、マウンテンビュー(Mountain View、CA)およびACLARAバイオサイエンス社(Mountain View、CA)から市販されている。
【0038】
本発明のある実施形態では、装置300の一部または全部をガラス、シリコン、石英またはその他任意の工学的に透明な材料等で、ラマン分光に用いられる励起および放射周波数で電磁波に透明である様に選ぶ事が出来る。タンパク質、ペプチド、核酸、ヌクレオチド等の様々な生物分子にさらされる液体を満たしたコンパートメントでは、例えば表面を疎水性から親水性表面に変える、および/または分子の表面への吸着を減少させるため、この様な分子にさらされる表面を被覆して修飾する事が出来る。ガラス、シリコン、石英および/またはPDMS等の通常のチップ材料の表面修飾は公知である(例えば米国特許第6,263,286号)。この様な修飾には市販のキャピラリー被覆剤(スペルコ(Sperco)、ベラフォンテ(Bellafonte)、PA)、ポリエチレンオキサイドまたはアクリルアミド等の様々な官能基を有するシランによる被覆、またはその他公知の任意の被覆が含まれる。
【0039】
ラマン分光
ラマン検出器
本発明のある実施形態では、ラマン分光の任意の公知の方法で被分析試料を検出および/または同定する事が出来る。この様な実施形態において、ラマン活性基板240、340を少なことも1個のラマン検出ユニットと動作可能に結合させる事が出来る。ラマン分光による被分析試料の検出の様々な方法は公知である(例えば米国特許第6,002,471号;第6,040,191号;第6,149,868号;第6,173,677号;第6,313,914号参照)。表面増強ラマン分光(SERS)、表面増強共鳴ラマン分光(SERRS)、ハイパーラマン分光およびコヒーレント反ストークスラマン分光(CARS)の変法が開示されている。SERSおよびSERRSでは、ラマン検出の感度は銀、金、白金、銅またはアルミニウム表面等の粗い金属表面に吸着された分子で106倍以上に増加する。
【0040】
ラマン検出ユニット360の非限定例は米国特許第6、002、471号に開示されている。励起光線390は波長532nmの周波数重複Nd:YAGレーザー370または波長365nmの周波数重複Ti:サファイアレーザーで発生する。パルスレーザー光線390が使用されても良いし、連続レーザー光線390が使用されても良い。励起光線390は共焦点光学系と顕微鏡対物レンズを通過し、少なくとも1個の被分析試料を含むラマン活性基板240、340上に焦点を結ぶ。被分析試料からのラマン発光は顕微鏡対物レンズと共焦点光学系で集められ、スペクトル分解のためのモノクロメーターと組み合わされる。共焦点光学系には二色性繊維、バリアフィルター、共焦点ピンホール、レンズおよび鏡の組み合わせが含まれ、バックグラウンド信号を低減する。標準的な全視野光学系が、共焦点光学系として同様に使用する事が出来る。ラマン発光信号は信号を計数しデジタル化するためのコンピュータ160、395で仲介されるアバランシェ(なだれ)フォトダイオードでなるラマン検出器380で検出される。
【0041】
ラマン検出ユニット360の他の例は米国特許第5,306,403号に開示され、その一つはガリウム砒素光電子倍増管を装備した、単一フォトン係数モードで作動するモデル1403仕様二重格子分光光度計(RCAモデルC31034またはバールインダストリーズ(BurleIndustries)モデルC3103404)である。励起光源は514.5nmのアルゴンイオンレーザー370(スペクトラルフィジクス(SpectralPhysics)社製モデル166)、または641.7nmのクリプトンイオンレーザー370(イノバ(Innova)70、コヒーレント)である。
【0042】
その他の励起光源には337nmの窒素レーザー370(レーザーサイエンス社(Laser ScienceInc.)、325nmのヘリウム-カドミウムレーザー370(リンコノックス(Linconox);米国特許第6,174,677号)、発光ダイオード、ND:YLFレーザー370、および/または様々なイオンレーザー370、および/または染料レーザー370が含まれる。励起光線390はバンドパスフィルター(コリオン(Corion))でスペクトル的に純化され、6X対物レンズ(ニューポート(Newport)、モデルL6X)を用いてラマン活性基板240、3402上に焦点を結ぶ。対物レンズはホログラフビームスプリッター(カイザーオプティカルシステムズ社(Kaiser OpticalSystemsInc.)、モデルHB647-26N18)を用い、励起光線390と発光ラマン信号を直角の位置関係にすることにより、被分析試料の励起と、ラマン信号の収集という2つの目的で使用される。ホログラフィックノッチフィルター(カイザーオプティカルシステムズ社)をレイリー散乱光を減少させるために使用できる。その他のラマン検出器380には赤色増強増幅電荷対装置(RE-ICCD)検出システム(プリンストンインスツルーメント社(PrincetonInstrument、Inc.))を装備したISAHR-230分光光度計がある。フーリエ変換分光光度計(ミカエルソン(Michaelson)干渉計に基づく)、電荷注入装置、フォトダイオードアレー、INGaAS検出器、電子倍増CCD、増幅CCDおよび/またはフォトトランジスターアレー等のその他のタイプの検出器380も使用し得る。
【0043】
任意の適当な形式または形態のラマン分光または公知の関連する技術を被分析試料の検出に使用でき、それらには通常のラマン散乱、共鳴ラマン散乱、表面増強ラマン散乱、表面増強共鳴ラマン散乱、コヒーレント反ストークスラマン分光(CARS)、誘導ラマン散乱、逆ラマン分光、誘導利得ラマン散乱、ハイパーラマン散乱、分子光学レーザー試験装置(MOLE)、またはラマンマイクロプローブ、ラマン顕微鏡、共焦点ラマン顕微鏡分光、三次元または走査ラマン、ラマン飽和分光、時間分解共鳴ラマン、ラマンデカップリング分光またはUV-ラマン顕微鏡が含まれるが、これらに限定されない。
【0044】
ラマン標識
本発明のある実施形態には、少なくとも1個の被分析試料に標識を付着させ、ラマン検出ユニット360による測定を助長する方法が含まれる。ラマン分光に使用し得る標識の非限定例にはTRIT(テトラメチルローダミンイソチオール)、NBD(7-ニトロベンズ-オキサ-1、3-ジアゾール)、テキサスレッド染料、フタル酸、テレフタル酸、イソフタル酸、クレジルファストバイオレット、クレジルブルーバイオレット、ブリリアンクレジルブルー、パラアミノ安息香酸、エリスロシン、ビオチン、ジオキシゲニン、5-カルボキシ-4’、5’-ジクロロ-2’、7’-ジメトキシフルオレッセイン、5-カルボキシ-2’、4’、5’、7’-テトラクロロフルオレッセイン、5-カルボキシフルオレッセイン、5-カルボキシローダミン、6-カルボキシローダミン、6-カルボキシメチルアミノフタロシアニン、アゾメチン、シアニン、キサンチン、スクシニルフルオレッセイン、アミノアクリジン、カンタムドット、カーボンナノチューブおよびフレーレンが含まれる。これらの、またはそれ以外のラマン標識は市販されている(例えばモレキュラープローブ社(MolecularProbe Inc.、オイゲン(Eugen)、OR);シグマアルドリッチケミカル社(Sigma Aldrich ChemicalCo.、セントルイス(St.Louice)、MO)および/または公知の方法で合成できる。
【0045】
公知の如く、多環式芳香族化合物がラマン標識として機能し得る。本発明の具体的実施形態で使用し得るその他の標識にはサイアナイド、チオール、塩素、臭素、メチル、燐および硫黄がある。ラマン分光における標識の使用は公知である(例えば米国特許第5,306,403号および第6,174,677号)。当業者は、使用したラマン標識が識別可能なラマンスペクトルを生成し、異なったタイプの被分析試料と特異的に結合または関連していなければならないことを認識するであろう。
【0046】
標識は被分析試料に直接結合するか、または様々なリンカー化合物を経由して結合し得る。解除された方法で有用な架橋剤およびリンカー化合物は公知である。被分析試料等の他の分子と共有結合反応する様に設計された反応性基を含むラマン標識は市販されている(例えばモレキュラープローブ社(MolecularProbe、オイゲン(Eugen)、OR)。標識被分析試料の調製法は公知である(米国特許第4,962,037号;第5,405,747号;第6,136,543号;第6,210,869号)。
【0047】
コンピュータ
本発明のある実施形態では、装置100、300はコンピュータ160、395を包含し得る。実施形態はコンピュータ160、395のタイプに限定されない。コンピュータ160、395の例には情報を伝達するバス、および情報を処理する演算装置が含まれる。ある実施形態では、演算装置はインテル社(Intel Corp.、サンタクララ(SantaClara)、CA)が市販するペンティアム(Pentium(登録商標))IIファミリー、ペンティアムIIIファミリーおよびペンティアムIVファミリーを含むペンティアムファミリーの演算装置から選ばれるが、それに限定されない。本発明の別な実施形態では、演算装置はセレロン(Celeron(登録商標))、イタニウム(Itanium(登録商標))、ペンティアムジオン(Pentium Xeon(登録商標))演算装置(インテル社(IntelCorp.)、サンタクララ(SantaClara)、CA)である。本発明の様々な他の実施形態では、演算装置はインテルIA-32またはインテルIA-64等のインテル基本設計に基づくものである。または、他の演算装置も使用し得る。
【0048】
コンピュータ160、395はさらにランダムアクセスメモリー(RAM)その他のダイナミック記憶装置、読み取り専用メモリー(ROM)その他の静的記憶装置、および磁気ディスクまたは光学ディスク等のデータ記憶装置および関連するドライブを有する。コンピュータ160、395はまた表示装置(例えばインキュベート曲線または液晶表示装置)、英数字入力装置(例えばキーボード)、カーソル制御装置(例えばマウス、トラックボールまたはカーソル方向キー)、および通信装置(例えばモデム、ネットワークインターフェース、またはイーサネット、トークンリングその他のネットワークに接続するインターフェース装置)を有する。
【0049】
本発明の具体的な実施形態では、ラマン検出ユニット360はコンピュータ160、395と連結して作動し得る。検出ユニット360からのデータは演算装置と主記憶装置に記憶されたデータで処理されて良い。標準被分析試料の発光プロフィルに関するデータも主記憶装置またはROMに記憶されて良い。演算装置はラマン活性基板240、340中の被分析試料からの発光スペクトルを比較し、試料中の被分析試料のタイプを同定する。演算装置は検出ユニット360からのデータを解析し、様々な被分析試料の同一性および/または濃度を決定する。ある種の作業では別個に装備されたコンピュータ160、395を使用することが好ましい。従って、システムの形態は本発明の異なった実施形態では変化し得る。
【0050】
本明細書に記載のプロセスはプログラムされた演算装置の制御で実行されるが、本発明の別な実施形態ではプロセス全体または一部を例えばフィールドプログラマブルゲートアレー(FPGA)、TTLロジックまたは特定用途向け集積回路(ASIC)等の任意のプログラム可能またはハードコードロジックで実行し得る。さらに、開示された方法はプログラムされた汎用コンピュータ160、395部品および/または顧客仕様ハードウエア部品の任意の組み合わせで実行し得る。
【0051】
データ収集操作後、典型例ではデータをデータ解析操作に送る。解析操作を容易にするため、検出ユニット360で得たデータを典型例ではデジタルコンピューター160、395を用いて上記の様に解析する。具体的には、コンピュータ160、395は検出ユニット360からのデータを受け取り記憶すると同時に、収集したデータを解析し報告するために適当にプログラムされる。
【0052】
本発明のある実施形態では、専用のソフトウエアパッケージを検出ユニット360からのデータを得るために使用し得る。本発明のまた別な実施形態では、コンピュータ160、395と公開されたソフトウエアパッケージを用いてデータ解析を行う事も出来る。
【実施例1】
【0053】
ラマン活性基板の作成
多孔質ナノ結晶性シリコンの作成
【0054】
ナノ結晶性多孔質シリコン基板110、210作成用の方法と装置100の例を図1に示す。ナノ結晶性多孔質シリコンの製造法は公知である(例えば米国特許第6,017,773号)。ペトロワ-コッホら(Appl.Phys.Lett.、61:943、1992)が開示した様に、ナノ結晶性多孔質シリコン層を電気化学的に生成する事が出来る。具体的な実施例では、エッチングに先立ってシリコンを高濃度または低濃度でp-ドープまたはn-ドープし、多孔質シリコン基板110、210の特性を制御する。単結晶シリコンインゴットを公知のチョクラルスキー(Czochralski)法で作成する(例えばhttp://www.msil.ab.psiweb.com/english/msilhist-e.html)。単結晶シリコンウェーハに希薄HF/エタノール150中で陽極エッチング処理を施し、ナノ結晶性多孔質シリコン基板110、210を作成する。または、陽極エッチングを行わず、ウェーハ150をHFおよび硝酸溶液中で化学エッチングする。
【0055】
エッチング前にウェーハをポリメチルメタクリレートレジストまたは公知の任意のレジスト化合物で被覆する。ナノ結晶性多孔質シリコン基板110、210のパターンを標準のフォトリゾグラフィー技術で作成する。本発明の異なった実施形態では、ナノ結晶性多孔質シリコン基板110、210は円形、溝、チャネルまたはその他任意に選ばれた形である。ある実施形態では、1枚のシリコンウェーハ上に複数の多孔質基板を作成し、ラマン分析用の複数のサンプリングチャネルおよび/またはチャンバーとする事が出来る。各サンプリングチャネルおよび/またはチャンバーは少なくとも1台のラマン検出器380に連結して作動する。
【0056】
レジスト被覆とリゾグラフィー後、図1に示す様にテフロン(登録商標)の電気化学セル120中でエタノールおよび/または蒸留水の約15~50重量パーセントHF溶液150中にウェーハ110をさらす。本発明の別の実施形態では、レジスト被覆ウェーハ110全体をHF溶液150に浸漬する。また別の実施形態では、ウェーハ110を例えば合成ゴムワッシャーを用いて電気化学セル120の所定の場所に保持し、ウェーハ110の一部のみがHF溶液150にさらされる様にする(米国特許第6、322、895号)。いずれの場合もウェーハ110は定電流電源130の陽極に電気的に接続され、電気化学セルの陽極110となる。セル120の陰極140は白金電極である。ウェーハ110は暗所で、陽極酸化電流密度5~250ミリアンペア/cm2で5秒~30分間エッチングされるが、この条件は選ばれた多孔質度に依存する。本発明の具体例では、多孔質度10%、15%、20%、25%、30%、35%、40%、45%、50%、60%、70%、80%または90%が選ばれる。公知の様に、多孔質シリコン基板110、120を生成するのに必要な陽極酸化電流密度は、部分的には基板が低濃度または高濃度でp-またはn-ドープされているかという、使用したシリコン基板のタイプに依存する。
【0057】
本発明のさらに別の実施形態では、ナノ結晶性多孔質シリコン基板110、210を、公知のチップ製造技術を用いて様々な検出器、センサー、電極、その他の電気部品、機械的アクチュエータ等でなるMEMS装置に組み込む事が出来る。ある実施形態では、この様な製造プロセスを多孔質シリコン基板110、210の作成および/またはラマン感受性金属被覆の前後で用いる事が出来る。
【0058】
金属被覆
多孔質シリコン110、210を公知の技術(ロペス(Lopez)およびフォーチェット(Fauchet)、2000)を用いて陰極電気移動により金属被覆する事が出来る。本実施例の目的では金属被覆に銀が使用されるが、金または白金等の他の金属も使用し得る。多孔質シリコン表面110、210を洗浄し、ロペスおよびフォーチェットの方法に従って電気移動で銀を被覆する(アプライド フィジクスレターズ(Appl.Phys.Lett.)、75:3989、1999)。当業者は、公知の任意の多孔質シリコン基板110、210上に薄膜を生成する技術を、本発明の様々な実施形態で使用し得ることを理解できると思われる。
【実施例2】
【0059】
被分析試料のラマン検出
上記に開示された様に作成したラマン活性金属被覆基板240、340を、図3に示す被分析試料のラマン検出、同定および/または定量用の装置300中に組み込む。基板240、340を例えば入り口320および出口350チャネルに接続されたフロースルーセル330に組み込む。入り口チャネル320を、試料インジェクター310および/または反応チャンバー310等の少なくとも1台の他の装置310に接続する。被分析試料はフローセル330に入りラマン活性基板340を横切って流れ、ラマン検出ユニット360で検出される。検出ユニット360はラマン検出器380とレーザー370等の光源370を備える。レーザー370は励起光線390を放射し、被分析試料を活性化しラマン信号を放射する。ラマン信号は検出器380で検出される。本発明のある実施形態では、検出器380はコンピュータ395と接続されて作動し、試料中に存在する被分析試料に関するデータを処理、解析、記憶および/または転送する。
【0060】
本発明の実施例では、励起光線390はチタン-サファイアレーザー370((スペクトラフィジクス(SpectralPhysics)によるツナミ(Tsunami))、により近赤外波長(750~950nm)で照射されるか、ガリウムアルミニウムダイオードレーザー370(PI-ECLシリーズ、プロセスインスツルーメント(ProcessInstrument))で785nmまたは830nmで照射される。パルスレーザー光線が使用されても良いし、連続レーザー光線が使用されても良い。励起光線は二色性ミラー(ホログラフノッチフィルター、カイザーオプティカル(KeiserOptical)、または干渉フィルター、クロマ(Chroma)またはオメガオプティカル(OmegaOprical)で収集光ビームと同一直線配置で反射される。反射光線390は顕微鏡対物レンズ(ニコン(Nikon)LUシリーズ)を通り、標的被分析試料が存在するラマン活性基板240、340上へ焦点を結ぶ。被分析試料からのラマン散乱光は同じ顕微鏡対物レンズで集められ、二色性ミラーを通ってラマン検出器380へ達する。ラマン検出器380は焦点レンズ、スペクトログラフおよびアレー検出器でなる。焦点レンズによりラマン酸乱光がスペクトログラフのエントランススリットを通って焦点を結ぶ。スペクトログラフ(ローパーサイエンティフィック(RoperScientific)は光をその波長で散乱させる格子を備える。散乱した光はアレー検出器(バック照明深空乏CCDカメラ(ローパーサイエンティフィック(RoperScientific))上で映像を形成する。アレー検出器はコントローラ回路と接続し、この回路はコンピュータ160、395に接続され検出器380機能のデータを転送し制御する。
【0061】
本発明の様々な実施形態では、検出器ユニット360は様々な被分析試料を高い感度で、単一分子の検出および/または同定レベルで検出、同定および/または定量する事が出来る。本発明のある実施形態では、被分析試料はラマンラベルされた、またはラベルされていない単一ヌクレオチド分子である。他の実施形態では、識別可能なラマン標識でラベルされた、またはラベルされない少なくとも1個のオリゴヌクレオチドプローブが試料中の標的核酸とハイブリダイゼーションする事が出来る。標的核酸の存在は相補的なオリゴヌクレオチドプローブとのハイブリダイゼーションと、ラマン検出により図3の装置300を用いることにより検出される。また、アミノ酸、ペプチドおよび/またはタンパク質も開示された方法と装置300を用いて検出および/または同定されても良い。当業者は、この方法と装置300は検出される被分析試料のタイプに制限されず、標識の有無に拘わらず、ラマン検出で検出される任意の被分析試料を所有権を主張する主題の範囲で分析し得ることを実現できることが当業者であれば認識するであろう。
【0062】
本発明のある実施形態では、少なくとも1個の“捕捉”分子をラマン活性基板240、340に共有結合または非共有結合で付着させ、被分析試料の感度および/または特異性を増進する事が出来る。例えば、選ばれた標的核酸に特異的なオリゴヌクレオチドを公知の技術で基板240、340の金属表面に付着させる事が出来る(例えばオリゴヌクレオチドを共有結合で修飾して、金被覆基板240、340と結合し得るスルフヒドリル基を含ませる事が出来る)。また、タンパク質、ペプチドその他の化合物に特異的な抗体を基板240、340に付着させる事も出来る。標的被分析試料の存在は、相補の核酸配列とのハイブリダイゼーションが行われ得る条件下の試料に基板240、340に付着したオリゴヌクレオチドをさらすことによって検出され得、その後洗浄され、結合した被分析試料が検出される。本発明のまた別な実施形態では、試料中の少なくとも1個の被分析試料をラマン活性基板240、340にさらす前に識別可能なラマン標識でラベルし、結合した被分析試料の検出を容易にする。同様な方法を抗体-光源ペア、リガンド-受容体ペアまたは相互に選択的および/または特異的結合を示すその他公知の被分析試料で用いる事が出来る。酸、水、有機溶剤または界面活性剤による洗浄、化学処理および/またはエキソヌクレアーゼおよび/またはプロテアーゼ等の加水分解酵素処理等、様々な試薬で処理し、結合した被分析試料および/または捕捉分子を除去することにより、基板240、340をリサイクルし再使用し得る。
【0063】
本明細書に開示され、所有権が主張される全ての方法および装置は、本明細書の開示と照合して不適切な実験を行うことなく実施し使用することが可能である。所有権が主張される主題の概念、精神および範囲から逸脱せず、本明細書に記載された方法と装置の変更を行い得ることは、当業者にとって明らかである。具体的には、化学的および生理学的に関連する試薬を本明細書に記載の試薬と置き換え、同一または類似の結果が得られることは明らかである。当業者に自明のこの様な類似の置換および変更の全ては、所有権が主張される主題の精神、範囲および概念の範囲内であると見なされる。

「特表2005-524857およびWO2003106943より引用」

|
|

2009年6月11日 (木)

[知財](特許法)「拒絶査定不服審判における特則」

Q: 「拒絶査定不服審判における特則」は、第何条から第何条までに記述があるか。

A: 第百五十八条から第百六十四条

|
|

2009年6月10日 (水)

[Claims] フォトマスク加工及び半導体処理において使用する増感され化学的に増幅されたフォトレジスト

※プリアンブルに「, which」有り

【特許請求の範囲】
【請求項1】
レチクルを製造することにおいて有用で、G線、H線又はI線パターニング放射線に露光されるフォトレジストであって、キャスティング溶媒と、変性フェノール置換樹脂と、光化学増幅化合物と、アントラセン又はナフタレン或いはその派生物からなる群から選択される増感剤とを有するフォトレジスト。
【請求項2】
前記増感剤が、アントラセン、9-フェノキシメチルアントラセン、1,4-ジメトキシアントラセン、9-アントラセンメタノール、9,10-ジメチルアントラセン、ナフタレン及び2-ヒドロキシル-1,4-ナフトキノンからなる群から選択される、請求項1に記載のフォトレジスト。
【請求項3】
前記増感剤が、前記フォトレジストの固体の約0.1重量%~約2.0重量%で存在する、請求項1又は2に記載のフォトレジスト。
【請求項4】
前記増感剤が、変性フェノール置換樹脂に結合され、アントラセンクロマホレ、ナフタレンクロマホレ、又はその組み合わせを含む、請求項1に記載のフォトレジスト。
【請求項5】
前記増感剤が、光化学増幅化合物に結合され、アントラセンクロマホレ、ナフタレンクロマホレ、又はその組み合わせを有する、請求項1に記載のフォトレジスト。
【請求項6】
前記キャスティング溶媒が、プロピレングリコールモノメチルエーテルアセテート、エトキシエチルプロピオネート、エチルセルソルブアセテート、ジグリム及びその組み合わせからなる群から選択される、請求項1又は2に記載のフォトレジスト。
【請求項7】
前記変性フェノール置換樹脂が、変性フェノール系樹脂、変性ノボラック樹脂及びその組み合わせからなる群から選択される、請求項1又は2、4、又は5に記載のフォトレジスト。
【請求項8】
前記光化学増幅化合物が、オニウム塩メタルハライド複合体、トリフリック及びその派生物、トシレート及びその派生物、並びに、メシレート及びその派生物からなる群から選択される、請求項7に記載のフォトレジスト。
【請求項9】
前記オニウム塩メタルハライド複合体がアリールスルホニウム塩である、請求項8に記載のフォトレジスト。
【請求項10】
G線、H線或いはI線撮像系においてDUVフォトレジストの使用を可能にする方法であって、アントラセン又はナフタレン或いはその派生物からなる群から選択される増感剤が前記DUVフォトレジストに添加される、方法。
【請求項11】
前記増感剤が、アントラセン、9-フェノキシメチルアントラセン、1,4-ジメトキシアントラセン、9-アントラセンメタノール、9,10-ジメチルアントラセン、ナフタレン及び2-ヒドロキシル-1,4-ナフトキノンからなる群から選択される、請求項10に記載の方法。
【請求項12】
前記増感剤が、前記フォトレジストの固体の約0.1重量%~約2.0重量%の範囲の濃度で存在する、請求項10又は11に記載の方法。
【請求項13】
前記増感剤が、変性フェノール置換された樹脂に結合され、アントラセンクロマホレ、ナフタレンクロマホレ、又はその組み合わせを含む、請求項10に記載のフォトレジスト。
【請求項14】
前記増感剤が、光化学増幅化合物に結合され、アントラセンクロマホレ、ナフタレンクロマホレ、又はその組み合わせを含む、請求項10に記載のフォトレジスト。
【請求項15】
前記増感剤が前記DUVフォトレジストと混合される、請求項10又は11に記載の方法。
【請求項16】
前記増感剤が前記DUVフォトレジストに含まれるベース樹脂、或いは添加樹脂に関連付けられる又は結合される、請求項13又は14に記載の方法。
【請求項17】
前記増感剤が光化学増幅化合物に関連付けられる又は結合される、請求項13又は14に記載の方法。
【請求項18】
集中及び分離した特徴を含むパターン化されたフォトレジストにおける光近接効果を減少する方法であって、
約364nm~約436nmの範囲の波長を有する撮像放射線の使用を可能にするようDUVフォトレジストと機能する増感剤を前記DUVフォトレジストに添加するステップを備える、方法。
【請求項19】
前記増感剤がアントラセン又はナフタレン或いはその派生物からなる群から選択される、請求項18に記載の方法。
【請求項20】
前記増感剤が、アントラセン、9-フェノキシメチルアントラセン、1,4-ジメトキシアントラセン、9-アントラセンメタノール、9,10-ジメチルアントラセン、ナフタレン及び2-ヒドロキシル-1,4-ナフトキノンからなる群から選択される、請求項15に記載の方法。
【請求項21】
前記増感剤が、前記DUVフォトレジスト中のベース樹脂に関連付けられ又は結合され、アントラセンクロマホレ、ナフタレンクロマホレ、或いはその組み合わせを備える、請求項18に記載の方法。
【請求項22】
前記増感剤が、前記DUVフォトレジスト中の光化学増幅化合物に関連付けられ又は結合され、アントラセンクロマホレ、ナフタレンクロマホレ、或いはその組み合わせを備える、請求項18に記載の方法。
【請求項23】
前記DUVフォトレジストが、置換ポリヒドロキシスチレン又はその共重合体を備える、請求項18又は19、21、又は22に記載の方法。
【請求項24】
前記光化学増幅化合物が、オニウム塩メタルハライド複合体、トリフリック及びその派生物、トシレート及びその派生物、並びに、メシレート及びその派生物からなる群から選択される、請求項18に記載の方法。
【請求項25】
前記DUVフォトレジストが、オニウム塩メタルハライド複合体である光化学増幅化合物を備える、請求項24に記載の方法。
【請求項26】
前記オニウム塩メタルハライド複合体がアリールスルホニウム塩である、請求項25に記載の方法。
【請求項27】
前記パターン化されたフォトレジストがレチクルを加工するために使用される、請求項18又は19、21、又は22に記載の方法。
【請求項28】
前記パターン化されたフォトレジストが7を加工するために使用され、前記パターン化されたフォトレジストが半導体基板に特徴を加工するために使用される、請求項18又は19、21、又は22に記載の方法。
【請求項29】
前記パターン化されたフォトレジストが、基板にMEMS特徴を加工するために使用される、請求項18又は19、21、又は22に記載の方法。

We claim: 1. A photoresist useful in the manufacture of reticles, which photoresist is exposed to G-line, H-line, or I-line patterning radiation, said photoresist comprising a casting solvent; a modified, phenol substituted resin; a photochemical amplifying compound; and a sensitizer selected from the group consisting of anthracene or naphthalene, or a derivative thereof.

2. A photoresist in accordance with Claim 1, wherein said sensitizer is selected from the group consisting of: anthracene; 9-phenoxymethylanthracene ; 1,4- dimethoxyanthracene ; 9-anthracene methanol ; 9, 10-dimethyl anthracene ; naphthalene; and 2-hydroxyl-1, 4-naphthaquinone.

3. A photoresist in accordance with Claim 1 or Claim 2, wherein said sensitizer is present at about 0. 1% by weight to about 2.0 % by weight of the solids of said photoresist.

4. A photoresist in accordance with Claim 1, wherein said sensitizer is attached to said modified, phenol substituted resin and comprises an anthracene chromaphore, a naphthalene chromaphore, or a combination thereof.

5. A photoresist in accordance with Claim 1, wherein said sensitizer is attached to said photochemical amplifying compound and comprises an anthracene chromaphore, a naphthalene chromaphore, or a combination thereof.

6. A photoresist in accordance with Claim 1 or Claim 2, wherein said casting solvent is selected from the group consisting of propylene glycol monomethyl ether acetate, ethoxy ethyl propyionate, ethyl cellosolve acetate, diglyme, and combinations thereof.

7. A photoresist in accordance with Claim 1 or Claim 2, or Claim 4, or Claim 5, wherein said modified, phenol substituted resin is selected from the group consisting of modified phenolic resin, modified novolak resin, and combinations thereof.

8.'A photoresist in accordance with Claim 7, wherein said photochemical amplifying compound is selected from the group consisting of an onium salt metal halide complex; triflic and derivatives thereof ; tosylate and derivatives thereof ; and mesylate and derivatives thereof.

9. A photoresist in accordance with Claim 8, wherein said onium salt metal halide complex is an aryl sulfonium salt.

10 A method of enabling the use of a DUV photoresist in a G-line, H-line, or I- line imaging system, wherein a sensitizer selected from the group consisting of anthracene or naphthalene, or a derivative thereof is added to said DUV photoresist.

11 A method in accordance with Claim 10 wherein said sensitizer is selected from the group consisting of : anthracene ; 9-phenoxymethylanthracene ; 1, 4-dimethoxyanthracene ; 9-anthracene methanol; 9, 10-dimethyl anthracene ; naphthalene; and 2-hydroxyl-1, 4-naphthaquinone.

12. A method in accordance with Claim 10 or Claim 11, wherein said sensitizer is present at a concentration ranging from about 0.1 % by weight to about 2.0 % by weight of the solids of said photoresist.

13. A photoresist in accordance with Claim 10, wherein said sensitizer is attached to said modified, phenol substituted resin and comprises an anthracene chromaphore, a naphthalene chromaphore, or a combination thereof.

14. A photoresist in accordance with Claim 10, wherein said sensitizer is attached to said photochemical amplifying compound and comprises an anthracene chromaphore, a naphthalene chromaphore, or a combination thereof.

15. A method in accordance with Claim 10 or Claim 11, wherein said sensitizer is mixed into said DUV photoresist.

16. A method in accordance with Claim 13 or Claim 14, wherein said sensitizer is associated with or bonded to a base resin or additive resin which is included in said DUV photoresist.

17. A method in accordance with Claim 13 or Claim 14, wherein said sensitizer is associated with or bonded to a photochemically amplifying compound.

18. A method of reducing optical proximity effects in a patterned photoresist which includes dense and isolated features, comprising: adding to a DUV photoresist a sensitizer which functions with said DW photoresist to enable the use of imaging radiation having a wavelength ranging from about 364 nm to about 436nm.

19. A method in accordance with Claim 18, wherein said sensitizer is selected from the group consisting of anthracene or naphthalene, or a derivative thereof.

20. A method in accordance with Claim 15, wherein said sensitizer is selected from the group consisting of: anthracene ; 9-phenoxymethylanthracene ; 1, 4-dimethoxyanthracene ; 9-anthracene methanol; 9,10-dimethyl anthracene; naphthalene; and 2-hydroxyl-1, 4-naphthaquinone.

21. A method in accordance with Claim 18, wherein said sensitizer is associated with or attached to a base resin in said DUV photoresist, and comprises an anthracene chromaphore, a naphthalene chromaphore, or a combination thereof.

22. A method in accordance with Claim 18, wherein said sensitizer is associated with or attached to a photochemical amplifying compound in said DUV photoresist, and comprises an anthracene chromaphore, a naphthalene chromaphore, or a combination thereof.

23. A method in accordance with Claim 18 or Claim 19, or Claim 21, or Claim 22, wherein said DUV photoresist comprises a substituted polyhydroxy styrene or a copolymer thereof.

24. A method in accordance with Claim 18, wherein said photochemical amplifying compound is selected from the group consisting of an onium salt metal halide complex; triflic and derivatives thereof; tosylate and derivatives thereof; and mesylate and derivatives thereof.

25 A method in accordance with Claim 24, wherein said DUV photoresist comprises a photochemical amplifying compound which is an onium salt metal halide complex.

26. A method in accordance with Claim 25, wherein said onium salt metal halide complex is an aryl sulfonium salt.

27. A method in accordance with Claim 18, or Claim 19, or Claim 21 or Claim 22, wherein said patterned photoresist is used to fabricate a reticle.

28. A method in accordance with Claim 18, or Claim 19, or Claim 21, or Claim 22, wherein said patterned photoresist is used to fabricate 7, wherein said patterned photoresist is used to fabricate a feature on a. semiconductor substrate.

29. A method in accordance with in accordance with Claim 18, or Claim 19, or Claim 21, or Claim 22, wherein said patterned photoresist is used to fabricate a MEMS feature on a substrate.

「特表2005-527003およびWO2004001797より引用」

|
|

フォトマスク加工及び半導体処理において使用する増感され化学的に増幅されたフォトレジスト

【発明の詳細な説明】
【発明の背景】
【0001】
[0003]1.発明の分野
[0004]本発明は、フォトマスク加工及び半導体製造において有用なフォトレジスト組成物に関わる。特に、フォトレジスト組成物は、集中した及び分離した特徴の組み合わせを表わすフォトマスクの製造を可能にする。フォトマスク加工に使用されるものと同じフォトレジストが半導体及び超小型電子機械系(MEM)処理に適用される。
【0002】
[0005]2.背景技術の簡単な説明
[0006]フォトレジスト組成物は、半導体装置構造の加工等において小型電子構成要素を作るため、小型機械系を作るため、及び、微生物学的構造を作るためにマイクロリソグラフィ処理において使用される。小型装置構造パターンは、パターン化されたマスキング層を通ってブランケット処理で実現され得るタイムエコノミーにより基板への直接的な書き込みよりも半導体又は他の基板の上にあるパターン化されたマスキング層からパターンを転写することで典型的には作成される。マイクロ装置処理に関して、パターン化されたマスキング層は、半導体装置構造又はパターン化されるべき他の基板の表面にあるパターン化されたフォトレジスト層又はパターン化された「硬質」マスキング層(典型的には無機材料或いは高温有機材料)でもよい。パターン化されたマスキング層は、フォトマスク又はレチクルとしてしばしば呼ばれる別のマスクを用いて典型的には生成される。レチクルは、典型的にはガラス又は石英板上に堆積される薄層の金属含有層(例えば、クロム含有、モリブデン含有、又は、タングステン含有材料等)である。レチクルは、半導体構造又は他の基板の上にあるマスキング層上に再生成されるべき個々の装置構造パターンの「ハードコピー」を含むようにパターン化される。
【0003】
[0007]レチクルは、レチクル上にパターンを書き込む方法によって幾つかの異なる技法で作成され得る。現在の半導体構造の寸法要件により、書き込み方法は一般的にレーザ又は電子ビームを用いて行われる。レチクルを生成する典型的なプロセスは、ガラス又は石英板を設ける、ガラス又は石英表面にクロム含有層を堆積する、クロム含有層上に反射防止膜(ARC)を堆積する、ARC層上にフォトレジスト層を適用する、所望のパターンを生成するためにフォトレジスト層に直接的に書き込む、フォトレジスト層にパターンを現像する、クロム層にパターンをエッチングする、残留するフォトレジスト層を除去することを含む。書き込み放射線によって接触されるフォトレジスト層の領域が現像中により除去され易くなると、フォトレジストはポジワーキングフォトレジストと呼ばれる。書き込み放射線によって接触されるフォトレジスト層の領域が現像中に除去し難くなると、フォトレジストはネガワーキングフォトレジストと呼ばれる。最新のレチクル製造材料は、例えば、クロミウム、酸化クロム、オキシナイトライドクロム、モリブデン、ケイ化モリブデン、及びケイ化モリブデンタングステンから選択される材料の層の組み合わせをしばしば含む。
【0004】
[0008]前述した通り、レチクル又はフォトマスクは、下にあるフォトレジストにパターンを転写するために使用され、このときレチクルの開領域を通ってフォトレジストの表面上に照射されるブランケット放射線にレチクルは露光される。次に、フォトレジストが現像され、パターン化されたフォトレジストは典型的にはプラズマドライエッチング処理を用いてパターンを下にある半導体構造に転写するために使用される。
【0005】
[0009]半導体基板に対する形状要件が小さくなり、半導体装置及びMEMSに対する新しい適用法が単一のチップ上で論理及び記憶特徴の両方を混在させることを必要とすると、レチクル製造及び半導体チップ製造の両方に関して新しい問題が生ずる。メモリ装置が密(集中)の特徴を有する傾向にあり、論理装置はまばら(分離)の特徴を有する傾向にある。その結果、レチクルの製造に使用されるフォトレジストのパターンニング中、及び、半導体基板へのパターン転写に使用されるフォトレジストのパターンニング中に近接効果が観察される。
【0006】
[0010]例えば、500nm(0.50μm)~2000nm(2.0μm)の範囲の特徴限界寸法(CD)を有するようパターン化されたレチクルの加工において、標準ノボラックベースのG線、H線、又はI線フォトレジストの撮像中に光近接効果が観察された。特に、日本国の東京にある東京応化工業株式会社から市販されているiP3600フォトレジストのような標準的なI線フォトレジスト、又は日本国の大阪にある住友から市販されているPF188Aフォトレジストに関して、近接効果に起因する約20nm~約40nmのCDエラーがレチクルにパターンを転写するために使用されるべきパターン化されたフォトレジストにおいて観察され、パターン化されたレチクルのクロムにおいて観察される。
【0007】
[0011]図1は、レチクルの加工において使用される典型的な初期構造100を示す。この初期構造は、レチクルを加工する本方法の開発中に試験片の準備で一般的に使用されるものである。初期構造100は、上から下に、ノボラックベースのフォトレジストiX1100P(Sommerville, NewJerseyのClarientCorp.から市販されている)の約5000Aの厚さの層108と、有機ARC、オキシナイトライドクロムの約500Aの厚さの層106と、本質的にはクロムであるマスク材料の約200Aの厚さの層104と、酸化珪素含有基板102とを含む積層体である。
【0008】
[0012]レチクル試験片の初期加工において観察された相当の大きさの20nm~40nmのCDエラーに鑑みて、連続的な書き込みレーザツールのパラメータを全て検査してどのパラメータがCDエラーの原因であるかを判断するために相当の努力が費やされた。その仕事の詳細は、レーザツールのパラメータがCDエラーの原因でないと判断されたためここでは詳述しない。CDエラーは、撮像及び現像中のフォトレジスト材料の挙動の結果として生じたものであることが判明した。
【0009】
[0013]図2Aは、図1に示すフォトレジスト層108の概略的な平面図200を示し、第1のパターン、特にバーパターン202がフォトレジスト層108の上部表面201に書き込まれる。バー線203と204の間の距離d1は、約2000nm(約2.0μm)であり、可能な限り厳しく制御されるCDを表す。各バー203及び204の厚さは約2.0μmである。距離d2は、バーパターン202の長さを表し、約5700μmである。
【0010】
[0014]図2Bは、図1に示すフォトレジスト層108の概略的な平面図220を示し、第2のパターン、特に、ステップパターン222がフォトレジスト層108の上部表面221に書き込まれる。ステップパターン222の各半分の間の距離d1は、約2.0μmであり、CDを表す。d2は、約5,700μmであり、各ステップの長さd3は約317μmであるが、上ステップ224だけは約2×317μmである。ステップパターン222の各端の端ステップ226の高さ(厚さ)d4は約6.5μmであり、中央ステップ224の高さd5は約512μmである。
【0011】
[0015]フォトレジストはポジフォトレジストであるため、フォトレジストを放射線に露光し、放射線によって生成されるパターンを現像して照射領域におけるフォトレジストを除去することで取り除かれた空間が発生される。図2Aを参照するに、バー203及び204は、約270nmの半強度ビーム径(スポットサイズ)の連続的な波レーザを用いる直接的な書き込みによって照射される。図2Bを参照するに、ステップパターン222の各半分は、同じ連続的なレーザを用いて直接的な書き込みによって照射され、このときレーザは照射されたパターンを発生させるためにフォトレジスト108の表面にわたって走査される。初期構造の表面にパターンを書き込んだ後、フォトレジスト層108におけるパターンは現像され、下にあるARC層106及びクロム層104を通って転写され、石英基板102の上部表面103上にクロムパターン(図示せず)が発生される。
【0012】
[0016]図3Aは、図2Aに示す「X」方向に移動される距離の関数として、石英基板108(図2Aによる)の上部表面上に発生されるクロム線の距離d1に対する平均CDを示す。CDにおけるばらつきは、X=0μmにおける約1753nmからX=2700μmにおける約1746nm、更には、X=5400μmにおける約1754nmまで変動する。CDにおける差は、クロム線の全長にわたって約7~8nmだけである。
【0013】
[0017]図3Bは、図2Bに示す「X」方向に移動される距離の関数として、石英基板108(図2Bによる)の上部表面上に発生されるクロム線の距離d4に対する平均CDを示す。CDにおけるばらつきは、X=0μmにおける約1780nmからX=2700μmにおける約1758nm、更には、X=5400μmにおける約1782nmまで変動する。CDにおける差は、クロム線の全長にわたって23nmである。
【0014】
[0018]図2A及び図2Bに例示される構造に対して得られる線のCD範囲における差は、線に隣接して放射線に露光されるフォトレジスト108の面積の大きさの差から結果として生ずる近接効果によるものである。これら近接効果は、フォトレジストローディング効果としばしば呼ばれる。
【0015】
[0019]特徴の一部分が集中し、別の部分が分離しているレチクルの加工をより可能にするため、フォトレジストローディングによるパターン化されたレチクルにわたって観察されるCDにおける変化を減少させることが非常に望ましいことは明らかであろう。
[0020]
【発明の概要】
【0016】
[0021]フォトレジスト組成物及びレチクルの加工においてフォトレジストを用いる方法は、パターン化されたレチクルの表面にわたってレチクル特徴の限界寸法におけるばらつきを減少するよう設計され、このとき限界寸法におけるばらつきは局所化されたレジストローディングによるものである。
【0017】
[0022]特に、レチクルの製造において有用なフォトレジストを開発し、フォトレジストは、G線、H線、I線パターニング放射線に露光される。フォトレジストは、キャスティング溶媒と、変性フェノール置換樹脂と、光化学増幅化合物と、アントラセン或いはその派生物、ナフタレン或いはその派生物、及びこれら材料の混合物からなる群から選択される増感剤とを含む。制限的ではなく例として、増感剤は典型的にはアントラセン、9-フェノキシメチルアントラセン、1,4-ジメトキシアントラセン、9-アントラセンメタノール、9,10-ジメチルアントラセン、ナフタレン及び2-ヒドロキシル-1,4-ナフトキノンからなる群から選択される。キャスティング溶媒は、プロピレングリコールモノメチルエーテルアセテート(PMA、PGMEA)、エトキシエチルプロピオネート、エチルセルソルブアセテート、ジグリム及びその組み合わせのような材料から選択される。
【0018】
[0023]フォトレジストのベース樹脂は、変性フェノール系樹脂、変性ノボラック樹脂、及びその組み合わせから選択される。一つの特に有利なベース樹脂は、置換ポリヒドロキシスチレン又はその共重合体である。
【0019】
[0024]光化学増幅化合物(PAC)は、制限的でなく例として、オニウム塩メタルハライド複合体、トリフリック酸及びその派生物、トシレート及びその様々な派生物、及びメシレート及びその様々な派生物等技術において公知のものである。しばしば使用されるPACの一つは、アニールスルホニウム塩である。
【0020】
[0025]上述の増感剤の使用により、G線=436nm、H線=405nm、及びI線=364nmの放射線波長で光活性化合物(PAC)から重合体のベース材料への効率的なエネルギー移動が可能となる。この効率的なエネルギー移動は、レジストを撮像するためにより少ないパワーを必要とすることを意味し、レジストへのパターンの直接的な書き込みは、連続的な書き込みレーザー(制限的でなく例として)によってより急速に行われる。より重要なことは、フォトレジストの(撮像の)照射中に生成される副生成物がフォトレジストをパターン化するために使用される現像剤と相互作用しないため、同じレチクル上に集中した及び分離した特徴の両方が存在する場合に起こる光近接効果が減少される。したがって、集中した及び分離した特徴の両方が存在する場合にレチクル表面上でのCDにおける変化を減少することが可能となる。増感剤は、PMA、変性フェノール置換樹脂、及び化学増幅液のようなキャスティング溶媒の組み合わせに単に混合されてもよく、或いは、増感剤はフェノール置換ベース樹脂又は化学増幅液に結合されてもよい。
【0021】
[0026]レチクル加工に適用されるのと同じコンセプトが半導体基板への上述のフォトレジストのパターニングにも適用される。正しい増感剤の添加により、例えば、I線撮像系がより制御された特徴の大きさを生成するために使用され、I線撮像系の能力をより小さい寸法特徴、例えば、約0.3μmの特徴の大きさまで広げることができる。G線及びH線撮像系において、撮像系の機能性もより小さい寸法特徴にまで広げることができる。
[0034]
【実施形態の詳細な説明】
【0022】
[0035]詳細な説明の前書きとして、本明細書及び添付の特許請求の範囲において、特に明確に記載しない限り構成要素は単数形及び複数形も含むことに注意する。
【0023】
[0036]電子産業における最近の進歩により、メモリ及び論理装置を同じチップ上に配置する必要性が生じた。このような装置を混在させることは、メモリ装置が集中配置特徴を利用する傾向にある一方で論理回路が分離特徴をしばしば利用するため、処理を困難にする。半導体基板にパターンを転写するために使用されるフォトレジストを撮像するために使用されるフォトマスク(レチクル)の処理は近接効果に影響を及ぼされる。更に、撮像されたフォトレジスト自体が同じ近接効果によって影響を及ぼされる。レチクルにパターンを生成するために使用されるフォトレジストにパターンを書き込むために使用されるツールの製造業者は、レチクルの表面上での特徴限界寸法(CD)におけるばらつきを減少させるために書き込みツールに悪影響を及ぼす変数を注意深く調査した。特に、オレゴン州ヒルスボロにあるETECシステムズ社は、レチクルにパターンを転写するために使用されるフォトレジストにパターンを書き込むために連続的な波直接書き込みレーザを用いる自身の商標ALTA3700ツールを調査した。相当量の調査の後、問題は書き込みツールではなく、フォトレジストを現像するために使用される現像器と、フォトレジスト上にパターンを書き込む際に生成される副生成化合物との間で行われる反応によるものであることがわかった。
【0024】
[0037]レチクルをパターン化するために使用されるフォトレジストが基本的には基板にパターンを転写するために半導体基板の表面に使用されるものと同じフォトレジストであるため、当業者は半導体処理に関してなぜこの問題が生じなかったのか疑問に思うであろう。その理由は、レチクル上の特徴が半導体基板上の特徴よりも典型的には約4倍大きいからである。180nm~250nmの範囲の半導体装置の特徴を得るために半導体処理業界は、フォトレジストの遠紫外線(DUV)248nm撮像まで行った。しかしながら、レチクル製造業者は、レチクルをパターン化するために使用されるフォトレジストを撮像するためにG線(=436nm)、H線(=405nm)、又はI線(=364nm)放射線をまだ使用することが可能である。
【0025】
[0038]ETECシステムズ社は、DUV撮像系ならびにI線撮像系を提供し、出願人は、現像器がフォトレジスト撮像中に生成される副生成物と反応する問題は、G線、H線、及びI線撮像系と組み合わせて使用されるノボラックベースの樹脂系に固有であることを発見した。出願人は、DUV撮像系と組み合わせて使用されるフォトレジストシステムに関しては同じ現像問題を観察していない。特に、化学的に増幅されたDUVフォトレジスト、DX1100が、例えば、I線パターニングに使用されるノボラックベースの樹脂で観察された厳密なフォトレジストローディング問題を示すことは観察されていない。しかしながら、DX1100フォトレジストは、基本的には、プロピレングリコールモノメチルエーテルアセテート(PMA、PGMEA,又は1-メトキシ-2-プロピルアセテート)キャスティング溶媒、変性フェノールポリマー、及び、オニウム塩メタルハライド複合体を化学増幅液として含む。このフォトレジストは、I線フォトレジストを撮像するために使用される放射線波長では実施されない。
【0026】
[0039]出願人は、DX1100DUVフォトレジストに増感剤を添加することでI線フォトレジストを撮像するために使用される364nm放射線に露光されるとき上記フォトレジストを実施することを可能にする。特に、アントラセンメタノール、アントラセン、又は、ジフェニルマライミドタイプの化合物のような増感剤がDUVフォトレジストに添加される。出願人の新しいフォトレジストは、次の固体を含む:約75重量%~約85重量%のプロピレングリコールモノメチルエーテルアセテートと、約20重量%~約30重量%の変性フェノール系樹脂と、約0.1重量%~約2重量%のアリールスルホニウム塩のような光活性剤と、約0.1重量%~約2重量%のアントラセン又はナフタレン或いはその派生物の形態にある増感剤とを含む。このような増感剤は、制限的でなく例として、アントラセン、9-フェノキシメチルアントラセン、1,4-ジメトキシアントラセン、9-アントラセンメタノール、9,10-ジメチルアントラセン、ナフタレン及び2-ヒドロキシル-1,4-ナフトキノンを含む。固体は典型的にはPGMEAインベンタ或いは等価溶媒に溶解される。一つの好ましい実施形態では、固体は13.4重量%のプロピレングリコールモノメチルエーテルアセテートと、約2.6重量%の変性フェノール系樹脂と、約2.0重量%のアリールスルホニウム塩と、約2.0重量%の上記に列挙した化合物から選択される増感剤とである。増感剤は、単にDUVフォトレジストと混合され、混合物は基板表面にスピンコーティングで適用される前にミリポア・フィルタでフィルタ処理される。増感剤がフォトレジストのベースポリマーと関連付けられる或いは結合される、又は、フォトレジストのPACと関連付けられる或いは結合されることが重要である。特に、当業者はアントラセン又はナフタレン或いはアントラセン/ナフタレンクロマホレが結合された状態でベースポリマー又はPACのいずれかを合成することができる。
【0027】
[0040]DUVレジスト製造業者によって現在推奨されている手順と同じものが基板へのフォトレジストの適用に使用され得る。フォトレジストの適用後、フォトレジストは、約7分から約15分の範囲の時間にわたって約110℃で適用後焼成(PAB)を受ける。フォトレジストの適用及び直接書き込み連続波UVレーザを用いる商標ALTA3700撮像系を使用するフォトレジストの撮像後、フォトレジストは、AZ300MIF現像器(ニュージャージー州ソマービル市にあるClariant社より市販されている)でスピン/スプレー処理を用いて現像される。この現像器は、2.38重量%溶液のテトラメチルアンモニア水酸化物(TMAH)である。この現像時間は、約60秒である。現像の施行の停止と同時に、CO2散布された又はCO2再イオン化された水が最初の15秒間約300rpmで、残りの15秒間1500rpmで約30秒間施行される。更に、スプレー-パドルを介してすすぎが開始され、続いて、2000~2500rpmで約90-120秒間スピン乾燥ステップが行われる。
【0028】
[0041]フォトレジスト中のパターンがドライエッチング処理を用いて下にあるフォトマスク構造に転写される。基板は、平均対目標偏差を減少する状況の下高密度プラズマでエッチングされ、好ましいCD均一性が維持される。プラズマエッチングは、デスカム/有機ARC(BARC)除去、オキシナイトライドクロム(無機ARC)/クロムエッチング、及びオーバーエッチングの3段階プロセスを用いて誘導結合プラズマ(ICP)エッチングツールで実施される。
【0029】
[0042]アプライドマテリアルズ社の(カリフォルニア州サンタクララ市にあるアプライドマテリアルズ社より市販されている)TETRA(R)DPSTMエッチングシステムのようなプラズマエッチングシステムが優れた結果を提供するために使用されてもよい。プラズマ生成及び基板バイアスに対して別々の電力の印加を可能にするプラズマ処理システムは、一般的にデカップルプラズマ源(DPS)と呼ばれる。基板バイアスは、プラズマから基板表面の方向にイオン及び他の高エネルギー種を引き付けるために使用され、それにより異方性エッチングが可能となる。プラズマ生成用の電力と基板バイアス用の電力の別々の印加により、プラズマ密度と基板の表面で生じる引力(DC電圧)を別々に制御することが可能となる。
【0030】
[0043]3段階エッチング工程のデスカム/ARC(BARC)除去段階は、現像後に開領域に残留する全ての残留フォトレジストを除去し、開領域から有機ARC(BARC)層を除去する。これは、酸素プラズマを用いて達成される。DCバイアスは、レチクル表面に対して垂直に高運動エネルギー及び方向性でレチクル表面に衝突するようプラズマからの酸素イオンを加速するためにデスカム/BARC除去段階中に基板に印加される。これには、酸素ガスが用いられ、低電極(レチクル板が上におかれる)だけに電力を供給し、プラズマとレチクルとの間に高DCバイアスを有した状態で容量結合プラズマを生成することで達成される。典型的には、このプロセスは、約3mトル~約8mトルの圧力でプロセスチャンバにおいて行われる。酸素のプラズマ源ガスが約17sccmの流速でプロセスチャンバに供給される。13.56MHZの周波数で約125WのRF電力が低電極(陰極)に印加される。これにより、約-310~-410ボルトのDCバイアスがフォトマスク基板表面に与えられる一方で、酸素プラズマがレチクル表面に供給される。レチクル板の温度は25℃の範囲であり、チャンバ壁の温度は40℃の範囲にある。デスカム/ARC(BARC)除去時間は約30秒である。デスカム/ARC(BARC)除去プロセスによるフォトレジストロスは約750Aである。
【0031】
[0044]オキシナイトライドクロム(無機ARC)/クロムマスク層エッチングは、塩素-酸素-ヘリウムガス混合物から生成されるプラズマを用いて行われる。塩素:酸素:ヘリウムガス混合物の分子比は約50:10:22である。合計ガス流速は約82sccmである。ICPコイルは、高密度プラズマを生成するために2MHZで約60Wに動力を供給される。低電極は、基板上で約-50VのDC電圧を生成するために約13.56MHZで約5Wに電力を供給される。レチクル板の温度は約25℃であり、プロセスチャンバの壁温度は約40℃である。エッチングのエンドポイントは、光学反射率によって検出され、約200秒で起こる。典型的には、より高い酸素濃度及びより低い圧力は、より高い平均対目標偏差及びより低い選択性を生じ、より好ましいCD均一性を支持する。当業者は、彼/彼女の特定の装置に対するプロセスを最適化することができる。
【0032】
[0045]典型的には、クロム層は、全ての開領域から残留クロムを取り除くためにエンドポイントを超えてオーバーエッチングされる。一般的に、オーバーエッチング段階は、上述のクロムエッチングプロセスの延長である。オーバーエッチング段階が長いと、平均対目標偏差が結果として高くなる。クロムスポット欠陥密度は、オーバーエッチングの長さによって影響を及ぼされ、オーバーエッチングプロセスが長いと欠陥密度が低くなる。
【0033】
[0046]クロム層エッチングの完成後、クロム層の表面から全ての残留する汚染物質を除去するために剥離及び洗浄プロセスが実施される。使用される剥離用化学物質は、硫黄過酸化物であり、これは約75℃に加熱され、基板板の表面に塗布される。硫黄過酸化物による処理の後、基板板はCO2再イオン化された又はCO2散布された脱イオン化された水ですすがれる。剥離後、基板板は、工業規格70:30H2SO4/H2O2溶液で酸洗浄され、続いて別の脱イオン化された水ですすがれる。剥離ステップは、カリフォルニア州サンタクララ市のSTEAG-HAMMATECH(R)から市販されているSteag ASC500ウェット化学処理ステーションで実施される。
【0034】
[0047]例1-比較例
[0048]以下の記載は制限的ではなく例として、I線フォトレジスト系に関するものである。
【0035】
[0049]前述したように、図2Aは、図1に示すフォトレジスト層108の概略的な平面図200を示し、第1のパターン、特にバーパターン202がフォトレジスト層108の上部表面201に書き込まれる。バー線203と204の間の距離d1は、約2000nm(約2.0μm)であり、可能な限り厳しく制御されるCDを表す。各バー203及び204の厚さは約2.0μmである。距離d2は、バーパターン202の長さを表し、約5700μmである。
【0036】
[0050]図2Bは、図1に示すフォトレジスト層108の概略的な平面図220を示し、第2のパターン、特に、ステップパターン222がフォトレジスト層108の上部表面221に書き込まれる。ステップパターン222の各半分の間の距離d1は、約2.0μmであり、CDを表す。距離d2は、約5,700μmであり、各ステップの長さd3は、約317μmであるが上ステップ224だけは約2×317μmである。ステップパターン222の各端の端ステップ226の高さ(厚さ)d4は約6.5μmであり、中央ステップ224の高さd5は約512μmである。
【0037】
[0051]フォトレジストはポジフォトレジストであるため、フォトレジストを放射線に露光し、放射線によって生成されるパターンを現像して照射領域におけるフォトレジストを除去することで取り除かれた空間が発生される。図2Aを参照するに、バー203及び204は、約270nmの半強度ビーム径(スポットサイズ)の連続的な波レーザを用いる直接的な書き込みによって照射される。図2Bを参照するに、ステップパターン222の各半分は、同じ連続的なレーザを用いて直接的な書き込みによって照射され、このときレーザは照射されたパターンを発生させるためにフォトレジスト108の表面にわたって走査される。初期構造の表面にパターンを書き込んだ後、フォトレジスト層108におけるパターンは現像され、下にあるARC層106及びクロム層104を通って転写され、石英基板102の上部表面103上にクロムパターン(図示せず)が発生される。
【0038】
[0052]図3Aは、図2Aに示す「X」方向に移動される距離の関数として、石英基板108(図2Aによる)の上部表面上に生成されるクロム線の距離d1に対する平均CDを例示するグラフ300を示す。平均CDは、図3Aの軸302に示され、「X」方向に移動される距離は軸304に示される。曲線306によって例示されるように、CDのばらつきは、X=0μmにおける約1753nmからX=2700μmにおける約1746nm、更には、X=5400μmにおける約1754nmまで変動する。CDにおける差は、クロム線の全長にわたって約7~8nmだけである。
【0039】
[0053]図3Bは、図2Bに示す「X」方向に移動される距離の関数として、石英基板108(図2Bによる)の上部表面上に生成されるクロム線の距離d4に対する平均CDを例示するグラフ302を示す。平均CDは、図3Bの軸322に示され、「X」方向に移動される距離は軸324に示される。曲線326によって例示されるように、CDのばらつきは、X=0μmにおける約1780nmからX=2700μmにおける約1758nm、更には、X=5400μmにおける約1782nmまで変動する。CDにおける差は、クロム線の全長にわたって23nmである。一連の同様の例は、約20nm~約40nmの範囲にあるCDにおける差を示す。
【0040】
[0054]例2:
[0055]図4は、図2Bに示す「X」方向に移動される距離の関数として、石英基板108(図2Bによる)の上部表面に生成されるクロム線の平均CDを例示するグラフ400を示す。しかしながら、試験片を加工するために使用されるフォトレジストは、以前I線撮像に使用されたジアゾキノン増感されたノボラックベースのフォトレジストではなく、出願人の9-アントラセンメタノール増感剤が添加されたDX1100DUVフォトレジストである。特に、図4は、平均CDを軸402に示し、「X」方向に移動される距離を軸404に示す。曲線406に例示するように、CDのばらつきは、X=0μmにおける約1755nmからX=2700μmにおける約1746nm、更には、X=5400μmにおける約1755nmまで変動する。CDにおける差は、クロム線の全長にわたって約9nmだけである。
【0041】
[0056]このCDにおけるばらつきは、図2Aを参照して説明したとおり、線202を定めるために2.0μmの厚さのバー202及び203だけを使用したときに得られたCDにおけるばらつきと非常に類似する。これは、ノボラックベースのフォトレジストではなく本発明によるフォトレジストがI線パターニングに使用されるとき、ノボラックベースのフォトレジストで生ずるフォトレジストローディング効果を回避することが可能であることを明確に示唆している。前述の通りに撮像され現像される本発明のフォトレジストは、最小限のCDエラーで集中した及び分離した特徴の両方を含むレチクルを生成することを当業者に可能にさせる。
【0042】
[0057]更に、本開示を読む当業者は、半導体基板をパターン化するためにジアゾキノンノボラックベースのフォトレジストではなく本発明のフォトレジストを用いることでより小さい特徴の生成までG線、H線、及びI線撮像系を拡大することができる。
【0043】
[0058]上述の典型的な実施例は、本発明の範囲を制限することを意図せず、当業者は、本開示に鑑みて上記実施例を添付の特許請求の範囲に開示する本発明の技術的内容に対応するよう展開し得るであろう。

[0001] SENSITIZED CHEMICALLY AMPLIFIED PHOTORESIST FOR USE IN PHOTOMASK FABRICATION AND SEMICONDUCTOR PROCESSING

[0002] BACKGROUND OF THE INVENTION

[0003] 1. Field of the Invention

[0004] The present invention pertains to a photoresist composition useful in photomask fabrication and in semiconductor production. In particular, the photoresist composition enables the fabrication of photomasks which exhibit a combination of dense and isolated features. The same photoresist which is used for the photomask fabrication may also be applied to semiconductor and micro electro mechanical systems (MEMS) processing.

[0005] 2. Brief Description of the Background Art

[0006] Photoresist compositions are used in microlithographic processes for making miniaturized electronic components, such as in the fabrication of semiconductor device structures ; for making miniaturized mechanical systems; and for making microbiological structures. The miniaturized device structure patterns are typically created by transferring a pattern from a patterned masking layer overlying the semiconductor or other substrate rather than by direct write on the substrate, because of the time economy which can be achieved by blanket processing through a patterned masking layer. With regard to the micro device processing, the patterned masking layer may be a patterned photoresist layer or may be a patterned "hard" masking layer (typically an inorganic material or a high temperature organic material) which resides on the surface of the semiconductor device structure or other substrate to be patterned. The patterned masking layer is typically created using another mask which is frequently referred to as a photomask or reticle. A reticle is typically a thin layer of a metal- containing layer (such as a chrome-containing, molybdenum-containing, or tungsten- containing material, for example) deposited on a glass or quartz plate. The reticle is patterned to contain a "hard copy" of the individual device structure pattern to be recreated on the masking layer overlying a semiconductor structure or other substrate.

[0007] A reticle may be created by a number of different techniques, depending on the method of writing the pattern on the reticle. Due to the dimensional requirements of today's semiconductor structures, the writing method is generally with a laser or e-beam.

A typical process for forming a reticle may include: providing a glass or quartz plate, depositing a chrome-containing layer on the glass or quartz surface, depositing an antireflective coating (ARC) over the chrome-containing layer, applying a photoresist layer over the ARC layer, direct writing on the photoresist layer to form a desired pattern, developing the pattern in the photoresist layer, etching the pattern into the chrome layer, and removing the residual photoresist layer. When the area of the photoresist layer contacted by the writing radiation becomes easier to remove during development, the photoresist is referred to as a positive-working photoresist. When the area of the photoresist layer contacted by the writing radiation becomes more difficult to remove during development, the photoresist is referred to as a negative-working photoresist. Advanced reticle manufacturing materials frequently include combinations of layers of materials selected from chromium, chromium oxide, chromium oxynitride, molybdenum, molybdenum silicide, and molybdenum tungsten silicide, for example.

[0008] As previously mentioned, the reticle or photomask is used to transfer a pattern to an underlying photoresist, where the reticle is exposed to blanket radiation which passes through open areas of the reticle onto the surface of the photoresist. The photoresist is then developed and the patterned photoresist is used to transfer the pattern to an underlying semiconductor structure, typically using a plasma dry etching process.

[0009] As the feature size requirements for a semiconductor substrate has become smaller, and as new applications for semiconductor devices and for MEMS devices are requiring the mixing of both logic and memory features on a single chip, new issues have arisen regarding both reticle fabrication and semiconductor chip production. While the memory devices tend to have features which are closely spaced (dense), the logic devices tend to have features which are sparsely spaced (isolated). As a result, proximity effects are observed during patterning of the photoresists used for fabrication of the reticle and during patterning of the photoresists used for pattern transfer to the semiconductor substrate.

[0010] For example, in the fabrication of a reticle patterned to have feature critical dimensions (CD) in the range of about 500 nm (0. 50, um) to about 2000 nm (2. ohm), optical proximity effects have been observed during imaging of a standard novolak- based G-line, H-line, or I-line photoresist. In particular, for a standard I-line photoresist, such as an iP3600 photoresist available from Tokyo Ohka, Tokyo, Japan, or a PF188A photoresist available from Sumitomo, Osaka, Japan, CD errors from about 20 nm to about 40 nm, which are attributable to proximity effects have been observed in the patterned photoresist which was to be used to transfer the pattern to the reticle, and were observed in the chrome of the patterned reticle.

[0011] Figure 1 shows a typical starting structure 100 used in the fabrication of a reticle. This starting structure was generally used in the preparation of test specimens during development of the present method of fabricating reticles. Starting structure 100 is a stack of layers which includes, from top to bottom, an approximately 5,000 A thick layer 108 of a novolak-based photoresist, iXHOOP (available from Clarient Corp. of Sommerville, New Jersey); an approximately 500 A thick layer 106 of an inorganic ARC, chrome oxynitride; an approximately 200 A thick layer 104 of a mask material which is essentially chrome; and a silicon oxide-containing substrate 102.

[0012] In view of the sizable 20 run to 40 nm CD error which was observed in the initial fabrication of reticle test specimens, a considerable amount of effort was spent examining all of the parameters of the continuous write laser tool to determine whether these parameters might be the cause of the CD errors. The details of that work will not be discussed here, since it was determined that the laser tool parameters were not responsible for the CD errors. It was discovered that the CD errors were generated as a result of the behavior of the photoresist material during imaging and development.

[0013] Figure 2A shows a schematic top view 200 of the photoresist layer 108 shown in Figure 1, where a first pattern, in particular a bar pattern 202 has been written on the upper surface 201 of photoresist layer 108. The distance dl between the bar lines 203 and 204 is about 2,000 nm (about 2. 0 =n), and represents the CD which is to be controlled as tightly as possible. The thickness of each bar, 203 and 204 was about 2. 0 um. The distance d2 represents the length of the bar pattern 202 and is about 5, 700pu.

[0014] Figure 2B shows a schematic top view 220 of the photoresist layer 108 shown in Figure 1, where a second pattern, in particular a steps pattern 222 has been written on the upper surface 221 of photoresist layer 108. The distance d between each half of the step pattern 222 is about 2. 0 um and represents the CD. The distance dz is about 5, 700 jum, with the length d3 of each step being about 317 pin, with the exception of the top step 224, which is about 2 x 317 pxn. The height (thickness) d4 of the end step 226 at each end of the steps pattern 222 is about 6.5 Sn, with the height d5 of the center step 224 being about 512 Sn.

[0015] Since the photoresist is a positive photoresist, a cleared space is produced by exposing the photoresist to radiation and then developing the pattern created by the radiation to remove the photoresist in the irradiated area. With reference to Figure 2A, bars 203 and 204 were irradiated by direct writing using a continuous wave laser having a half-intensity beam diameter (spotsize) of about 270 nm. With reference to Figure 2B, each half of the steps pattern 222 was irradiated by direct writing using a the same continuous laser, where the laser was scanned over the surface 221 of photoresist 108 to produce the irradiated pattern. After writing of the pattern on the surface of starting structure, the pattern in photoresist layer 108 was developed and then transferred through underlying ARC layer 106 and chrome layer 104, to produce a chrome pattern (not shown) on the upper surface 103 of quartz substrate 102.

[0016] Figure 3A shows the average CD for the distance d, of the a chrome line which was produced on the upper surface of the quartz substrate 108 (in accordance with Figure 2A), as a function of the distance traveled in direction"X"as shown in Figure 2A.

The variation in CD ranged from about 1753 nm at X = 0 pm to about 1746 nm at X = 2700 ym, to about 1754 nm at X=5, 400 um. The difference in CD was only about 7-8 nm over the entire length of the chrome line.

[0017] Figure 3B shows the average CD for the distance d4 of a chrome line which was produced on the upper surface of quartz substrate 108 (in accordance with Figure 2B), as a function of the distance traveled in direction"X"as shown in Figure 2B. The variation in CD ranged from about 1780 nm at X = 0 tun to about 1758 nm at X = 2700 , um, to about 1782 nm at X=5, 400 um. The difference in CD was 23 nm over the length of the chrome line.

[0018] The difference in the CD range of the line obtained for the structures illustrated in Figures 2A and 2B is attributed to proximity effects which resulted from the difference in the size of the surface area of the photoresist 108 which was exposed to radiation adjacent to the line. These proximity effects are frequently referred to as photoresist loading effects.

[0019] Clearly, it would be highly desirable to be able to reduce the change in CD which is observed across a patterned reticle due to photoresist loading, as this would better enable the fabrication of a reticle where a portion of the features is dense, while another portion of the features is isolated.

[0020] SUMMARY OF THE INVENTION [0021] The photoresist composition and method of using the photoresist in the fabrication of reticles is designed to reduce the variation in critical dimension of reticle features across a surface of a patterned reticle, where the variation in critical dimension is a result of localized resist loading.

[0022] In particular, we have developed a photoresist useful in the manufacture of reticles, where the photoresist is exposed to G-line, H-line, or I-line patterning radiation.

The photoresist comprises a casting solvent, a modified phenol-substituted resin, a photochemical amplifying compound, and a sensitizer selected from the group consisting of anthracene or a derivative thereof, naphthalene or a derivative thereof, and mixtures of these materials. By way of example, and not by way of limitation, typically the sensitizer is selected from a group consisting of anthracene; 9-phenoxymethylanthracene; 1, 4-dimethoxyanthracene ; 9-anthracene methanol; 9,10-dimethyl anthracene; naphthalene; and 2-hydroxyl-1, 4-naphthaquinone. Frequently, the casting solvent is selected from materials such as propylene glycol monomethyl ether acetate (PMA, PGMEA), ethoxy ethyl propyionate, ethyl cellosolve acetate, diglyme and combinations thereof.

[0023] The base resin for the photoresist is selected from a modified phenolic resin, a modified novolak resin, and combinations thereof. One particularly advantageous base resin is a substituted polyhydroxy styrene or a copolymer thereof.

[0024] The photochemical amplifying compound (PAC) may be one of those known in the art, such as an onium salt metal halide complex, triflic acid and derivatives thereof, tosylate and various derivatives thereof, and mesylate and various derivatives thereof, for example and not by way of limitation. One of the frequently used PACs is an aryl sulfonium salt.

[0025] Use of the. sensitizer described above permits efficient energy transfer from the photoactive compound (PAC) to the polymeric base material at the radiation wavelengths for G-line, 436 nm; H-line, 405 nm; and I-line, 364 nm. This efficient energy transfer means that less power is required to image the resist and direct writing of a pattern on the resist can be carried out more rapidly, by a continuous writing laser (for example and not by way of limitation). Even more importantly, since the byproducts created during irradiation of (imaging of) the photoresist do not tend to inter react with the developing agent used to pattern the photoresist, there is a reduction in the optical proximity effects which tend to occur when both dense and isolated features are present on the same reticle. This enables a reduction in the change in CD over a reticle surface when both dense and isolated features are present. The sensitizer may be simply mixed into the combination of casting solvent such as PMA, modified phenol-substituted resin, and chemical amplifier, or the sensitizer may be attached to the phenol-substituted base resin or to the chemical amplifier.

[0026] The same concepts which apply to reticle fabrication also apply to the patterning of the photoresist described above on a semiconductor substrate. By the addition of the proper sensitizer, an 1-line imaging system, for example can be used to generate a more controlled feature size, extending the ability of an I-line imaging system into smaller dimension features, for example, feature sizes down to about 0. 3 tun. In the case of G-line and H-line imaging systems, the functionality of the imaging systems can be extended into smaller dimension features as well.

[0027] BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 shows a typical starting structure 100 used in the fabrication of a reticle.

[0029] Figure 2A shows a schematic top view 200 of the photoresist layer 108 shown in Figure 1, where a first pattern, in particular a bar pattern 202 has been written on the upper surface 201 of photoresist layer 108.

[0030] Figure 2B shows a schematic top view 220 of the photoresist layer 108 shown in Figure 1, where a second pattern, in particular a steps pattern 222 has been written on the upper surface 221 of photoresist layer 108.

[0031] Figure 3A shows the average CD for the distance di of the a chrome line which was produced on the upper surface of the quartz substrate 108 (in accordance with Figure 2A), as a function of the distance traveled in direction"X"as shown in Figure 2A.

[0032] Figure 3B shows the average CD for the distance d4 of a chrome line which was produced on the upper surface of quartz substrate 108 (in accordance with Figure 2B), as a function of the distance traveled in direction"X"as shown in Figure 2B.

[0033] Figure 4 shows the average CD for the distance d4 of a chrome line which was produced on the upper surface of quartz substrate 108 (in accordance with Figure 2B), as a function of the distance traveled in direction"X"as shown in Figure 2B. However, with reference to Figure 4, the photoresist of the present invention was used to form photoresist layer 108 rather than a diazoquinone sensitized, novolak-based photoresist of the kind described in the Background Art section of this disclosure.

[0034] DETAILED DESCRIPTION OF EMBODIMENTS

[0035] As a preface to the detailed description, it should be noted that, as used in this specification and the appended claims, the singular forms "a", "an", and "the" include plural referents, unless the context clearly dictates otherwise.

[0036] Recent advances in the electronics industry have created a need to place memory and logic devices on the same chip. It happens that mixing of such devices leads to processing difficulties, since memory devices tend to make use of densely placed features, while logic devices frequently make use of isolated features. Processing of the photomasks (reticles) used to image photoresists which are used to transfer patterns to a semiconductor substrate is affected by proximity effects. Further, processing of the imaged photoresists themselves is affected by the same proximity effects. Manufacturers of the tools used to write patterns on a photoresist used to create the pattern on the reticle have very carefully investigated the variables which affect the writing tool, to reduce the variation in feature critical dimension (CD) across the reticle surface. In particular ETEC Systems of Hillsboro, Oregon investigated their ALTA 3700 Tool which employs a continuous wave direct write laser to write a pattern on a photoresist which is used to transfer the pattern to the reticle. After considerable investigation, it was discovered that the problem was not the writing tool, but was instead a reaction which was going on between the developer used to develop the photoresist and byproduct compounds which were generated during writing of the pattern on the photoresist.

[0037] Since the photoresist used to pattern the reticle is basically the same photoresist as the one used on the surface of a semiconductor substrate to transfer a pattern to the substrate, one skilled in the art might wonder why this problem has not been observed with respect to semiconductor processing. The reason is that the features on the reticle are typically about four times larger than the features on the semiconductor substrate. In order to obtain semiconductor device features in the 180nm to 250nm range, the semiconductor processing industry has gone to deep ultraviolet radiation (DW) 248 nm imaging of photoresists. However, reticle manufacturers are still able to use G-line 436 nm) or H-line (405 nm) or I-line (364 nm) radiation for imaging photoresists used to pattern a reticle.

[0038] ETEC Systems provides DUV imaging systems as well as 1-line imaging systems, and applicants discovered that the problem of the developer reacting with the byproducts created during photoresist imaging is unique to the novolak-based resin systems which have been used in the combination with the G-line, H-line, and 1-line imaging systems. Applicants did not observe the same development problem with respect to the photoresist system used in combination with DUV imaging systems. In particular, a chemically amplified DUV photoresist, DXI 100 was not observed to exhibit the severe photoresist loading problems which had been observed with the novolak-based resin systems used for I-line patterning, for example. However, the DX1100 photoresist consists basically of a propylene glycol monomethyl ether acetate (PMA, PGMEA, or 1-methoxy-2-propyl acetate) casting solvent; a modified phenolic polymer; and an onium salt metal halide complex as a chemical amplifier. This photoresist does not perform at the radiation wavelength used to image an I-line photoresist.

[0039] Applicants have been able to add a sensitizer to the DXI 100 DLTV photoresist which enables this photoresist to perform when exposed to 364 nm radiation used to image an I-line photoresist. In particular, a sensitizer such as anthracene methanol, anthracene, or a diphenyl malaimide type compound has been added to the DUV photoresist. Applicants'new photoresist comprises the following solids : From about 75 % by weight to about 85 % by weight of the propylene glycol monomethyl ether acetate ; about 20 % by weight to about 30 % by weight of modified phenolic resin, about 0.1 % by weight to about 2 % by weight of a photoactive agent such as an aryl sulfonium salt, and about 0.1 % by weight to about 2 % by weight of a sensitizer in the form of anthracene or naphthalene, or a derivative thereof. Such sensitizers include, by way of example and not by way of limitation : anthracene ; 9-phenoxymethylanthracene ; 1,4- dimethoxyanthracene; 9-anthracene methanol; 9, 10-dimethyl anthracene; naphthalene; and 2-hydroxyl-1, 4-naphthaquinone. The solids are typically dissolved in PGMEA Inventors or an equivalent solvent. In one preferred embodiment, the solids are 13.4 % by weight propylene glycol monomethyl ether acetate, 2.6 % by weight modified phenolic resin, about 2.0 % by weight aryl sulfonium salt, and about 2.0 % of a sensitizer selected from the compounds listed above. The sensitizer was simply mixed in with the DUV photoresist, and the mixture was filtered using a millipore filter prior to application by spin coating to the substrate surface. It is important to mention that the sensitizer could be associated with or bonded to the base polymer of the photoresist or associated with or bonded to the PAC of the photoresist. In particular, one skilled in the art could synthesize either a base polymer or a PAC with an anthracene or naphthalene or anthracene/naphthalene chromaphore attached.

[0040] The same procedures as those currently recommended by the DUV resist manufacturer may be used for application of the photoresist to the substrate. After application of the photoresist, the photoresist is post apply baked (PAB) at about 110 °C for a time period of ranging between about 7 minutes and about 15 minutes. After application of the photoresist and imaging of the photoresist using an ALTAR 3700 imaging system, which employs a direct write continuous wave UV laser, the photoresist was developed using a spin/spray process with an AZ 300 MIF developer (available from AZ Cariant Corp. of Somerville, NJ). This developer is a 2.38 wt % solution of tetramethyl ammonium hydroxide (TMAH). The development time was about 60 seconds. Concurrent with cessation of develop dispense, CO2-sparged or C02-reionized water was dispensed for approximately 30 seconds at about 300 rpm for the first 15 seconds and at 1,500 rpm for the remaining 15 seconds. Further rinse via spray-puddle was commenced, followed by a spin-dry step of approximately 90-120 seconds at 2,000 - 2, 500 rpm.

[0041] The pattern in the photoresist was then transferred to the underlying photomask structure using a dry etch process. The substrate was etched in a high density plasma under conditions that reduced mean-to-target deviation, while still maintaining good CD uniformity. The plasma etch was performed in an inductively coupled plasma (ICP) etch tool using a three step process: descum/organic ARC (BARC) removal; chrome oxynitride (inorganic ARC)/chrome etch ; and overetch.

[0042] Plasma etch systems such as the Applied Materials, Inc. TETRA DPSTM etch system (available from Applied Materials, Inc., of Santa Clara, California) may be used to provide excellent results. A plasma processing system which permits separate power application for plasma generation and for substrate biasing is commonly referred to as a Decoupled Plasma Source (DPS). Substrate biasing is used to attract ions and other high energy species from the plasma toward the substrate surface, enabling anisotropic etching. Separate application of power for plasma generation and power for substrate biasing permits separate control of the plasma density and the attractive forces (DC voltage) generated on the surface of the substrate.

[0043] The descum/ARC (BARC) removal step of the three step etch process removes any residual photoresist remaining on open areas after development, and removes the organic ARC (BARC) layer from the open areas. This is accomplished using an oxygen plasma. A DC bias is applied to the substrate during the descum/BARC removal step to accelerate oxygen ions from the plasma so that they impinge upon the reticle surface with high kinetic energy and directionality, normal to the reticle surface.

This is accomplished using oxygen gas and by powering only the lower electrode (upon which the reticle plate rests), creating a capacitively coupled plasma with a high DC bias between the plasma and the reticle. Typically the process is carried out in a process chamber at a pressure of about 3 mTorr to about 8 mTorr. A plasma source gas of oxygen was fed into the processing chamber at a flow rate of about 17 sccm. RF power of about 125 W at a frequency of 13.56 MHZ was applied to the lower electrode (cathode). This provided a DC bias of about-340 to about-410 Volts on the photomask substrate surface, while providing an oxygen plasma over the reticle surface. The temperature of the reticle plate was in the range of 25 °C, with a chamber wall temperature in the range of 40 °C. The descum/organic ARC (BARC) removal time was about 30 seconds. The photoresist loss due to the descum/organic ARC (BARC) removal process is about 750 A.

[0044] The chrome oxynitride (inorganic ARC)/chrome mask layer etch was done using a plasma generated from a chlorine-oxygen-helium gas mixture. The molecular ratio of the chlorine: oxygen : helium gas mixture was about 50: 10: 22. The total gas flow rate was about 82 sccm. The ICP coil was powered to about 60 W at 2 MHZ to generate a high density plasma. The lower electrode was powered to about 5 W at about 13.56 MHZ, to generate a DC voltage of about-50 V on the substrate. The temperature of the reticle plate was about 25 °C, while the wall temperature of the process chamber was about 40 °C. The etch end point was detected by optical reflectance, and occurred in about 200 seconds. Typically, higher oxygen concentrations and lower pressures cause higher mean-to-target deviation and lower selectivities, while favoring better CD uniformity. One skilled in the art can optimize the process for his/her particular apparatus.

[0045] Typically the chrome layer is overetched beyond endpoint to clear residual chrome from all open regions. Generally the overetch step is an extension of the chrome etch process described above. Longer overetch steps result in higher mean-to-target deviations. Chrome spot defect densities can be affected by the length of overetch, with lower defect densities for longer overetch processes.

[0046] After completion of the chrome layer etch, a strip and clean process is performed to remove any residual contaminants from the surface of the chrome layer.

The strip chemical used was sulfuric peroxide which was heated to about 75 °C and applied over the surface of the substrate plate. After treatment with sulfuric peroxide, the substrate plate is rinsed with CO2-reionized, or C02-sparged deionized water. After strip, the substrate plate was subjected to an acid clean using an industry standard 70: 30 H2SO4/H202 solution, followed by another deionized water rinse. The strip step was performed on a Steag ASC 500 wet chemical processing station available from STEAG- HAMMATECH (D, Santa Clara, California.

[0047] Example One-Comparative Example:

[0048] The description below is with respect to an I-line photoresist system, by example, and not by way of limitation.

[0049] As previously described, Figure 2A shows a schematic top view 200 of the photoresist layer 108 shown in Figure 1, where a first pattern, in particular a bar pattern 202 has been written on the upper surface 201 of photoresist layer 108. The distance di between the bar lines 203 and 204 is about 2,000 nm (about 2. 0 um), and represents the CD which is to be controlled as tightly as possible. The thickness of each bar, 203 and 204 was about 2. 0 um. The distance d2 represents the length of the bar pattern 202 and is about 5, 700ecum.

[0050] Figure 2B shows a schematic top view 220 of the photoresist layer 108 shown in Figure 1, where a second pattern, in particular a steps pattern 222 has been written on the upper surface 221 of photoresist layer 108. The distance da between each half of the step pattern 222 is about 2. 0 um and represents the CD. The distance d2 is about 5,700 , an, with the length d3 of each step being about 317 E. cm, with the exception of the top step 224, which is about 2 x 317 pm. The height (thickness) d4 of the end step 226 at each end of the steps pattern 222 is about 6.5 n, with the height d5 of the center step 224 being about 512/mi.

[0051] Since the photoresist is a positive photoresist, a cleared space is produced by exposing the photoresist to radiation and then developing the pattern created by the radiation to remove the photoresist in the irradiated area. With reference to Figure 2A, bars 203 and 204 were irradiated by direct writing using a continuous wave laser having a half-intensity beam diameter (spotsize) of about 270 nm. With reference to Figure 2B, each half of the steps pattern 222 was irradiated by direct writing using a the same continuous laser, where the laser was scanned over the surface 221 of photoresist 108 to produce the irradiated pattern. After writing of the pattern on the surface of starting structure, the pattern in photoresist layer 108 was developed and then transferred through underlying ARC layer 106 and chrome layer 104, to produce a chrome pattern (not shown) on the upper surface 103 of quartz substrate 102, in the manner described in detail above.

[0052] Figure 3A shows a graph 300 which illustrates the average CD for the distance d, of the a chrome line which was produced on the upper surface of the quartz substrate 108 (in accordance with Figure 2A), as a function of the distance traveled in direction"X"as shown in Figure 2A. The average CD is shown on axis 302 of Figure 3A, while the distance traveled in direction"X"is shown on axis 304. As illustrated by curve 306, the variation in CD ranged from about 1753 nm at X = O, um to about 1746 nm at X = 2700 um, to about 1754 nm at X=5,400 ym. The difference in CD was only about 7-8 nm over the entire length of the chrome line.

[0053] Figure 3B shows a graph 320 which illustrates the average CD for the distance d4 of a chrome line which was produced on the upper surface of quartz substrate 108 (in accordance with Figure 2B), as a function of the distance traveled in direction "X"as shown in Figure 2B. The average CD is shown on axis 322 of Figure 3B, while the distance traveled in direction"X"is shown on axis 324. As illustrated by curve 326, the variation in CD ranged from about 1780 nm at X= O llm to about 1758 nm at X = 2700 um, to about 1782 nm at X=5,400 pin. The difference in CD was 23 nm over the length of the chrome line. A series of similar Examples has shown differences in CD ranging from about 20 mu to about 40 nm.

[0054] Example Two :

[0055] Figure 4 shows a graph 400 which illustrates the average CD of a chrome line which was produced on the upper surface of quartz substrate 108 (in accordance with Figure 2B), as a function of the distance traveled in direction"X"as shown in Figure 2B. However, the photoresist used to fabricate the test specimen was the DX1100 DUV photoresist with applicants'9-anthracene methanol sensitizer added rather than the diazoquinone sensitized, novolak-based photoresist previously used for I- line imaging. In particular, Figure 4 shows the average CD on axis 402, while the distance traveled in direction"X"is shown on axis 404. As illustrated by curve 406, the variation in CD ranged from about 1755 nm at X = 0 pm to about 1746 nm at X = 2700 Am, to about 1755 nm at X=5,400 gm. The difference in CD was about 9 nm over the length of the chrome line.

[0056] This variation in CD is very comparable with the variation in CD obtained when only the 2. 0 um thick bars 202 and 203 were used to define a line 202, as described with reference to Figure 2A. This is a clear indication that when the photoresist of the present invention is used for I-line patterning, rather than a novolak-based photoresist, it is possible to avoid the photoresist loading effects which occur with the novolak-based photoresist. The photoresist of the present invention, imaged and developed in the manner described above, enables one skilled in the art to produce a reticle which contains both dense and isolated features with minimal CD error.

[0057] In addition, one skilled in the art reading this disclosure will be able to extend G-line, H-line, and I-line imaging systems for use in production of smaller features, by using the photoresist of the present invention rather than a diazoquinone, novolak-based photoresist for patterning semiconductor substrates.

[0058] The above described exemplary embodiments are not intended to limit the scope of the present invention, as one skilled in the art can, in view of the present disclosure expand such embodiments to correspond with the subject matter of the invention claimed below.

「特表2005-527003およびWO2004001797より引用」

|
|

[AMENDED CLAIMS] ウエハレベルのスプリングを有するプローブカードアセンブリおよびパッケージの構造および製造工程

【特許請求の範囲】
【請求項1】
集積回路ウエハ用のテスト装置であって、
下面および上面と、前記下面から前記上面まで延在する複数の電気導体とを有するマザーボード基板と、
プローブ面およびコネクタ面と、前記プローブ面上の複数のプローブスプリングと、前記コネクタ面上の複数の電気接触子と、複数のプローブチップ電気接続部と、を有するプローブチップ基板であって、前記プローブスプリングのそれぞれが、少なくとも一つのプローブチップ電気接続部を通じて少なくとも一つの接触子に電気的に接続されている、前記プローブチップ基板と、
前記マザーボード基板と前記プローブチップ基板との間に位置する少なくとも一つの中間コネクタであって、前記中間コネクタが、前記プローブチップ基板上の前記複数の電気接触子のそれぞれと、前記マザーボード基板の前記下面上の前記電気導体のそれぞれとの間の少なくとも一つの導電性接続部を有する、前記中間コネクタと、
を有する、少なくとも一つの集積回路デバイス用のテスト装置であって、
前記プローブチップ基板がコネクタ面から支持され、前記マザーボード基板との関係において配置される、
テスト装置。
【請求項2】
前記複数のプローブスプリングが、前記プローブチップ基板に付着している固定部分と自由部分とを有し、前記自由部分が、最初は前記プローブチップ基板に付着しており、引き離し時に、それぞれのスプリングプローブが有する複数の層の間に画成されている固有の応力勾配の結果として前記プローブチップ基板から離れるように延在する、請求項1に記載のテスト装置。
【請求項3】
前記複数のプローブスプリングが、薄膜工程、またはMEMS(微小電気機械システム)工程などのロット型工程において作製されたプローブ先端を有する、請求項1に記載のテスト装置。
【請求項4】
前記複数のプローブスプリングが、対応性を持つフレキシブル導電性接触子構造を有する、請求項1に記載のテスト装置。
【請求項5】
前記複数のプローブスプリングが、フォトリソグラフィによってパターニングされる、請求項1に記載のテスト装置。
【請求項6】
少なくとも一つの中間コネクタに前記プローブチップ基板の前記コネクタ面上の接触子を固定的に取り付けることによって、プローブチップ基板が、コネクタ面から支持され、前記マザーボード基板との関係において配置される、請求項1に記載のテスト装置。
【請求項7】
前記プローブチップ基板の前記コネクタ面上の接触子を少なくとも一つの中間コネクタに固定的に取り付けるための前記手段が、異方性導電膜を有する、請求項6に記載のテスト装置。
【請求項8】
前記プローブチップ基板の前記コネクタ面上の接触子を少なくとも一つの中間コネクタに固定的に取り付けるための前記手段が、はんだジョイントを有する、請求項6に記載のテスト装置。
【請求項9】
前記プローブチップ基板の前記コネクタ面の前記外側周辺部に固定的に取り付けられ、対応性を持つ部材を有するプローブチップ保持器によって、プローブチップ基板がコネクタ面から支持され、前記マザーボード基板との関係において配置される、請求項1に記載のテスト装置。
【請求項10】
前記対応性を持つ部材が、膜である、請求項9に記載のテスト装置。
【請求項11】
前記膜が、ポリイミドを有する、請求項10に記載のテスト装置。
【請求項12】
前記マザーボード基板の前記上面に固定的に取り付けられている、堅い材料から作製された補強プレートをさらに有する、請求項1に記載のテスト装置
【請求項13】
前記補強プレートが、ステンレス鋼を有する、請求項12に記載のテスト装置。
【請求項14】
少なくとも一つの構成要素くぼみが、前記マザーボード基板の前記上面に近い前記補強プレートに画成されており、かつ、前記装置が、前記構成要素くぼみの中に前記マザーボード基板から延在している少なくとも一つの構成要素をさらに有する、請求項12に記載のテスト装置。
【請求項15】
前記中間コネクタが少なくとも一つ、第一面上の第一の複数の対応性を持つ電気接触子と、前記第一面と反対の第二面上の第二の複数の対応性を持つ電気接触子とを有するインターポーザーを有する、請求項1に記載のテスト装置。
【請求項16】
前記プローブチップ基板の前記コネクタ面に固定的に取り付けられている少なくとも一つの支持棒をさらに有する、請求項1に記載のテスト装置。
【請求項17】
前記プローブチップ基板上に、組立済み構成要素として組み込まれている少なくとも一つの構成要素をさらに有する、請求項1に記載のテスト装置。
【請求項18】
前記少なくとも一つの構成要素が、前記プローブチップ基板の前記コネクタ面上、または、プローブ面上のいずれかに実装されている、請求項17に記載のテスト装置。
【請求項19】
前記少なくとも一つの構成要素が、キャパシタである、請求項14、または、請求項17に記載のテスト装置。
【請求項20】
前記プローブチップ上のプローブスプリング接触子先端の平面性が前記マザーボード基板に対して調整可能である平面性調整機構をさらに有する、請求項1に記載のテスト装置。
【請求項21】
前記少なくとも一つの中間コネクタが、プリント基板を有し、かつ、前記複数の導電接続部が、前記プローブチップ基板との電気接続のための手段と、前記マザーボード基板との電気接続のための手段とを有するビアを有する、請求項1に記載のテスト装置。
【請求項22】
いずれの前記マザーボード基板、または、プローブチップ基板との電気接続のための前記手段も、インターポーザーを有する、請求項1、または、請求項21に記載のテスト装置。
【請求項23】
前記少なくとも一つの中間コネクタが、Zブロックを有し、前記Zブロックが、下面および上面を有する垂直変換基板と、当該基板の前記下面から前記上面まで延在する複数の導電接続部とを有し、前記導電接続部のそれぞれが、少なくとも一つの導電ビアを有する、請求項1に記載のテスト装置。
【請求項24】
前記Zブロックと前記いずれのプローブチップ基板、または、マザーボード基板との間のインターポーザーをさらに有する、請求項23に記載のテスト装置。
【請求項25】
前記プローブチップ基板が前記少なくとも一つの中間コネクタに取り付けられ、これによって、前記プローブチップ基板上の前記複数のプローブスプリングを機械構造的に支持する、請求項1に記載のテスト装置。
【請求項26】
前記いずれのプローブチップ基板と前記少なくとも一つの中間コネクタが少なくとも部分的にセラミック、ガラスまたは石英、シリコン、有機基板 (organic board)、多層セラミックまたはこれらの組み合わせから成るグループから選択された材料によって作製されている、請求項1、または、請求項25に記載のテスト装置。
【請求項27】
前記いずれのプローブチップ基板と前記少なくとも一つの中間コネクタが貫いて画成されている複数のホールをさらに有し、かつ、前記複数のプローブチップ基板の接続部と中間コネクタの電気接続部のそれぞれが、前記複数のホールの少なくとも一つのホールの中に位置する少なくとも一つの導電ビアである、請求項1、請求項25、または、請求項26に記載のテスト装置。
【請求項28】
前記プローブチップ基板、または、前記少なくとも一つの中間コネクタを貫く前記複数のホールが、レーザーまたは機械的な穿孔工程を使用して作製される、請求項27に記載のテスト装置。
【請求項29】
いずれの前記マザーボード基板と、前記少なくとも一つの中間コネクタ、前記プローブチップ基板が、インピーダンス整合されている少なくとも一つの導電経路をさらに有する、請求項1に記載のテスト装置。
【請求項30】
前記少なくとも一つの中間コネクタが標準化された構成要素であり、かつ、プローブチップ基板が前記少なくとも一つのデバイス、すなわち電気的相互接続位置に対応する前記コネクタ面上の前記複数のプローブスプリングの電気的相互接続位置を含む、受け取った相互接続仕様に基づいて作製された、請求項1に記載のテスト装置。

AMENDED CLAIMS [ (received by the International Bureau on 10 December 2004 (10.12. 04); original claims 1-179 replaced by new claims 1-30 (5 pages)] 1. A test apparatus for at least one integrated circuit device, comprising: a m otherboard substrate h aving a bottom surface and a top s urface, and a plurality of electrical conductors extending from the bottom surface to the top surface; a probe chip substrate comprising a probe surface and a connector surface, a plurality of probe springs on the probe surface, a. plurality of electrical contacts on the connector surface, and a plurality of probe chip electrical connections, wherein each of the probe springs is electrically connected to at least one contact through at least one probe chip electrical connection; at least one intermediate connector located between the motherboard substrate and the probe chip substrate, the intermediate connector comprising at least one electrically conductive connection between each of the plurality of electrical contacts on the probe chip substrate and each of the electrical conductors on the bottom surface of the motherboard substrate; wherein the probe chip substrate is supported from the connector surface and positioned in relation to the motherboard substrate.

2. The test apparatus of Claim 1, wherein the plurality of probe springs comprises a fixed portion attached to the probe chip substrate and a free portion, initially attached to the probe chip substrate, which upon release, extend away from the probe chip substrate as a result of an inherent stress gradient between a plurality of layers within each of the probe springs.

3. The test apparatus of Claim 1, wherein the plurality of probe springs comprises probe tips fabricated using batch mode thin film or MEMS processes.

4. The test apparatus of Claim 1, wherein the plurality of probe springs comprises flexible, compliant, electrically conductive contact structures.

5. The test apparatus of Claim 1, wherein the plurality of probe springs is photolithographically patterned.

6. The test apparatus of claim 1, wherein the probe chip substrate is supported from the connector surface and positioned in relation to the motherboard substrate by fixedly attaching the contacts on the connector surface of the probe chip substrate to the at least one intermediate connector.

7. The test apparatus of Claim 6, wherein the means for fixedly attaching the contacts on the connector surface of the probe chip substrate to the at least one intermediate connector comprises a solder ball array.

8. The test apparatus of Claim 6, wherein the means for fixedly attaching the contacts on the connector surface of the probe chip substrate to the at least one intermediate connector comprises solder joints.

9. The test apparatus of claim 1, wherein the probe chip substrate is supported from the connector surface and positioned relative to the motherboard substrate by a probe chip carrier comprising a compliant member fixedly attached to the outer periphery of the connector surface of the probe chip substrate.

10. The test apparatus of Claim 9, wherein the compliant member is a film.

11. The test apparatus of Claim 10, wherein the film comprises polyimide.

12. The test apparatus of Claim 1, further comprising : a stiffener plate, fabricated from a rigid material, fixedly attached to the top surface of the motherboard substrate.

13. The test apparatus of Claim 12, wherein the stiffener plate comprises stainless steel.

14. The test apparatus of Claim 12, wherein at least one component recess is defined in the stiffener plate proximate the top surface of the motherboard substrate, and wherein the apparatus further comprises: at least one component extending from the motherboard substrate within the component recess.

15. The test apparatus of Claim 1, wherein the at least one intermediate connector comprises an interposer having a first plurality of compliant electrical contacts on a first surface and a second plurality of compliant electrical contacts on a second surface opposite the first surface.

16. The test apparatus of Claim 1, further comprising: at least one standoff fixedly attached to the connector surface of the probe chip substrate.

17. The test apparatus of Claim 1, further comprising: at least one component incorporated as an assembled component on the probe chip substrate.

18. The test apparatus of Claim 17, wherein the at least one component is mounted on either t he connector surface o r t he p robe s urface of t he p robe chip substrate.

19. The test apparatus of C laim 14 or Claim 17, wherein the at least one component is a capacitor.

20. The test apparatus of Claim 1, further comprising: a planarity adjustment mechanism in which the planarity of the probe spring contact tips on the probe chip is adjustable relative to the motherboard substrate.

21. The test apparatus of Claim 1, wherein the at least one intermediate connector comprises a printed wiring board, and wherein the plurality of electrically conductive connections comprise vias having means for electrical connection to the probe chip substrate and means for electrical connection to the motherboard substrate.

22. The test apparatus of Claim 1 or Claim 21, wherein the means for electrical connection to any of the motherboard substrate and the probe chip substrate comprises an interposer.

23. The test apparatus of Claim 1, wherein the at least one intermediate connector comprises a Z-block, comprising a vertical translation substrate having a lower surface and an upper surface, and a plurality of electrically conductive connections which extend from the lower surface to the upper surface thereof, each of the electrically conductive connections comprising at least one electrically conductive via.

24. The test apparatus of Claim 23, further comprising an interposer between the Z-block and any of the probe chip substrate and the motherboard substrate.

25. The test apparatus of Claim 1, wherein the probe chip substrate is attached to the at least one intermediate connector, thereby providing mechanical structural support for the plurality of probe springs on the probe chip substrate.

26. The test apparatus of Claim 1 or Claim 25, wherein any of the probe chip substrate and the at least one intermediate connector is fabricated at least in part from a material selected from a group consisting of ceramic, glass, quartz, silicon, organic board, multi-layer ceramic or combinations thereof.

27. The test apparatus of Claim 1, Claim 25, or Claim 26, wherein any of the probe chip substrate and the at least one intermediate connector further comprises a plurality of holes therethrough, and wherein each connection of the plurality of probe chip substrate and intermediate connector electrical connections is at least one electrically conductive via located within at least one hole of the plurality of holes.

28. The test apparatus of Claim 27, wherein the plurality of holes through the probe chip substrate or the at least one intermediate connector are fabricated using laser or mechanical drilling processes.

29. The test apparatus of Claim 1, wherein any of the motherboard substrate, the at least one intermediate connector, and the probe chip substrate further comprises at least one electrically conducting path with matched impedance.

30. The test apparatus of Claim 1, wherein the at least one intermediate connector is a standardized component and wherein the probe chip substrate is produced based on a received interconnection specification containing electrical interconnection locations for the at least one device, the plurality of probe springs on the probe surface corresponding to the electrical interconnection locations.

「特表2006-507479およびWO2004001807より引用」

|
|

ウエハレベルのスプリングを有するプローブカードアセンブリおよびパッケージの構造および製造工程

※以下の明細書の原文(英文)では、
[Name, U.S. Patent No. X,XXX,XXX (XX July 200X) disclose...]
のパターンとあり、和訳されたものは、
【特許文献1】米国特許第X,XXX,XXX号「タイトル(英語タイトル)」(氏名)(200X年7月XX日)
のパターンとなっている。
以下では、和訳の明細書から【特許文献X】と【非特許文献X】のがリストのように並んでいる箇所を省略した。(【発明の開示】の直前)

--------
【発明の詳細な説明】
【技術分野】
【0001】
本発明は、プローブカードアセンブリシステムの分野に関する。より詳細には、本発明は、フォトリソグラフィによってパターニングされるスプリング接触子における改良と、集積回路のテストまたはバーンインにおける使用を目的とする、フォトリソグラフィによってパターニングされるスプリング接触子を有する強化されたプローブカードアセンブリとに関する。
【背景技術】
【0002】
従来の集積回路(IC)ウエハ用のプローブカードにおいては、プローブカードと集積回路ウエハとの間の電気的接触は、一般にはタングステンの針プローブによって形成される。しかしながら、先進の半導体技術では、タングステン針プローブでは不可能である、より多くのピン数と、より小さいパッドピッチと、より高いクロック周波数とがしばしば要求される。
【0003】
新たに登場した技術によって、様々なプロービングアプリケーション用のスプリングプローブが提供されているが、ほとんどのプローブには本質的な制約があり、例えば、ピッチの制限、ピン数の制限、可曉性レベルの変動、プローブ先端形状の制限、材料の制限、高い製作コストなどである。
【0004】
特許文献1には、ランナー(runner)および基板のアセンブリが開示されている。このアセンブリは、「基板に接着されている複数の導電ランナーを有し、少なくともいくつかの導電ランナーの一部が、所定の応力を受けたときに基板から導電ランナーを選択的に引き離すための、基板と共に平面でない領域を有する」。
【0005】
特許文献2には、ランナーおよび基板のアセンブリとが開示されている。このアセンブリは、「基板に接着されている複数の導電ランナーを有し、少なくともいくつかの導電ランナーの一部において、所定の応力を受けたときに基板から導電ランナーが選択的に引き離されるように、接着性が低い」。
【0006】
特許文献3には、ベアダイ段階において集積回路(IC)をテストするテスト装置が開示されている。この装置は、「多層相互接続構造の相互接続トレース終端部上に導電材料のマイクロバンプが配置されているテストステーションを含んでおり、これらの終端部が、テストするダイ上の接触パッドのパターンに対応するパターンに分布している。ウエハから分離する前のダイのテストをマイクロバンプを使用して容易にするため、別に設けられている、相互接続構造への接続部および相互接続からの接続部が低い輪郭高さを有する」。
【0007】
特許文献4には、複数の電子デバイスの間に一つの回路を電気的に接続するためのコネクタが開示されており、「このコネクタは、カンチレバー型スプリングアームがそこから斜めに延在するプラットフォーム(platform)を含んでいる。スプリングアームには、高くなっている接触面が含まれており、一つの実施例においては、アームの幾何学形状によって、曲がるときに化合物の拭き取り機能が得られる。
【0008】
特許文献5には、「それぞれリードピンのセットが設けられている側縁部分を有する集積回路(IC)チップをテストするためのテストデバイスが開示されている。このテストデバイスは、ソケット基部と、それぞれが接触子支持部材とソケット接触子部材とを含んでいる接触子ユニットと、それぞれが弾性絶縁シートと導電部材とを含んでいる異方性導電シートアセンブリとを有する。異方性導電シートアセンブリは、接触子ユニットのソケット接触子部材の一つに接触している各導電部材を保持するように構成されている。このテストデバイスは、異方性導電シートアセンブリの導電部材とソケット接触子部材との間に電気的な通信を確立するため、ソケット接触子部材を異方性シートアセンブリに接触させる目的で、ソケット基部上に脱着式に取り付けられている接触子保持器をさらに有する。接触子ユニットのそれぞれは、ソケット接触子部材の一部が疲労したときに新しい接触子ユニットに交換することができ、これによって、テストデバイスの保守を容易にすることが可能である。さらには、異方性導電シートアセンブリの導電部材とソケット接触子部材の一部とによって形成される最短経路で、ICチップのリードピンをテスト回路基板に電気的に接続することができる。」と開示されている。
【0009】
特許文献6には、「回路基板の一つの主面において露出している導電材料のパッドを有し、かつ、回路基板の接触パッドに対して所定の位置にある位置決め形状を有する回路基板に、接触パッドを有する基板構造が実装される。基板構造には、基板構造の接触パッドに電気的に接続されており、かつ、カンチレバー状に基板構造から突き出しているリード部、が設けられている。位置決め要素は、プレート部分を有し、さらに、このプレート部分付近に分散している位置決め形状も有する。位置決め形状は、回路基板の位置決め形状と係合可能であり、そのように係合しているときに、回路基板の主平面に平行な動きに抗して位置決め要素を維持する。基板構造は、リード部が回路基板の位置決め形状に対して所定の位置にあり、かつ、位置決め要素がこの位置のときに、基板構造のリード部が回路基板の接触パッド上に位置するように、位置決め要素のプレート部分に取り付けられている。クランプ部材は、リード部を回路基板の接触パッドと圧力により導電的に接触している状態に維持する。」ことが、開示されている。
【0010】
特許文献7には、「プリント基板上に制御可能な接着型導体パターンを印刷するための有用な合成物は、細かく粉砕されている銅パウダーと、スクリーニング剤と、結合剤とを含んでいる。結合剤は、熱応力に応答して層が基板から持ち上がることができるように、基板への焼結後に形成される銅層の制御可能な接着性が得られるように設計されている。さらに、結合剤は、銅層が破断することなくリフトオフに耐えることができるように、銅層に良好な機械的強度を与える目的で、銅粒子間の良好な結束性を促進する役割を果たす。」ことが、開示されている。
【0011】
特許文献8には、「薄膜多層技術を使用して、抵抗の小さい金属-金属型接触子と、明確に異なるオン/オフ特性とを有する超小型電気機械スイッチが構築される。電熱的に作用するこのスイッチは、薄膜回路を製造するのに採用される工程と両立する工程を使用して、従来のハイブリッド回路基板上に作製される。好ましい形態においては、このようなスイッチは、金属(例:ニッケル)加熱要素が接着されている堅い絶縁性材料(例:窒化珪素)の弾性的に曲がることのできる帯状片を有するカンチレバー型アクチュエータ部材を含んでいる。カンチレバー型部材の自由端は、金属接触子を担持しており、この接触子は、加熱要素に印加される電流を介して部材を制御式に曲げることによって動き、下層の固定接触子に密着する(または離れる)。」ことが、開示されている。
【0012】
特許文献9には、「多層パッケージにおいて、複数のセラミック薄層それぞれが導電パターンを有しており、かつ、パッケージの内部空隙が存在しており、この空隙に、一つのチップ、または、チップアレイを形成するように相互接続されている複数のチップが接着されている。チップまたはチップアレイは、様々な薄層レベルにおいて短いワイヤボンドによって接続されて、各レベルに導電パターンが金属被覆され、各薄層レベルが特定の導電パターンを有する。それぞれの薄層上の導電パターンは、金属被覆された基板上に実装されているセラミックパッケージの下面における複数のパッドに最終的に接続されるように、金属被覆材料で満たされているトンネル状貫通穴によって、または、縁部に形成されている金属被覆のいずれかによって相互接続されている。構成要素の高い密度が達成されるが、接続リード部が「互い違いに配置されている」、またはまったく異なるパッケージレベルの交互のポイントにおいて接続されているため、ワイヤボンドランド(wire bond land)の10 milの間隔と10 milのサイズとを維持することが可能である。この結果として、構成要素の密度がさらに増すが、各ワイヤボンドが互いに干渉することがなく、この干渉要因は、多層セラミックパッケージにおいて高密度の構成要素の回路網を達成するうえでの従来の制限要因である。」ことが、開示されている。
【0013】
特許文献10には、集積回路をテストするためのプローブアセンブリが開示されている。このアセンブリは、「中央の開口を有する絶縁材料のプローブカードと、プローブカードに取り付けられているより小さい開口を有する長方形の枠と、導電性接地面シートを有するフレキシブル積層部材をそれぞれが有する四つの個別のプローブ羽根部と、接地面に接着されている接着性誘電体膜と、誘電体膜上のスプリング合金銅のプローブ羽根部トレースと、を有する。各プローブ羽根部は、中央の開口内に延在しており、かつ、プローブ羽根部トレースのそれぞれの終端部によって形成されている、整列している個々のプローブフィンガー(probe finger)のグループにおいて終了している、カンチレバー型リーフスプリング部分を有する。プローブフィンガーは、実質的に直線に沿って配置されており、かつ、テストするICの縁部に沿う各接触パッドの間隔に対応する間隔で配置されている先端部、を有する。四つのスプリングクランプそれぞれは、リーフスプリング部分の一つに対して調整可能な拘束力が与えられるように、各プローブ羽根部のリーフスプリング部分に接触するカンチレバー型部分を有する。スプリングクランプのそれぞれによって各プローブ羽根部に対してかかる拘束圧力を個別に調整するための、四つの個別のスプリングクランプ調整手段がある。個別のスプリングクランプ調整手段は、各プローブ羽根部上のプローブフィンガー先端部の位置合わせを達成する目的で、スプリングクランプを動かして任意の望ましい方向に向けることができるように、それぞれが三本のねじとスプリングワッシャとによって枠部材に取り付けられているスプリング式の台を有する。」
【0014】
特許文献11と、特許文献12には、テストするベアICまたはベアウエハとの接続を確立するためのテストプローブ構造が開示されている。この構造は、「必要な接続を確立する目的で下側に一列のマイクロバンプを有するMCM-D型基板を先端に担持している多層プリント回路プローブアーム、を有する。プローブアームは、デバイスまたはウエハの表面に浅い角度で支持されており、MCM-D型基板は、テスト対象のデバイスとのインタフェースとして機能するための必要な受動構成要素を有するように形成されている。テスト対象のデバイスの各辺に一本ずつ、四本のこのようなプローブアームを設けることができる。
【0015】
特許文献13、特許文献14、特許文献15、特許文献16、特許文献17、および特許文献18には、「半導体ウエハからダイが切り離される(分離される)前に、弾性接触子構造が半導体ダイ上のボンドパッドに直接実装される。これによって、表面上に複数の端子が配置されている回路基板などを有する半導体ダイと接続することによって、半導体ダイの動作試験を行う(例:テストする、および/またはバーンインする)ことが可能となる。その後、半導体ダイを半導体ウエハから切り離し、その後、同じ弾性接触子構造を使用して、半導体ダイと別の電子コンポーネント(例えば、配線基板、半導体パッケージなど)との間の相互接続を有効にすることができる。本発明の、すべて金属から成る相互接続複合要素を弾性接触子構造として使用することにより、バーンインを少なくとも150℃の温度において実行することができ、かつ、60分以内に完了することができる。」と開示されている。B. Eldridgeらによって開示されているこの接触子端構造では、弾性接触子構造が提供されるが、この構造は、半導体ダイ上のボンドパッド上にそれぞれ個々に実装されるため、複雑かつコストのかかる製造が要求される。さらに、この接触子端構造はワイヤから作製されており、このことによって、接触子の先端の幾何学形状がしばしば制約される。さらに、このような接触子端構造は、ピッチの小さなアプリケーション(例:一般には、周辺部型プローブカード(peripheral probe card)の場合の50 μmのオーダーの間隔、または、エリアアレイの場合の75 μmのオーダーの間隔)のニーズを満たすことができていない。
【0016】
特許文献19には、「表面実装式の、下側はんだソケット(solder-down socket)によって、半導体パッケージなどの電子コンポーネントを、取り外し可能な状態で回路基板に実装することが可能となる。弾性接触子構造が、支持基板の上面から延在しており、はんだボール(または他の好適な)接触子構造が、支持基板の下面上に配置されている。支持基板の上部に配置される弾性接触子構造として、相互接続複合要素が使用される。好適な方法においては、支持基板の上部の弾性接触子構造のうちの選択された一つが、支持基板を介して、支持基板の下面上の接触子構造のうちの対応する一つに接続されている。LGA型の半導体パッケージを受け入れるように意図されている実施例においては、弾性接触子構造と、半導体パッケージの外部接続ポイントとの間に、支持基板の上面にほぼ垂直な接触力による圧力接点が形成される。BGA型の半導体パッケージを受け入れるように意図されている実施例においては、弾性接触子構造と、半導体パッケージの外部接続点との間に、支持基板の上面にほぼ平行な接触力による圧力接点が形成される。」ことが、開示されている。
【0017】
上記以外の新しい技術においては、薄膜工程、またはMEMS(微小電気機械システム)工程などのロット型工程において作製される、スプリングによるプローブチップが開示されている。
【0018】
特許文献20、特許文献21、および特許文献22には、フォトリソグラフィによってパターニングされるスプリング接触子が開示されており、この接触子は、「基板上に形成されており、二つのデバイス上の接触パッドを電気的に接続する。また、このスプリング接触子は、熱的および機械的な変動、その他の環境的要因の補正も行う。スプリング接触子における固有な応力勾配に起因して、スプリングの自由部分が基板から上に曲がって離れる。固定部分は、基板に固定されたままであり、基板上の第一接触パッドに電気的に接続されている。スプリング接触子は、弾性材料から成り、自由部分が第二接触パッドに対応的に (compliantly)接触し、これによって、二つの接触パッドが接触する。」 Smithらによって開示されている、フォトリソグラフィによってパターニングされるスプリングは、ICプロービングの多くのニーズを満たすことができるが、スプリングが小さく、現在の多くのICプローバシステムを高い信頼性で動作させるうえで必要な平面性の対応性に対処するための垂直対応性がほとんど提供されない。多くのプロービングシステムでは、垂直対応性は、一般には0.004”~0.010”のオーダーであり、多くの場合、タングステン針プローブを使用する必要がある。
【0019】
さらには、最大で数千本のピンを含むプローブをテスターと相互接続し、その一方で平面性の要件に効果的に対応する方法は、過去において開示されていない。先進の集積回路デバイスが複雑化し、同時に小型化が進むに伴い、そのようなデバイスと高い信頼性で相互接続する目的で使用できるプローブカードアセンブリを提供することは、有利であろう。
【0020】
プローブ先端のアレイとテスト対象ウエハ上の表面パッドとの間の平面性の差異に対応する目的で、中心の周りに少しだけ自由に回転できるプローブ基板を提供することは有利であろう。しかしながら、そのようなシステムでは、依然として、基板をX、Y、およびθ方向には位置的に安定に保持しながら、接点を密着させるための正確に制御される力を与えなければならない。さらに、基板の裏側から多数の(例:数千の)ワイヤまたは信号が出ており、支持部が基板の周辺部に位置しているアプリケーションの場合、これらの支持部材がファンアウトの出口経路を妨げてはならない。さらに、信号ワイヤが、基板の回転を妨げてはならず、かつ、テスト対象デバイス(DUT)にスプリングを密着させる目的で制御的に印加される力を妨げてもならない。
【0021】
多数のピン数と、小さなピッチと、費用効率の高い製造と、カスタマイズ可能なスプリング先端とが可能である、改良された可曉性プローブスプリングの方法および装置を提供することは有利であろう。また、そのような可曉性プローブスプリングを使用するプローブカードアセンブリであって、テスト対象および/またはバーンイン対象の半導体デバイスとの平面性の対応性を提供し、その一方で、軸方向およびθ方向に正確に位置決めすることのできる、プローブカードアセンブリを提供することも有利であろう。
【0022】
同様に、集積回路パッケージは、集積回路チップ44がテストシステムの残りの部分のインタフェースとして機能できるように、集積回路チップICとマザーボードとの間のパワー信号および伝送信号の接続を提供する。
【0023】
マイクロプロセッサデバイスは、今日のICパッケージの制約が最も重大に影響するICデバイスである。今後のマイクロプロセッサは、10,000以上のI/Oを必要とし、20 GHz以上で動作するであろう。
【0024】
従来のICパッケージにおいては、信号、パワー、および接地の接続は、一般にはワイヤボンドまたははんだボールのいずれかによって達成される。ワイヤボンドまたははんだボールによる接続を使用する従来のパッケージでは、信号と電力の両方について寄生が生じ、これらは性能に影響する。現在のマイクロプロセッサはクロック周波数が2 GHz以上であるが、近い将来には20 GHz以上の周波数に進歩するであろう。現在のワイヤボンド技術とはんだボール技術では、20 GHz範囲における信号を維持することができない。
【0025】
ワイヤボンドまたははんだボールによる接続を使用するパッケージでは、信号、電力、および接地の寄生が生じ、これらは性能に影響する。従って、新しいソリューションが必要である。Intel社のBBUL(バンプレスビルドアップレイヤ)パッケージング技術などの先進のパッケージでは、マイクロプロセッサ上にパッケージを構築し、これにより、このような寄生を減少させることができる。また、BBULパッケージングは、同じパッケージ内に複数のチップを密に結合する目的にも使用することができ、これは「チップファースト(chips-first)」またはマルチチップモジュール(MCM)と称される。 BBUL構造に関する詳細は、ASME国際機械工学会議&エクスポ(IMECE)(ニューヨーク、2001年11月12日)においてIntel社のComponents Research研究所によって発表された非特許文献1と、Advanced Metallization Conference(カナダのモントリオール、2001年10月9日)においてIntel社によって発表された非特許文献2に記載されている。
【0026】
ICの上部の高密度のビルドアップ層は、従来のパッケージング手法よりも性能がずっと良好である。チップとの相互接続が、はんだバンプより短く、ワイヤボンドよりずっと短く、この結果、インダクタンスが大幅に低い。キャパシタをICのより近くに位置させることができ、これにより、良好なパワー供給が可能になる。信号距離が短いため、ICをより低い電圧で動作させることが可能であり、電気的クロストークが低減し、電力消費量も減少する。高密度相互接続(HDI)によって、シリコンからの相互接続の数をはんだバンプまたはワイヤボンドの場合よりも多くすることができる。多くの場合、HDIの相互接続を通じて送られる信号の遅延およびクロストークは、ICの相互接続を通じて送られる信号の遅延よりも小さい。信号をICの相互接続の代わりにHDIの相互接続を通じて送ることによって、HDIにおいては伝播遅延とクロストークがICの場合よりも良好であるため、より高い性能を達成することができる。BBUL パッケージは従来より薄く、同じパッケージ内に複数のICを埋め込むことが可能である。
【0027】
このBBUL手法およびMCM手法は、製造が複雑であることと、モジュール内の各チップが良好であることを保証する必要があることとに起因して、非常に高価となる傾向にある。チップ間のHDIの製造において不良なチップまたは欠陥が生じると、すべてのチップとそのパッケージが不合格となる。不良なチップとは、性能要件を満たしていないチップである。従って、BBUL/MCM手法によるパッケージは、ICを個々にパッケージングする場合よりも、一般には大幅に高価である。過去においては、この「チップファースト(chips-first)」手法は、コストが上昇してもサイズと重量の小型化が求められる宇宙用途の衛星に使用されるMCMを構築する目的でしか使用されない。
【0028】
集積回路を取り付ける前にテストできるパッケージを提供することは、有利であろう。このようなパッケージは、一つの大きな技術的進歩であろう。さらには、複数の電気配線層を有するスルーホールが設けられており、かつ、高度な高密度インタフェース(HDI)機能(例えば、フリップチップパッケージまたはワイヤボンドパッケージにおいて達成可能であるよりも高密度のI/O接続、あるいは、より薄いパッケージの中でのICとの高い相互接続性能)が提供される、パッケージ、を提供することは有利であろう。このようなパッケージは、さらなる技術的進歩であろう。

【発明の開示】
【課題を解決するための手段】
【0029】
MEMS方式で製造されるプローブと薄膜方式で製造されるプローブの両方のタイプのスプリングプローブ構造を、半導体ウエハ上の一つ以上の集積回路をテストする目的に使用できるように、これらのタイプのプローブの機械的な対応性が拡張されている、強化された集積回路プローブカードアセンブリのいくつかの実施例が開示されている。プローブカードアセンブリのいくつかの実施例として、信号パッドのピッチが密であるときの対応性が提供される、および/または、市販のウエハプロービング装置における高いレベルの並列テストが可能となる、実施例が開示されている。いくつかの好ましい実施例においては、プローブカードアセンブリ構造は、分離可能な標準的な構成要素を含んでおり、これによって、アセンブリの製造コストと製造時間とが低減される。これらの構造およびアセンブリでは、ウエハ形態における高速テストが可能である。これらのプローブには、基板上のMEMSまたは薄膜方式で製造されるスプリング先端およびプローブレイアウト構造と、集積回路の両方に対する機械的な保護手段も組み込まれている。代替のカードアセンブリ構造は、プローブチップ基板に接着式に取り付けられるデカル(decal)またはスクリーンなど、対応性を持つ保持器の構造を有する。
【発明を実施するための最良の形態】
【0030】
図1は、基板16から離す前の、フォトリソグラフィによってパターニングされたスプリング14a~14nの直線状アレイ12の平面図10である。導電スプリング14a~14nは、一般には、半導体業界において周知であるように、低エネルギおよび高エネルギのプラズマ蒸着工程の後、フォトリソグラフィによってパターニングするなどにより、連続する蒸着金属層によって基板層16の上に形成される。連続する層は、それぞれ固有の応力レベルが異なる。次いで、基板16の引き離し(release)領域18にアンダーカットエッチング(undercut etching)処理を施し、これによって、蒸着金属層間の固有の応力の結果として、スプリング接触子14a~14nのうち、引き離し領域18上の引き離し層(図3)に位置している部分が、基板16から離れるように、基板16から離れるように突出する(すなわち曲がる)。蒸着金属トレースの固定領域 15(図3と図4)は、基板16に付着したままであり、一般には、スプリング接触子14a~14nからの配線(すなわちファンアウト処理)に使用される。図2は、基板16から離れた後の、フォトリソグラフィによってパターニングされたスプリング14a~14nの直線状アレイ12の透視図22である。スプリング接触子14a~14nは、微細なピッチ20の高密度アレイに形成することができ、このピッチは現在では0.001 inchのオーダーである。
【0031】
図3は、フォトリソグラフィによってパターニングされた短い長さ28aの第一スプリング14の側面図26aであり、この場合、パターニングされたスプリング14が基板16の引き離し領域18aから離れて平面的な固定領域から離れた後に、第一有効スプリング角度30aと、スプリング半径31aと、スプリング高さ32aとが画成されるように形成されている。図4は、フォトリソグラフィによってパターニングされた、長いスプリング長さ28bの第二スプリング14 の側面図26bであり、この場合、パターニングされたスプリング14が基板16の引き離し領域18bから離れた後に、第二の大きな有効スプリング角度 30bと、スプリング半径31bと、スプリング高さ32bとが画成されるように形成されている。形成されるスプリング先端14の有効形状は、目的のアプリケーションに基づいて大幅にカスタマイズ可能である。さらに、このスプリング先端は、一般にはフレキシブルであり、このため多くのアプリケーションに使用することができる。
【0032】
パターニングされたプローブスプリング14は、スプリング間のピッチ20を非常に小さくすることができ、これによって、複数のプローブスプリング14を、集積回路デバイス44(図13)のパワーパッドまたは接地パッドに接触させる目的に使用することができ、これにより電流容量が向上する。さらに、プローブスプリング14のアレイ12を有するプローブカードアセンブリの場合、テスト対象集積回路デバイス44(DUT)のI/Oパッドをプロービングする目的に、複数のプローブスプリング14を使用することができ、従って、テスト対象ウエハ92にスプリング接触子14を密着させた後、すべての接触子14の導通性を確認することができ、これによって、テスト手順を開始する前に、プローブカードアセンブリとデバイス44との間の完全な電気的接触が確保される。
【0033】
小型スプリング用の改良された構造。図5は、交互配置型スプリング先端パターンを有する、フォトリソグラフィによる対向するスプリング34a,34bの第一透視図であり、スプリングが基板から離れる前の図である。図6は、フォトリソグラフィによる対向する交互配置型スプリング34a,34bの透視図であり、スプリングが基板から離れた後の図である。
【0034】
フォトリソグラフィによる交互配置型スプリング34a,34bのそれぞれは、複数のスプリング接点24を有する。スプリング接触子が集積回路デバイス44 のパワーまたは接地のトレース46もしくはパッド47との接続に使用されるときには、この接点に大きな電気抵抗が生じる。従って、複数の接点24を有する交互配置型スプリング接触子34では、スプリング接触子34およびトレース46またはパッド47との間の抵抗が本質的に減少する。上述されているように、複数の交互配置型プローブスプリング34は、集積回路デバイス44またはプローブカードアセンブリ60の高品質の電気的接続用など、多くの用途に使用することができ、例えば、テスト時に集積回路デバイス44をプロービングする目的などである。
【0035】
図7は、テスト対象集積回路デバイス(DUT)44の一本のトレース46に接触している、フォトリソグラフィによる対向する対の交互配置型スプリング 34a,34bの透視図42である。この対の交互配置型スプリング接触子34a,34bでは、それぞれが複数の接点24を有するスプリング34a,34b の両方が、同じトレース46に接触することができる。図5に示されているように、基板16上の二本のスプリング34a,34bの間にジグザグ状の隙間38 が形成されて、各スプリング34a,34bに複数の先端24が確立される。交互配置型スプリングプローブ34a,34bが基板16から離れる前は、交互配置型接点24は、重なり合っている交互配置領域36の中に位置している。交互配置型スプリングプローブ34a,34bが基板16から離れたとき、交互配置型スプリング接点24は、スプリング34a,34bの間に定義される接触領域40内で互いに近接したままである。従って、テスト対象デバイス44の場合などでは、交互配置型スプリングプローブ34a,34bの両方が同じトレース46に接触して高い信頼性が提供されるように、対の交互配置型スプリング接触子 34a,34bを位置させることができる。さらに、交互配置型スプリング34a,34bのそれぞれが複数のスプリング接点24を含むため、トレース46との接触が増し、その一方で、過熱する、あるいは複数の接点間に電流アークが生じる可能性が最小である。
【0036】
図8は、フォトリソグラフィによる平行かつ対向する単一ポイント型スプリング14の上面図であり、スプリング14が基板16から離れる前の図である。交互配置型スプリング34a,34bの場合について上述されているように、平行なスプリング14も、複数のスプリングのスプリング先端24がデバイス44の一本のトレース46に接触するように配置することができる。さらに、引き離し領域18において基板16から離れたときにスプリング先端24が互いに近接位置にあるように、対向するスプリングプローブ14を基板16上に互いに重ねることができる。図9は、フォトリソグラフィによる平行かつ対向する単一ポイント型スプリング14の上面図であり、スプリング14が基板16から離れた後の図であり、この場合、フォトリソグラフィによる平行かつ対向する単一ポイント型スプリング14が、集積回路デバイス44上の一つのパッド47に接触している。
【0037】
図10は、ショルダー54から突き出しているポイント52を有する、フォトリソグラフィによるショルダーポイント型スプリング50の正面図である。図11 は、フォトリソグラフィによるショルダーポイント型スプリング50の一部断面の側面図であり、集積回路デバイス上のトレース46に接触した状態の図である。図12は、フォトリソグラフィによる複数ショルダーポイント型スプリング50の透視図である。一般に、単一ポイント型スプリングプローブ14は、多くの場合、トレース46またはパッド47上に存在する酸化物層を一つの尖ったプローブ先端24によって貫通することによって、集積回路デバイス44の導電トレース46と物理的に良好に接触する。しかしながら、トレース46またはパッド47が薄い、または相対的に柔らかい半導体ウエハ92または集積回路デバイスの場合、一つの長いプロープ先端24が、トレース46の厚みを超えてIC基板48またはその他の回路内まで貫通することがある。
【0038】
従って、フォトリソグラフィによるショルダーポイント型スプリング50は、一つ以上の突き出したポイント52と、ショルダー54とを含んでおり、ポイント 52は、貫通することによってトレース46との良好な電気接点を形成し、ショルダー54は、スプリング50がデバイス44またはウエハ92に深く貫通しすぎることを防止する。プローブスプリング50の形状は、フォトリソグラフィによるスクリーニングとエッチング工程とによって大幅に制御できるため、フォトリソグラフィによるショルダーポイント型スプリング50の詳細な幾何学形状が容易に達成される。
【0039】
改良されたプローブカードアセンブリ。図13は、プローブカードアセンブリ60aの断面図58であり、複数の導電プローブ先端61a~61nが、基板16の下側のプローブ面62aに配置されている。複数のフレキシブル導電接続部64a~64nは、基板16の上側のコネクタ面62bに配置されており、それぞれ、対応する電気接続部66a~66n によって複数の導電スプリングプローブ先端61a~61nに接続されている。
【0040】
基板16は、一般には固体の板であり、セラミック、セラミックガラス、ガラス、またはシリコンなど、熱膨張係数(TCE)が小さい材料であることが好ましい。導電スプリングプローブ先端61a~61nは、プローブカードアセンブリ60aと半導体ウエハ92とが一体に配置されているときに、プローブカードアセンブリ60と半導体ウエハ92との間の電気的接触を確立する。
【0041】
スプリングプローブ先端61a~61nは、単一ポイント型スプリング14、交互配置型スプリング34、またはショルダーポイント型スプリング50など、様々な先端形状を有してよく、一般には、薄膜方式またはMEMS製造方式を使用して基板16上に作製され、小さな製造コストと、良好に制御された均一性と、非常に微細なパッドピッチ20と、大きなピン数とが達成される。
【0042】
プローブ先端61a~61nは、好ましくは基板16内の金属被覆されたビア66a~66nを通じて、フレキシブル電気接続部64a~64nに電気的に接続されている。複数のフレキシブル電気接続部64a~64nのそれぞれは、プリント基板プローブカード68に電気的に接続されており、このプリント基板プローブカード68は、一般には、金属製のリングまたはフレーム支持構造70によって所定の位置に保持されている。好ましい金属被覆されたビアの電気接続部 66a~66n(例:Micro Substrate社(アリゾナ州テンプ)製の接続部)は、一般には、レーザーまたはその他の穿孔方法を使用して最初に基板16にホールを作成することによって形成する。次いで、めっきまたは押し出し(extrusion)などによって、このホールに導電性材料を満たす、または導電性材料によってメッキする。導電ビア66a~66nが形成された後、一般には、平らでなめらかな表面が得られるようにビアを研磨する。
【0043】
図14は、プローブカードアセンブリ60aの一部を拡大した断面図79であり、基板16およびプリント基板プローブカード68における段階的なピッチおよびファンアウトを示している。プローブ先端61a~61nは、一般には、微細なスプリングピッチ20で基板のプローブ面62aに配置されている。固定トレース部分15は、一般には基板ピッチ81で配置されている金属被覆されたビア66a~66nにファンアウトされている。導電接続部64a~64nは、基板16の上側コネクタ面62bに位置し、ビア66a~66nに接続されており、一般には接続部ピッチ83(これは基板ピッチ81に位置を合わせることができる)で配置されている、または、基板16の上側コネクタ面62bにさらにファンアウトされていることが好ましい。
【0044】
プリント基板プローブカード68の下側の導電パッド77a~77nは、一般には、基板16の上側コネクタ面62bに位置する導電接続部64a~64nと位置が合うようなパッドピッチ85で配置されている。導電パッド77a~77nは、一般にはプローブカードピッチ87で配置されている導電経路 78a~78nにファンアウトされていることが好ましい。導電接続部72a~72nは、プリント基板プローブカード68の上面に位置し、導電経路 78a~78nに接続されており、一般には、プローブカード接続部ピッチ89(これはプローブカードピッチ87に位置を合わせることができる)で配置されている、または、プリント基板プローブカード68の上面にさらにファンアウトされていることが好ましい。プローブカード接続部ピッチ89は、導電接続部 72a~72nが、テストヘッド76上に配置されているテストヘッドコネクタ74a~74n(一般にはテストヘッドピッチ91で配置されている)に位置が合うように選択されることが好ましい。
【0045】
フレキシブル電気接続部64a~64nは、一般には、約4~10 milの対応性が提供されるように、プローブ先端61a~61nより長いスプリング長さ28を使用して製作される。いくつかの実施例においては、フレキシブル電気接続部64a~64nは、上述されているように、あるいは、米国特許第5,848,685号または米国特許第5,613,861号に開示されているように、フォトリソグラフィによるスプリングによる対応性が生じるように構築され、これらの特許文書は本文書に参照文献として組み込まれている。
【0046】
フレキシブル接続部64a~64nは、永久的に(例:はんだまたは導電性エポキシによって)、または非永久的に(例:フレキシブル接続部スプリング 64a~64nの先端24と対になる対応する金属パッドによって)、プリント基板(PWB)プローブカード68に接続されている。プリント基板(PWB)プローブカード68においては、パッド72a~72nに信号がファンアウトされており、これらのパッドは、テストヘッド76上に一般にはテストヘッドピッチ91で配置されている標準的なポゴピン接触子74a~74nに適するパッドピッチ89を有する。
【0047】
フレキシブル接続部64a~64nは、例えば、1.00 mmまたは1.27 mmなどのアレイピッチ83を有するエリアアレイ内に配置されていることが好ましく、この構造では、プリント基板プローブカード68のめっき処理済スルーホール(PTH)78の密度(すなわち、プローブカードピッチ87)が適正なものとなり、ブラインド導電ビア78a~78nが含まれている高度なプリント基板プローブカード68を使用する必要なしに、プリント基板プローブカード68の中の複数の層に信号をファンアウトすることが可能となる。
【0048】
フレキシブル導電接続部64a~64nは、プリント基板プローブカード68の下側の導電パッド77a~77nに接触して、プリント基板プローブカード68 と基板16との間の電気的接続を維持する。その一方で、基板16は、z軸84沿いに上下にわずかに動くことができ、さらに、その中心の周りに傾くことができる。フレキシブル接続部64a~64nは、熱膨張係数がそれぞれ異なる基板16とプリント基板プローブカード68(例:基板16の熱膨張係数が小さく、プリント基板プローブカード68の熱膨張係数が相対的に高い場合など)との間の横方向の対応性も提供する。
【0049】
これに代えて、基板16は、メンブレンバンプ接触子64a~64nを通じてプリント基板プローブカード68に接続されるメンブレンプローブカードなどのアセンブリでもよい。プローブカードアセンブリの代替実施例においては、接続部64a~64nは、分離型コネクタ132(図18)によって形成されているか、または好ましくは、FCI Electronics社(ペンシルベニア州エターズ)のMEG-Array(登録商標)コネクタ162(図24)によって形成されている。これらの場合、コネクタ132,162の対向する上面および下面に配置されているボールグリッドはんだアレイが、図14に見られるような基板16およびプリント基板プローブカード68上の対応する導電パッドにはんだ付けされており、かつ、コネクタ132,162の対向する二分割部分が、複数のスプリングプローブ先端 61a~61nのそれぞれと、プリント基板プローブカード68の下側の複数の導電パッド77a~77nのそれぞれとの間の複数の対の電気的接続を提供するように、導電パッドそれぞれがエリアアレイパターン内に配置されている。
【0050】
集積回路デバイス44のサイズがますます小さくなり、その設計がますます複雑になるにつれて、小型スプリングプローブ先端61a~61nによって形成される微細ピッチ20(図2)が、ますます重要になる。さらに、集積回路44と必要なプローブカードテストアセンブリの両方の小型化に伴って、多数のスプリングプローブ61a~61nを含んでいる基板16と集積回路44との間の平面性の差異が重大になる。
【0051】
プローブカードアセンブリ60aは、数千のスプリングプローブ先端61a~61nを含みうる基板16との電気的な相互接続を提供する一方で、一般的な集積回路テストのプロービング環境においてプローブカードアセンブリ60aが効果的に機能するための十分な機械的支持を提供する。プローブカードアセンブリ 60aは、非常に多数のピン数、狭いピッチ、または高い周波数を必要とするアプリケーションに容易に使用される。さらに、プローブカードアセンブリ60a は、集積回路ダイ44の中央領域へのアクセスが要求されるテストプローブアプリケーションの場合に、集積回路デバイスのすべてのトレース46(図7)と入力および出力パッド47(図7、図9)との電気的接触が提供されるように、容易に適合化される。
【0052】
プローブカードアセンブリ60aは、一般的には、図13に示されているように、一般にソー経路(saw streets) 94によって隔てられている一つ以上の集積回路44を有する半導体ウエハ92との関係において配置される。X軸80とY軸82は、一般に、半導体ウエハ92またはデバイス44上の横方向におけるプローブカードアセンブリ60の位置を定義するのに対し、Z軸は、ウエハ92の表面とプローブカードアセンブリ60との間の垂直距離を定義する。テストヘッド76とプローブカードアセンブリ60aとの関係におけるテスト対象ウエハ92の位置は、X軸80、Y軸 82、およびZ軸84と、Z軸84の周りのZ軸回転(すなわちθ)位置90に関して正確に位置決めする必要がある。
【0053】
しかしながら、半導体ウエハ92とプローブカードアセンブリが、X軸回転86および/またはY軸回転88におけるわずかな変動などによって、互いに平面性がわずかにずれている場合にも、プローブカードアセンブリが平面的な半導体ウエハ92との接触を提供できるようにすることが、ますます重要である。
【0054】
図13に示されているプローブカードアセンブリ60aにおいては、プローブ先端61a~61nはフレキシブルであり、これによって、基板16と半導体ウエハ92との間の平面性の対応性が内在的に提供される。さらに、フレキシブル接続部64a~64nは、同じくフレキシブル導電スプリング14, 34, 50であることが好ましく、基板16と半導体ウエハ92との間の平面性のさらなる対応性を提供する。従って、プローブカードアセンブリ60aは、基板16と集積回路デバイス44との間の平面性の対応性(すなわちX軸回転86および/またはY軸回転88などによる対応性)を提供する。さらに、プローブカードアセンブリ60aは、基板16(一般にはセラミック、セラミックガラス、ガラス、またはシリコンから成る)と、プリント基板プローブカード68(一般にはガラスエポキシ材料から成る)との間の熱膨張係数(TCE)の差異にも対応する。
【0055】
一般には小さなピッチ20を有するプローブ先端61a~61nからの信号トレースは、基板16の面62a,62bの片面または両面上の配線トレースを使用して、一般には大きなピッチを有するフレキシブル接続部64a~64nにファンアウトされていることが好ましい。
【0056】
フレキシブル接続部64a~64nは、標準化されたレイアウトパターンに配置されていることが好ましく、標準化されたレイアウトパターンは、プリント基板プローブカード68上の電力および接地の標準化されたパッドパターン(すなわちアサインメント)に合致させることができる。従って、同じプリント基板プローブカード68を、複数の異なる集積回路デバイス44と対になるように配置されている基板16に対して使用することが可能になる。プリント基板プローブカード68は、様々な異なるデバイス44をテストするための専用の基板16用に適合させることができるため、プリント基板プローブカード68の運用コストが低減する。
【0057】
高周波パワーデカップリング(power decoupling)において役立てる目的で、AVX社(サウスカロライナ州マートルビーチ)のLICA(登録商標)シリーズのキャパシタなど、キャパシタ172(図24)が、基板16の上面62b上に実装されていることが好ましい。これに代えて、基板16の中で、配線トレース層の未使用領域上に形成される平面と基準面との間に、平行板キャパシタを形成することができる。基板16がシリコンから成る実施例の場合、シリコン基板16の中に内蔵キャパシタ (integral capacitor) 67(例:内蔵バイパスキャパシタなど)を形成することが好ましい。このようなキャパシタは、集積回路の場合のように、金属-誘電体-金属の構造か、金属-誘電体-高濃度ドーピングされた半導体、またはp-n接合を含んでいる。非半導体基板の場合、金属-誘電体-金属構造のキャパシタを、集積回路製造手法を使用して基板上または基板内に作製することができる。
【0058】
一般には、ウエハチャックと基板16とを位置合わせする目的で、見上げ/見下ろし型カメラ(look up and look down camera)が使用され、このカメラによって、半導体ウエハ92上に位置するテスト対象デバイス44の接触パッド47またはトレース46に、プローブ先端20が位置合わせされる。この位置合わせは、一般には、スプリング先端24か、または基板16に印刷されている位置合わせマーク125を見ることによって達成される。
【0059】
このようなカメラを持たないプローブカードアセンブリの場合、基板16は半透明または透明な材料(例:ガラスセラミックまたはガラス)から成ることが好ましく、これによって、テストオペレータは、上から見て位置合わせする方法(view-through-the-top alignment methods)を実行することができる。ウィンドウ165(図24)は、プリント基板プローブカード68に画成されていることが好ましいが、位置合わせマーク125(図17)および185(図26)は、基板および/またはテスト対象ウエハ92上に位置していることが好ましい。この場合、テストオペレータは、カメラまたは顕微鏡を使用してウィンドウを通じて位置合わせマーク125を見て、基板16とウエハ92を位置合わせする。
【0060】
プローブの接触が維持されている間、半導体ウエハ92の表面へのアクセスが要求されるアプリケーションの場合(例:集積回路44の製造時に電圧コントラスト式電子ビームプロービングを行う場合など)、ICの中心の基板領域16にウィンドウ123(図17)が画成されていることが好ましく、これによって、ダイ92における信号を観察するためのアクセスが可能になる。ウィンドウ123は、ダイの縁部に沿ってI/Oパッドが配置されている集積回路デバイス44の場合に最適に機能し、ウエハ92上に位置する集積回路デバイス44の直接的なプロービングが可能になる。現在のところ、半導体ウエハダイ92を最初に切断しなくてはならず、個別の集積回路デバイス44をパッケージにワイヤボンドした後、テストが行われる。
【0061】
基板16に画成される開口(すなわち、ウィンドウ123)は、DRAMなどのデバイスを現場で電子ビームによって修復する目的にも使用されることが好ましく、この場合、プローブカードアセンブリ60は所定の位置のままでよい。従って、テスト、修復、再テストを、ウエハ92を動かすことなく同じステーションにおいて実行することができる。
【0062】
プローブカードアセンブリ60aの構造では、プローブ先端61a~61nの間の電気的距離が非常に短く、かつ、プリント基板プローブカード68内に、インピーダンスが制御された環境が形成され、これによって、プローブカードアセンブリ60aを高周波アプリケーションに使用することが可能になる。基板16の面62a, 62bの一方または両面上のトレースをインピーダンス制御する必要のある実施例の場合、基板16の中のトレースの上、トレースの下、またはトレースの上下に一つ以上の導電基準面を追加することができる。超高周波アプリケーションの場合、遮蔽された同軸伝送ライン環境260を効果的に形成する目的で、ビア 266(図37、図38)を使用して一定の間隔で基準面262(図37、図38)の間に接続されている交互接地基準トレース(alternating ground reference traces)を、基板16に含めることができる。いくつかの実施例においては、接地面トレースが信号ラインの片側に配置される。
【0063】
高い対応性のプローブアセンブリ。上述されているように、プローブカードアセンブリ構造60では、横方向であるXおよびY方向と、Z軸84の周りの回転90に関して、プリント基板プローブカード68に対して基板16が固定的に支持される。
【0064】
フレキシブルスプリングプローブ61a~61nと、フレキシブル接続部64a~64nとによって、プローブカードアセンブリ60と半導体ウエハ92またはデバイス44との間のある程度の平面性の対応性が提供されるが、プローブカードアセンブリ60の別の好ましい実施例では、より大きな平面性の対応性が提供される。
【0065】
高密度の接続と微細なピッチ20とを提供する目的で、プローブスプリング61a~61nは極めて小さいことが要求されることが多いため、相当に大きな平面性の対応性が要求されるプローブカードアプリケーションにおいては、プローブスプリング61a~61nのみによって提供される対応性では十分でないことがある。従って、プローブカードアセンブリ60のいくつかの好ましい実施例においては、テスト対象半導体ウエハ92に対する平面性のより大きな対応性を提供する目的で、プローブカードアセンブリ60においては、基板16がその中心の周りに回転する(すなわち、 X軸回転86および/またはY軸回転88において変動する)ことができる。このような用途においては、基板16の下面62a上に位置するプローブスプリング接触子61a~61nを半導体ウエハ92に密着させるための、Z方向84における下向きの制御された力が、プローブカードアセンブリ60によって依然としてかけられる必要がある。
【0066】
プローブカードアセンブリ60の多くの実施例の場合、基板16の中央領域119(図17)は、基板16とプリント基板プローブカード68との間の電気接続部64a~64n用に使用され、従って、基板16はその周辺部127(図17)に沿って支持される必要がある。
【0067】
基板16がその中心の周りに回転できるようにすることと、プローブ先端61a~61nを密着させる力を印加することとを目的として、プローブカードアセンブリの中央領域内、基板支持構造の裏側に、玉継手支持構造(ball joint fulcrum structure)を配置することができる。しかしながら、このような構造では、一般にはワイヤリードまたはその他の電気接続部が妨害され、プローブカードアセンブリの中央領域の上に出す必要がしばしば生じる。さらに、このような可動結合では、一般に、基板16のθ回転90が確実には抑制されない。
【0068】
図15は、ブリッジ/リーフスプリング懸架型プローブカードアセンブリ60bの第一部分断面図96aである。図16は、図15に示されている、ブリッジ/リーフスプリング懸架型プローブカードアセンブリ60bの第二部分断面図96bであり、このアセンブリは、プローブカードアセンブリ60bと同一平面にない。可能性のある半導体ウエハ92上の一つ以上の集積回路デバイス44との平面性の対応性を提供している。図17は、ブリッジ/スプリング懸架型プローブカードアセンブリ60bの一部を拡大した組立図124である。
【0069】
リーフスプリング98は、ブリッジ構造100を通じて基板16に結合されている。リーフスプリング98とブリッジ構造100とによって、基板16に回転の自由(すなわちわずかなX軸回転86とY軸回転88)が与えられ、Z方向84、X方向80、Y方向82、およびZ軸回転(θ)方向90における動きは制御されている。好ましい実施例においては、プリント基板プローブカード68bに対する基板16の初期面およびZ位置を正確に設定することと、リーフスプリング98の予荷重力(pre-load force)を設定することとを目的とする手段として、予荷重アセンブリ121(図15)が使用される。例えば、図15および図16に示されている実施例においては、予荷重アセンブリ121は、ブリッジシム122と組み合わせて使用される締結具118を有する。代替実施例においては、予荷重アセンブリ 121は、キャリブレーションねじアセンブリ122、緩衝パッド120、および/またはその他の支持器116を有することができる。
【0070】
図15および図16に示されているように、リーフスプリング99の外縁部は、プリント基板プローブカード68に、その外縁部に沿って取り付け枠107によって固定されている。リーフスプリング98の中心は、一つ以上の締結具108と、上側ブリッジスペーサ104と、下側ブリッジスペーサ106とによって、ブリッジ100に結合されている。リーフスプリング98とブリッジ100との間のZ距離を変える(これによってリーフスプリング98によってブリッジ 100にかかる下向きの力の予荷重が変化する)などを目的として、ブリッジ予荷重シム110が追加されていることが好ましい。ブリッジ100は、中央からの支持を隅に移し、複数の(一般には三個以上)のブリッジ脚102によって基板16に結合されている。ブリッジ脚102は、プリント基板プローブカード 68に画成されている脚開口111を通って突き出しており、接着性または機械的な結合112などによって基板16に固定的に取り付けられている。
【0071】
リーフスプリング98は、一般にはステンレス鋼またはスプリング鋼の板から作製され、一般には化学的エッチング法を使用してパターニングされる。下向きの力は、スプリングの剛性と、スプリングスペーサ104,106の直径と、リーフスプリング98のサイズの関数である。
【0072】
図16に示されているリーフスプリング98は十字形状であるが、下向きの力、傾斜の自由、X、Y、およびθ方向の並進運動に対する抑制を提供する目的で、別の幾何学形状を使用してもよい。例えば、十字形状のリーフスプリング98に、任意の数の羽根99を含めることができる。さらに、羽根99は、外縁部から中心に向かって幅が変化する非対称形状でもよい。また、リーフスプリング98の安定性をさらに高める目的で、リーフスプリング98の外縁部をリング内に結合することができる。
【0073】
ブリッジ/リーフスプリング懸架型プローブカードアセンブリ60bの可動質量を最小にする目的で、ブリッジ100とスペーサ104,106は、アルミニウムまたはチタンなど軽量かつ強度のある金属から成ることが好ましい。
【0074】
基板16は、一般には、エポキシまたははんだなどの接着剤112を使用して、ブリッジ100の脚102に取り付けられている。基板を交換できることが必要である場合、図18に示されているものなど、取り外し可能な結合器130を使用することができる。
【0075】
基板16の下側62aには、基板16がテスト対象ウエハ92に接触することを防止する下側支持器114が使用されることが好ましい。この下側支持器114は、テスト対象の半導体ウエハ92への損傷を回避するため、ポリイミドなど相対的に柔らかい材料から作製されていることが好ましい。さらに、半導体ウエハ92内の能動回路44への損傷もさらに回避するため、下側支持器114は、プローブカードアセンブリ60が半導体ウエハ92上のデバイス44と位置が合っているときに、この支持器が半導体ウエハ92上のソー経路94(図13)(ここには能動デバイス44もテスト構造も存在していない)と位置が合うように配置されていることが好ましい。さらに、下側支持器114の高さは、スプリングプローブ61a~61nの最大圧縮が制限され、従ってスプリングプローブ61a~61nの損傷が防止されるように選択されることが好ましい。
【0076】
基板16の上面62bにも、上側のフレキシブル電気接続部64a~64nの損傷を防止する目的で、上側支持器116が使用されることが好ましい。この上側支持器116は、LEXAN(登録商標)、シリコン、またはプラスチックなど、中程度の堅さの絶縁性材料から作製されていることが好ましい。
【0077】
図15、図16、および図17に示されている好ましい実施例においては、基板16の初期面を設定することと、フレキシブル接続部64a~64nが過度に伸びることによって損傷しないように、下向きの停止機構を基板16に提供することとを目的として、調整可能なブリッジねじ118とブリッジシム122とが使用されている。
【0078】
プリント基板プローブカード68bは、一般には相対的に柔らかい材料(例:ガラスエポキシなど)から作製されているため、接触が何度も繰り返される中で調整ねじ118の先端がプリント基板プローブカード68bの中に沈むことを防止する目的で、プローブカード68b上の調整ねじ118の下に緩衝パッド120 が配置されていることが好ましい。また、基板16とプリント基板プローブカード68bとの間の初期距離および平面性を正確に設定できるように、調整ねじ 118と一緒に締結具シム122も使用されることが好ましい。
【0079】
リーフスプリング98によってブリッジ100にかかる下向きの力の初期予荷重を制御する目的で、予荷重シム110が使用されることが好ましい。設定される予荷重によって、基板16の振動が防止され、テスト対象半導体ウエハ92と基板16との間の接触特性が向上する。
【0080】
図18は、プリント基板プローブカード基板68bに脱着式に接続されている中間ドーターカード134を有し、かつ、スプリングプローブ基板16が脱着式にブリッジ構造100に結合されている、代替のブリッジ/リーフスプリング懸架型プローブカードアセンブリ60cの第一部分断面図126aである。図19 は、図18に示されている代替のブリッジ/リーフスプリング懸架型プローブカードアセンブリ60cの第二部分断面図126bである。このアセンブリは、最初の時点ではプローブカードアセンブリ60cと同一平面にない、半導体ウエハ92上の一つ以上の集積回路デバイス44との平面性の対応性を提供している。
【0081】
図18に示されている代替のブリッジ/リーフスプリング懸架型プローブカードアセンブリ60cにおいては、ドーターカード134との電気的接続を提供するフレキシブル接続部64a~64nなどの電気接続部が、プローブチップ基板16とドーターカード134との間に設けられている。
【0082】
図18に見られるように、脱着式の二分割部分133a,133bを有する分離型コネクタ132は、中間ドーターカード134とプリント基板プローブカード基板68bとの間の脱着式の接続を提供し、これによって、基板16とドーターカード134を交換することができる。基板取り付け締結具130(例えば、ねじ、ただしこれに限定されない)は、ブリッジ脚128を貫いて延在することが好ましく、基板16の上面62bに実装されている基板支柱128に、ブリッジ 100を取り外し式に結合することができる。
【0083】
プローブカードアセンブリ60の一つの実施例においては、好ましい分離型コネクタ132は、FCI Electronics社(ペンシルベニア州エターズ)のMEG-Array(登録商標)コネクタである。分離型コネクタ132の片側は、一般にはプリント基板プローブカード68にはんだ付けされているのに対して、対になる側は、一般にはドーターカード134にはんだ付けされ、これによって、ドーターカード134をプリント基板プローブカード68bに取り外し可能な状態に結合することができ、その一方で、信頼性の高い多数の電気的接続が提供される。ドーターカード134には、フレキシブル接続部64a~64nの一般的なピッチである約1 mmから、分離型コネクタ132の一般的なピッチである約1.27 mmへの、電気接続部のさらなるファンアウトが設けられていることが好ましい。
【0084】
図20は、ワイヤ/スプリング支柱懸架型プローブカードアセンブリ60dの断面図136である。複数(例:一般には三本以上)の鋼線138によって、基板 16のZ方向の動き84が可能になる。スプリング支柱枠140は、一般にはプリント基板プローブカード68cにはんだ付けまたは接着されており、一般には、一本以上のスプリング支柱141を含んでおり、この支柱は、Z方向の下向きの力を提供することと、移動距離を制限することとを目的として使用されることが好ましい。
【0085】
図21は、分離型コネクタの二分割部分133a,133bを有する分離型の(すなわち、取り外し可能な)コネクタ132と、支持部149とによってプリント基板プローブカード68に脱着式に結合されている中間ドーターカード134を有する、懸架型プローブカードアセンブリ60eの断面図142である。フレキシブル接続部64a~64nは、スプリング14,34,50から作製されていることが好ましく、プリント基板プローブカード68との電気的接続と、プリント基板プローブカード68とドーターカード134との間の機械的な結合の両方を提供する。プローブカードアセンブリ60eにおいては、フレキシブル接続部64a~64nは、ドーターカード134上の導電パッド143a~143nにはんだまたは導電性エポキシ127のいずれかを使用して永久的に接続されている。フレキシブル接続部64a~64nは、下側のプローブスプリング61a~61nすべてを完全に圧縮する(例えば、2~10 milの範囲に圧縮する)のに必要な力よりも、トータルで大きな力を提供するように設計されていることが好ましい。さらに、フレキシブル接続部 64a~64nは、フレキシブル接続部64a~64nが圧縮されるときに基板16がX方向80、Y方向82、およびZ軸回転(θ)90の方向(図13)に移動しないように、配置されていることが好ましい。
【0086】
フレキシブル接続部64a~64nを保護するため、ドーターカード134に対する基板16のZ軸方向の最大移動距離を制限する目的で、上側基板支持器 116が使用されることが好ましい。この上側基板支持器116は、わずかな予荷重がフレキシブル接続部64a~64nに存在するように調整可能であることも好ましく、予荷重は、基板16をドーターカード134から離すように作用することにより、運用時の基板16の振動およびチャタリングを軽減する。基板 16の振動、動揺、チャタリングを防止する目的で、基板16とドーターカード134との間の一つ以上の位置に減衰材料145(例:ゲルなど)も配置されていることが好ましい。
【0087】
分離型コネクタ132(例:FCI社製コネクタ132など)は、対になる基板の共面性の要件が寛大であることが好ましく、これによって、ドーターカード 134とプリント基板プローブカード68との間の良好な平面性の対応性が提供される。ドーターカード134とプリント基板プローブカード68との間には、一般に、支持部149も使用され、例えば、締結具166、スペーサ164、ナット168、シム170(図24)、および/または接着接合である(ただしこれらに限定されない)。懸架型プローブカードアセンブリ60eのいくつかの実施例においては、支持部149は調整可能である。このことも、プローブチップの平面性の調整を容易にする。
【0088】
図22は、プローブスプリング基板16が分離型アレイコネクタ147によってプローブカード基板に取り付けられている、プローブカードアセンブリ60fの断面図146である。このプローブカードアセンブリ60fは、基板16とテスト対象半導体ウエハ92との間の小さな非平面性をスプリングプローブ61a~61nのみによって吸収できる小さな基板16に適する。
【0089】
図23は、ポゴワイヤ懸架型プローブカードアセンブリ60gの断面図148であり、このアセンブリにおいては、基板16とプリント基板プローブカード68 との間に保持できるように固定されているLGA(large grid array)。インターポーザーコネクタ150によって、ナノスプリング基板16がプリント基板プローブカード基板68に接続されている。一つの実施例においては、LGAインターポーザーコネクタ150は、AMP社(ペンシルベニア州ハリスバーグ)製のAMPIFLEX(登録商標)コネクタである。別の実施例においては、インターポーザーコネクタ150は、W.L. Gore and Associates社(ウィスコンシン州オークレア)製のGOREMATE(登録商標)コネクタである。別の代替実施例においては、プリント基板プローブカード68に存在するポゴピン152を基板16上の電気接続部66a~66nに接続する目的に、ポゴピンインターポーザー150が使用される。基板16 は、複数のポゴ懸架鋼ワイヤ154によって保持されており、このワイヤ154は、わずかに上向きの力がかかるように偏って配置されていることが好ましく、これによって、インターポーザーコネクタ150が保持される一方で、アセンブリ60gの振動とチャタリングが防止される。
【0090】
小テスト面積プローブアセンブリ。図24は、小面積スプリングプローブ基板16に取り付けられるドーターカード134と、メインのプリント基板プローブカード68との間に一つ以上のエリアアレイコネクタを有する小テスト面積プローブカードアセンブリ60hの断面図である。
【0091】
上述されているプローブカードアセンブリ60の多くは、プローブスプリング基板16に大きな平面性の対応性を提供するが、いくつかのプローブカードアセンブリは、テスト対象デバイスの表面積が相対的に小さいアプリケーションに使用される。例えば、少数の集積回路44(例:2個のICなど)を含んでいるウエハ92の場合、対になる基板16のサイズも相対的に小さくすることができる(例:2 cm2以下など)。
【0092】
従って、このような実施例においては、テスト対象ウエハ92に対する基板16の平面性は、表面積が大きい場合よりも重要度が低くなり、プローブスプリング 61a~61nのみによって提供される対応性でテスト環境を十分に補正できることがよくある。プローブスプリング61a~61nによって提供される対応性は、従来の針スプリングと比較して相対的に小さいが、このようなアプリケーションは、フォトリソグラフィによって形成される、またはMEMSによって形成されるスプリングプローブ61a~61nを有するプローブカードアセンブリ60に非常に適している。
【0093】
従って、プローブカードアセンブリ60hは、多層プローブカードアセンブリの設計よりも本質的に複雑度が低く、一般にはコストがより小さい。基板16のコストは基板16の表面積に強く関係するため、基板16のサイズが小さいことにより、プローブカードアセンブリ60hのコストが低減する。
【0094】
プローブスプリング61a~61nは、上述されている薄膜またはMEMS工程方式のいずれかを使用して、堅い基板16の下面62a上に作製されている。プローブスプリング61a~61nからの信号は、表面62a,62bの一方または両面上の金属トレースと、基板16を貫く導電ビア66a~66nとを使用して、基板16の上面62bに位置している金属パッドのアレイ 182,184,186(図26)にファンアウトされている。上側パッドは、一般には0.5 mmなどのアレイピッチにおける一般的なマイクロボールグリッドはんだアレイを使用してドーターカード134に接続されている。ドーターカード134は、ドーターカード134の反対面のピッチ約0.050 inchのパッドに、アレイのピッチをさらに拡大する。ピッチ0.050 inchのパッドアレイをプリント基板プローブカード68に接続する目的には、FCI Electronics社(ペンシルベニア州エターズ)のMEG-Array(登録商標)コネクタなどのエリアアレイコネクタ162が使用されている。低インピーダンスのパワーフィルタリングを提供する目的で、AVX社(サウスカロライナ州マートルビーチ)のLICA(登録商標)キャパシタなどのパワーバイパスキャパシタ172が、基板のマイクロBGAパッド182,184,186の近くに、ドーターカード134に追加されることが好ましい。
【0095】
小テスト面積プローブカードアセンブリ60hは、プリント基板プローブカード基板68とドーターカード134との間の機械的な結合を提供する手段を含むことが好ましい。図24に示されているプローブカードアセンブリ60hの実施例においては、一つ以上のスペーサ164とスペーシングシム170とによって、ドーターカード134とプリント基板プローブカード基板68とが位置合わせされ、間に制御された隔たり距離が形成されており、その一方で、一つ以上の締結具166とナットによって、機械的な取り付け手段が提供されている。図24には、スペーサ164と、シム170と、締結具166と、ナット168の組み合わせが示されているが、小テスト面積プローブカードアセンブリ60hの代替実施例では、ドーターカード134とプリント基板プローブカード基板68とを取り付ける目的で、スプリング式締結具、接着性支持器、またはその他の取り付けハードウェアの組み合わせなど(ただしこれらに限定されない)、手段を任意に組み合わせて使用することができる。
【0096】
一般には(スプリング先端61a~61nを除いて)基板16上の他の形状よりも高さのある下側基板支持器114は、基板16の下面62aに配置されていることが好ましく、また、テスト対象半導体ウエハ92上のソー経路94に一致していることが好ましく、これによって、テスト対象ウエハ92が基板16に衝突することを防止し、かつ、半導体ウエハ92上の能動領域への損傷を防止することが好ましい。
【0097】
プローブカードアセンブリ60hがウエハ92上に位置している状態で半導体ウエハ92へのアクセスが提供されるように(例えば目視による位置合わせ、または電子ビームによるプロービングを目的として)、図24に示されているように、基板16はアクセスウィンドウ123(図17)を含んでいることが好ましく、その一方で、ドーターカード134もドーターカードアクセスホール163 を含んでいることが好ましく、かつ、プリント基板プローブカード68はプローブカードアクセスホール165を含んでいることが好ましい。プローブカードアセンブリ60のいずれにおいても、アクセスホール123,163,165が使用されることが好ましい。
【0098】
図25は、複数のマイクロボールグリッドアレイ型スプリングプローブ接触子チップ基板16が上に配置されている基板ウエハ174の上面図である。表面積 175が小さいスプリングプローブ基板16の場合、一般には、いくつかのスプリングプローブ接触子チップ基板16を一つのウエハ174から作製することができる。例えば、図25に示されているように、標準的な4 inchの丸い開始ウエハ174上に、幅176と長さ178とを有する(例:一辺14 mmの正方形)サイト(site)を24個も形成することができる。さらに、開始ウエハ174上に異なる基板(例:16a,16b)を作製することができ、これによって、マスキングのコストと処理工程のコストなど、複数の異なるスプリングプローブ基板16の製造コスト(これは重要である)を共有することができる。従って、複数の異なる基板16a,16bの製造コストを大幅に(例:最大1/10以下などに)小さくすることができる。
【0099】
図26は、14 mm四方のスプリングプローブ接触子チップ(NSCC)16bの、0.5 mmピッチの一つのマイクロボールグリッドアレイ180の上面図である。マイクロBGAパッド182,184,186は、標準のピッチ(例:0.5 mm)であることが好ましい。外側の5列のパッド182と中央のパッド184とによって341個の信号接続が形成されており、内側の2列186によって 96個の専用のパワー/接地接続が形成されている。スプリングプローブ61a~61nへの配線トレースをカスタマイズすることによって、テスト対象集積回路44に一致させるための固有のパワー/接地スプリングの位置を、一層の配線によって対応することができる。
【0100】
支持器114は、テスト対象デバイス44上の能動デバイス44への損傷を防止する目的で、ウエハ92上の非能動領域(スクライブライン94上など)に一致する位置に配置されていることが好ましい。一つ以上の位置合わせマーク185も、基板ウエハ174上に配置されていることが好ましい。マイクロBGAパッドアレイ180、ドーターカード134、およびプリント基板プローブカード68のフットプリントを標準化することによって、プローブカードアセンブリ60 の製造コストとターンアラウンドタイムを大幅に改善することができる。マイクロBGAパッドアレイ180に加えて、基板16,134,68上に位置するパッドのパワー/接地パッドアサインメントを標準化することによって、ベース基板174におけるビア66a~66nのパターンを標準化することができる。
【0101】
プローブカードアセンブリ60の上記以外の構成要素を標準化することにより、プリント基板プローブカード68(および、いくつかの実施例においてはドータカード134)を、複数の異なる基板16および集積回路デバイス44用に使用できることがしばしばあり、この場合、基板16の配線のみがカスタマイズされる。
【0102】
ビア66a~66nの標準化されたパターンを有する開始基板174(図25)を使用することにより、開始基板174を大量に発注、保管、および使用することも可能になり、従って、開始基板174のコストが低減し、開始基板174が得られるまでのリードタイムもしばしば短縮される。
【0103】
プローブスプリングの代替アプリケーション。フォトリソグラフィまたはMEMSによるスプリングプローブ61,14,34,50は、上記の説明に代えて、ベアダイ・バーンインソケット、例えば、Texas Instruments社(マサチューセッツ州マンスフィールド)製のDieMate(登録商標)バーンインソケット、あるいは、Aehr Test社(カリフォルニア州フレモント)を通じて入手可能
なDie(登録商標)Pakバーンインソケット用として使用することができる。縁部に沿って基板16に接触するベアダイ・バーンインソケットの場合、プローブスプリング61のスプリングおよびファンアウトの金属化が必要であるのは、基板 16の片面(例:プローブ面62a)のみである。基板16のサイズは、基板16の縁部のパッドまで配線するのに必要なI/O信号の数に基づいて必要なファンアウトによって決まる。これに代えて、上述されているような基板16におけるビア66を使用して、基板16の反対の面62bのパッドのアレイまでI/O 信号を配線することができ、これによって基板をより小さくすることができ、従って製造コストが低減する。
【0104】
タイル型プローブアセンブリ。図27は、プローブ帯状片長さ198とプローブ帯状片幅200とを有する一般的なタイル型プローブ帯状片192の平面図190である。タイル型プローブ帯状片192は、それぞれが複数のスプリングプローブ61a~61nを有する、複数のプローブ帯状片接触子領域194a~194nを有する。さらに、図示されている実施例においては、スプリングプローブ61a~61nは、縦方向に整列しているプローブ領域196a,196b内に配置されている。プローブカードアセンブリにおいて一つ以上のタイル型プローブ帯状片192を使用することにより、半導体ウエハ92上の隣接する集積回路デバイスサイト44をテストする場合など、複数の集積回路デバイス44との同時の電気的接触が可能となる。複数のプローブ帯状片接触子領域194a~194nは、ウエハ92上の対称的な複数の集積回路デバイス44と位置が合うように、タイル型プローブ帯状片192の長手方向に沿って対称的に配置されていることが好ましい。理解すべき点として、フォトリソグラフィによってロット製造される応力金属スプリングプローブ (stress metal spring probe)をタイル型プローブ帯状片基板上に有するタイル型プローブ帯状片の使用は、メモリ、ロジック、マイクロプロセッサなど各種のICのテストに適用可能である。
【0105】
さらに、スプリングプローブ61a~61nを有するタイル型プローブ帯状片192は、一般には、図13、図14、および/または図21のプローブチップ基板16に見られるような、導電ビア66a~66nと、電気接続部のアレイ64a~64nとを含んでいる。スプリングプローブ61a~61nは、一般には、テスト対象の特定のデバイス44に合致するように配置することができるが、タイル型プローブ帯状片192は、標準的な導電ビア66a~66n、および/または、電気接続部のアレイ64a~64nを含む。例えば、図28および図29に示されているプローブカードアセンブリ202においては、タイル型プローブ帯状片192のそれぞれは、はんだ接続の標準的なボールグリッドアレイ160を含んでいる。従って、タイル型プローブ帯状片192の好ましい実施例は、テスト対象の特定のデバイス44に合致するように配置されているスプリングプローブ61a~61nを含むことができるが、標準化されているドーターカード 204および/または標準化されている中間コネクタ(例:分離型コネクタ132など)にタイル型プローブ帯状片192を取り付けることができ、これによって、タイル型プローブアセンブリ202を生産するための技術開発コストが最小になる。
【0106】
図28は、支持基板204に取り付けられている複数のタイル型プローブ帯状片192を有するタイル型プローブヘッド202の部分下面図であり、支持基板 204は導電ビア205のアレイ207(図29)を含んでいる。図29は、プローブカード16またはドーターカード204に取り付けられている複数のタイル型プローブ帯状片192の側面図であり、半導体ウエハ92上に位置する複数の集積回路デバイス44に接触する目的で使用されている。タイル型プローブヘッド202は、一般には、半導体ウエハ92上に位置する複数の集積回路デバイス44に接触する目的に使用される。タイル型プローブ帯状片192は、ウエハ92上の複数の対称的な集積回路デバイス44と位置が合うように、基板204上において対称的に配置されていることが好ましい。
【0107】
支持基板204は、熱膨張係数(TCE)が小さいことが好ましく、シリコンの熱膨張係数と同程度であることが好ましい。さらに、基板204においては、一般には、基板204の反対の面209b上のコネクタに多数の信号トレース46(図7)がファンアウトされている。一つの実施例においては、基板204は、シリコンウエハであり、ビア205(例:ピッチ0.056 inchで配置されるビアなど)と、基板の表面209a, 209bの一方または両面上の薄膜配線46とを含んでいる。
【0108】
図28および図29に示されているタイル型プローブヘッド202においては、タイル型プローブ帯状片192は、プローブスプリング61のグループを含んでおり、このグループは、テスト対象デバイス44の対向する側(例:集積回路デバイスサイト44の右側と左側など)に位置しているパッド47を有する集積回路デバイス44上のパッド47の列(図7)に接触する目的で使用される。図示されているタイル型プローブヘッド202においては、一般にタイル型プローブ帯状片192のうちの一つが、一つの回路デバイスサイト44の右側に(例えば、図27におけるプローブ接触子領域196aを使用して)接触し、さらに、隣接する回路デバイスサイト44の左側に(例えば、図27におけるプローブ接触子領域196bを使用して)接触するように、タイル型プローブ帯状片192が配置されている。従って、図28に示されている実施例は、複数のタイル型プローブ帯状片192と複数の集積回路デバイス44との間の同時の接触を提供する一方で、隣接するタイル型プローブ帯状片192の間の十分な間隔が確保され、タイル型プローブ帯状片192の側縁部は集積回路デバイス44のソー経路上に配置されていることが好ましい。例えば、ウエハ92上の隣接するデバイス44の間のソー経路94(図13)は、一般に4~8 milのオーダーの幅とすることができ、これによって、タイル型プローブカードアセンブリ202におけるタイル型プローブ帯状片192の間に、この幅に近い隙間を設けることができる。
【0109】
タイル型プローブヘッドアセンブリ202の代替実施例においては、集積回路デバイスサイト44のすべてのパッド47(図9)に、一つのプローブ帯状片192からのプローブによって接触させることができる。
【0110】
バーンイン構造。図30は、複数の集積回路デバイス44をバーンインボード212に一時的に接続することができるバーンイン構造210の部分断面図である。複数の集積回路デバイス44と外部のバーンイン回路(図示されていない)との間の電気的接続を提供するマイクロボールグリッドアレイ216などによって、プローブスプリング(すなわち、ナノスプリング)接触子チップ(NSCC)214のアレイがバーンインボード212上に実装されている。バーンインボード212には、ボード真空ポート218が画成されていることが好ましく、その一方で、NSCC基板214には接触子チップ真空ポート220が画成されていることが好ましく、この場合、(例えば、ボード真空ポート218を通じて適用される真空が、位置合わせされている接触子チップ真空ポート220にも適用されるように)ボード真空ポート218は接触子チップ真空ポート220に位置が合っている。適用される真空がマイクロBGAボールアレイ216を通じて失われることを防止する目的で、ナノスプリング接触子チップ214それぞれの周辺部の周囲に空気シール222(例:エポキシなど)が配置されていることが好ましい。
【0111】
集積回路デバイス44は、ナノスプリング接触子チップ214上に(例えば、「ピックアンドプレイス(pick and place)」マシンなどによって)最初に配置されるため、バーンインボード212のボード真空ポート218と、このポートに位置が合っている、ナノスプリング接触子チップ214の接触子チップ真空ポート220とに適用される真空によって、配置されている集積回路デバイス44がその配置位置から動くことが防止される。
【0112】
バーンイン処理時に集積回路デバイス44を所定の位置に保持することと、プローブスプリング61a~61nが集積回路デバイス44への電気的接続を提供できるようにすることとを目的として、すべての集積回路デバイス44が対応する接触子チップ214上に配置されたときに、クランププレート224が集積回路デバイス44に接触した状態に配置されることが好ましい。クランププレート224とバーンインボード212の平面性の裕度が与えられるように、テスト対象の集積回路デバイス44を押し付ける目的で、個々のスプリングパッド226も使用することができる。クランププレート224が集積回路デバイス44に接触した状態に配置されたときにクランププレート224がバーンインボード212に固定され、また、適用されている真空をオフに切り替えることができるように、バーンイン構造210は、クランププレート224を保持するための手段217を含んでいることが好ましい。
【0113】
改良されたスプリングプローブの保護コーティングの工程。上述されているように、スプリングプローブ61は、微細なピッチ、多数のピン数、およびフレキシブルであるという利点を有するため、幅広いアプリケーションに使用することができる。しかしながら、これらの一般に小さなスプリングプローブ61が、半導体ウエハ92上などの集積回路デバイス44(トレース46に酸化物層がしばしば含まれている)のトレース46に接触する目的で使用される場合、スプリングプローブ61が酸化物層を貫通して、金属トレースまたは導電パッドとの十分な電気的接触を確立することが要求されることがよくある。スプリングプローブ61 は何度も使用されることが多いため、保護されていない小さなスプリング先端24は、磨耗することがある。従って、プローブスプリング61の接触子先端24 に導電性の耐磨耗コーティングを施すことは有利であろう。しかしながら、このような保護コーティングは、スプリング先端24の両面全体を覆うことが要求される。
【0114】
上述されているように、プローブスプリング61は、例えば、米国特許第5,848,685号と米国特許第5,613,861号に開示されているように、プラズマ化学蒸着法とフォトリソグラフィ工程とによって形成することができる。この場合、導電性材料の連続する層が基板に形成され、その後、非平面的なスプリングが形成される。しかしながら、このような工程においては、蒸着工程時に形成される保護コーティングでは、形成される非平面的なプローブスプリングの表面全体を隙間なくコーティングすることができない。代替実施例においては、プローブスプリング61は、めっき工程によって形成することができる。
【0115】
プローブスプリング61は、基板から離れた後、基板表面と同一平面にない。従って、スプリング61が引き離し層18から離れた後に、保護コーティングを塗布することができる。図31は、スプリングプローブアセンブリのコーティング工程の第一ステップ230の図である。このステップにおいては、一本以上の非平面プローブスプリング61を有するスプリングプローブアセンブリ基板16のプローブ面に保護コーティング232が塗布される。スプリングプローブアセンブリのこのコーティング工程では、非平面的なプローブスプリング61上に保護層が形成される。このコーティング工程は、各種の非平面構造に使用できるが、特に、薄膜方式およびMEMS方式によるプローブスプリング接触子61の工程に有用である。図31においては、塗布される導電性保護コーティングは、窒化チタン、ロジウム、タングステン、またはニッケルなど、堅い導電性材料であることが好ましい。また、塗布される導電性保護コーティングは、不活性材料であることが好ましく、これによって、スプリングプローブ61のプローブ先端24に潤滑特性(すなわち低い摩擦係数)が与えられ、従って、テスト対象装置とスプリングプローブ61の両方において磨耗が最小になる。
【0116】
基板16とプローブ61とに保護コーティング233が塗布される(232)と、この保護コーティング233は、基板16の露出した表面62の平面領域と非平面的領域の両方を覆う。コーティングステップ230の間、スプリングプローブ61が保護コーティング233によって覆われるが、基板構造上のすべてのトレースは、塗布された導電性コーティング233によって互いに電気的に短絡している。従って、相異なるプローブスプリング61とそれぞれのトレースとの間の電気的絶縁が回復するように、導電性コーティング233がパターニングされるか、または部分的に除去される必要がある。集積回路の工程のほとんどでは、窒化チタンコーティングなどの導電性コーティングをエッチングにより選択的に除去する目的で、一般には従来のフォトマスク方式の工程が使用されるが、このようなフォトマスク方式の工程は、平面構造に使用される。
【0117】
図32は、スプリングプローブアセンブリのコーティング工程の第二ステップ234の図であり、この工程においては、好ましくはディッピング支持器 238(例:高さ約30ミクロン)を有する第二基板236に一層のフォトレジスト材料240(例:深さ約10ミクロン)が塗布される。フォトレジスト材料 240は、プローブスプリングの非平面部分に塗布されている保護層233を保護する目的に使用される。図33は、スプリングプローブアセンブリのコーティング工程の第三ステップの図であり、このステップにおいては、コーティングされたスプリングプローブアセンブリが第二基板236上のフォトレジスト材料 240の中に制御可能な方法で部分的に浸される。残る保護コーティング233は、最終的に、塗布されるフォトレジスト材料240の深さによって制御される。基板16が、フォトレジスト材料240の中の望ましい深さまで下げられ、この深さは、一般には、第二基板236上のフォトレジスト材料240の塗布深さと、ディッピング支持器20の高さとによって制御される。これに代えて、塗布される深さをオペレータによって制御することができ、例えば、処理装置の軸方向の動きを制御することによって、フォトレジスト材料240の中への基板16の動きを制御する。
【0118】
図34は、スプリングプローブアセンブリのコーティング工程の第四ステップの図であり、このステップにおいては、コーティングされて部分的に浸されているスプリングプローブアセンブリが、第二基板16上のフォトレジスト材料240から引き上げられ(246)、ソフトにベイクされ(soft baked)、保護コーティング(233)されたプローブスプリング61の一部が、ベイクされたフォトレジスト層248に覆われている状態となる。図35 は、スプリングプローブアセンブリのコーティング工程の第五ステップの図であって、このステップにおいては、コーティングされて浸されたスプリングプローブアセンブリ16,61がエッチングされ(250)、これによって、基板16の一部(すなわち基板16のフィールド領域)と、プローブスプリング61のうちベイクされたフォトレジスト層248に浸されていない部分とから、保護コーティング233が除去される。図36は、スプリングプローブアセンブリのコーティング工程の第六ステップの図であって、このステップにおいては、フォトレジスト層248に覆われているプローブスプリング61の部分からフォトレジスト層248が除去され、これによって保護コーティング233が露出する。
【0119】
従って、非平面プローブスプリングのこのコーティング工程では、プローブスプリングの先端24に保護コーティングが施され、その一方で、基板面16と、スプリングプローブ61のうちフォトレジスト層248によってコーティングされていない部分とにおける不要な保護コーティングがエッチングされる。
【0120】
超高周波アプリケーション用のスプリングプローブ基板。上述されているように、プローブカードアセンブリ60の構造では、プローブ先端61a~61nの間の距離が非常に短く、また、プリント基板プローブカード68内に、インピーダンスが制御された環境が形成され、これらによって、プローブカードアセンブリ 60を高周波アプリケーションに使用することが可能になる。さらに、スプリングプローブ基板16は、超高周波アプリケーション用に修正されることが好ましい。
【0121】
図37Aは、異なる固有な応力レベルを持つ複数の層17a~17nを有するプローブスプリング61を有する超高周波スプリングプローブ基板16aの部分断面図260aを示している。スプリング61は、一般には、引き離し層19の上に形成され、この場合、例えば、リリース領域18(図3、図4)内で引き離し層19を選択的に引き離すことによって、自由な非平面部分61を基板16aの表面から突出させることができる。基板16の表面62a,62bの片面または両面上のトレースをインピーダンス制御する必要がある実施例の場合、一つ以上の導電基準面262a,262bを、基板16の中または基板16上に、トレース270の上、トレースの下、またはトレースの上下に追加することができる。基板16には、遮蔽された同軸伝送ライン環境268を効果的に形成する目的で、一つまたは二つの基準面262a,262bに接続されている交互接地基準トレース266a,266bを含めることもできる。スプリングプローブ基板 16は、一般にはセラミック材料であるが、基準面の間の層264は、一般には、有機誘電体材料など誘電体材料である。
【0122】
図37Bは、一つまたは二つの基準面262a,262bに接続されている接地基準トレース266bから延在する接地基準トレース266dをさらに有する、代替の超高周波スプリングプローブ基板16の部分断面図260bを示している。接地基準トレース266dは、中央の導電ビア領域78を囲んでおり、誘電体領域259によって隔てられている。ビア領域を囲んでいる接地基準トレース266dによって、遮蔽された同軸伝送ライン環境268が、コネクタ面62bから基板16を貫いてプローブ面62aまで効果的に延在する。導電層266dと誘電体膜259は、好ましくは化学蒸着(CVD)技術(原子層化学蒸着法など)によって連続的に蒸着される。次いで、ビアの中央領域内の導電層が、CVD、物理蒸着法、または電子溶着(電解メッキまたは無電解めっき)、またはこれらの組み合わせによって蒸着/堆積される。導電層を堆積させる目的に電解メッキが使用される実施例においては、従来のいずれかの成膜手法(CVDなど)によって、誘電体膜259の表面上にシード層を堆積させることができる。
【0123】
図38は、代替の超高周波スプリングプローブ基板16の部分断面図261である。例えば、高周波アプリケーションの場合など、基板16上のスプリングプローブ61と基板内の関連する電気導体273[320],78,322とがインピーダンス制御されている(整合インピーダンス)必要のある実施例の場合、基板16の中または基板上のいずれかに、一つ以上の導電基準面262a,262b,262c,262dとビア266a,266b,266cが追加されることが好ましい。さらに、インピーダンス制御面262a,262b,262c,262dは、図38に示されている平面に限定されない。高い性能を目的としてビア78を貫く導電部にインピーダンス整合が要求される実施例においては、ビアは図37Bのような構造を持ち、この詳細については上述されている。
【0124】
インピーダンス制御された同軸接続部を形成する目的で、絶縁層267上に導電層262dを蒸着することができる。例えばプローブスプリング61に近くにデカップリングキャパシタを必要とする実施例の場合には、導電性材料262と誘電体材料263の交互の層を基板16に組み込むことが好ましい。導電性材料、またはシリコンなどの半導体材料である基板16の場合、基板16と導電基準面262c との間に、酸化物層または誘電体層269を蒸着または形成することによってキャパシタ構造271を形成することが好ましく、このキャパシタ構造をデカップリングキャパシタとして使用することができる。同様に、いくつかの実施例においては、基板16の反対の面に、または両面にキャパシタ構造を形成することもできる。この場合、接地面262bを蒸着する前に、表面上に誘電体層(例:酸化物)を蒸着または形成することができる。導電基板16との電気的接続は、誘電体269および接地面膜のいずれによっても覆われていない、基板16の表面領域(図示されていない)によって提供される。導電基板との接続を確立するためのこのような領域を開く目的には、フォトリソグラフィ手法を使用することができる。このように基板16が導電性材料である場合、電気的な短絡を防止する目的で、ビアまたはホールの壁に電気絶縁性の膜(例:酸化物)を蒸着または形成することができる。さらに、受動構成要素265(例:一般にはキャパシタ、抵抗器、および/またはインダクタ)、または能動構成要素デバイス265など、一つ以上の組立済み構成要素265を、基板16の面62a,62bのいずれかに組み込むことができる。
【0125】
スプリングプローブ61の固定部分15は、一般には、基板16上において相対的に短い距離だけ延在する。基板16の表面上に位置するトレース60は、スプリングプローブ61の固定部分15に電気的に接続されており、プローブスプリング61をビア78に電気的に接続している。このトレースは、スプリングプローブ61とは異なる材料から作製することができ、導電性の高い金属(例:銅、金など)から作製されていることが好ましい。
【0126】
高度なマルチダイプローブカードシステムアーキテクチャ。図39は、プローブカードアセンブリ60(例:図14の60a)、または強化されたプローブカードアセンブリ300(図40)によって、チャック282上に位置するマルチダイウエハ92とテストヘッド76との間のインタフェースが提供されているウエハテストシステム280の線図である。図39に示されているプローブカードアセンブリ60,300は、プローバヘッドプレート286に取り付けられている支持リング284によって、ウエハテストシステム280に取り付けられている。プローブカードアセンブリ60,300は、プローブインタフェースボード 288を通じてテストヘッド76とのインタフェースとして機能する。ウエハテストシステム280は、チャック282とテストヘッド76を互いに対して動かすことのできるプローバ駆動機構292も有する。
【0127】
マルチダイプロービングでは、ダイ44のそれぞれ(図13、図40)とテストヘッド76(図39)との間に多数の並列な接触経路(例:電気的接触経路など)が必要となる。図40に見られるような多数の並列接触経路307の場合には、一般には、基準面に対する(例:IC 44に対する)接触子端61,412(図45)の均一性と平面性が要求される。さらに、シグナルインテグリティと実動作速度でのテストを目的として、信号トレース307はファンアウトされている必要がある。また、多数の並列接触経路307では、プローブの力の合計が、ダイ44のそれぞれに有効に接触するものであることも要求される。さらに、動的なテスト時にすべてのテストチャネルの遅延を等しくするなどを目的として、トレース307の長さが各サイト44間で一致していることが好ましい。
【0128】
プローブカードアセンブリ60a(図14)、またはデカルインターポーザーアセンブリ300a(図41)など、複数のダイ44に接触するプローブカードアセンブリ60,300は、好ましくは、プローブ保持器(probe carrier)16の熱膨張係数(TCE)がウエハ92(例:シリコン)と同程度であり、その一方で、マザーボードプリント基板304への信頼性の高い接続と電気的整合性が維持されることが好ましい。
【0129】
複数のダイ44に接触するプローブカードアセンブリ60,300は、各ダイ44用に独立したパワーサプライを有することが好ましく、かつ、各ダイ44と接続するための複数の(例:2~3本以上の)パワーレールを備えていることが好ましい。さらに、プローブカードアセンブリ60は、テスト対象の各デバイス(DUT)44のできるだけ近くに、複数のバイパスキャパシタ172を含んでいることが好ましい。
【0130】
プローブカードアセンブリ60では、テスト対象デバイス44およびテストヘッド76からの信号およびパワーの接続がZ軸84沿いに伝送され、その一方で、好ましくはインピーダンス制御が提供される。また、プローブカードアセンブリでは、信号およびパワーの接続がX方向80およびY方向82(図13)にも伝送され、信号トレース307は集積回路44のピッチ20からテストヘッド76のピッチ91(図14)にファンアウトされている。プローブアセンブリ60のマザーボード304からプローブ先端61,412までの合計のプローブオフセット(total probe depth offset)は、一般には、支持リング284(図39)の厚さによって定義される。
【0131】
上述されているように、ブリッジ/リーフスプリング懸架型プローブカードアセンブリ60b(図15)、ワイヤ/スプリング支柱懸架型プローブカードアセンブリ60d(図20)、およびポゴワイヤ懸架型プローブカードアセンブリ60g(図23)など、プローブカードアセンブリ60のいくつかの実施例には、ウエハ92に対する対応性と平面性を提供する目的で、プローブチップ基板16を傾斜させる手段が設けられている。
【0132】
図40は、段階式インタフェースプローブカードアセンブリ300の基本的な線図であり、このアセンブリは、Zブロックプリント基板(PWB)342(図 42)などによって、全体として電気トレース経路307の垂直方向の変換を行う。この段階式インタフェースプローブカードアセンブリ300は、一般に、プローブチップ基板16,310のプローブ面62a(図43)上に位置している(例えば高さ180 &micro;mの)プローブスプリング412(図46、図47)を有し、これによって、改良された対応性がアセンブリ300に提供される。マザーボードプリント基板304とプローブチップ基板16,310との間には、上側インタフェース構造308と、中間コネクタアセンブリ306と、下側インタフェース構造312とをそれぞれ貫いている、電気接続経路307が設けられている。マザーボードプリント基板304には、補強プレート302も取り付けられていることが好ましい。図40に示されている上側インタフェース構造308と下側インタフェース構造312は、電気的なインタフェースおよび/またはハードウェアを有する。
【0133】
段階式インタフェースプローブカードアセンブリ300のいくつかの実施例においては、強化された対応性がプローブスプリング412によって完全に提供され、この場合、プローブチップ基板16,310は、マザーボードプリント基板304に対して動かないように保持されている。プローブチップ基板310の下面には、一般に下側支持器114が設けられており、この支持器は、プローブチップ基板とウエハ92との間の最小垂直距離を制限する。下側支持器114の高さは、一般には、プローブスプリング61,412が曲がってウエハ92上の一つ以上のデバイス44との対応性を持つ接続を提供できるように、プローブスプリング61,412の自然な状態の、すなわち接触していない状態の高さよりも低い。図40に見られるように、プローブカードアセンブリ300のいくつかの実施例においては、中間コネクタ306またはマザーボード304との関係における、プローブチップ基板310のコネクタ面311aの垂直移動距離を制限するなどの目的で、上側支持器116(図19)も使用される。
【0134】
図39に見られるように、金属の支持リング、すなわちリングインサート284は、ポゴタワーコネクタ290(図39)によってかかる下向きの圧力に対して、マザーボードプリント基板304を機械的に支持する。
【0135】
図40に示されている中間コネクタ306は、一般には、垂直変換ブロック342(図43)、ピンブロック742(図64~69)、または拡張ピンブロック(図70、図71)を有する。中間コネクタ306は、アセンブリ300にZ方向の変換を提供し、これによって、アセンブリ300においてプローブをオフセットさせることができる。このオフセットは、一般には約0.300"であり、この値は、ほとんどのプローブアセンブリにおいて、金属の支持リングインサート284をクリアするための要件である。
【0136】
上側インタフェース308は、様々なインタフェースを有することができ、例えば、両側スプリング521(図52)を有する上側インターポーザー344(図 43)、はんだボールアレイ756またはZ積層(図65)、ピングリッドアレイ745(図64)、またはその他のインターポーザーアセンブリ 344(Circuit Components社(アリゾナ州テンペ)から入手可能なISOCON(登録商標)コネクタなど)である。
【0137】
下側インタフェース312も、様々なインタフェースを有することができ、例えば、下面349aと上面349bの両方にスプリングを持つインターポーザー 348(図42、図43)である。これに代えて、下側インタフェース312は、はんだボールアレイ756、プローブチップ基板310の上面311b上に位置するスプリング64a~64n(図13、図14)、または、プローブチップ基板16,310とのその他の直接接続、を有することができる。
【0138】
いくつかの実施例において、垂直変換ブロックVTB 342(図43)が、その下面343とシステムの残りとの間の横方向の対応性が提供されるように作製されている場合には、下側インタフェース312は固定接続または永久接続でよい。横方向の対応性の一つの例は、金属ピンブロック742を使用している図67に示されており、この場合、対応性を持つピンテンプレート748,752によってピンがわずかに動くことができる。
【0139】
垂直変換ブロックVTB 342の代替実施例は、インターポーザー、またはプローブチップ310の上面62b上に作製されているスプリング64a~64n(図67)など、複数の非固定型の接続部を有する。非固定型の接続を使用することによって、プローブチップ310は垂直変換ブロックVTB 342に対して動く、および/または、拡張する(expand)ことができる。
【0140】
図40に示されている段階式インタフェースプローブカードアセンブリ300は、マザーボードプリント基板302を平らに維持する役割りを果たす上側補強プレート302をさらに有する。
【0141】
図41は、デカルインターポーザーアセンブリ300aの部分断面図である。マザーボードプリント基板304は、複数の締結具322などによって、上側補強プレート302に固定的に取り付けられている。補強プレート302は、ステンレス鋼など堅い材料から成ることが好ましい。図41に示されている補強プレート302は、マザーボードプリント基板304の上面305b(図43)に構成要素(例:キャパシタ172)を実装する、または上面305bから延在させることができるように、複数の構成要素くぼみ325をさらに有する。
【0142】
補強プレートのいくつかの実施例では、アセンブリの取り扱いを容易にするなどの目的で、ハンドル837,839(図73、図74、図75)が組み込まれている。さらに、補強プレート302のいくつかの実施例では、プレートを貫通するウィンドウ840(図73)が定義されており、これにより、例えば、顧客向けのカスタマイズにおける電気ジャンパなどを目的として、マザーボード304を組立後に変更することができる。
【0143】
図41に示されているように、マザーボード304は、プローブチップ16,310の上面62b(図43)とマザーボード304の下面305a(図43)との間に位置している中間インターポーザー150を通じて、プローブチップウエハ16に電気的に接続されている。インターポーザー150は、一般にはインターポーザー位置合わせピン330によって、マザーボード304に位置合わせされていることが好ましく、この位置合わせピンは、マザーボード304に実装されてマザーボード304から延在しており、インターポーザー基板348の位置合わせホール347(図43)および/または縁部351(図43)に対応している。
【0144】
図79は、対応性を持つメンブレン326aに実装されているプローブチップ310の透視図である。図80は、対応性を持つデカル326bに実装されているプローブチップ310の透視図である。図81は、対応性を持つシート326cに実装されているプローブチップ310の透視図である。図82は、対応性を持つスクリーン326dに実装されているプローブチップ310の透視図である。プローブチップ310は、対応性を持つ部材326によって所定の位置に保持されており、対応性を持つ部材は、一般には、フレキシブルメンブレン326a(図79)、デカル326b(図80)、シート326c(図81)、またはメッシュ構造326d(図82)を有する。図41に見られるように、対応性を持つ部材326は、締結具334によってマザーボード304に取り付けられている外側周囲リング328に、固定的に取り付けられているか、または実装されている。外側リング328と、取り付けられている対応性を持つ部材は、リング位置合わせピン332などによってアセンブリ300aに位置合わせされていることが好ましい。対応性を持つ部材は、周囲リング328への取り付け時に伸ばされ、このため張力がかかった状態に保持されている。
【0145】
対応性を持つ部材326は、プローブチップ310をX軸80およびY軸82には所定の位置に保持するが、プローブチップ310はZ方向84には動く、または浮動することができる。対応性を持つ部材326は、対応性を持つ部材またはデカル326の制御された曲がりによって、押し付けられている状態にインターポーザー150を保持し、この曲がりは、インターポーザー150の下面305aと対応性を持つ部材326の実装位置との間のZ面84のデルタに起因する。
【0146】
対応性を持つ部材326は、プローブチップ16,310とデカルリング328との熱膨張係数(TCE)の差に対する対応性も提供する。例えば、デカルリング328の熱膨張係数がプローブチップ16,310よりも大きいデカルプローブアセンブリ300の場合、対応性を持つ部材326は、高い温度においてプローブチップ12,300とデカルリング328との間で容易に曲がる、すなわち伸びる。
【0147】
図42は、Zブロックデカルインターポーザーアセンブリ300bの部分断面図340である。図43は、Zブロックデカルインターポーザーアセンブリ 300bの拡大組立図360である。それぞれプローブチップ16,310の上面311bとマザーボード304の下面305aとの間に位置している、上側インターポーザー344と、プリント基板Zブロック342と、下側インターポーザー348とによって、マザーボード304はプローブチップウエハ 16,310に電気的に接続されている。下側補強プレート346は、マザーボードプリント基板304の対応するホールまたはスロット309にピン354を介して位置合わせされている。Zブロック垂直変換ブロック(VTB)342は、プローブスプリング61a~61nのプローブ先端がプローブリング 284(図38)をクリアするように、マザーボードプリント基板304からプローブチップ16,310までの信号およびパワーのZ方向の変換を提供する。図43に示されているZブロック342は、上面343bおよび/または下面343aから延在する三組の位置合わせピン330,333,350を持つ。下側インターポーザー348と上側インターポーザー344は、縁部351または一つ以上の位置合わせホール347などによって、それぞれ、ピン330、ピン 333に位置合わせされている。Zブロック342自体は、ピン350を介して下側補強プレート346に位置合わせされている。
【0148】
対応性を持つ部材326は、X軸80およびY軸82上の所定の位置にプローブチップ310を保持し、その一方で、Zブロック342が熱変動に起因して膨張するときに、プローブチップ310がZ方向84に動くかまたは浮動することを可能にする。対応性を持つ部材326は、下側インターポーザー348の下面 349aと対応性を持つ部材326の取り付け位置との間のZ面84のデルタに起因する、対応性を持つ部材またはデカル326の制御された曲がりによって、押し付けられている状態に下側インターポーザー348を保持する。対応性を持つ部材326は、プローブチップ16,310とデカルリング328との熱膨張係数(TCE)の差に対する対応性も提供する。
【0149】
図42に見られるように、Zブロック342は、下側補強リング346に取り付けられており、このリングは、ステンレス鋼など堅い材料から成ることが好ましい。下側補強リング346は、Zブロック342を平らな、すなわち平面的な状態に保持する一方で、この補強リング346の厚さによって、対応性を持つメンブレンまたはデカル326の曲がり量が制限、すなわち制御される。
【0150】
Zブロック保持クリップアセンブリ352は、Zブロック342を下側補強リング346に保持し、かつ、Zブロックが高温においてZ方向に膨張できるようにする。下側補強リング346は、一般には中間上側インターポーザー344を有するマザーボード304の方向に、Zブロック342を保持および支持する。
【0151】
マザーボード304とプローブチップウエハ16との間には、一つ以上の平面性調整ねじおよび/またはシムなどによる平面性調整器326が設けられることが好ましい。デカルインターポーザーアセンブリ300のいくつかの実施例においては、平面性を提供する手段324は、複数の差動ねじアセンブリ824(図 73、図74、図75)を有する。
【0152】
Zブロックデカルインターポーザーアセンブリ300bのいくつかの実施例においては、下側補強リング346は、一つ以上の平面性調整ねじ 324,824(図73、図74、図75)などによる、平面性を調整する手段をさらに有し、この手段によって、マザーボード304の下面305aに対する接触子先端の間の平面性を調整することができる。下側補強リング346は、スプリングの力によって平面性調整ねじ324,806に密着した状態に保持されており、これによって、ねじがアセンブリ内で縦に動くとき、下側補強リングはねじに接触している状態を維持する。平面性調整システム326は、各種の構造を有することができ、例えば、締結具、差動ねじ、ガイド、ショルダーボルト、あるいは、圧縮ワッシャおよび/または対応性を持つOリングなどの位置偏向 (biasing)ハードウェアなどである。
【0153】
図43に見られるように、下側インターポーザー348は、プローブチップ310とZブロック342との間の電気的な接続を提供し、また、プローブチップ 310を取り外し可能な副構成要素とすることができる。下側インターポーザー348は、熱膨張係数の差に対応する目的で、プローブチップ310とZブロック342との間の摺動界面(sliding interface)も提供する。
【0154】
対応性を持つメンブレンまたはメッシュ326は、X軸80およびY軸82上の所定の位置にプローブチップ310,16を保持し、その一方で、Zブロック 342が熱変動に起因して膨張するときに、プローブチップ310,16がZ方向84に動くかまたは浮動することを可能にする。対応性を持つメンブレンまたはメッシュ326は、下側インターポーザー348の下面とデカルリング328の取り付け位置との間のZ面のデルタに起因するデカル326の制御された曲がりによって、下側インターポーザー348を常に完全に押し付けられている状態に保持する。対応性を持つメンブレンまたはメッシュ326は、プローブチップ 310とデカルリング328との間の熱膨張係数の差に対する対応性を提供する。
【0155】
強化されたプロープチップ。図44は、フィンガーが持ち上がる前のプローブチップ310の断面図370である。図45は、フィンガーが持ち上がり(404)、めっきした(406)後のプローブチップ310の断面図400である。フィンガーが持ち上がる前、プローブ面62aの一部は、一般に、セラミック基板372上に形成されている引き離し層376(チタンを有することが好ましい)を有する。次に、引き離し層376上に複合層380が形成され、この複合層380は、一般には、複数の、好ましくは5層(ただしこれに限定されない)の金属膜層17a~17n(図37)を有し、この金属膜層においては、スプリングが持ち上がる(404)(図 45)前において少なくとも二つの隣接する層の固有の応力レベルが異なっている。このような複合層は、スパッタリング蒸着などの手法によって形成されることが好ましい。いくつかの実施例においては、このような複合層は、電着手法によっても形成される。図44に示されているプローブチップ構造310においては、複合層380の上に導電シード層384(金を有することが好ましい)が形成されることが好ましい。次いで、複合層380の一部の上に、一般にはフォトリソグラフィによるパターニングによって、導電トレース382が制御可能な方法で形成される。
【0156】
いくつかの実施例においては、次いで、一般には信号トレースの遮蔽を提供する目的で、一般には第一ポリイミド(PMID)層384と、接地面388と、第二ポリイミド(PMID)層390と、ハードマスク層392とを有する上側層が、トレース領域の上に形成される。
【0157】
図44および図45は、基板372のコネクタ側に形成されている模範的な接続構造393も示している。図44および図45に見られるように、この接続構造 393には、例えば図43に見られるようなインターポーザー348との接続と、パッド445(図49)とビア374との間の接続トレースとの接続とを目的とする接触パッド396が設けられている。
【0158】
接続構造393は、一層以上の金属層376,394,396の積層から成り、この積層には、インターポーザーの接触子またはスプリングとの信頼性の高い接触を目的とする、金を有する外側層396を一般に含んでいる。プローブチップ310のいくつかの実施例においては、接続構造393は、第一チタン金属層 376と、第二ニッケル層394と、第三金金属層396とを有する。プローブチップ310の別の実施例においては、接続構造393は、第一Cr層および Cu金属層376と、第二ニッケル層394と、第三金金属層396とを有する。層376,394,396の厚さは、一般には、電気的シート抵抗が減少するように制御する。この接続構造のいくつかの実施例においては、第一Cr/Cu層374の厚さが1~4μmであり、および/または、外側の金層の厚さが 1~4μmである。
【0159】
いくつかの実施例においては、上面396に構成要素(例:バイパスキャパシタ)をはんだ付けするためのはんだマスク398としての役割りを果たすこと、または、インターポーザー348内の望ましくない位置への短絡に対する絶縁部としての役割りを果たすことなどを目的として、積層構造393の上に、ポリイミド、フォトレジスト、エポキシ、またはKAPTONなどの誘電体層を追加することができる。
【0160】
図45に見られるように、フィンガーが持ち上がった(404)後、プローブスプリング61が、複数の膜層における固有の応力勾配の結果として、引き離し領域386(図44)内で基板372のプローブ面62aから離れる。一般には、プローブスプリングの自由な非平面部分は、離れた後には実質的に弧の形状をとり、上側蒸着層であるMo-Cr膜380と金シード層384は凹形状をとる。この結果として、非常に小さなピッチを有するプローブスプリングのアレイを、フォトリソグラフィなどの集積回路製造手法を使用して一括して製造するのに望ましい、三次元のスプリング構造が形成される。次いで、プローブスプリングのめっき層402が、突き出した、すなわち非平面的なプローブスプリング61上に形成されることが好ましい(406)。プローブスプリングのいくつかの実施例においては、めっき層402によって、強化されたスプリング力、高い耐磨耗性、高い強度、および/または、高い導電性が得られる。いくつかの実施例においては、ニッケル、パラジウム合金(パラジウム-コバルト)、ロジウム、金などの異なる材料を有する複数の膜が、持ち上がったフィンガーに連続的にめっきされる。
【0161】
プローブスプリング61,412の代替実施例においては、スプリング先端は、二次の接触子先端細部、すなわち、スプリング先端ボタン接触子をさらに有することができる。この接触子は、米国仮出願第60/365625号「耐破損性の高いナノスプリング(Nanospring with Increased Resistance to Failure)」(出願日:2002年3月18日)に開示されているように、一般にはロジウム、パラジウム、またはコバルトを有する。この文書は本出願に参照文献として組み込まれている。
【0162】
プローブチップ構成要素の機能。プローブチップ16,310は、プローブチップ基板372のプローブ側62a上において、フォトリソグラフィによって画成されているスプリング61など、テスト対象ウエハ92に接触するためのプローブスプリングを支持している。プローブチップ16,310は、図40に見られるような厚いマザーボードプリント基板304によって達成可能なグリッドへの信号およびパワーのファンアウトを提供する。図44および図45に見られるように、プローブチップ16,310は、一般にはマザーボード304上のめっき処理済スルーホールのアレイのピッチに合致するピッチ配列において、基板 372を貫いて接続面62b上のパッドまたははんだボール398のアレイまでの信号接続を提供する。
【0163】
プローブチップ16,310では、ファンアウトトレース307(図40)がインピーダンス制御されていることも好ましく、また、バイパスキャパシタを実装するための領域(例えば、プローブチップ16,310のコネクタ面311b上など)が設けられていることも好ましい。
【0164】
プローブチップ16,310のいくつかの好ましい実施例は、冗長性により製造歩留まりが高まるように、電気経路あたり二本以上の導電ビア374を有する。同様に、プローブチップのいくつかの好ましい実施例は、電気経路あたり二本以上のプローブスプリング61(図9)を有する。
【0165】
プローブカードアセンブリのシーケンス。図43に見られるように、ウエハ92上の一つ以上のテスト対象デバイスとの強化された接続を提供する目的で、強化されたプローブカードアセンブリ310が容易に組み立てられる。
【0166】
図43に見られるように、アセンブリの模範的な工程においては、マザーボードプリント基板304は、一般には、ねじ324などの締結具を使用して上側補強プレート302に取り付けられており、第一サブアセンブリ361aが形成されている。いくつかの実施例においては、ホール309はスルーホールとすることができ、このスルーホール(図示されていない)に挿入されるピンを使用して、より寸法の大きな上側補強プレートがマザーボードに取り付けられる。
【0167】
第二サブアセンブリ361bは、一般には、位置合わせピン330,350をZブロックプリント基板342に押し込むことと、位置合わせピン350、Zブロック固定クリップリング352、およびねじ353などによって、Zブロック342を下側補強リング346の中に取り付けることとによって形成される。
【0168】
次いで、上側インターポーザー344がZブロック342上に制御可能な方法で配置されて、かつインターポーザー位置合わせピン333に位置が合った状態となるように、第二サブアセンブリ361bが一時的な支持構造上に配置される。マザーボードプリント基板304には、ピンの露出部分を挿入するための空間を形成する目的で、ピン333に対応する小さなブラインドホール(図示されてない)が設けられている。
【0169】
次いで、補強リング346(図43)の上面347b(図43)上の補強リング位置合わせピン354を、マザーボードプリント基板304(図43)の下面 305a(図43)に画成されている位置合わせホール309(図43)に位置合わせし、かつ、上側インターポーザー344上の上側インターポーザースプリングを静かに圧縮しながら、第一サブアセンブリ361aが第二サブアセンブリ361bの上に配置される。次いで、第一サブアセンブリ361aと第二サブアセンブリ361bが、ねじ834(図75)などによって一つに固定され、第三サブアセンブリ361cが形成される。
【0170】
次いで、一般には、第三サブアセンブリ361cの上下が逆にされて、第二の一時支持構造上に配置され、この場合、下側インターポーザー348は、制御可能な方法でZブロック342の下面343a上に配置され、かつインターポーザー位置合わせピン330によって位置合わせされる。次いで、デカルサブアセンブリ361dが、下側インターポーザー348上に配置され、これは、いくつかの実施例においては、位置合わせピン354によってガイドされる。一時的アセンブリ支持取り付け器を使用することにより、サブアセンブリ361cと361dとの間が確実に位置合わせされ、デカルサブアセンブリ361dをインターポーザー348上にゆっくりと降ろすことができ、従って、外側リング328がさらに押されて下側補強器346に接触する一方で、メンブレン326が伸びる。締結具356が締め付けられて、アセンブリ取り付け器がリング328を所定の位置に保持する。
【0171】
上記のアセンブリの説明は、一般的な実施例の場合であることは、理解されるべきである。上記に概説されている基本的な概念に基づくアセンブリ取り付け器およびその工程におけるバリエーションは、本発明の範囲内である。例えば、インターポーザー348または344には、図43に示されているようなホールを必ずしも形成する必要はない。代わりに、インターポーザーを所定の位置に保持する目的で、多数の追加のピン(例:333,330)を設けることができる。同様に、図43における外側リング328がメンブレン326上に配置されることが好ましい。
【0172】
プローブチップ構造の細部。図46は、プローブチップ16,310のスプリング側の面311aの平面図410である。図47は、プローブチップ16,310のスプリング側の面 311aの詳細な部分的配置図430である。テスト対象ウエハ92上の一つ以上のデバイスとの複数の接続を提供する目的で、フォトリソグラフィによって形成されたプローブスプリング61a~61nなど、プローブチップのプローブスプリング412は、一般にプローブスプリンググループ422の中に配置されている。フォトリソグラフィによって形成されたプローブスプリング61a~61nなどのプローブスプリング412は、テスト対象ウエハ92上の接触パッド 47に一致している。図46に示されているプローブスプリンググループ422の模範的な配置構成は、32個のダイ位置に同時に接触することを目的として、スプリンググループ422が4×8セットである。
【0173】
ビア414は、スプリング側の面311aからプローブチップ基板372を貫いてコネクタ側、すなわちインターポーザー側の面311bまで延在する。プローブチップ16,310の製造歩留まりを高めること、および/または、特にパワートレースの導電性を高めることなどを目的として、ビアは、冗長なビアの対 416に配置されていることが好ましい。図46および図47に示されているビアの対416は、ビア行418およびビア列420とを有するビアグリッドアレイ417に配置されている。ビアグリッドアレイ417は、マザーボード304のめっき処理済スルーホール(PTH)グリッドに一致していることが好ましい。スプリング412がビア414,416に一致している場合、ビア414,416は隣接する行418に移動し、トレース424は、一般には、プローブチップ16,310の反対のインターポーザー側311b上の、元のビア414,416が位置していたはずの場所まで配線される。
【0174】
図48は、プローブチップ16,310のインターポーザー側の面311bの平面図432である。図49は、プローブチップ16,310のインターポーザー面311bの詳細な部分的配置図444である。接触パッド434は、インターポーザー側の面311b上に位置しており、一般にはパッドグリッドアレイ 436に配置されており、このアレイは、一般にはビア214(図49)に一致する。ビア214は、インターポーザー348との電気的接触を提供する目的で、冗長なビアの対416に配置されていることが好ましい。プローブチップ16,310のいくつかの実施例においては、パッドグリッドアレイ436は、接続されるインターポーザー348上の類似する接触ピッチに一致するように、0.056"のピッチを有する。図48に見られるように、インターポーザー側の面311b内には、一般には中央領域438が画成されており、この領域内には、接触パッド434とプローブチップ上側トレース442が一般にデバイス領域 440内に配置されている。
【0175】
接着領域437は、プローブチップ16,310のインターポーザー側の面311bの外側周辺部に沿って位置しており、これによって、プローブチップ16,310と対応性を持つ部材またはデカル326との間に接着剤329が付着または塗布される。
【0176】
図49に見られるように、プローブチップ16,310のインターポーザー側の面62b,311bには、信号接触子(S)445と、能動接地接触子(S)446と、パワー接触子(P)448と、接地接触子(G)450とが設けられている。ビア414,416とのスプリング位置の衝突などに起因してビア414,416を再配置できるように、トレース442が、接触子445,446,448および/または450の配線を提供する。図48および図49に示されているプローブチップ16,310は、能動接地接触子446とパワー接触子448との間、能動接地接触子446と接地接触子450との間、および/または、パワー接触子448と接地接触子450との間などに、キャパシタ452をさらに有する。
【0177】
プローブチップアセンブリ構造。図50は、デカル/プローブチップアセンブリ取り付け器の拡大組立図である。取り付け器基部462は、内側プローブスプリングリリーフ領域466を囲んでいるプローブチップ棚464を有する。このプローブチップ棚464には、プローブチップ16,310を正確に受け入れる目的で、一般に一つ以上のプローブチップウエハ位置合わせピン468が配置されている。この取り付け器基部462は、同様に、プローブチップ棚464を囲んでいるデカルリング空隙470を有する。
【0178】
マッチングプレート位置合わせホール(matching plate alignment hole) 476を有する上側積層プレート474、または、ステンシル位置合わせホール479を有するデカル切除ステンシル478のいずれかを正確に受け入れる目的で、一般には、一つ以上の積層プレート/ステンシル位置合わせピン472がデカルリング空隙470に配置されている。後述されているように、上側積層プレート474は、プローブチップ16,310に対応性を持つ部材326を接着剤329などによって層状に取り付けるときに使用される。
【0179】
デカル切除ステンシル478は、ステンシルテンプレート開口480をさらに有する。デカル切除ステンシル478が、対応性を持つ部材またはデカル326に接着式に取り付けられているプローブチップ310を有するアセンブリの上、取り付け器基部462の上に位置している状態で、ステンシルテンプレート開口 480が、一般にはプローブチップ310の外側接着領域に対して位置合わせされ、従って、デカル326の内側領域を切断して除去することによって、プローブチップ16,310の表面へのアクセスが提供される。
【0180】
デカルアセンブリのシーケンス。図51は、デカルプローブチップアセンブリの工程のフローチャートである。完成したプローブチップ310(持ち上がって(402)めっきされている(404))が、積層取り付け器460の下側プレート462上に、下向きに配置され(492)、この場合、取り付け器はプローブチップ310を接着領域の下のみを支持する。空隙 466は、スプリング61を保護する目的で設けられている。プローブチップ310は、取り付け器基部462上の位置合わせピン468に位置合わせされる。
【0181】
次いで、接着シート329(図43)(プローブチップ16,310上の望ましい接着領域の形状にあらかじめ切断されていることが好ましい)が、プローブチップ16,310の周辺部に貼られる(494)。接着シート329は、プローブチップ16,310上の形状に位置合わせされた後、一般には熱い鉄 (hot iron)によって貼り付けられる。アセンブリ300の一つの実施例においては、接着シート329は、Bステージ接着シート329である。
【0182】
次いで、指定の張力にあらかじめ張った状態でデカルリング328に取り付けられている対応性を持つ部材またはデカル326が、取り付け器462上、接着剤 329の上に配置される(496)。積層取り付け器460のいくつかの実施例においては、デカル326のリング側は、取り付け器460の中でリング側が上の状態に配置される。積層取り付け器460の代替実施例においては、デカル326のリング側は、取り付け器460の中でリング側が下の状態に配置される。デカルリング328は、位置合わせピン472を使用して取り付け器462に位置合わせされる。
【0183】
次いで、取り付け器上側プレート474が、デカル326の上に配置される(498)。このプレートは、取り付け器基部462の所定の位置に位置合わせピン472を介して保持され、デカル326を接着シート329に押し付ける。
【0184】
次いで、アセンブリ取り付け器460を使用して接着剤329を硬化させる。この手順は、一般には、アセンブリ460をオーブンに入れるステップ500と、取り付け器上側プレート474に重りまたはその他の圧縮力をかけるステップ502と、ベイクして接着剤329を硬化させるステップ504と、オーブンから重りとアセンブリ460とを取り除くステップ506とを有する。
【0185】
対応性を持つデカル326の内側部分を除去する手順には、デカル326の上にデカル切除テンプレート478を配置するステップ508が含まれる。デカル切除テンプレート478は、取り付け器の下側プレート上の位置合わせピン472に位置合わせされる。デカル326の内側部分、すなわち接着領域319の内側が、ナイフによって分離または切除される(510)。
【0186】
インターポーザーの構造。図52は、上側インターポーザー344または下側インターポーザー348など、インターポーザー構造520の部分断面図である。
【0187】
インターポーザースプリング521(フォトリソグラフィによって形成されたプローブスプリング521など)は、複数の標準化された接続部を提供する目的で、一般に、インターポーザーグリッドアレイ内に配置されている。例えば、図43に示されている上側インターポーザー344においては、インターポーザースプリング521は、マザーボード304とZブロック342との間の接続を提供している。同様に、図43に示されている下側インターポーザー348においては、インターポーザースプリング521は、Zブロック342とプローブチップ16,310との間の接続を提供している。
【0188】
インターポーザービア524は、第一面523aから基板522を貫いて第二面523bまで延在する。インターポーザービア524は、インターポーザー 520,344,348の製造歩留まりを高めること、および/または、特にパワートレースの導電性を高めることなどを目的として、冗長なビアの対に配置されていることが好ましい。
【0189】
対向する面523a,523bは、一般には、引き離し層526(例えば、チタンを有する)と、複合層530,532(一般には固有の応力レベルが異なる複数の導電層を有する)とから成る。インターポーザービア524(例えば、CuWまたは金によって満たされている)は、中央基板522(一般にはセラミック)を貫いて延在しており、引き離し層526の間の導電接続を提供している。複合層530,532は、一般にはMoCrを有し、この層には、インターポーザープローブスプリング521がパターニングされており、後から引き離し領域528内で持ち上がる。
【0190】
複合層530,532の上には、シード層534(例えば、厚さ0.5~1 μmの金の層)が形成されていることが好ましい。スプリングフィンガー532の少なくとも先端には、耐磨耗性および/または信頼性の高い接触を提供するなどの目的で、制御可能な方法で先端コーティング540(例えば、ロジウムまたはパラジウム合金)が形成されている。トレース536は、一般には銅を有し、抵抗を小さくするなどを目的として、図示されているように構造520の上へのめっきによって選択的に形成される。さらに、一般には、スプリングフィンガー持ち上がり領域を画成する目的で、図示されているように、ポリイミドPMID層538が構造520上に形成される。フィンガー521のシート抵抗が低減するように、厚い金の層534は、持ち上がったフィンガー521上に残る。
【0191】
図53は、インターポーザー520の平面配置図550である。スプリングセットアレイ552は、インターポーザー基板522の内側接触子領域554の中に配置されており、複数のマルチフィンガーセット560を有する。このマルチフィンガーセットは、図52および図54に見られるように、複数のフィンガースプリング521(図52、図54)と、複数の冗長型インターポーザービア524とによって、接続の冗長性を提供することが好ましい。インターポーザー基板 522の周辺部付近には、外側支持領域556が存在している。画成されているソー境界558の中には、インターポーザー基板522の内側接触子領域554 が存在している。
【0192】
図54は、インターポーザー520の一部の詳細な配置図562である。インターポーザー基板上には、マルチフィンガーセット560がピッチ566で配置されていることが好ましく、このピッチは、マザーボード304上の対応する接続部のピッチに一致していることが好ましい。マルチフィンガーセット560のそれぞれは、共通の導電領域564を有し、この領域から複数の冗長型スプリングフィンガー521が形成されて、持ち上がり領域528の上に基板522から延在する。複数の冗長型インターポーザービア524は、共通の導電領域564に電気的に接続されており、例えば、第一面523aから反対の面523b(図 52)までなど、インターポーザー基板522を貫いて延在する。図54に示されているインターポーザー520においては、マルチフィンガーセット560のそれぞれは、四本の冗長型フィンガー521と四本の冗長型インターポーザービア524とを有する。代替のインターポーザー520においては、例えば、二本または三本のフィンガー521および/またはインターポーザービア524など、任意の数の冗長型フィンガー521および/またはインターポーザービア 524を設けることができる。図54に示されているマルチフィンガーセット560は、一様にクローバーまたは十字の形状に配置されているが、様々な幾何学形状を好適に使用することができる。
【0193】
インターポーザー520のいくつかの実施例は、組み込み型キャパシタ576を有し、この厚さは、一般にインターポーザー基板522の厚さよりも小さい。図 55は、キャパシタ空隙領域574内にスプリングが存在しないインターポーザー520の一部の平面図570である。図56は、バイパスキャパシタ576が埋め込まれているインターポーザー520の一部の平面図580である。マルチフィンガーセット560は、空白領域572が形成されるように容易に配置または修正され、この領域内にスルーホール空隙または切除領域574が配置されている。図56に見られるように、ビア582(例:骨状ビア(dog bone vias)など)は、マルチフィンガーセット560と埋め込み型キャパシタ576との間の電気的な接触を提供している。骨状ビア582を有する材料は、基板の切断時、レーザーによる影響を受けない。露出したビアの表面は、導電性エポキシまたははんだによるキャパシタ576との接続部に適した面が形成されるように、めっき処理されることが好ましい。
【0194】
フィンガーが持ち上がる前に、インターポーザー基板522の片側に接着性テープが貼られる。キャパシタ576が、基板522の反対側 523(例:523b)から空隙574内にピックアンドプレイスされ、接着性テープによって所定の位置に一時的に保持される。次いで、キャパシタ576と骨状ビア582との間を完全に接触させる目的で、両端にエポキシが塗布される。エポキシが硬化した後、接着性テープが除去される。次いで、フィンガー 521が基板522から持ち上がる。
【0195】
インターポーザーの代替の構造および工程。図57は、スプリングが持ち上がる前の代替のデカルインターポーザー構造520bの一部の断面図590である。図58は、スプリングが持ち上がった後の代替のデカルインターポーザー構造520bの一部の断面図620である。
【0196】
代替のデカルインターポーザー構造520bは、反対向きのスプリングの対521を有し、このスプリングの対は、基部において結合されており、支持枠610 の内側にドラム状に(like a drum)支持されているメンブレン602の開口の縁部に取り付けられている。スプリングの対521は、メンブレン602のホールを貫いて突き出しているメンブレン602の面から離れるように延在する。メンブレン602は、スプリングの対521の位置が一定に保持されるように、枠610の内側に張力がかかった状態で保持されていることが好ましい。
【0197】
このインターポーザー構造520bは、ビアを有する基板を必要としないことと、一回のフォトリソグラフィステップによって両側をパターニングできるため製造コストが大幅に低減するという利点を有する。さらに、プローブチップ310のコネクタ側62bに受動構成要素(例えばバイパスキャパシタなど)が実装されており、かつ、プローブチップ310のコネクタ側62bとプリント基板304との間のインタフェースとして機能させる目的に代替インターポーザー520 が使用される実施例においては、構成要素がインターポーザー520内に突き出すための開口は、フォトリソグラフィによるステップにおいて容易に形成してメンブレン602をパターニングすることができ、高価なレーザーステップを使用してインターポーザー基板522にホールを形成する必要はない。
【0198】
代替インターポーザー520bは、フレキシブルメンブレン602上に懸架されているため、インタフェース面が平らではないアプリケーションにも使用することができる。また、コネクタとしての代替インターポーザー520bの厚さは、わずか数ミクロンとすることができる。また、代替インターポーザー520bはプローブチップ基板310のビアピッチによって制限されないため、コネクタピッチをずっと小さくすることができる。このピッチは、フォトリソグラフィ工程によってスプリングをパターニングできる最小の長さとすることができる。また、相互接続も形成する目的で、持ち上がらない金属部分をメンブレン602上にパターニングすることもできる。支持枠610が金属などの延性材料から作製されている場合には、支持枠610も曲げることができる。これによって、メンブレンインターポーザー520bをあらゆる位相形状に成形することができる。
【0199】
図59は、インターポーザーを製造する模範的な工程を示すフローチャートである。厚さ約2000オングストロームの第一引き離し層594(一般にはチタンを有する)が、一般にはシリコン、アルミニウム、セラミック、またはガラスを有する基板592上に、スパッタリングによって形成される(632)。例えば図37および/または図38に見られるものなど、複数の応力層17a~17nを有し、結果的に下向きの剥離応力を有するMoCr応力サンドイッチ層596 が、第一引き離し層594上にスパッタリングによって形成される(634)。第一応力サンドイッチ層596内の複数の応力層17a~17nは、一般には、下向きの剥離応力を有する固有の応力勾配が生じるように、第一張力層17aと最終圧縮層17nとを有する。
【0200】
互いに離れる方向に持ち上がるフィンガーの領域を定義するパターンを有する第二層598(一般にはチタンを有する)が、第一応力サンドイッチ層の上に形成される。留意すべき点として、この引き離し層が除去されるフィンガーの領域597が存在する。この領域597は、上側および下側フィンガー521の間の接触領域を形成している。第二引き離し層598は、一般には厚さが約1000オングストロームであり、スパッタリングによって形成され(636)、接続領域 597がフィンガー持ち上がり領域528の端部をほぼ定義するように、パターニングされる(638)。次いで、例えば、図37および/または図38に見られるものなど、複数の応力層17a~17nを有し、結果的に上向きの剥離応力を有する第二MoCr層600が、第二引き離し層598の上にスパッタリングによって形成される(640)。第二応力サンドイッチ層600内の複数の応力層17a~17nは、一般には、上向きの、すなわち逆方向の剥離応力を有する固有の応力勾配が生じるように、第一圧縮層17aと最終張力層17nとを有する。
【0201】
次いで、スプリング521のフィンガーレイアウトを画成するため、フォトレジストがスピンコーティングされる(642)。スプリング521が持ち上がる領域と、ホールがメンブレン内に必要である他の領域(例えば、構成要素が突き出すための領域)とを開く目的で、ウェットまたはドライエッチング工程のいずれかなどによって、第二MoCr層600がエッチングされる(644)。次いで、例えば、第一MoCr層596におけるフィンガー521をさらにアンダーカットする(undercut)一方で、第一引き離し層594は残されて覆うように、第二引き離し層598がエッチングされ(646)、第一MoCr張力応力層596がエッチングされる(648)。
【0202】
模範的な厚さ18 &micro;m~25 &micro;mのポリイミド層602がスピンコーティングされ、持ち上がりウィンドウ(lift window)とフィンガー521の平らな基部領域とが開くように、ウェットエッチング工程によってパターニングされる(650)。次いで、すべてのフィンガー521の間が短絡し、ただし持ち上がりが妨げられないように、シード層604(一般には金を有する)がスパッタリングされてパターニングされる(652)。
【0203】
次いで、フィンガー521のみがめっきされるように、陽極処理アルミニウムなどのめっきマスク606がシード層604の上にパターニングされる(654)。めっきマスク606は、基板592が除去された後に曲がらずに耐えることができ、かつフィンガー521を攻撃することなく容易に除去される材料から成る。
【0204】
支持リング610(一般にはステンレス鋼を有する)は、デカル支持リングとしての役割りを果たす目的で、ウエハ基板592の外側にエポキシ接着剤608などによって取り付けられる(656)。取り付けられる基板上の領域には、シード層604とめっきマスク606のいずれも含まれていない。プローブスプリングがプリント基板に接触する場合、支持リング用にはステンレス鋼が好ましい材料である。ピンがシリコンデバイスに接触することが要求されるいくつかの実施例においては、支持リングは、シリコンの熱膨張係数に相対的に近い低い熱膨張係数を持つ材料(例えば、モリブデン)から成る。
【0205】
次いで、アセンブリに対して、フィンガーを持ち上げるためのエッチングが行われ(658)、これに起因して、層596,600によって画成されているフィンガー521a,521bがそれぞれ反対方向に持ち上がると、メンブレン全体が基板592から剥離する。フィンガーを持ち上げるためのエッチングステップ 658では、支持リング610の下に位置する領域においてより長い時間が必要となることがある。
【0206】
これに代えて、基板592は、機械的手段または化学的手段によって除去することのできる材料(例えば、アルミニウム)から作製することができる。
【0207】
図58に見られるように、持ち上がったフィンガー521a,521b上には、必要に応じて一層以上のめっき層622a,622bが形成されることが好ましい(660)。インターポーザーのいくつかの実施例520bにおいては、めっき層622a,622bは、アンダーコートとしての1~10 μmのニッケル層と、0.2~5 μmの接触磨耗層(例えば、ロジウム、パラジウム、パラジウム-コバルト、または金など)とを有する。次いで、めっきマスク606が除去され(662)、その後、露出しているシード層604が除去される(664)。
【0208】
デカルインターポーザーアセンブリ300の代替実施例においては、例えば、Circuit Components社(アリゾナ州テンペ)から入手可能なISOCON(登録商標)コネクタなどが、上側インターポーザー344として使用される。 ISOCON(登録商標)コネクタでは、一般に電気的な接触を確立するのに大きな力が必要であるが、この必要な力をZブロックとマザーボードプリント基板 304との間で達成できるため、上側インターポーザー344用として好適なコスト効果の高い相互接続がしばしば提供される。Zブロック342とプローブチップ16,310との間の接続用には、一般には、要求される接触力が相対的に低いインターポーザーが選択される。
【0209】
Zブロックのアーキテクチャ。図60は、Zブロックプリント基板342の平面配置図670である。図61は、Zブロックプリント基板342の一部の詳細図 680である。Zブロック342は、Zブロック基板672の対向する面343a,343bの間に延在する複数の電気接続部(例えば、めっき処理済スルーホール674)を有する。強化されたプローブアセンブリ300内での位置合わせを目的として、図60に見られるように、インターポーザー位置合わせピン 330と補強リング位置合わせピン350が、Zブロック基板672を貫いて配置されている。図60に示されているZブロック基板672は、Zブロックを保持クリップ352(図43)に位置合わせするための保持クリップくぼみ676も有する。図61に見られるように、複数の電気接続部は、一般には、デバイスのVccおよびVss接続を含む信号接触子サイト682と、システム接地GND接触子サイト684とを有する。図43に示されている模範的なZブロック 342においては、高周波システム環境用のインピーダンス整合構造を提供するなどを目的としてインピーダンスとクロストークの性能を向上させるために、各信号682のめっき処理済スルーホール674は、四個の接地684のめっき処理済スルーホール674によって囲まれている。
【0210】
代替の強化されたプローブアセンブリ。図62は、マザーボードのプリント基板304とZブロックとの間に永久的なインタフェース702を有するZブロックデカルインターポーザーアセンブリ300cの一部の断面図700である。
【0211】
図42および図43に見られるような、上側インタフェース308(図40)がインターポーザー344を有するZブロックデカルインターポーザーアセンブリ 300bとは対照的に、Zブロックデカルインターポーザーアセンブリ300cにおける上側インタフェース308は、Zブロック342をマザーボード304 に固定することを目的とする永久的なインタフェース702(例えば、はんだボールアレイ756、異方性導電膜、または導電ピン)を有する。
【0212】
Zブロックデカルインターポーザーアセンブリ300cのいくつかの実施例においては、マザーボード304と下側補強リング346との間に位置するシム 827(図72)によって、平面性調整機能が提供されている。プローブの力は、下側補強リング346によってではなく、Zブロック342とマザーボード 304との間に位置する上側インタフェース接続部702によって支持されている。Zブロックデカルインターポーザーアセンブリ300cでは、Z軸の変換と平面性の対応性の両方が提供され、その一方で、相対的に安価な永久的上側インタフェース702が提供され、上側インターポーザー344のコストが削減される。
【0213】
図63は、テスター側スプリング64a~64nを有するプローブチップ16,310を有するZブロックデカルインターポーザー300dの一部の断面図である。下側インタフェース312(図40)がインターポーザー348を有するZブロックデカルインターポーザーアセンブリ300cとは対照的に、Zブロックデカルインターポーザーアセンブリ300dにおける下側インタフェース308は、プローブチップ16,310の上面311b上に、Zブロック342に直接的に接触するスプリング接続部64a~64nを有する。
【0214】
Zブロックデカルインターポーザーアセンブリ300dのいくつかの実施例においては、プローブチップ16,310の上面311b上のスプリング接続部 64a~64nは、冗長型スプリング64a~64nである。図63に示されているZブロック342は、プローブチップ16,310上に実装されるバイパスキャパシタ724のための空間を提供する空隙722を有することが好ましく、この空隙は、一般にはフライス加工によって形成される。
【0215】
Zブロックデカルインターポーザーアセンブリ300dでは、Zブロックデカルインターポーザーアセンブリ300cと類似する方式において、マザーボード 304とプローブリング346との間に位置しているシム827によって、または差動ねじアセンブリ824(図73)などの他の平面性調整機構のいずれかによって、平面性調整手段324が提供される。このZブロックデカルインターポーザーアセンブリ300dでは、Z軸方向の変換と平面性の対応性の両方が提供され、その一方で、下側インターポーザー348のコストが削減される。このZブロックデカルインターポーザーアセンブリ300dも、Zブロックの空隙 722内に位置するバイパスキャパシタ724を有し、この空隙は、インターポーザーの切除領域574(図55、図56)(一般にはレーザー切断によって形成される)よりも一般には低コストである。
【0216】
図64は、はんだボールアレイ756(図65)を有するピンブロック742を有するプローブカードアセンブリ300eの一部の断面図740である。図65 は、はんだボールアレイ756を有するピンブロックの詳細な線図750である。プローブカードアセンブリ300eにおいては、上側インタフェース 308(図40)は、ピングリッドアレイ745を有し、下側インタフェース312(図40)は、はんだボールアレイ756を有し、プローブチップ310 は、下面311a上に位置している片側スプリング61a~61nを有する。ZIF作動テンプレート743は、金属ピンブロック742とマザーボード304 との間に位置している。システムの平面性は、三つの差動ねじアセンブリ745などによって提供されることが好ましい。
【0217】
ピンブロック742は、複数の誘電体ホール746を有し、ピンコネクタ744のアレイ745がこのホールを貫いて延在する。ピンテンプレート 748,752は、一般には、KAPTON(登録商標)から成り、ピンブロック742の対向する面上に位置している。ホール746内の接地接触子758 は、一本以上のピン744とピンブロック742との間の接地接続を提供する。ピングリッドアレイ745は、マザーボード304におけるピンソケットアレイ 755に電気的に接触している。
【0218】
図66は、ピンブロック742を有するプローブカードアセンブリ300fの一部の断面図760であり、このアセンブリにおいては、プローブチップカード 16,310が裏側スプリング64a~64nを有する。図67は、プローブチップカード16,310が裏側スプリング64a~64nを有する、ピンブロック742の一部の詳細な線図770である。ピンブロック742は、複数の誘電体ホール746を有し、ピンコネクタ744のアレイ745がこのホールを貫いて延在する。ピングリッドアレイ745は、マザーボード304を貫いて延在しており、マザーボード304とピンブロック742との間の機械的結合と電気的接続の両方ははんだジョイント762によって提供されている。プローブチップ16,310は、撓める状態で対応性を持つ部材またはデカル326によってアセンブリから懸架されており、その一方で、上側スプリング64a~64nは対応性を持つ電気的インタフェース312(図40)を提供している。従って、プローブチップ16,310は、ピンブロック742に対して可動であり、プローブカードアセンブリ300fに対応性を提供する。
【0219】
図67、図69、および図71に見られるように、一つ以上のキャパシタ724(例えばバイパスキャパシタ724)は、目的のプロービング環境に基づいて、ピンブロック742上に配置することができる。
【0220】
図68は、ピンブロック742を有するプローブカードアセンブリ300gの一部の断面図780であり、このアセンブリはプレスばめピン接続部782を有し、プローブチップカード16,310が裏側スプリング64a~64nを有する。図69は、ピンブロック742の詳細な線図790であり、このアセンブリはプレスばめピン接続部782を有し、プローブチップカード16,310が裏側スプリング64a~64nを有する。ピングリッドアレイ745は、マザーボード304を貫いて延在しており、マザーボード304とピンブロック742との間の機械的結合と電気的接続の両方はプレスばめピン接続部782によって提供されている。プローブチップ16,310は、撓める状態で対応性を持つ部材またはデカル326によってアセンブリから懸架されており、その一方で、上側スプリング64a~64nは対応性を持つ電気的インタフェース312(図40)を提供している。従って、プローブチップ16,310は、ピンブロック 742に対して可動であり、プローブカードアセンブリ300gに対応性を提供する。
【0221】
図70は、SMTはんだと上側インターポーザー344とを有するピンブロック742を有するプローブカードアセンブリ300hの一部の断面図である。図71は、SMTはんだと上側インターポーザーとを有するピンブロック742の詳細な線図である。
【0222】
対応性を持つ保持器型のプローブカードアセンブリの平面性調整機構。上述されているように、デカルインターポーザーアセンブリ300の多くの実施例は、平面性調整機構324を有する。図72は、平面性調整機構324を有するZブロックデカルインターポーザーアセンブリ300iの一部の断面図であり、この調整機構は、マザーボードプリント基板304の下面305aにシムプレート825と取り付けねじ823とによって取り付けられている一つ以上のシム827を有する。マザーボードプリント基板304の周辺部付近の複数の位置に一つ以上のシム827を使用することによって、プローブリング284(図39)に対するインターポーザーアセンブリ300i全体の平面性を調整することができる。
【0223】
図73は、平面性調整機構324を有するZブロックデカルインターポーザーアセンブリ300の拡大側面組立図830であり、この調整機構324は、複数の差動ねじアセンブリ824と複数のショルダーボルトアセンブリ832とを有する。図74は、平面性調整機構を有するZブロックデカルインターポーザーアセンブリ300の拡大透視組立図850である。図75は、複数の差動ねじアセンブリ824を有する平面性調整機構324を有するZブロックデカルインターポーザーアセンブリ300の一部の断面図である。複数の差動ねじアセンブリ824は、図75に見られるようにアセンブリ824の先端部(例:プランジャ)によって平面が容易に定義されるように、一般には三つのアセンブリ824を有する。図74に見られるように、マザーボードプリント基板304は、複数の締結具322によって上側補強プレート302に固定的に取り付けられている。差動ねじアセンブリ824は、下側補強リング346から上側補強リング302まで延在している。従って、この差動ねじアセンブリ824を調整することによって、下側補強リング346とアセンブリ300iの上側部分871(図75)との間の平面性が調整される。
【0224】
図73に見られるように、差動ねじアセンブリ824は、プランジャ826と、差動ねじ828と、薄ナット830とを有する。図75に見られるように、プランジャ826の先端部872は、マザーボードプリント基板304内に画成されているマッチングスロット827内に回転式に固定されている。差動ねじ828 は、プランジャ826とねじ式に係合しており(874)、かつ、上側補強リング302ともねじ式に係合している(876)。図75に示されているねじ界面 874,876は、異なる回転ピッチ(すなわち、ねじ山間隔(thread gage))を有し、従って、差動ねじ828を回転させることによってプランジャ826が軸方向に動き(878)、これによって下側補強リング346とマザーボードプリント基板304との間の隔たり838(図75)が調整される。マザーボードプリント基板304に対するプローブチップの平面性を精密に調整できるように、ピッチ差877によってプランジャの調整878が可能となる。
【0225】
図73および図75に見られるように、ショルダーボルトアセンブリ832のそれぞれは、ショルダーボルトねじ834と、対応性を持つOリングまたはスプリングワッシャ836とを有する。上述されているように、差動ねじアセンブリ824を調整すると、下側補強リング346がマザーボード304に対して動く。下側補強リング346はプランジャ826の先端部872と連動的に取り付けられており、その一方で、ショルダーボルトねじ834と、対応性を持つOリングまたはスプリングワッシャ836とによって補強リング346がプランジャ826に接触した状態に維持されているため、マザーボード304に対する下側補強リング346の平面性、すなわち隔たり838が調整可能である。
【0226】
高性能スプリング接触子パッケージ。図 76は、集積回路44用の高性能スプリングパッケージ902aの一部の断面図900である。図77は、多層配線924をさらに有する、集積回路44用の代替の高性能スプリングパッケージ902bの一部の断面図920である。図78は、集積回路44用の高性能スプリングパッケージ902の上側図940である。
【0227】
図76に見られるように、パッケージ基板903は、第一面906aと第二面906bとを有する第一基板904を有し、プローブスプリング61は、第一面 906a上に位置しており、第一面906aから第二面906bまで延在する電気接続部908(例:めっき処理済スルーホール)まで延在する。図77に見られるように、パッケージ基板903は、第一基板904の第一面906a上に位置している追加の配線層922をさらに有することができ、この場合、プローブスプリング61は、この配線層922の外面923a上に位置しており、多層配線924によって電気接続部908に接続されている。
【0228】
高性能スプリングパッケージ902は、単一のICまたはMCMパッケージの場合など、基板903上のスプリング61を利用してパッケージを構築するための構造を有する。極小ピッチ(例:10~50 μ)の極めて小型のプローブスプリングのアレイを作製することに加えて、低い製造コストおよび良好に制御された均一性が達成されるように、プローブ先端 61は、薄膜またはICまたはMEMSベースの工程方式を使用して基板903上に作製される。
【0229】
プローブスプリング61は、第一基板904(図76)の第一面906a上、または、多層配線層924の外面923a上のいずれかに、上述されているように薄膜またはICまたはMEMSベースの工程方式を使用して作製される。プローブスプリング61からの信号は、接続されている集積回路44から、好ましくは多層配線924を使用して送られる。パッケージ902a,902bの反対側906bは、電気接触子910を有し、これは例えば、一般には1.0 mmなどのアレイピッチでの一般的なマイクロボールグリッドはんだアレイパッド918(図76)か、または、一般にはプリント基板912a,912bのホール936にはんだ付け可能であるブレイズドオンピン(braised on pin)またはパッケージピン928を有する接触子のいずれかである。従って、パッケージ902a,902bは、最終製品934などのプリント基板 912a,912bに電気接触子910を通じて接続することができる。
【0230】
一枚の基板ウエハ(複数の高性能スプリングパッケージ902を有することができる)を、第一基板ウエハ904上に構築することができ、これによってコスト効果の高い製造が得られる。表面積の小さなスプリングプローブ基板904の場合、一般には、一枚のウエハ92(図13)から、いくつかのスプリングプローブ接触子パッケージを作製することができる。例えば、標準的な4 inchの丸型開始ウエハ上に、24個ものサイトを確立することができる。
【0231】
図78に見られるように、基板アセンブリ902上に位置しているマイクロBGAパッド944のアレイ942は、標準のピッチ(例:0.5mm、1mm、または1.27mmのピッチなど)で配置されていることが好ましい。図76および図77に見られるように、高性能スプリングパッケージ902は、一般には高周波パワーデカップリングにおいて役立てる目的で、キャパシタ932をさらに有することができる。キャパシタ932は、基板904の面906a,906b のいずれかに実装されるか、または、基板904内、一般には基準面と、基板904の未使用トレース領域上に形成されている平面との間に、例えば平行板キャパシタ932として形成することができる。第一基板904がシリコンから成る実施例の場合、シリコン基板904の中に内蔵キャパシタ932が形成されることが好ましく、このキャパシタは、一般には、集積回路の製造手法を使用しての、金属-誘電体-金属の構造か、金属-誘電体-高濃度ドーピングされた半導体、またはp-n接合を有する。第一基板904が非半導体材料から成る実施例の場合、金属-誘電体-金属構造のキャパシタが、集積回路製造手法を使用して基板904の上または中に作製されることが好ましい。基板903が導電性材料または半導体材料(例えばドーピングされたシリコン)から成る実施例の場合には、製造工程は、図38に関して先に説明されているプローブチップの変形と類似して変更される。この場合の変更としては、プローブチップ基板の製造の場合のように、基板面上と、基板を貫いて延在するビアの壁とに、電気絶縁性膜(例:酸化物)を蒸着または形成することが含まれる。
【0232】
上述されているように、プローブカードパッケージ902の構造では、プローブ先端61の間の電気的距離が非常に短く、かつ、インピーダンスが制御された環境が形成される。これによって、高性能スプリングパッケージ902を高周波アプリケーションに使用することが可能になる。図76に示されているように、高性能スプリングパッケージ902では、ピン916を囲む接地経路を形成することによって、遮蔽された高周波ピン916へのアクセスが提供される。図77に見られるように、接地は、スプリング61の下の層924内の配線トレース938によって提供されている。さらに、スプリングプローブ基板903は、超高周波アプリケーション用に修正されることが好ましい。
【0233】
基板903の片面または両面上のトレースがインピーダンス制御される必要のある実施例の場合、基板の中のトレースの上、トレースの下、またはトレースの上下に一つ以上の導電基準面を追加することができる。基板903には、一つまたは二つの基準面に接続されている交互接地基準トレースを含めることもでき、これによって、遮蔽された同軸伝送ライン環境が効果的に形成される。第一基板904は、一般にはセラミック材料を有するが、層状基板922は、例えば、有機材料または無機材料などの誘電体材料の中に導電トレースを有する。その他のいくつかの実施例の場合、図37Bに示されており上述されているビアを構築することによって、基板903内の導電ビアまたはスルーホールに、インピーダンスが制御された環境を形成することができる。このようなビアには、誘電体膜によってコアの電気導体から隔てられている接地面が設けられている。
【0234】
高性能スプリングパッケージの利点。上述されているように、スプリング61,412を作製する目的に、MEMSまたはIC工程方式を使用することができる。高性能スプリングパッケージ902は、「チップファースト」手法および/またはBBULパッケージの利点のすべてを有し、その一方で、デバイスIC 44を交換することもできる。IC 44を取り付ける前に高性能スプリングパッケージ902をテストすることができ、これによって、一つのパッケージ902に一つまたは複数のIC 44を配置するときのコストおよびリスクが大幅に減少する。
【0235】
スルーホール908を有する基板構造は、高密度相互接続(HDI)の機能を提供する目的で、複数の電気配線層が上に構築されていることが好ましい。この場合のHDIの機能としては、1)フリップチップパッケージまたはワイヤボンドパッケージにおいて達成可能であるよりもI/O接続の密度がずっと高いこと、 2)より高い相互接続性能をIC上で達成できること、3)パッケージがより薄いこと、があげられる。
【0236】
このスプリングベースのパッケージ902では、多層配線トレースに対してほぼ平らにスプリングが押されることにより、パワーおよび信号インダクタンスが低く維持される。スプリングの真下の配線は、国際公開第09623号パンフレットに記載されているように、制御されたインピーダンスが維持されるように構築することができる。基板を貫く距離は、極めて短く維持することができ、セラミック基板904ではRF周波数動作がサポートされる。
【0237】
高性能スプリングパッケージ902は、高価なIC 44をパッケージ902に装着する前にテストするなどの目的で、「テストパッケージ」として使用することができる。高性能スプリングパッケージ902のいくつかの実施例は、プローブカードの実施例の場合について上述されているものに類似する、デカップリングとインピーダンス制御に関する特徴を有する。プローブスプリング61と、IC 44上の電気接触パッドとの間に永久的な接続が望まれる場合には、テストを行って良好なチップであることが確認された後、パッケージングに使用されている従来の結合手法(例えば、接触パッドにおけるはんだのリフロー、または接着剤による)を使用して、永久的な接続を行うことができる。
【0238】
図76に見られるように、高性能スプリングパッケージ902は、集積回路デバイスをパッケージに接触した状態に保持する手段、例えば、基板903を貫いて画成されている真空引き込みポート905、および/または、一時的な蓋またはプッシュダウンプランジャ909、をさらに有することができる。
【0239】
高性能スプリングパッケージ902は、無線周波数において個々のIC 44をテストするための効果的なデバッギングツールであり、この場合、パッケージ902の負荷は、パッケージング後の最終的なICの性能に影響する。従って、高性能スプリングパッケージ902を使用することは、MCMパッケージにおいて使用される新しい高性能デバイスのタイムトゥーマーケット特性 (time-to-market characterization)を短縮するうえで役立ち、なぜなら、実負荷テストに合格するまで集積回路デバイス44を最終パッケージ902に装着する必要がないためである。このテストパッケージでは、一つ以上の無線信号と数百または数千のデジタル信号とを混合する手段も提供される。高性能スプリングパッケージ902では、テストボリュームおよび/または製造ボリュームが多いとき、類似する周波数能力を有するBBUL手法と比較して、パッケージコストが大幅に低減する。
【0240】
クイックターン型のプローブアセンブリ。図83は、プローブアセンブリのクイックターン製造工程960のフローチャートである。上述されているように、プローブカードアセンブリ60、強化されたプローブカードアセンブリ300、および高性能スプリングパッケージ902の多数の実施例は、一般に、標準化された構成要素を相当に有し、このような構成要素は、ウエハ92上の一つ以上のデバイス44との接続を目的として容易に再利用できる
図83に示されているプローブアセンブリ60,300は、マスタースライス構造962を有し、この構造は、マザーボード基板304(図40)などの標準化された部分と、少なくとも一つの中間コネクタ306と、プローブチップ310の標準化された部分とを有する。プローブチップ310の標準化された部分には、一般には、プローブチップ基板とコネクタ面電気接続部とが含まれており、一般には、プローブチップ基板310のコネクタ面62bからプローブ面62a まで延在する標準化された電気接続部を含めることもできる。理解すべき点として、中間コネクタは、インターポーザーおよび/またはZブロックなどの一つ以上の構成要素を含むことができる。
【0241】
図83に見られるように、クイックターン工程960、すなわち、ウエハ上の少なくとも一つのデバイスと接続するためのプローブアセンブリを製造する方法は、一般には、マスタースライス962を構築するステップを有し、当該ステップが、
下面および上面と、前記上面から前記下面まで延在する複数の電気導体とを有するマザーボード基板304を作製するステップ964と、
上側インタフェースと下側インタフェースとを有する少なくとも一つの中間コネクタ306を作製するステップ966であって、前記上面が、前記マザーボード基板304の前記下面の近くに配置可能であり、前記中間コネクタ306が、前記マザーボード基板304の前記下面上の前記電気導体のそれぞれに対応する、前記上側インタフェースと前記下側インタフェースとの間の少なくとも一つの導電接続部を有する、前記ステップ966と、
コネクタ面62bと、前記コネクタ面62bの反対のプローブ面62aと、固定レイアウトに配置されてる前記コネクタ面上の複数の接触子と、を有するプローブチップ基板を設計するステップ968であって、前記コネクタ面62bが、前記中間コネクタ306の前記下面の近くに配置可能である、前記ステップ968 と、
を有する。
【0242】
図83に示されているクイックターン工程960は、次いで、ウエハ92上の前記少なくとも一つのデバイス44の相互接続仕様972を受け取るステップであって、前記相互接続仕様972がデバイス44の相互接続位置を有する、前記ステップを含む。
【0243】
次いで、ステップ974において、標準化されている情報968と、受け取った、すなわちカスタマイズされた情報970の両方に基づいて、プローブチップ基板310が作製され、この場合に、プローブチップ基板は、ウエハ92上の相互接続位置972に対応する複数のスプリングプローブ接触子先端412(例えば図40に見られるもの)をプローブ面62a上に有し、スプリングプローブ接触子先端のそれぞれが、コネクタ面62a上の少なくとも一つの接触子に電気的に接続されている。
【0244】
システムの利点。上述されているように、プローブカードアセンブリ60、強化されたプローブカードアセンブリ300、インターポーザー構造520,520b、および高性能スプリングパッケージ902は、従来のプローブおよびパッケージ技術に優るいくつかの利点を持つ。
【0245】
例えば、プローブカードアセンブリ60、強化されたプローブカードアセンブリ300、インターポーザー構造520,520b、および高性能パッケージ 902の多数の実施例は、フォトリソグラフィによってパターニングされるスプリング61,412,521を有し、一般には、応力金属膜ロット工程によって作製され、この工程は、機械的工程またはMEMSベースの工程のいずれよりも本質的にコストが低い。スプリングは、三次元構造が形成される二次元工程方式を使用することによって形成される。対照的に、他の従来のスプリング工程では、三次元のスプリングを形成するための追加の工程ステップが必要となる。代替の工程実施例においては、スプリングは、めっき層の間の差応力勾配を形成するための二次元のめっき処理方式を使用することによって形成され、この勾配によって三次元構造が形成される。
【0246】
本文書に説明されているアセンブリ手法では、フォトリソグラフィなどのIC工程手法によって各基板上にすべてロット製造される応力金属プローブ要素、すなわちスプリングを有する構成要素または基板が使用されている。その結果として、この工程では、現在使用されているいかなる技術によっても製造できない、極めて短く(例:長さ100~200ミクロン、幅10~20ミクロン)、かつ超微細なピッチ(例:10~50ミクロン)のプローブ要素のアレイを使用するプローブカードアセンブリおよびパッケージを製造することができる。さらに理解すべき点として、本文書に開示されているアセンブリおよびパッケージング手法の用途は、本文書に提示されている技術を使用して、または現在利用可能な任意の他の方法によって製造できる、より大きな寸法とより広いピッチとを有するプローブ要素のアレイを有する基板または構成要素にも適用可能である。
【0247】
さらに、プローブアーキテクチャは、一般には、他方の側に直接接続するためのスルーホールを有する基板を有し、このアーキテクチャにおいては、スプリングの配置と、ホールからスプリングまでの経路接続の両方が、フォトリソグラフィによる方法によって画成され、これによって、より単純な開始材料を使用することと、工程時間を短縮することとが可能となる。
【0248】
さらに、接続対象である様々なデバイス44およびウエハ92用として、プローブカード60,300および/またはパッケージアセンブリ902の大部分を実施することができるように、開示されているプローブおよびパッケージアーキテクチャ60,300,902は、標準化されたマスタースライス962を使用することによって、再利用可能な構成要素、すなわち標準構成要素を有することが好ましい。マスタースライス962は、特定の相互接続仕様972に容易に合致し、これによって、特定の顧客に対して必要な「総所有コスト」が減少する。
【0249】
上述されているように、プローブカードアセンブリ60、強化されたプローブカードアセンブリ300、インターポーザー構造520,520b、および高性能パッケージ902は、一つ以上の基板を有し、この基板は、セラミック、ガラスまたは石英、シリコン、有機基板(organic board)、および/または多層セラミックなど、様々な開始材料を要件に応じて使用して構築することができる。アセンブリ基板は、配線および性能の向上などを目的として、多層金属層をさらに有することができる。
【0250】
さらに、プローブカードアセンブリ60、強化されたプローブカードアセンブリ300、インターポーザー構造520,520b、および高性能パッケージ 902は、プローブ61,412を保護する、デバイス44および/またはウエハ92を保護する、または固定式ウエハレベルカートリッジまたはカセットアプリケーションにおけるパッドのスクラブ(pad scrub)の量を制御するなどを目的として、これらの任意の基板上に支持棒(stand off)をさらに有することができる。支持棒は、インターポーザー520,520bから延在するスプリングを保護するなどを目的として、プローブチップ 16,310のコネクタ側62bにも適用することができる。
【0251】
開示されているプローブカードアセンブリシステムと、改良された非平面スプリングプローブと、製造方法は、本文書においては、集積回路テストプローブ、プローブカード、および/またはパッケージとの関連において説明されているが、このシステムおよび手法は、電子構成要素または電子デバイス、バーンインデバイス、MEMSデバイス、またはこれらの任意の組み合わせにおける基板と集積回路との間の相互接続など、必要に応じて他のデバイスと一緒に実施することができる。
【0252】
さらに、本文書に開示されているアセンブリ構造および方法を、シリコンおよびIII-V半導体デバイスの両方、液晶ディスプレイパネル、固体センサーアレイ(例えば、バイオセンサー、環境センサー、表面弾性波デバイスセンサーなど)など、様々な小型集積固体回路のテスト、バーンイン、またはパッケージングを目的として使用できることは、本発明の範囲内である。
【0253】
従って、本発明は、特定の好ましい実施例を参照しながら詳細に説明されているが、本発明が属する分野における通常の技術を有する者には、添付されている請求項の精神および範囲から逸脱することなく、様々な変更および改良を行えることが理解されるであろう。

CONSTRUCTION STRUCTURES AND MANUFACTURING PROCESSES FOR PROBE CARD ASSEMBLIES AND PACKAGES HAVING WAFER LEVEL SPRINGS

FIELD OF THE INVENTION The invention relates to the field of probe card assembly systems. More particularly, the invention relates to improvements in photolithography-patterned spring contacts and enhanced probe card assemblies having photolithography patterned spring contacts for use in the testing or burn-in of integrated circuits.

BACKGROUND OF THE INVENTION In conventional integrated circuit (IC) wafer probe cards, electrical contacts between the probe card and an integrated circuit wafer are typically provided by tungsten needle probes. However, advanced semiconductor technologies often require higher pin counts, smaller pad pitches, and higher clock frequencies, which are not possible with tungsten needle probes.

While emerging technologies have provided spring probes for different probing applications, most probes have inherent limitations, such as limited pitch, limited pin count, varying levels of flexibility, limited probe tip geometries, limitations of materials, and high costs of fabrication.

K. Banerji, A. Suppelsa, and W. Mullen tt, Selectively Releasing Conductive Runner and Substrate Assembly Having Non-Planar Areas, U.S. Patent No. 5,166, 774 (24 November 1992) disclose a runner and substrate assembly which comprises "a plurality of conductive runners adhered to a substrate, a portion of at least some of the conductive runners have non-planar areas with the substrate for selectively releasing the conductive runner from the substrate when subjected to a predetermined stress".

A. Suppelsa, W. Mullen III and G. Urbish, Selectively Releasing Conductive Runner and Substrate Assembly, U.S. Patent No. 5,280, 139 (18 January 1994) disclose a runner and substrate assembly which comprises "a plurality of conductive runners adhered to a substrate, a portion of at least some of the conductive runners have a lower adhesion to the substrate for selectively releasing the conductive runner from the substrate when subjected to a predetermined stress".

D. Pedder, Bare Die Testing, U.S. Patent No. 5,786, 701 (28 July 1998) disclose a testing apparatus for testing integrated circuits (ICs) at the bare die stage, which includes "a testing station at which microbumps of conductive material are located on interconnection trace terminations of a multilayer interconnection structure, these terminations being distributed in a pattern corresponding to the pattern of contact pads on the die to be tested. To facilitate testing of the die before separation from a wafer using the microbumps, the other connections provided to and from the interconnection structure have a low profile".

D. Grabbe, I. Korsunsky and R. Ringler, Surface Mount Electrical Connector, U.S. Patent No. 5,152, 695 (06 October 1992) disclose a connector for electrically connecting a circuit between electronic devices, in which"the connector includes a platform with cantilevered spring arms extending obliquely outwardly therefrom. The spring arms include raised contact surfaces and in one embodiment, the geometry of the arms provide compound wipe during deflection".

H. Iwasaki, H. Matsunaga, and T. Ohkubo, Partly Replaceable Device for Testing a Multi-Contact Integrated Circuit Chip Package, U.S. Patent No. 5,847, 572 (08 December 1998) disclose"a test device for testing an integrated circuit (IC) chip having side edge portions each provided with a set of lead pins. The test device comprises a socket base, contact units each including a contact support member and socket contact numbers, and anisotropic conductive sheet assemblies each including an elastic insulation sheet and conductive members. The anisotropic conductive sheet assemblies are arranged to hold each conductive member in contact with one of the socket contact members of the contact units. The test device further comprises a contact retainer detachably mounted on the socket base to bring the socket contact members into contact with the anisotropic sheet assemblies to establish electrical communication between the socket contact members and the conductive members of the anisotropic conductive sheet assemblies. Each of the contact units can be replaced by a new contact unit if the socket contact members partly become fatigued, thereby making it possible to facilitate the maintenance of the test device. Furthermore, the lead pins of the IC chip can be electrically connected to a test circuit board with the shortest paths formed by part of the socket contact members and the conductive members of the anisotropic conductive sheet assemblies".

W. Berg, Method of Mounting a Substrate Structure to a Circuit Board, U.S. Patent No. 4,758, 9278 (19 July 1988) discloses "a substrate structure having contact pads is mounted to a circuit board which has pads of conductive material exposed at one main face of the board and has registration features which are in predetermined positions relative to the contact pads of the circuit board. The substrate structure is provided with leads which are electrically connected to the contact pads of the substrate structure and project from the substrate structure in cantilever fashion. A registration element has a plate portion and also has registration features which are distributed about the plate portion and are engageable with the registration features of the circuit board, and when so engaged, maintain the registration element against movement parallel to the general plane of the circuit board. The substrate structure is attached to the plate portion of the registration element so that the leads are in predetermined position relative to the registration features of the circuit board, and in this position of the registration element the leads of the substrate structure overlie the contact pads of the circuit board. A clamp member maintains the leads in electrically conductive pressure contact with the contact pads of the circuit board".

D. Sarma, P. Palanisamy, J. Heam and D. Schwarz, Controlled Adhesion Conductor, U.S. Patent No. 5,121, 298 (09 June 1992) disclose"Compositions useful for printing controllable adhesion conductive patterns on a printed circuit board include finely divided copper powder, a screening agent and a binder. The binder is designed to provide controllable adhesion of the copper layer formed after sintering to the substrate, so that the layer can lift off the substrate in response to thermal stress. Additionally, the binder serves to promote good cohesion between the copper particles to provide good mechanical strength to the copper layer so that it can tolerate lift off without fracture".

R. Mueller, Thin-Film Electrothermal Device, U.S. Patent No. 4,423, 401 (27 December 1983) discloses"A thin film multilayer technology is used to build micro miniature electromechanical switches having low resistance metal-to-metal contacts and distinct on- off characteristics. The switches, which are electrothermally activated, are fabricated on conventional hybrid circuit substrates using processes compatible with those employed to produce thin-film circuits. In a preferred form, such a switch includes a cantilever actuator member comprising a resiliently bendable strip of a hard insulating material (e.g. silicon nitride) to which a metal (e.g. nickel) heating element is bonded. The free end of the cantilever member carries a metal contact, which is moved onto (or out of) engagement with an underlying fixed contact by controlled bending of the member via electrical current applied to the heating element".

S. Ibrahim and J. Eisner, Multi-Layer Ceramic Package, U.S. Patent No. 4,320, 438 (16 March 1982) disclose") n a multi-layer package, a plurality of ceramic lamina each has a conductive pattern, and there is an internal cavity of the package within which is bonded a chip or a plurality of chips interconnected to form a chip array. The chip or chip array is connected through short wire bonds at varying lamina levels to metalized conductive patterns thereon, each lamina level having a particular conductive pattern. The conductive patterns on the respective lamina layers are interconnected either by tunneled through openings filled with metalized material, or by edge formed metallizations so that the conductive patterns ultimately connect to a number of pads at the undersurface of the ceramic package mounted onto a metalized board. There is achieved a high component density; but because connecting leads are"staggered"or connected at alternating points with wholly different package levels, it is possible to maintain a 10 mil spacing and 10 mil size of the wire bond lands. As a result, there is even greater component density but without interference of wire bonds one with the other, this factor of interference being the previous limiting factor in achieving high component density networks in a multi layer ceramic package".

F. McQuade, and J. Lander, Probe Assembly for Testing Integrated Circuits, U.S. Patent No. 5,416, 429 (16 May 1995) disclose a probe assembly for testing an integrated circuit, which"includes a probe card of insulating material with a central opening, a rectangular frame with a smaller opening attached to the probe card, four separate probe wings each comprising a flexible laminated member having a conductive ground plane sheet, an adhesive dielectric film adhered to the ground plane, and probe wing traces of spring alloy copper on the dielectric film. Each probe wing has a cantilevered leaf spring portion extending into the central opening and terminates in a group of aligned individual probe fingers provided by respective terminating ends of said probe wing traces. The probe fingers have tips disposed substantially along a straight line and are spaced to correspond to the spacing of respective contact pads along the edge of an IC being tested. Four spring clamps each have a cantilevered portion which contact the leaf spring portion of a respective probe wing, so as to provide an adjustable restraint for one of the leaf spring portions. There are four separate spring clamp adjusting means for separately adjusting the pressure restraints exercised by each of the spring clamps on its respective probe wing. The separate spring clamp adjusting means comprise spring biased platforms each attached to the frame member by three screws and spring washers so that the spring clamps may be moved and oriented in any desired direction to achieve alignment of the position of the probe finger tips on each probe wing".

D. Pedder, Structure for Testing Bare Integrated Circuit Devices, European Patent Application No. EP 0 731 369 A2 (Filed 14 February 1996), U.S. Patent No. 5,764, 070 (09 June 1998) discloses a test probe structure for making connections to a bare IC or a wafer to be tested, which comprises"a multilayer printed circuit probe arm which carries at its tip an MCM-D type substrate having a row of microbumps on its underside to make the required connections. The probe arm is supported at a shallow angle to the surface of the device or wafer, and the MCM-D type substrate is formed with the necessary passive components to interface with the device under test. Four such probe arms may be provided, one on each side of the device under test".

B. Eldridge, G. Grube, I. Khandros, and G. Mathieu, Method of Mounting Resilient Contact Structure to Semiconductor Devices, U.S. Patent No. 5,829, 128 (03 November 1998), Method of Making Temporary Connections Between Electronic Components, U.S. Patent No. 5,832, 601 (10 November 1998), Method of Making Contact Tip Structures, U.S. Patent No. 5,864, 946 (02 February 1999), Mounting Spring Elements on Semiconductor Devices, U.S. Patent No. 5,884, 398 (23 March 1999), Method of Buming-In Semiconductor Devices, U.S. Patent No. 5,878, 486 (09 March 1999), and Method of Exercising Semiconductor Devices, U.S. Patent No. 5,897, 326 (27 April 1999), disclose "Resilient contact structures are mounted directly to bond pads on semiconductor dies, prior to the dies being singulated (separated) from a semiconductor wafer. This enables the semiconductor dies to be exercised (e.g. tested and/or burned-in) by connecting to the semiconductor dies with a circuit board or the like having a plurality of terminals disposed on a surface thereof. Subsequently, the semiconductor dies may be singulated from the semiconductor wafer, whereupon the same resilient contact structures can be used to effect interconnections between the semiconductor dies and other electronic components (such a wiring substrates, semiconductor packages, etc. ). Using the all-metallic composite interconnection elements of the present invention as the resilient contact structures, bum-in can be performed at temperatures of at least 150° C., and can be completed in less than 60 minutes". While the contact tip structures disclosed by B. Eldridge et al. provide resilient contact structures, the structures are each individually mounted onto bond pads on semiconductor dies, requiring complex and costly fabrication. As well, the contact tip structures are fabricated from wire, which often limits the resulting geometry for the tips of the contacts. Furthermore, such contact tip structures have not been able to meet the needs of small pitch applications (e.g. typically on the order of 50 um spacing for a peripheral probe card, or on the order of 75 pm spacing for an area array).

T. Dozier I1, B. Eldridge, G. Grube, l. Khandros, and G. Mathieu, Sockets for Electronic Components and Methods of Connecting to Electronic Components, U.S. Patent No. 5,772, 451 (30 June 1998) disclose"Surface-mount, solder-down sockets permit electronic components such as semiconductor packages to be releasably mounted to a circuit board. Resilient contact structures extend from a top surface of a support substrate, and solder-ball (or other suitable) contact structures are disposed on a bottom surface of the support substrate. Composite interconnection elements are used as the resilient contact structures disposed atop the support substrate. In any suitable manner, selected ones of the resilient contact structures atop the support substrate are connected, via the support substrate, to corresponding ones of the contact structures on the bottom surface of the support substrate. In an embodiment intended to receive an LGA-type semiconductor package, pressure contact is made between the resilient contact structures and external connection points of the semiconductor package with a contact force which is generally normal to the top surface of the support substrate. In an embodiment intended to receive a BGA-type semiconductor package, pressure contact is made between the resilient contact structures and external connection points of the semiconductor package with a contact force which is generally parallel to the top surface of the support substrate".

Other emerging technologies have disclosed probe tips on springs which are fabricated in batch mode processes, such as by thin-film or micro electronic mechanical system (MEMS) processes.

D. Smith and S. Alimonda, Photolithographically Pattemed Spring Contact, U.S. Patent No. 5,613, 861 (25 March 1997), U.S. Patent No. 5,848, 685 (15 December 1998), and International Patent Application No. PCT/US 96/08018 (Filed 30 May 1996), disclose a photolithography patterned spring contact, which is"formed on a substrate and electrically connects contact pads on two devices. The spring contact also compensates for thermal and mechanical variations and other environmental factors. An inherent stress gradient in the spring contact causes a free portion of the spring to bend up and away from the substrate. An anchor portion remains fixed to the substrate and is electrically connected to a first contact pad on the substrate. The spring contact is made of an elastic material and the free portion compliantly contacts a second contact pad, thereby contacting the two contact pads". While the photolithography patterned springs, as disclosed by Smith et al., are capable of satisfying many IC probing needs, the springs are small, and provide little vertical compliance to handle the planarity compliance needed in the reliable operation of many current IC prober systems.

Vertical compliance for many probing systems is typically on the order of 0.004"- 0.010", which often requires the use of tungsten needle probes.

Furthermore, no one has taught a way to interconnect such a probe containing up to several thousand pins to a tester, while effectively dealing with planarity requirements.

As advanced integrated circuit devices become more complex while decreasing in size, it would be advantageous to provide a probe card assembly which can be used to reliably interconnect to such devices.

To accommodate for planarity differences between an array of probe tips and the surface pads on a wafer under test, it may be advantageous to provide a probe substrate which can pivot freely by a small amount about its center. For such a system, however, an accurately controlled force must still be provided to engage the contacts, while holding the substrate positional stable in the X, Y, and theta directions.

Furthermore, for applications in which the substrate includes a large number (e.g. thousands) of wires or signals exiting its backside, wherein supports are located at the periphery of the substrate, these supports must not hinder the fan out exit pathways.

As well, the signal wires must not hinder the pivoting of the substrate, nor should they hinder the controlled force provided to engage the springs against a device under test (DUT).

It would be advantageous to provide a method and apparatus for improved flexible probe springs, which are capable of high pin counts, small pitches, cost-effective fabrication, and customizable spring tips. It would also be advantageous to provide probe card assemblies using such flexible probe springs, which provide planarity compliance to semiconductor devices under testing and/or bum-in, while providing accurate axial and theta positioning.

Similarly, integrated circuit packages provide connections for power signals and transport signals, between an integrated circuit chip IC and a motherboard, so that the integrated circuit chip 44 can interface to the rest of a test system.

Microprocessor devices are some of the IC devices which are most severely limited by today's IC packages. Future microprocessors will need over 10, 000 I/Os and will operate at over 20 Ghz.

In conventional IC packages, the signal, power and ground connections are typically achieved through either wire bonds or solder balls. Conventional packages using wire bonds or solder ball attachments have both signal and power parasitics, which impact performance. Current microprocessors have clock frequencies over 2 GHz but will be advancing to frequencies over 20 GHz in the near future. Current wire bond and solder ball technologies cannot maintain signals in the 20GHz range.

Packages using wire bonds and solder balls attachments have signal, power and ground parasitics that impact performance. Therefore, new solutions are needed. Advanced packages, such as the Intel Bumpless Build-up Layer (BBUL) Packaging Technology (BBUL), build the package on top of the microprocessor, which can help to reduce such parasitics. BBUL packaging can also be used to tightly couple multiple chips in the same package which is referred to as a"chips-first"or Multi-Chip Module (MCM).

Various details of BBUL structures are described in S. Towle, H. Braunisch, C. Hu, R.

Emory, and G. Vandentop, Bumpless Build-Up Layer Packaging, Intel Corporation, Components Research, Presented at ASME International Mechanical Engineering Congress and Exposition (IMECE), New York, 12 November 2001; and R. Emory, S.

Towle, H. Braunsich, C. Hu, G. Raiser, and G. Vendentop, Novel Microelectronic Packaging Method for reduced Thermomechanical Stresses on Low Dielectric Constant Materials, Intel Corporation, presented at Advanced Metallization Conference, Montreal, Canada, 09 October 2001.

The high-density build up layer on top of ICs has much better performance than traditional packaging approaches. The interconnections to the chip are shorter than solder bumps and much shorter than wire bonds, resulting in far lower inductance. Capacitors can be located closer to the IC, which enables better power delivery. The shorter signal distances should allow the IC to run at lower voltages, reducing electrical cross talk and also reducing power consumption. The high density interconnect (HDI) allows more interconnects from the silicon than solder bumps or wire bonds. In many cases, the delays and cross talk of the signals carried in the interconnect of the HDI is lower than delays of signals carried in the interconnect of the IC. Higher performance can be attained by having signals leave the IC interconnect and travel through the HDI interconnect since the propagation delays and cross talk are better in the HDI than on the IC. BBUL packages are thinner and capable of embedding multiple ICs in the same package.

This BBUL and MCM approaches tend to be very expensive due to fabrication complexity, and the need to guarantee that every chip in the module is good. Any bad chip or defect in manufacturing of the HDI between chips will cause all of the chips and the package to be rejected. A bad chip is any chip that does not meet performance requirements. Resultant BBUUMCM packages are therefore typically significantly more expensive to manufacture than individually packaged ICs. In the past, the"chips-first" approach was only used to build MCMs used in satellites for space applications, where the smaller size and weight justified the higher cost.

It would be advantageous to provide a package which can be tested prior to attaching integrated circuits. Such a package would constitute a major technical advance. Further more, it would be advantageous to provide a package which provides through holes comprising multiple electrical routing layers, and provides advanced high density interface (HDI) functions, such as higher densities of 1/0 connections than attainable in flip-chip or wire bonded packages, high interconnect performance to an IC, within a thinner package. Such a package would constitute a further technical advance.

SUMMARY OF THE INVENTION Several embodiments of enhanced integrated circuit probe card assemblies are disclosed, which extend the mechanical compliance of both MEMS and thin-film fabricated probes, such that these types of spring probe structures can be used to test one or more integrated circuits on a semiconductor wafer. Several embodiments of probe card assemblies, which provide tight signal pad pitch compliance and/or enable high levels of parallel testing in commercial wafer probing equipment, are disclosed. In some preferred embodiments, the probe card assembly structures include separable standard components, which reduce assembly manufacturing cost and manufacturing time. These structures and assemblies enable high speed testing in wafer form. The probes also have built in mechanical protection for both the integrated circuits and the MEMS or thin film fabricated spring tips and probe layout structures on substrates.

Alternate card assembly structures comprise a compliant carrier structure, such as a decal or screen, which is adhesive attached to the probe chip substrate.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a plan view of a linear array of photolithographically patterned springs, prior to release from a substrate; Figure 2 is a perspective view of a linear array of photolithographically patterned springs, after release from a substrate; Figure 3 is a side view of a first, short length photolithographically patterned spring, having a first effective radius and height after the short length spring is released from a substrate; Figure 4 is a side view of a second, long length photolithographically patterned spring, having a second large effective radius and height after the long length spring is released from a substrate; Figure 5 is a perspective view of opposing photolithographic springs, having an interleaved spring tip pattern, before the springs are released from a substrate; Figure 6 is a perspective view of opposing photolithographic springs, having an interleaved spring tip pattern, after the springs are released from a substrate; Figure 7 is a top view of opposing pairs of interleaved multiple-point photolithographic probe springs, in contact with a single trace on an integrated circuit device; Figure 8 is a plan view of opposing single-point photolithographic probe springs, before the springs are released from a substrate; Figure 9 is a top view of parallel and opposing single-point photolithographic probe springs, after the springs are released from a substrate, in contact with a single pad on an integrated circuit device; Figure 10 is a front view of a shoulder-point photolithographic probe spring; U c,,..

Figure 11 is a partial cross-sectional side view of a shoulder-point photolithographic spring in contact with a trace on an integrated circuit device; Figure 12 is a perspective view of a multiple shoulder-point photolithographic probe spring; Figure 13 is a cross-sectional view of a probe card assembly, wherein a plurality of photolithographic spring probes on a lower surface of a substrate are electrically connected to flexible connections on the upper surface of the substrate, and wherein the flexible connections are connected to a printed wiring board probe card; Figure 14 is a partial expanded cross-sectional view of a probe card assembly, which shows staged pitch and fan-out across a substrate and a printed wiring board probe card; Figure 15 is a first partial cross-sectional view of a bridge and leaf spring suspended probe card assembly; Figure 16 is a second partial cross-sectional view of a bridge and leaf spring suspended probe card assembly in contact with a device under test (DUT); Figure 17 is a partially expanded assembly view of a bridge and leaf spring suspended probe card assembly ; Figure 18 is a first partial cross-sectional view of a bridge and leaf spring suspended probe card assembly, having an intermediate daughter card detachably connected to the probe card substrate, and wherein the probe spring substrate is detachably connected to the bridge structure; Figure 19 is a second partial cross-sectional view of the bridge and leaf spring suspended probe card assembly shown in contact with a device under test (DUT); Figure 20 is a cross-sectional view of a wire and spring post suspended probe card assembly; Figure 21 is a cross-sectional view of a suspended probe card assembly having an intermediate daughter card detachably connected to the probe card substrate, and wherein the probe spring substrate is mechanically and electrically connected to the bridge structure by flexible interconnections; Figure 22 is a cross-sectional view of a probe card assembly, wherein a nano-spring substrate is directly connected to a probe card substrate by an array connector; Figure 23 is a cross-sectional view of a wire suspended probe card assembly, wherein a nano-spring substrate is connected to a probe card substrate by an LGA interposer connector; Figure 24 is a cross-sectional view of a small test area probe card assembly, having one or more connectors between a probe card and a daughter card, in which the daughter card is attached to a small area probe spring substrate by a micro ball grid solder array; Figure 25 is a top view of a substrate wafer, upon which a plurality of micro ball grid array probe spring contactor chip substrates are laid out; Figure 26 is a top view of a single pitch micro ball grid array nano-spring contactor chip; Figure 27 is a plan view of a tiled probe strip having a plurality of probe strip contact areas; Figure 28 is a bottom view of a plurality of tiled probe strips attached to a probe card support substrate; Figure 29 is a side view of a plurality of tiled probe strips attached to a probe card support substrate; Figure 30 is a cross-sectional view of a structure which allows a plurality of integrated circuits to be temporarily connected to a bum-in board, through a plurality of probe spring contacts; Figure 31 is a view of a first step of a spring probe assembly coating process, in which a protective coating is applied to a probe surface of a spring probe assembly ; Figure 32 is a view of a second step of a spring probe assembly coating process, in which a layer of photoresistive material is applied to a second substrate; Figure 33 is a view of a third step of a spring probe assembly coating process, in which a coated spring probe assembly is partially dipped into photoresistive material on a second substrate; Figure 34 is a view of a fourth step of a spring probe assembly coating process, in which a coated and partially dipped spring probe assembly is removed from the second substrate; Figure 35 is a view of a fifth step of a spring probe assembly coating process, in which the coated and dipped spring probe assembly is etched, thereby removing the protective coating from portions of the substrate not dipped in the photo-resist; Figure 36 is a view of a sixth step of a spring probe assembly coating process, in which photo-resist is stripped from the spring tips on the spring probe assembly, exposing the protective coating; Figure 37A is a partial cross-sectional view of a reference plane layered spring probe substrate; Figure 37B is a partial cross-sectional view of an alternate ultra high frequency spring probe substrate comprising a shielded coaxial transmission line environment which extends through the substrate; Figure 38 is a partial cross-sectional view of an alternate reference plane layered spring probe substrate; Figure 39 is a partial schematic view of a typical wafer test system; Figure 40 is a schematic diagram of an enhanced probe interface assembly; Figure 41 is a partial cross sectional view of a decal interposer assembly ; Figure 42 is a partial cross sectional view of a Z-block decal interposer assembly ; Figure 43 is an expanded assembly view of a Z-block decal interposer assembly; Figure 44 is a cross sectional view of a probe chip prior to finger lift ; Figure 45 is a cross sectional view of a probe chip after finger lift and plate ; Figure 46 is a plan view of a spring side surface of a probe chip; Figure 47 is a detailed partial layout view of a spring side surface of a probe chip; Figure 48 is a plan view of an interposer side surface of a probe chip; Figure 49 is a detailed partial layout view of an interposer surface of a probe chip; Figure 50 is an expanded assembly view of a decal to probe chip assembly fixture; Figure 51 is a flow chart of a decal probe chip assembly process; Figure 52 is a partial cross sectional view of an interposer structure; Figure 53 is a plan layout view of an interposer; Figure 54 is a partial detailed layout view of an interposer; Figure 55 is a partial plan view of an interposer having vacated springs within a capacitor cavity region; Figure 56 is a partial plan view of an alternate interposer having embedded bypass capacitors and dog bone vias; Figure 57 is a partial cross sectional view of an alternate decal interposer structure before spring lift ; Figure 58 is a partial cross sectional view of an alternate decal interposer structure after spring lift ; Figure 59 is a flow chart showing an exemplary interposer construction process; Figure 60 is a plan layout view of a Z-block printed wiring board; Figure 61 is a partial detailed view of a Z-block printed wiring board; Figure 62 is a partial cross sectional view of a decal interposer assembly having a permanent interface between a motherboard PWB and a Z-block ; Figure 63 is a partial cross sectional view of a decal interposer assembly having a permanent interface; Figure 64 is a partial cross sectional view of a probe card assembly having a pin block with a solder ball array; Figure 65 is a detailed partial schematic view of a pin block having a solder ball array; Figure 66 is a partial cross sectional view of a probe card assembly having a pin block with backside springs; Figure 67 is a detailed partial schematic view of a pin block having backside springs; Figure 68 is a partial cross sectional view of a probe card assembly having a pin block with backside springs and press fit pins; Figure 69 is a detailed partial schematic view of a pin block having backside springs and press fit pins; Figure 70 is a partial cross sectional view of a probe card assembly having a pin block with SMT solder and top interposer; Figure 71 is a detailed partial schematic view of a pin block with SMT solder and top interposer; Figure 72 is a partial cross sectional view of a Z-block decal interposer assembly having a planarity adjustment mechanism; Figure 73 is an expanded side assembly view of a Z-block decal interposer assembly having planarity adjustment; Figure 74 is an expanded perspective assembly view of a Z-block decal interposer assembly having planarity adjustment; Figure 75 is a partial cross sectional view of a Z-block decal interposer assembly having a differential screw planarity ; Figure 76 is a partial cross sectional view of a high performance spring package for integrated circuits; Figure 77 is a partial cross sectional view of an alternate high performance spring package for integrated circuits; Figure 78 is a topside view of a high performance spring package for integrated circuits; Figure 79 is a perspective view of a probe chip mounted to a compliant flexible membrane; Figure 80 is a perspective view of a probe chip mounted to a compliant decal ; Figure 81 is a perspective view of a probe chip mounted to a compliant sheet; Figure 82 is a perspective view of a probe chip mounted to a compliant screen; and Figure 83 is a flowchart for a quick-turn probe assembly fabrication process.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS Figure 1 is a plan view 10 of a linear array 12 of photolithographically patterned springs 14a-14n, prior to release from a substrate 16. The conductive springs 14a 14n are typically formed on the substrate layer 16, by successive layers of deposited metal, such as through low and high energy plasma deposition processes, followed by photolithographic patterning, as is widely known in the semiconductor industry. The successive layers have different inherent levels of stress. The release regions 18 of the substrate 16 are then processed by undercut etching, whereby portions of the spring contacts 14a-14n located on a release layer 19 (FIG. 3) over the release region 18, are released from the substrate 16 and extend (i.e. bend) away from the substrate 16, as a result of the inherent stresses between the deposited metallic layers. Fixed regions 15 (FIG. 3, FIG. 4) of the deposited metal traces remain affixed to the substrate 16, and are typically used for routing (i.e. fanning-out) from the spring contacts 14a-14n. Figure 2 is a perspective view 22 of a linear array 12 of photolithographically patterned springs 14a-14n, after release from a substrate 16. The spring contacts 14a-14n may be formed in high density arrays, with a fine pitch 20, currently on the order of 0.001 inch.

Figure 3 is a side view 26a of a first photolithographically patterned spring 14 having a short length 28a, which is formed to define a first effective spring angle 30a, spring radius 31 a, and spring height 32a, after the patterned spring 14 is released from the release region 18a of the substrate 16, away from the planar anchor region 15. Figure 4 is a side view 26b of a second photolithographically patterned spring 14, having a long spring length 28b, which is formed to define a second large effective spring angle 30b, spring radius 31 b and spring height 32b, after the patterned spring 14 is released from the release region 18b of the substrate 16. The effective geometry of the formed spring tips 14 is highly customizable, based upon the intended application. As well, the spring tips are typically flexible, which allows them to be used for many applications.

Patterned probe springs 14 are capable of very small spring to spring pitch 20, which allows multiple probe springs 14 to be used to contact power or ground pads on an integrated circuit device 44 (FIG. 13), thereby improving current carrying capability. As well, for a probe card assembly having an array 12 of probe springs 14, multiple probe springs 14 may be used to probe)/0 pads on an integrated circuit device 44 under test (DUT), thus allowing every contact 14 to be verified for continuity after engagement of the spring contacts 14 to the wafer 92 under test, thereby ensuring complete electrical contact between a probe card assembly and a device 44, before testing procedures begin.

Improved Structures for Miniature Springs. Figure 5 is a first perspective view of opposing photolithographic springs 34a, 34b, having an interleaved spring tip pattern, before spring to substrate detachment. Figure 6 is a perspective view of opposing interleaved photolithographic springs 34a, 34b, after spring to substrate detachment.

The interleaved photolithographic springs 34a, 34b each have a plurality of spring contact points 24. When spring contacts are used for connection to power or ground traces 46 or pads 47 of an integrated circuit device 44, the greatest electrical resistance occurs at the point of contact. Therefore, an interleaved spring contact 34, having a plurality of contact points 24, inherently lowers the resistance between the spring contact 34 and a trace 46 or pad 47. As described above, multiple interleaved probe springs 34 may be used for many applications, such as for high quality electrical connections for an integrated circuit device 44, or for a probe card assembly 60 (FIG. 13), such as for probing an integrated circuit device 44 during testing.

Figure 7 is a perspective view 42 of opposing interleaved photolithographic spring pairs 34a, 34b in contact with single traces 46 on an integrated circuit device under test (DUT) 44. The interleaved spring contact pair 34a and 34b allows both springs 34a and 34b, each having a plurality of contact points 24, to contact the same trace 46. As shown in Figure 5, when a zig-zag gap 38 is formed between the two springs 34a, 34b on a substrate 16, multiple tips 24 are established on each spring 34a, 34b. Before the interleaved spring probes 34a, 34b are released from the substrate 16, the interleaved points 24 are located within an overlapping interleave region 36. When the interleaved spring probes 34a, 34b are detached from the substrate 16, the interleaved spring points 24 remain in close proximity to each other, within a contact region 40, which is defined between the springs 34a, 34b. The interleaved spring contact pair 34a and 34b may then be positioned, such that both interleaved spring probes 34a and 34b contact the same trace 46, such as for a device under test 44, providing increased reliability. As well, since each interleaved spring 34a, 34b includes multiple spring points 24, contact with a trace 46 is increased, while the potential for either overheating or current arcing across the multiple contact points 24 is minimized.

Figure 8 is a top view of parallel and opposing single-point photolithographic springs 14, before the springs 14 are released from a substrate 16. As described above for interleaved springs 34a, 34b, parallel springs 14 may also be placed such that the spring tips 24 of multiple springs contact a single trace 46 on a device 44. As well, opposing spring probes 14 may overlap each other on a substrate 16, such that upon release from the substrate 16 across a release region 18, the spring tips 24 are located in close proximity to each other. Figure 9 is a top view of parallel and opposing parallel single-point photolithographic springs 14, after the springs 14 are released from the substrate 16, wherein the parallel and opposing parallel single-point photolithographic springs 14 contact a single pad 47 on an integrated circuit device 44.

Figure 10 is a front view of a shoulder-point photolithographic spring 50, having a point 52 extending from a shoulder 54. Figure 11 is a partial cross-sectional side view of a shoulder-point photolithographic spring 50, in contact with a trace 46 on an integrated circuit device. Figure 12 is a perspective view of a multiple shoulder-point photolithographic spring 50. Single point spring probes 14 typically provide good physical contact with conductive traces 46 on an integrated circuit device 22, often by penetrating existing oxide layers on traces 46 or pads 47 by a single, sharp probe tip 24. However, for semiconductor wafers 92 or integrated circuit devices having thin or relatively soft traces 46 or pads 47, a single long probe tip 24 may penetrate beyond the depth of the trace 46, such as into the IC substrate 48, or into other circuitry.

Shoulder-point photolithographic springs 50 therefore include one or more extending points 52, as well as a shoulder 54, wherein the points 52 provide desired penetration to provide good electrical contact to traces 46, while the shoulder 54 prevents the spring 50 from penetrating too deep into a device 44 or wafer 92. Since the geometry of the probe springs 50 are highly controllable by photolithographic screening and etching processes, the detailed geometry of the shoulder-point photolithographic spring 50 is readily achieved.

Improved Probe Card Assemblies. Figure 13 is a cross-sectional view 58 of a probe card assembly 60a, wherein a plurality of electrically conductive probe tips 61 a- 61 n are located on a lower probe surface 62a of a substrate 16. A plurality of flexible, electrically conductive connections 64a-64n are located on the upper connector surface 62b of the substrate 16, and are each connected to the plurality of electrically conductive springs probe tips 61 a-61 n, by corresponding electrical connections 66a-66n.

The substrate 16 is typically a solid plate, and is preferably a material having a low thermal coefficient of expansion (TCE), such as ceramic, ceramic glass, glass, or silicon.

The electrically conductive spring probe tips 61 a-61 n establish electrical contact between the probe card assembly 60 and a semiconductor wafer 92, when the probe card assembly 60a and the semiconductor wafer 92 are positioned together.

The spring probe tips 61a-61n may have a variety of tip geometries, such as single point springs 14, interleaved springs 34, or shoulder point springs 50, and are fabricated on the substrate 16, typically using thin-film or MEMS processing methods, to achieve low manufacturing cost, well controlled uniformity, very fine pad pitches 20, and large pin counts.

The probe tips 61a-61n are electrically connected to flexible electric connections 64a- 64n, preferably through metalized vias 66a-66n within the substrate 16. Each of the plurality of flexible electric connections 64a-64n are then electrically connected to a printed wiring board probe card 68, which is then typically held in place by a metal ring or frame support structure 70. The preferred metalized via electrical connections 66a- 66n (e.g. such as produced by Micro Substrate Corporation, of Tempe, Arizona), are typically formed by first creating holes in the substrate 16, using laser or other drilling methods. The holes are then filled or plated with conductive material, such as by plating or by extrusion. After the conductive vias 66a-66n are formed, they are typically polished back, to provide a flat and smooth surface.

Figure 14 is a partial expanded cross-sectional view 79 of a probe card assembly 60a, which shows staged pitch and fan-out across a substrate 16 and a printed wiring board probe card 68. The probe tips 61a-61n are typically arranged on the probe surface 62a of the substrate, with a fine spring pitch 20. The fixed trace portions 15 are then preferably fanned out to the metalized vias 66a-66n, which are typically arranged with a substrate pitch 81. The electrically conductive connections 64a-64n, which are located on the upper connector surface 62b of the substrate 16 and are connected to the vias 66a-66n, are typically arranged with a connection pitch 83, which may be aligned with the substrate pitch 81, or may preferably be fanned out further on the upper connector surface 62b of the substrate 16.

The conductive pads 77a-77n on the underside of the printed wiring board probe card 68 are typically arranged with a pad pitch 85, such that the conductive pads 77a-77n are aligned with the electrically conductive connections 64a-64n located on the upper connector surface 62b of the substrate 16. The conductive pads 77a-77n are then preferably fanned out to conductive paths 78a-78n, which are typically arranged with a probe card pitch 87. The electrically conductive connections 72a 72n, which are located on the upper surface of the printed wiring board probe card 68 and are connected to the conductive paths 78a-78n, are typically arranged with a probe card connection pitch 89, which may be aligned with the probe card pitch 87, or may preferably be fanned out further on the upper surface of the printed wiring board probe card 68. The probe card connection pitch 89 is preferably chosen such that the electrically conductive connections 72a-72n are aligned with the test head connectors 74a-74n located on the test head 76, which are typically arranged with a test head pitch 91.

The flexible electric connections 64a-64n are typically fabricated using a longer spring length 28 than the probe tips 61 a-61 n, to provide a compliance of approximately 4-10 mils. In some embodiments, the flexible connections 64a-64n are typically built in compliance to photolithographic springs, such as described above, or as disclosed in either U.S. Patent No. 5,848, 685 or U.S. Patent No. 5,613, 861, which are incorporated herein by reference.

The flexible connections 64a-64n are connected to the printed wiring board (PWB) probe card 68, either permanently (e.g. such as by solder or conductive epoxy) or non- permanently (e.g. such as by corresponding metal pads which mate to the tips 24 of flexible connection springs 64a-64n). The printed wiring board probe card 68 then fans out the signals to pads 72a-72n, on a pad pitch 89 suitable for standard pogo pin contactors 74a-74n typically arranged with a test head pitch 91 on a test head 76.

The flexible connections 64a-64n are preferably arranged within an area array, having an array pitch 83 such as 1.00 mm or 1.27 mm, which provides a reasonable density (i.e. probe card pitch 87) for plated through-holes (PTH) 78 on the printed wiring board probe card 68, and allows signal fan-out on multiple layers within the printed wiring board probe card 68, without resorting to advanced printed wiring board probe cards 68 containing blind conductive vias 78a-78n.

The flexible conductive connections 64a-64n, which contact conductive pads 77a 77n on the underside of the printed wiring board probe card 68, maintain electrical connection between the printed wiring board probe card 68 and the substrate 16, while the substrate 16 is allowed to move up and down slightly along the Z-axis 84, as well as tilt about its center. The flexible connections 64a-64n also provide lateral compliance between a substrate 16 and a printed wiring board probe card 68 having different thermal coefficients of expansion (e.g. such as for a low TCE substrate 16 and a relatively high TCE printed wiring board probe card 68).

Alternately, the substrate 16 may be an assembly, such as a membrane probe card, which connects to the printed wiring board probe card 68 through membrane bump contacts 64a-64n. In alternate embodiments of the probe card assembly, connections 64a-64n are provided by a separable connector 132 (FIG. 18), or preferably by a MEG-Array connector 162 (FIG. 24), from FCI Electronics, of Etters, PA, wherein ball grid solder arrays located on opposing halves of the connector 132,162 are soldered to matching conductive pads on the substrate 16 and printed wiring board probe card 68, such as seen in Figure 14, and wherein the conductive pads are each arranged within an area array pattern, such that the opposing halves of the connector 132,162 provide a plurality of mating electrical connections between each of the plurality of spring probe tips 61a-61n and each of the plurality of conductive pads 77a-77n on the underside of the printed wiring board probe card 68.

As the size and design of integrated circuit devices 44 becomes increasingly small and complex, the fine pitch 20 (FIG. 2) provided by miniature spring probe tips 61a-61n becomes increasingly important. Furthermore, with the miniaturization of both integrated circuits 44 and the required probe card test assemblies, differences in planarity between an integrated circuit 44 and a substrate 16 containing a large number of spring probes 61 a-61 n becomes critical.

The probe card assembly 60a provides electrical interconnections to a substrate 16, which may contain thousands of spring probe tips 61a-61n, while providing adequate mechanical support for the probe card assembly 60a, to work effectively in a typical integrated circuit test probing environment. The probe card assembly 60a is readily used for applications requiring very high pin counts, for tight pitches, or for high frequencies. As well, the probe card assembly 60a is easily adapted to provide electrical contact for all traces 46 (FIG. 7) and input and output pads 47 (FIG. 7, FIG. 9) of an integrated circuit device, for test probe applications which require access to the central region of an integrated circuit die 44.

As shown in Figure 13, the probe card assembly 60a is typically positioned in relation to an a semiconductor wafer 92, having one or more integrated circuits 44, which are typically separated by saw streets 94. An X-axis 80 and a Y-axis 82 typically defines the location of a probe card assembly 60 across a semiconductor wafer 92 or device 44, while a Z-axis defines the vertical distance between the surface of the wafer 92 and the probe card assembly 60. Position of the wafer 92 under test, in relation to the test head 76 and the probe card assembly 60a is required to be precisely located in relation to the X-axis 80, the Y-Axis 82, and the Z-axis 84, as well as rotational Z-axis (i.e. theta) location 90 about the Z-axis 84.

However, it is increasingly important to allow probe card assemblies to provide contact with a planar semiconductor wafer 92, wherein the semiconductor wafer 92 and the probe card assembly are slightly non-planar to each other, such as by a slight variation in X-axis rotation 86 and/or Y-axis rotation 88.

In the probe card assembly 60a shown in Figure 13, the probe tips 61a-61n are flexible, which inherently provides planarity compliance between the substrate 16 and the semiconductor wafer 92. As well, the flexible connections 64a-64n, which are also preferably flexible conductive springs 14,34, 50, provide further planarity compliance between the substrate 16 and the semiconductor wafer 92. The probe card assembly 60a therefore provides planarity compliance between a substrate 16 and an integrated circuit device 44 (i.e. such as by X-axis rotation 86 and/or Y-axis rotation 88). As well, the probe card assembly 60a also accommodates differences in thermal coefficients of expansion (TCE) between the substrate 16 (which is typically comprised of ceramic, ceramic glass, glass, or silicon) and the printed wiring board probe card 68 (which is typically comprised of glass epoxy material).

The signal traces from the probe tips 61 a-61 n, typically having a small pitch 20, are preferably fanned out to the flexible connections 64a-64n, typically having a larger pitch, using routing traces on one or both surfaces 62a, 62b of the substrate 16.

The flexible connections 64a-64n are preferably laid out on a standardized layout pattern, which can match standardized power and ground pad patterns (i.e. assignments) on the printed wiring board probe card 68, thus allowing the same printed wiring board probe card 68 to be used for substrates 16 laid out to mate to different integrated circuit devices 44. As a printed wiring board probe card 68 may be adapted to specialized substrates 16, for the testing of a variety of different devices 44, the operating cost for a printed wiring board probe card 68 is reduced.

To aid in high frequency power decoupling, capacitors 172 (FIG. 24), such as LICAw series capacitors, from AVX Corporation, of Myrtle Beach SC, are preferably mounted on the top surface 62b of the substrate 16. Alternately, a parallel plate capacitor may be formed within the substrate 16, between the reference plane and a plane formed on the unused areas of the routing trace layer. For embodiments in which the substrate 16 is composed of silicon, an integral capacitor 67 (e.g. such as an integral bypass capacitor) may preferably be formed within the silicon substrate 16. As in the integrated circuits, such capacitors include metal-dielectric-metal construction, metal-dielectric-heavily doped semiconductors, or p-n junctions. For non-semiconductor substrates, metal- dielectric-metal capacitors may be fabricated on or within the substrates, using integrated circuit fabrication techniques.

A look up and look down camera is typically used to align the wafer chuck to the substrate 16, whereby the probe tips 20 are aligned to the contact pads 47 or traces 46 on a device under test 44 located on a semiconductor wafer 92. Alignment is typically achieved, either by looking at spring tips 24, or at alignment marks 125 printed on the substrate 16.

For probe card assemblies without such a camera, the substrate 16 is preferably comprised of translucent or transparent material (e.g. such as glass ceramic or glass), thereby allowing view-through-the-top alignment methods to be performed by a test operator. A window 165 (FIG. 24) is preferably defined in the printed wiring board probe card 68, while alignment marks 125 (FIG. 17), 185 (FIG. 26) are preferably located on the substrate and/or the wafer 92 under test. A test operator may then use a camera or microscope to view the alignment marks 125 through the window, and align the substrate 16 and wafer 92.

For applications where access to the surface of the semiconductor wafer 92 is required while probe contact is maintained (e.g. such as for voltage contrast electron beam probing during development of the integrated circuit device 44), a window 123 (FIG.

17) in the substrate region 16 over the IC center is preferably defined, allowing access to observe signals in the die 92. Windows 123 work best for integrated circuit devices 44 having 1/0 pads located along the die edge, enabling direct probing of integrated circuit devices 44 located on a wafer 92. Currently, the semiconductor wafer dies 92 must be diced first, wherein separate integrated circuit devices 44 are wire bonded into a package, and are then tested.

Defined openings (i.e. windows 123) within the substrate 16 are also preferably used for in-situ e-beam repair of devices such as DRAMs, in which the probe card assembly 60 may remain in place. Testing, repair and retesting may thus be performed at the same station, without moving the wafer 92.

The structure of the probe card assembly 60a provides very short electrical distances between the probe tips 61a-61n and the controlled impedance environment in the printed wiring board probe card 68, which allows the probe card assembly 60a to be used for high frequency applications. For embodiments wherein the traces on one or both surfaces 62a, 62b of the substrate 16 are required to be impedance controlled, one or more conductive reference planes may be added within the substrate 16, either on top of the traces, below the traces, or both above and below the traces. For ultra high- frequency applications, the substrate 16 may contain alternating ground reference traces, which are connected between reference planes 262 (FIG. 37, FIG. 38) at regular intervals using vias 266 (FIG. 37, FIG. 38), to effectively provide a shielded coaxial transmission line environment 260. In some embodiments, ground plane traces are placed on one side of the signal line.

High Compliance Probe Assemblies. As described above, a probe card assembly structure 60 fixedly supports a substrate 16, relative to the printed wiring board probe card 68, in the lateral X and Y directions, as well as rotationally 90 in relation to the Z axis 84.

While the flexible spring probes 61a-61n, as well as flexible connections 64a-64n, provide some planarity compliance between a probe card assembly 60 and a semiconductor wafer 92 or device 44, other preferred embodiments of the probe card assembly 60 provide enhanced planarity compliance.

Since probe springs 61a-61n are often required to be very small, to provide high density connections and a fine pitch 20, in some probe card applications which require substantial planarity compliance, the compliance provided by the probe springs 61a- 61 n alone may not be sufficient. Therefore, in some preferred embodiments of the probe card assembly 60, the probe card assembly 60 allows the substrate 16 to pivot about its center (i.e. vary in X-axis rotation 86 and/or Y-axis rotation 88), to provide increased planarity compliance to a semiconductor wafer 92 under test. In such applications, the probe card assembly 60 must still exert a controlled downward force in the Z direction 84, for engaging the probe spring contacts 61 a-61 n located on the bottom surface 62a of the substrate 16 against a semiconductor wafer 92.

For many embodiments of the probe card assembly 60, the central region 119 (FIG. 17) of the substrate 16 is used for electrical connections 64a-64n between the substrate 16 and the printed wiring board probe card 68, thus requiring that the substrate 16 be supported along the periphery 127 (FIG. 17) of the substrate 16.

A ball joint fulcrum structure may be located within the central region of a probe card assembly on the back side of the substrate support structure, to allow the substrate 16 to pivot about the center, and to provide force to engage the probe tips 61 a 61 n.

However, such a structure would typically impede wire leads or other electrical connections, which often need to exit over the central region of the probe card assembly. Moreover, such a movable joint does not typically restrict theta rotation 90 of the substrate 16 reliably.

Figure 15 is a first partial cross-sectional view 96a of a bridge and leaf spring suspended probe card assembly 60b. Figure 16 is a second partial cross-sectional view 96b of the bridge and leaf spring suspended probe card assembly 60b shown in Figure 15, which provides planarity compliance with one or more integrated circuit devices 44 on a semiconductor wafer 92, which may be non-coplanar with the probe card assembly 60b. Figure 17 is a partial expanded assembly view 124 of major components for a bridge and spring probe card suspension assembly 60b.

A leaf spring 98 connects to the substrate 16 through a bridge structure 100. The leaf spring 98 and bridge structure 100 provide pivoting freedom for the substrate 16 (i.e. slight X-axis rotation 86 and Y-axis rotation 88), with controlled movement in the Z direction 84, X direction 80, Y direction 82 and Z-Axis rotation (theta) 90 directions. In preferred embodiments, a preload assembly 121 (FIG. 15) is used as a means for accurately setting the initial plane and Z position of the substrate 16 in relation to the printed wiring board probe card 68b, and to set the pre-load force of the leaf spring 98.

For example, in the embodiment shown in Figure 15 and Figure 16, the preload assembly 121 comprises fasteners 118, which are used in conjunction with bridge shims 122. In alternate embodiments, the preload assembly 121 may comprise calibration screw assemblies 122, crash pads 120, and/or other standoffs 116.

As shown in Figure 15 and Figure 16, the outer edges of a leaf spring 99 are fixed to the printed wiring board probe card 68 along its outside edges by attachment frame 107. The center of the leaf spring 98 is connected to the bridge 100, by one or more fasteners 108, an upper bridge spacer 104, and a lower bridge spacer 106. Bridge preload shims 110 are preferably added, such as to vary the Z distance between the leaf spring 98 and the bridge 100, which varies the pre-load of the downward force exerted by the leaf spring 98 on the bridge 100. The bridge 100 translates the support from the center out to the comers, and connects to the substrate 16 by a plurality (typically three or more) bridge legs 102. The bridge legs 102 protrude through leg openings 111 defined in the printed wiring board probe card 68, and are fixedly attached to the substrate 16, such as by adhesive or mechanical connections 112.

The leaf spring 98 is typically fabricated from a sheet of stainless steel or spring steel, and is typically patterned using chemical etching methods. The downward force is a function of the stiffness of the spring, the diameter of the spring spacers 104 and 106, as well as the size of the leaf spring 98.

While the leaf spring 98 shown in Figure 16 has the shape of a cross, other geometric shapes may be used to provide downward force, tilting freedom, and X, Y, and theta translation resistance. For example, a leaf spring 98 having a cross shape may include any number of wings 99. As well, the wings 99 may have asymmetrical shapes, which vary in width as they go from the outside edge towards the center. Also, the outside edge of the leaf spring 98 may be connected into a ring, to provide further stability of the leaf spring 98.

The bridge 100 and the spacers 104 and 106 are preferably comprised of light and strong metals, such as aluminum or titanium, to minimize the moveable mass of the bridge and leaf spring suspended probe card assembly 60b.

The substrate 16 is typically attached to the legs 102 of the bridge 100, using an adhesive 112, such as an epoxy, or solder. Where substrate replaceability is needed, detachable connections 130, such as shown in Figure 18, can be used.

On the bottom side 62a of the substrate 16, lower standoffs 114 are preferably used, which prevent the substrate 16 from touching a wafer under test 92. The lower standoffs 114 are preferably made of a relatively soft material, such as polyimide, to avoid damage to the semiconductor wafer under test 92. In addition, to further avoid damage to active circuits 44 in the semiconductor wafer 92, the standoffs 114 are preferably placed, such that when the probe card assembly 60 is aligned with a device 44 on a semiconductor wafer 92, the standoffs are aligned with the saw streets 94 (FIG. 13) on the semiconductor wafer 92, where there are no active devices 44 or test structures.

Furthermore, the height of the lower standoffs 114 are preferably chosen to limit the maximum compression of the spring probes 61 a-61 n, thus preventing damage to the spring probes 61 a-61 n.

On the upper surface 62b of the substrate 16, upper standoffs 116 are also preferably used, to prevent damage to the topside flexible electrical connections 64a-64n. The upper standoffs 116 are preferably made of a moderately hard insulative material, such as LEXAN, silicone, or plastic.

In the preferred embodiment shown in Figure 15, Figure 16 and Figure 17, adjustable bridge screws 118 and bridge shims 122 are used to set the initial plane of the substrate 16, as well as to provide a downward stop to the substrate 16, so that the flexible connections 64a-64n are not damaged by over-extension.

Since printed wiring board probe cards 68b are typically made of relatively soft materials (e.g. such as glass epoxy), crash pads 120 are preferably placed on the probe card 68b, under the adjusting screws 118, to prevent the tip of the adjusting screws 118 from sinking into the printed wiring board probe card 68b over repeated contact cycles. Fastener shims 122 are also preferably used with the adjusting screws 118, such that the initial distance and planarity between the substrate 16 and the printed wiring board probe card 68b may be accurately set.

The preload shims 110 are preferably used to control the initial pre-load of the downward force exerted by the leaf spring 98 onto the bridge 100. The set preload prevents vibration of the substrate 16, and improves contact characteristics between the substrate 16 and to the semiconductor wafer under test 92.

Figure 18 is a first partial cross-sectional view 126a of an alternate bridge and spring suspended probe card assembly 60c, having an intermediate daughter card 134 detachably connected to the printed wiring board probe card substrate 68b, and wherein the spring probe substrate 16 is detachably connected to the bridge structure 100. Figure 19 is a second partial cross-sectional view 126b of the alternate bridge and spring suspended probe card assembly 60c shown in Figure 18, which provides planarity compliance with one or more integrated circuit devices 44 on a semiconductor wafer 92, which is originally non-coplanar with the probe card assembly 60c.

In the alternate bridge and spring suspended probe card assembly 60c shown in Figure 18, electrical connections are provided between the probe chip substrate 16 and the daughter card 134, such as between flexible connections 64a-64n which provide electrical connection to the daughter card 134.

As seen in Figure 18, a separable connector 132, having detachable halves 133a, 133b, provides a detachable connection between the intermediate daughter card 134 and the printed wiring board probe card substrate 68b, which allows replacement of the substrate 16 and the daughter card 134. Substrate attachment fasteners 130, e.g. such as but not limited to screws, preferably extend through bridge legs 128, and allow the bridge 100 to be removably connected to substrate posts 128, which are mounted on the upper surface 62b of the substrate 16.

In one embodiment of the probe card assembly 60, the preferred separable connector 132 is a MEG-ArrayT" connector, manufactured by FCI Electronics, of Etters, PA. One side of the separable connector 132 is typically soldered to the printed wiring board probe card 68, while the mating side is typically soldered to the daughter card 134, whereby the daughter card 134 may be removably connected from the printed wiring board probe card 68b, while providing a large number of reliable electrical connections.

The daughter card 134 preferably provides further fanout of the electrical connections, from a typical pitch of about 1 mm for the flexible connections 64a-64n, to a common pitch of about 1.27 mm for a separable connector 132.

Figure 20 is a cross-sectional view 136 of a wire and spring post suspended probe card assembly 60d. A plurality of steel wires 138 (e.g. typically three or more) allow Z movement 84 of the substrate 16. The spring post frame 140, which is typically soldered or epoxied 139 to the printed wiring board probe card 68c, typically includes one or more spring posts 141, which are preferably used to provide downward Z force, as well as to limit travel. Figure 21 is a cross-sectional view 142 of a suspended probe card assembly 60e having an intermediate daughter card 134 detachably connected to the printed wiring board probe card 68 by a separable, i.e. demountable, connector 132, comprising separable connector halves 133a, 133b, and by supports 149. The flexible connections 64a 64n are preferably made with springs 14,34, 50, and provide both electrical connections to the printed wiring board probe card 68, as well as a mechanical connection between the printed wiring board probe card 68 and the daughter card 134.

In the probe card assembly 60e, the flexible connections 64a-64n are permanently connected to conductive pads 143a-143n on the daughter card 134, using either solder or conductive epoxy 127. The flexible connections 64a-64n are preferably designed to provide a total force larger than that required to compress all the bottom side probe springs 61 a-61 n fully, such as when compressed in a range of 2 to 10 mils. As well, the flexible connections 64a-64n are preferably arranged, such that the substrate 16 does not translate in the X direction 80, Y direction 82, or Z-Axis rotation (theta) 90 directions (FIG. 13) as the flexible connections 64a-64n are compressed.

Upper substrate standoffs 116 are preferably used, to limit the maximum Z travel of the substrate 16, relative to the daughter card 134, thereby providing protection for the flexible connections 64a-64n. The upper standoffs 116 are also preferably adjustable, such that there is a slight pre-load on the flexible connections 64a-64n, forcing the substrate 16 away from the daughter card 134, thereby reducing vibrations and chatter of the substrate 16 during operation. A damping material 145 (e.g. such as a gel) may also preferably be placed at one or more locations between the substrate 16 and the daughter card 134, to prevent vibration, oscillation or chatter of the substrate 16.

The separable connector 132 (e.g. such as an FCI connector 132) preferably has forgiving mating coplanarity requirements, thereby providing fine planarity compliance between the daughter card 134 and the printed wiring board probe card 68. Supports 149, e.g. such as but not limited to fasteners 166, spacers 164, nuts 168, shims 170 (FIG. 24), and/or an all glued connection, are also typically used between the daughter card 134 and the printed wiring board probe card 68. In some embodiments of the suspended probe card assembly 60e, the supports 149 are adjustable. This also facilitates planarity adjustment of probe chips.

Figure 22 is a cross-sectional view 146 of a probe card assembly 60f, in which the probe spring substrate 16 is attached to a printed wiring board probe card 68 through a separable array connector 147. The probe card assembly 60f is suitable for small substrates 16, wherein a small non-planarity between the substrate 16 and a semiconductor wafer under test 92 can be absorbed by the spring probes 61 a 61 n alone.

Figure 23 is a cross-sectional view 148 of a pogo wire suspended probe card assembly 60g, wherein a nano-spring substrate 16 is connected to a printed wiring board probe card substrate 68 by a large grid array (LGA) interposer connector 150, which is retainably fixed between the substrate 16 and the printed wiring board substrate 68. In one embodiment, the LGA interposer connector 150 is an AMPIFLEX connector, manufactured by AMP, Inc., of Harrisburg PA. In another embodiment, the interposer connector 150 is a GOREMATEw connector, manufactured by W. L. Gore and Associates, Inc., of Eau Clare, Wl. In another alternate embodiment, a pogo pin interposer 150 is used to connect overlying pogo pins 152 on the printed wiring board probe card 68 to electrical connections 66a-66n on the substrate 16. The substrate 16 is held by a plurality of steel pogo suspension wires 154, which are preferably biased to provide a slight upward force, thereby retaining the interposer connector 150, while preventing vibration and chatter of the assembly 60g.

Small Test Area Probe Assemblies. Figure 24 is a cross-sectional view of a small test area probe card assembly 60h, having one or more area array connectors 162 located between the main printed wiring board probe card 68 and a daughter card 134, which is attached to a small area spring probe substrate 16.

While many of the probe card assemblies 60 described above provide large planarity compliance for a probe spring substrate 16, some probe card assemblies are used for applications in which the device under test comprises a relatively small surface area. For example, for wafers 92 which include a small number of integrated circuits 44 (e.g. such as two ICs), the size of a mating substrate 16 can also be relatively small (e.g. such as less than 2 cm square).

In such embodiments, therefore, the planarity of the substrate 16 to the wafer under test 92 may become less critical than for large surface areas, and the compliance provided by the probe springs 61 a-61 n alone is often sufficient to compensate for the testing environment. While the compliance provided by the probe springs 61 a 61 n may be relatively small, as compared to conventional needle springs, such applications are well suited for a probe card assembly 60 having photolithographically formed or ME MS formed spring probes 61 a-61 n.

The probe card assembly 60h is therefore inherently less complex, and typically more affordable, than multi-layer probe card assembly designs. The small size of the substrate 16 reduces the cost of the probe card assembly 60h, since the cost of a substrate 16 is strongly related to the surface area of the substrate 16.

The probe springs 61 a-61 n are fabricated on the lower surface 62a of a hard substrate 16, using either thin-film or MEMS processing methods, as described above. Signals from the probe springs 61a-61n are fanned out to an array of metal pads 182,184, 186 (FIG. 26), located on the upper surface 62b of the substrate 16, using metal traces on one or both surfaces 62a, 62b, and conductive vias 66a-66n through the substrate 16.

The top side pads are connected to a daughter card 134, using common micro-ball grid solder array pads, typically at an array pitch such as 0.5 mm. The daughter card 134 further expands the pitch of the array, to pads having an approximate pitch of 0.050 inch on the opposing surface of the daughter card 134. An area array connector 162, such as a MEG-Array connector, from FCI Electronics Inc. of Etters PA, is used to connect the 0.050 inch pitch pad array to the printed wiring board probe card 68. Power bypass capacitors 172, such as LICA capacitors from AVX Corporation of Myrtle Beach SC, are preferably added to the daughter card 134, close to the substrate micro-BGA pads 182,184, 186, to provide low impedance power filtering.

The small test area probe card assembly 60h preferably includes a means for providing a mechanical connection between the printed wiring board probe card substrate 68 and the daughter card 134. In the probe card assembly 60h embodiment shown in Figure 24, one or more spacers 164 and spacing shims 170 provide a controlled separation distance and alignment between the daughter card 134 and the printed wiring board probe card substrate 68, while one or more fasteners 166 and nuts provide a means for mechanical attachment. While a combination of spacers 164, shims 170, fasteners 166, and nuts 168 are shown in Figure 24, alternate embodiments of the small test area probe card assembly 60h may use any combination of means for attachment between the daughter card 134 and the printed wiring board probe card substrate 68, such as but not limited to spring loaded fasteners, adhesive standoffs, or other combinations of attachment hardware.

Lower substrate standoffs 114, which are typically taller than other features on the substrate 16 (except for the spring tips 61a-61n), are preferably placed on the lower surface 62a of the substrate 16, preferably to coincide with the saw streets 94 on a semiconductor wafer 92 under test, thereby preventing the wafer under test 92 from crashing into the substrate 16, and preventing damage to active regions on the semiconductor wafer 92.

As shown in Figure 24, the substrate 16 preferably includes an access window 123 (FIG. 17), while the daughter card 134 also preferably includes a daughter card access hole 163, and the printed wiring board probe card 68 preferably includes and a probe card access hole 165, such that access to a semiconductor wafer 92 is provided while the probe card assembly 60h is positioned over the wafer 92 (e.g. such as for visual alignment or for electron beam probing). Access holes 123, 163, 165 may preferably be used in any of the probe card assemblies 60.

Figure 25 is a top view of a substrate wafer 174, upon which a plurality of micro ball grid array spring probe contactor chip substrates 16 are laid out. For spring probe substrates 16 having a small surface area 175, several spring probe contactor chip substrates 16 may typically be fabricated from a single wafer 174. For example, as shown in Figure 25, as many as twenty four sites having a width 176 and a length 178 (e.g. 14 mm square), may be established on a standard four inch round starting wafer 174. As well, different substrates (e.g. 16a, 16b) may be fabricated across a starting wafer 174, whereby the cost of production (which may be significant) for different spring probe substrates 16 may be shared, such as for masking costs and processing costs.

Therefore, the cost of development for different substrates 16a, 16b may be lowered significantly (e.g. such as by a factor of up to 10 or more).

Figure 26 is a top view of a single 0.5 millimeter pitch micro ball grid array 180 for a 14 mm square spring probe contactor chip (NSCC) 16b. The micro BGA pads 182,184, 186 are preferably on a standard pitch (e.g. 0.5 mm). The outer five rows of pads 182 and the center pads 184 provide 341 signal connections, and the inside two rows 186 provide ninety six dedicated power and ground connections. By customizing the routing traces to the spring probes 61 a-61 n, specific power/ground spring positions to match the integrated circuit 44 under test can be accommodated with a single layer of routing.

Standoffs 114, are preferably placed in locations matching inactive regions on the wafer 92, such as on the scribe lane 94, to prevent damage to active devices 44 on the device under test 44. One or more alignment marks 185 are also preferably located on the substrate wafer 174. The production cost and turnaround time for a probe card assembly 60 can be significantly improved, by standardizing the footprints of the micro BGA pad array 180, the daughter card 134, and the printed wiring board probe card 68.

Standardization of the micro-BGA pad array 180, as well power/ground pad assignments for the pads located on the substrates 16,134, 68, allows a standardized pattern of vias 66a-66n in the base substrate 174.

Standardization of other componentry for probe card assemblies 60 often allows printed wiring board probe cards 68 (and in some embodiments daughter cards 134), to be used for different substrates 16 and integrated circuit devices 44, wherein only the routing of the substrate 16 is customized.

The use of a starting substrate 174 (FIG. 25) having a standardized pattern of vias 66a- 66n also allows starting substrates 174 to be ordered, stored and used in quantity, thus reducing the cost of starting substrates 174, and often reducing the lead-time to obtain the starting substrates 174.

Alternate Applications for Probe Springs. Photolithographic or MEMS spring probes 61,14, 34,50 may alternately be used for bare die bum-in sockets, such as for DieMateT" bum-in sockets, manufactured by Texas Instruments Inc., of Mansfield MA, or for Die TM Pak bum-in sockets, available through Aehr Test, Inc. of Fremont CA. For bare die bum-in sockets which contact the substrate 16 around the edges, the probe springs 61 springs and fanout metalization are needed only on one surface (e.g. probe surface 62a) of the substrate 16. The required fanout is used to determine the size of the substrate 16, based on the number of the 1/0 signals needed to be routed to pads on the edge of the substrate 16. Alternately, vias 66 in the substrate 16, as described above, can be used to route the I/0 signals to an array of pads on the opposite surface 62b of the substrate 16, allowing the substrate to be smaller, and thereby reducing the cost of fabrication.

Tiled Probe Assemblies. Figure 27 is a plan view 190 of a typical tiling probe strip 192, having a probe strip length 198 and a probe strip width 200. The tiling probe strip 192 has a plurality of probe strip contact areas 194a-194n, each having a plurality of spring probes 61a-61n. As well, in the embodiment shown, the spring probes 61 a- 61 n are laid out in longitudinally aligned probe regions 196a, 196b. Use of one or more tiling probe strips 192 in a probe card assembly allows simultaneous electrical contact with a plurality of integrated circuit devices 44, such as for testing adjoining integrated circuit device sites 44 on a semiconductor wafer 92. The plurality of probe strip contact areas 194a-194n are preferably located symmetrically along the length of the tiling probe strip 192, such that they align with a symmetrical plurality of integrated circuit devices 44 on a wafer 92. It is to be appreciated that use of tiling probe strips, comprising the photolithographically batch fabricated stress metal spring probes on the tiling probe strip substrates, are applicable for the testing of various IC-s, such as memory, logic, and microprocessors.

As well, the tiling probe strips 192, having spring probes 61a-61n, typically include electrical vias 66a-66n and an array of electrical connections 64a-64n, such as seen in the probe chip substrate 16 of Figure 13, Figure 14, and/or Figure 21. While the spring probes 61a-61n may typically be laid out to match specific devices 44 under test, the tiling probe strips 192 include standard electrical vias 66a-66n and/or arrays of electrical connections 64a-64n. For example, in the probe card assembly 202 shown in Figure 28 and Figure 29, each of the tiling probe strips 192 includes a standard ball grid array 160 of solder connections. Therefore, while preferred embodiments of tiling probe strips 192 may include spring probes 61a-61n which are laid out to match specific devices 44 under test, the tiling probe strips 192 may be attached to standardized daughter cards 204 and/or standardized intermediate connectors (e.g. such as a separable connector 132), thus minimizing engineering development costs to produce a tiled probe assembly 202.

Figure 28 is a partial bottom view of tiled probe head 202 comprising a plurality of tiling probe strips 192 attached to a support substrate 204, which includes an array 207 (FIG.

29) of electrically conductive vias 205. Figure 29 is a side view of a plurality of tiling probe strips 192 attached to a probe card 16 or daughter card 204, which are used to contact a plurality of integrated circuit devices 44 located on a semiconductor wafer 92.

The tiled probe head 202 is typically used to contact a plurality of integrated circuit devices 44 located on a semiconductor wafer 92. The tiling probe strips 192 are preferably located symmetrically across the substrate 204, such that they align with a symmetrical plurality of integrated circuit devices 44 on a wafer 92.

The support substrate 204 preferably has a low thermal coefficient of expansion (TCE), and is preferably matched to silicon. As well, the substrate 204 typically fans out a large number of signal traces 46 (FIG. 7), to connectors on the opposite surface 209b of the substrate 204. In one embodiment, the substrate 204 is a silicon wafer, which includes vias 205, e.g. such as arranged on a 0.056 inch pitch, and thin film routing 46 on one or both substrate surfaces 209a, 209b.

In the tiled probe head 202 shown in Figure 28 and Figure 29, the tiling probe strips 192 include groups of probe springs 61 which are used to contact rows of pads 47 (FIG. 7) on integrated circuit devices 44 having pads 47 located on opposing sides of a device under test 44 (e.g. such as on the right and left sides of an integrated circuit device site 44). In the tiled probe head 202 shown, the tiling probe strips 192 are arranged such that one of the tiling probe strips 192 typically contacts the right side of one circuit device site 44 (e.g. such as using probe contact region 196a in Figure 27), in addition to contacting the left side of a neighboring circuit device site 44 (e.g. such as using probe contact region 196b in Figure 27). The embodiment shown in Figure 28 therefore provides simultaneous contact between the plurality of tiling probe strips 192 and a plurality of integrated circuit devices 44, while allowing adequate tolerances between adjoining tiling probe strips 192, wherein the side edges of the tiling probe strips 192 may preferably be placed over the saw streets of the integrated circuit device sites 44. For example, saw streets 94 (Fig. 13) between adjoining devices 44 on a wafer 92 may commonly be on the order of 4 to 8 mils wide, thereby providing a similar gap between tiled probe strips 192 in the tiled probe card assembly 202.

In alternate embodiments of the tiled probe head assembly 202, all pads 47 (Fig. 9) for an integrated circuit device site 44 may be contacted by probes from a single probe strip 192.

Burn-In Structures. Figure 30 is a partial cross-sectional view of a bum-in structure 210 which allows a plurality of integrated circuit devices 44 to be temporarily connected to a bum-in board 212. An array of probe spring (i.e. nano-spring) contactor chips (NSCC) 214 are mounted onto a bum-in board 212, such as by micro ball grid arrays 216, which provide electrical connections between the plurality of integrated circuit devices 44 and external bum-in circuitry (not shown). Board vacuum ports 218 are preferably defined in the bum-in board 212, while contactor chip vacuum ports 220 are preferably defined in the NSCC substrate 214, wherein the board vacuum ports 218 are generally aligned to the contactor chip vacuum ports 220 (e.g. such that an applied vacuum through the board vacuum ports 218 is also applied to the generally aligned contactor chip vacuum ports 220). An air seal 222 (e.g. such as an epoxy), is preferably dispensed around the periphery of each nano-spring contactor chip 214, to prevent the loss of applied vacuum through the micro BGA ball array 216.

As integrated circuit devices 44 are initially placed on nano-spring contactor chips 214 (e.g. such as by a"pick and place"machine), an applied vacuum to the board vacuum ports 218 on the bum-in board 212 and generally aligned contactor chip vacuum ports 220 on the nano-spring contactor chips 214 prevents the placed integrated circuit devices 44 from shifting from their placed positions.

When all of the integrated circuit devices 44 are placed onto the corresponding contactor chips 214, a clamp plate 224 is preferably placed in contact with the integrated circuit devices 44, to retain the integrated circuit devices 44 in place during bum-in operation, and to allow the probe springs 61 a-61 n to provide electrical connections to the integrated circuit devices 44. Individual spring pads 226 may also be used, to push on the integrated circuit devices 44 under test, to allow for planarity tolerances of the clamp plate 224 and the bum-in board 212. The bum-in structure 210 preferably includes means 217 for retaining the clamp plate 224, such that once the clamp plate 224 is placed in contact with the integrated circuit devices 44, the clamp plate 224 is attached to the burn-in board 212, and the applied vacuum may be switched off.

Protective Coating Processes for Improved Spring Probes. As described above, since spring probes 61 provide advantages of high pitch, high pin count, and flexibility, they may be used for a wide variety of applications. However, when these typically small spring probes 61 are used to contact traces 46 on integrated circuit devices 44, such as on semiconductive wafers 92, wherein the traces 46 often contain an oxide layer, the spring probes 61 are often required to break through oxide layers and establish adequate electrical contact with metal traces or conductive pads. As the spring probes 61 are often used many times, the small, unprotected spring probe tips 24 may become worn. Therefore, it would be advantageous to provide an electrically conductive wear coating on the contact tips 24 of the probe springs 61. However, such a protective coating is required to cover both the entire surface of the spring tip 24.

As described above, the probe springs 61 may be formed by a plasma chemical vapor deposition and photolithographic processes, such as disclosed in U.S. Patent No. 5,848, 685 and U.S. Patent No. 5,613, 861, wherein successive layers of conductive material are applied to a substrate, and wherein non-planar springs are subsequently formed. In such processes, however, a protective coating applied during the deposition process would not inherently provide a continuous coating on all surfaces of the formed non-planar probe springs. In alternate embodiments, the probe springs 61 may be formed by plating processes.

The probe springs 61, after their release, are not planar to the substrate surface.

Therefore, a protective coating may be applied after the springs 61 have been released from the release layer 18. Figure 31 is a view of a first step 230 of a spring probe assembly coating process, in which a protective coating 232 is applied to a probe surface of a spring probe assembly substrate 16, having one or more non planar probe springs 61. The spring probe assembly coating process forms a protective layer on the non-planar probe springs 61. While the coating process may be used for a wide variety of non-planar structures, it is specifically useful for the processing of thin film and MEMS probe spring contacts 61. In Figure 31, the applied electrically conductive protective coating is preferably a hard electrically conductive material, such as titanium nitride, rhodium, tungsten, or nickel. The applied electrically conductive protective coating is also preferably an inert material, thereby providing lubricative characteristics (i.e. a low coefficient of friction) for the probe tips 24 on the spring probes 61, thus minimizing wear to both devices under test and to the spring probes 61.

When the protective coating 233 is applied 232 to the substrate 16 and probes 61, the protective coating 233 covers both the planar and non-planar regions on the exposed surface 62 of the substrate 16. While the spring probes 16 are covered with the protective coating 233 during the coating step 230, all the traces on the substrate structure are electrically shorted together, from the applied conductive coating 233. The conductive coating 233 is therefore required to be patterned, or partially removed, to restore electrical isolation between different probe springs 61 and their respective traces. While conventional photo-masking processes are typically used in the majority of integrated circuit processing, to selectively etch away conductive coatings, such as titanium nitride coatings, such photo-masking processes are used for planar structures.

Figure 32 is a view of a second step 234 of a spring probe assembly coating process, in which a layer of photoresistive material 240 (e.g. approximately 10 microns deep) is applied to a second substrate 236, which preferably has dipping standoffs 238 (e.g. approximately 30 microns high). The photoresistive material 240 is used to protect the applied protective layer 233 on non-planar portions of the probe springs. Figure 33 is a view of a third step of a spring probe assembly coating process, in which a coated spring probe assembly is partially and controllably dipped 242 into photoresistive material 240 on the second substrate 236. The depth of applied photoresistive material 240 eventually controls the remaining protective coating 233. The substrate 16 is lowered to a desired depth in the photoresistive material 240, which is typically controlled the applied depth of the photoresistive material 240 on the second substrate 236, and the height of the dipping standoffs 20. The applied depth may altemately controlled by an operator, such as by controlled axial movement of a processing apparatus, to control the movement of the substrate 16 into the photoresistive material 240.

Figure 34 is a view of a fourth step of a spring probe assembly coating process, in which a coated and partially dipped spring probe assembly is removed 246 from the photoresistive material 240 on second substrate 16 and soft baked, leaving a portion of the protectively 233 coated probe springs 61 covered in a baked photo resist layer 248. Figure 35 is a view of a fifth step of a spring probe assembly coating process, in which the coated and dipped spring probe assembly 16,61 is etched 250, thereby removing the protective coating 233 from portions of the substrate 16 (i.e. the field area of the substrate 16) and probe springs 61 not dipped covered in a baked photo-resist layer 248. Figure 36 is a view of a sixth step of a spring probe assembly coating process, in which photo-resist layers 248 are stripped from the portions of the probe springs 61 which were covered in a photo-resist layer 248, thereby exposing the protective coating 233.

The non-planar probe spring coating process therefore provides a protective coating to the tips 24 of the probe springs, while etching the unwanted protective coating in the substrate surface 16 and portions of the spring probes 61 which are not coated with photo-resist layers 248.

Spring Probe Substrates for Ultra High Frequency Applications. As described above, the structure of the probe card assemblies 60 provides very short electrical distances between the probe tips 61 a-61 n and the controlled impedance environment in the printed wiring board probe card 68, which allows the probe card assemblies 60 to be used for high frequency applications. As well, the spring probe substrate 16 may preferably be modified for ultra high frequency applications.

Figure 37A shows a partial cross-sectional view 260a of an ultra high frequency spring probe substrate 16a, having a probe spring 61 comprising a plurality of layers 17a-17n having different inherent levels of stress. The spring 61 is typically formed over a release layer 19, wherein selective removal of the release layer 19, such as within a release region 18 (FIG. 3, FIG. 4), allows a free, nonplanar portion 61 to extend from the surface of the substrate 16a. For embodiments wherein the traces on one or both surfaces 62a, 62b of the substrate 16 are required to be impedance controlled, one or more conductive reference planes 262a, 262b may be added within or on the substrate 16, either on top of the traces 270, below the traces 270, or both above and below the traces 270. The substrate 16 may also contain alternating ground reference traces 266a, 266b, which are connected to the one or two reference planes 262a, 262b, to effectively provide a shielded coaxial transmission line environment 268. While the spring probe substrate 16 is typically a ceramic material, the layer 264 between reference planes is typically a dielectric material, such as an organic dielectric material.

Figure 37B shows a partial cross-sectional view 260b of an alternate ultra high frequency spring probe substrate 16, further comprising a ground reference trace 266d extending from ground reference trace 266b, which is connected to the one or two reference planes 262a, 262b. The ground reference trace 266d surrounds the central conductive via region 78, and is separated by a dielectric region 259. The surrounding ground reference trace 266d effectively extends a shielded coaxial transmission line environment 268 through the substrate 16, from the connector surface 62b to the probe surface 62a. The conducting layer 266d and dielectric film 259 are sequentially deposited preferably by chemical vapor deposition (CVD) techniques, including atomic layer chemical vapor deposition technique. The electrically conducting layer in the central region of the via is then deposited by CVD or physical vapour deposition or electro- deposition (either electroplating or electro-less) method or combination thereof. In embodiments where electroplating is used for depositing the electrically conducting layer, a seed layer may be deposited on the surface of the dielectric film 259 by any of the conventional film deposition techniques, such as CVD.

Figure 38 shows a partial cross-sectional view 261 of an alternate ultra high frequency spring probe substrate 16. For embodiments wherein a spring probe 61 and related electrical conductors 273 [320], 78,322 on and through the substrate 16 are required tohave controlled impedance (matched impedance), e.g. for high frequency applications, one or more conductive reference surfaces 262a, 262b, 262c, 262d and vias 266a, 266b, 266c may preferably be added, either within or on the substrate 16.

As well, the impedance control surfaces 262a, 262b, 262c, 262d are not limited to the planar surfaces shown in Figure 38. In some embodiments, where electrical conduction through the via 78 requires impedance matching for enhanced performance, the via is constructed as in Fig. 37B, the details of which has been explained above.

A conductive layer 262d may be deposited on top of the insulating layer 267, to provide a coaxial, controlled impedance connection. Alternate layers of conductive materials 262 and dielectric materials 263 can preferably be integrated with the substrate 16, such as for embodiments which require decoupling capacitors in close proximity to a probe spring 61. For a substrate 16 which is a conductive or semiconductive material, such as silicon, an oxide or dielectric layer 269 may preferably be deposited or formed between the substrate 16 and a conductive reference plane 262c, thereby forming a capacitance structure 271, which may be used as a decoupling capacitor. Similarly, in some embodiments, the capacitor structure can also be formed on the opposite surface, or both surfaces of the substratel 6. In this case, a dielectric layer, e.g. oxide, may be deposited or formed on the surface prior to the deposition of the ground plane 262b..

Electrical connection to the conducting substrate 16 is provided through regions (not shown) on the surface of the substrate 16 which are not covered by the dielectric 269 or ground plane films. Photolithographic techniques can be used to open up such regions for establishing contacts to the conducting substrate. In such cases where the substrate 16 is a conducting material, electrically insulative films, e.g. oxide, may be deposited or formed on the walls of the vias or holes to prevent electrical shorting. As well, one or more assembled components 265, such as passive components 265 (e.g. typically capacitors, resistors, and/or inductors), or active component devices 265, may be incorporated on either surface 62a, 62b of the substrate 16.

The fixed portions 15 of the spring probes 61 typically extend a relatively short distance across the substrate 16. Traces 60 located on the surface of the substrate 16 are electrically connected to the fixed portions 15 of the spring probes 61, and electrically connect the probe springs 61 to the vias 78. The traces may be comprised of a different material than the spring probes 61, and are preferably comprised of metals having high electrical conductivity (e.g. such as copper or gold).

Advanced Multi-Die Probe Card System Architectures. Figure 39 is a schematic view of a wafer test system 280, in which a probe card assembly 60, e.g. 60a (FIG.

14), or an advanced probe card assembly 300 (FIG. 40), provides an interface between a test head 76 and a multi-die wafer 92 located on a chuck 282. The probe card assembly 60,300 shown in Figure 39 is attached to the wafer test system 280, through a support ring 284, to a prober head plate 286. The probe card assembly 60,300 interfaces with the test head 76 through a probe interface board 288. The wafer test system 280 also comprises a prober drive mechanism 292, by which the chuck 282 and the test head 76 are moveable in relation to each other.

Multi-Die probing requires a high number of parallel contacts, e.g. such as electrical contacts, between the test head 76 (FIG. 39) and each of the die 44 (FIGS. 13,40).

The high number of parallel contacts 307, such as seen in Figure 40, typically require uniformity and planarity of the contact tips 61,412 (FIG. 45) to a reference surface, e.g. such as to an IC 44. Furthermore, signal traces 307 are required to fan-out, for signal integrity and for at-speed testing. The high number of parallel contacts 307 also require a total probe force to effectively contact each of the dies 44. In addition, it is preferable that the length of the traces 307 be matched from site 44 to site 44, such as to provide an equal delay for all test channels during dynamic testing.

Probe card assemblies 60,300 such as a probe card assembly 60a (FIG. 14), or a decal interposer assembly 300a (FIG. 41), which contact multiple die 44 preferably match the thermal coefficient of expansion (TCE) of the probe carrier 16 to the wafer 92, e.g. silicon, while maintaining dependable connections and electrical integrity to the motherboard PWB 304.

Probe card assemblies 60,300 which contact multiple die 44 preferably comprise independent power supplies for each die 44, and preferably provide a plurality of power rails, e.g. 2-3 or more, for connection to each die 44. As well, the probe card assemblies 60 preferably include multiple bypass capacitors 172, as close as possible to each device under test DUT 44.

The probe card assembly 60 provides transfer of signal and power connections from the devices under test 44 and the test head 76, along the Z-axis 84, while preferably providing controlled impedance. The probe card assembly also transfers signal and power connections in the X-Y directions 80,82 (Fig. 13), which fans out the signal traces 307 from the pitch 20 of the integrated circuits 44 to the pitch 91 (FIG. 14) of the test head 76. The total probe offset from the mother board 304 to the probe tip 61,412 of the probe assembly 60 is typically defined by the thickness of the support ring 284 (FIG. 39).

As described above, some embodiments of the probe card assemblies 60, such as the bridge and leaf spring suspended probe card assembly 60b (FIG. 15), the wire and spring post suspended probe card assembly 60d (FIG. 20), and the pogo wire suspended probe card assembly 60g (FIG. 23), provide means for tilting the probe chip substrate 16 to provide compliance and planarity to a wafer 92.

Figure 40 is a basic schematic diagram of a staged interface probe card assembly 300, which integrally provides vertical translation of electrical trace paths 307, such as through a Z-Block printed wiring board (PWB) 342 (FIG. 42). The staged interface probe card assembly 300 typically comprises probe springs 412 (FIG. 46, FIG. 47), e.g. such as having a 180um height, located on the probe surface 62a (FIG. 43) of the probe chip substrate 16,310, which provide enhanced compliance for the assembly 300. Electrical connections 307 are provided between the motherboard PWB 304 and the probe chip substrate 16,310, through an upper interface arrangement 308, an intermediate connector assembly 306, and a lower interface arrangement 312, respectively. A stiffener plate 302 is also preferably affixed to the motherboard PWB 304. The upper interface arrangement 308 and the lower interface arrangement 312 shown in Figure 40 comprise electrical interfaces and/or hardware.

In some embodiments of the staged interface probe card assembly 300, enhanced compliance is provided entirely by probe springs 412, wherein the probe chip substrate 16,310 is held stationary with respect to the motherboard PWB 304. Lower standoffs 114 are typically provided on the lower surface of the probe chip substrate 310, which limit the minimum vertical distance between the probe chip substrate and a wafer 92. The height of the lower standoffs 114 is typically less than the resting, ie. non-contacting height of the probe springs 61,412, such that the probe springs 61,412 are allowed to flex and provide a compliant connection to one or more devices 44 on a wafer 92. As seen in Figure 40, upper standoffs 116 (Fig. 19) are also used in some embodiments of the probe card assembly 300, such as to provide a vertical travel limit for the connector surface 311 a of the probe chip substrate 310, such as in relation to the intermediate connector 306 or to the motherboard 304.

As seen in Figure 39, the metal support ring, i.e. ring insert 284, provides mechanical support for the motherboard PWB 304 against the downward pressure exerted by the pogo tower connector 290 (FIG. 39).

The intermediate connector 306 shown in Figure 40 may typically comprise a vertical translation block 342 (FIG. 43), a pin block 742 (FIGS. 64-69) or an expanded pin block (FIG. 70, FIG. 71). The intermediate connector 306 provides Z-translation for the assembly 300, which allows the assembly 300 to provide a probe offset, which is typically about 0.300", which is a requirement in most probe assemblies, to clear the metal support ring insert 284.

The upper interface 308 may comprise a variety of interfaces, such as an upper interposer 344 (FIG. 43) having two sided springs 521 (FIG. 52), a solder ball array 756 or Z-Laminate (FIG. 65), a pin grid array 745 (FIG. 64), or any other interposer assembly 344, such as an ISOCONz Connector, available through Circuit Components, Inc., of Tempe AZ.

The lower interface 312 may also comprise a variety of interfaces, such as an interposer 348 (FIG. 42, FIG. 43) having springs located on both lower and upper surfaces 349a, 349b. The lower interface 312 may alternately comprise a solder ball array 756, springs 64a-64n (FIG. 13, FIG. 14) located on the upper surface 311b of the probe chip substrate 310, or other direct connections to the probe chip substrate 16,310.

The lower interface 312 can be a fixed or permanent connection in some embodiments, such as a solderball interface, if the vertical translation block VTB 342 (Fig. 43) is made to provide lateral compliance between the lower surface 343 of the vertical translation block 342 and the rest of the system. One example of lateral compliance is shown in Fig. 67 using metal pin block 742 where the compliant pin template 748 and 752 allows the pin to move slightly.

Alternate embodiments of the vertical translation block VTB 342 comprise a plurality of non-fixed connections, such as an interposer or springs 64a-64n (FIG. 67), 412 which are fabricated onto the upper surface 62b of the probe chip 310. The use of non-fixed connections allow the probe chip 310 to move and/or expand in relation to the vertical translation block VTB 342.

The staged interface probe card assembly 300 shown in Figure 40 further comprises a top stiffener plate 302, which helps to keep the motherboard PWB 302 flat.

Figure 41 is a partial cross sectional view of a decal interposer assembly 300a. A motherboard PWB 304 is fixedly attached to a top stiffener plate 302 such as by a plurality of fasteners 322. The stiffener plate 302 is preferably comprised of a rigid material, such as stainless steel. The stiffener plate 302 shown in Figure 41 further comprises a plurality of component recesses 325, such that componentry, e.g. capacitors 172, can be mounted to or extend from the upper surface 305b (FIG. 43) of the motherboard PWB 304.

Some embodiments of the stiffener plate incorporate handles 837,839 (FIG. 73, FIG.

74, FIG. 75), such as to aid handling of the assembly. Furthermore, windows 840 (FIG.

73) are defined through some embodiments of the stiffener plate 302, which allow post- assembly modification of the motherboard 304, e.g. such as for electrical jumpers for customer customization.

As seen in Figure 41, the motherboard 304 is electrically connected to a probe chip wafer 16, through an intermediate interposer 150, which is located between the upper surface 62b (FIG. 43) of the probe chip 16,310 and the lower surface 305a (FIG. 43) of the motherboard 304. The interposer 150 is preferably aligned to the motherboard 304, typically through interposer alignment pins 330, which are mounted to and extend from the motherboard 304, and correspond to alignment holes 347 (FIG. 43) and/or the edge 351 (FIG. 43) of the interposer substrate 348.

Figure 79 is a perspective view of a probe chip 310 mounted to a compliant membrane 326a. Figure 80 is a perspective view of a probe chip 310 mounted to a compliant decal 326b. Figure 81 is a perspective view of a probe chip 310 mounted to a compliant sheet 326c. Figure 82 is a perspective view of a probe chip 310 mounted to a compliant screen 326d. The probe chip 310 is held in place by a compliant member 326, which typically comprises a flexible membrane 326a (FIG. 79), decal 326b (FIG. 80), sheet 326c (FIG. 81), or mesh structure 326d (FIG. 82). As seen in Figure 41, the compliant member 326 is fixedly attached or mounted to an outer surrounding ring 328, which is attached to the motherboard 304, by fasteners 334. The outer ring 328 and the affixed compliant member are preferably aligned to the assembly 300a, such as by ring alignment pins 332. The compliant member is stretched during attachment to the surrounding ring 328 and thus held under tension.

The compliant member 326 holds the probe chip 310 in position on the X-Y axis 80,82 while allowing it to move or ride in the Z-Direction 84. The compliant member 326 retains the interposer 150 in a compressed position, through a controlled flexion in the compliant member or decal 326, which is caused by any Z plane 84 delta between the bottom surface 305a of the interposer 150 and the mounted position of the compliant member 326.

The compliant member 326 also provides compliance for differences in thermal coefficients of expansion (TCE) between the probe chip 16,310 and the decal ring 328. For example, for decal probe assembly 300 in which a decal ring 328 which has a higher coefficient of expansion than a probe chip 16,310, the compliant member 326 readily flexes, i.e. stretches, at elevated temperatures, between the probe chip 12,300 and the decal ring 328.

Figure 42 is a partial cross sectional view 340 of a Z-Block decal interposer assembly 300b. Figure 43 is an expanded assembly view 360 of a Z-block decal interposer assembly 300b. The motherboard 304 is electrically connected to the probe chip wafer 16,310 through an upper interposer 344, a PWB Z-block 342, and a bottom interposer 348, respectively, which are located between the upper surface 311 b of the probe chip 16,310 and the lower surface 305a of the motherboard 304. The bottom stiffener plate 346 is aligned to corresponding holes or slots 309 in the motherboard PWB 304 via pins 354. The Z-Block vertical translation block (VTB) 342 provides Z translation of signals and power from the motherboard PWB 304 to the probe chip 16,310, so that the probe tips of the probe springs 61 a-61 n will clear the probe ring 284 (FIG. 38). The Z-Block 342 shown in Figure 43 has three sets of alignment pins 330,333, 350, extending from the top surface 343b and/or bottom surface 343a. The bottom interposer 348 and top interposer 344 are aligned to the pins 330,333 respectively, such as by an edge 351 or one or more alignment holes 347. The Z-Block 342 itself is aligned to the bottom stiffener 346 via pins 350.

The compliant member 326 holds the probe chip 310 in position on the X-Y axis 80,82 while allowing it to move or ride in the Z-Direction 84, as the Z-Block 342 expands due to thermal variation. The compliant member 326 retains the bottom interposer 348 in a compressed position, through a controlled flexion in the compliant member or decal 326, which is caused by any Z plane 84 delta between the bottom surface 349a of the bottom interposer 348 and the mounted position of the compliant member 326. The compliant member 326 also provides compliance for differences in thermal coefficients of expansion (TCE) between the probe chip 16,310 and the decal ring 328. As seen in Figure 42, the Z-block 342 is mounted to a bottom stiffener ring 346, which is preferably comprised of a rigid material, such as stainless steel. The bottom stiffener ring 346 retains the Z-block 342 in a flat, i.e. planar, position, while the thickness of the stiffener ring 346 limits, i.e. controls, the deflection amount of the compliant membrane or decal 326.

Z-Block retaining clip assemblies 352 hold the Z-block 342 to the bottom stiffener ring 346, and allow the Z-block to expand in the Z-direction at elevated temperatures. The bottom stiffener ring 346 holds and supports the Z-block 342 toward the motherboard 304, typically with an intermediate upper interposer 344.

Planarity adjustment 326 is preferably provided between the motherboard 304 and the probe chip wafer 16, such as by one or more planarity adjustment screws and/or shims.

In some embodiments of the decal interposer assembly 300, means for providing planarity 324 comprises a plurality of differential screw assemblies 824 (FIG. 73, FIG.

74, FIG. 75).

In some embodiments of the Z-Block decal interposer assembly 300b, the bottom stiffener ring 346 further comprises means for planarity adjustment, such as by one or more planarity adjustment screws 324,824 (FIG. 73, FIG. 74, FIG. 75), whereby the planarity between contact tips are adjustable relative to the lower surface 305a of the motherboard 304. The bottom stiffener ring 346 is held under spring force against the planarity adjustment screws 324,806, whereby the bottom stiffener ring stays in contact with the screw as it is moved vertically in the assembly. The planarity adjustment system 326 may comprise a wide variety of structures, such as fasteners, differential screws, guides, shoulder bolts, as well as biasing hardware, such as compression washers and/or compliant o-rings.

As seen in Figure 43, the bottom interposer 348 provides electrical connections between the probe chip 310 and the Z-block 342, and allows the probe chip 310 to be a removable sub-component. The bottom interposer 348 also provides a sliding interface between the probe chip 310 and the Z-block 342, to accommodate TCE differences.

The compliant membrane or mesh 326 retains the probe chip 310,16 in position on the X axis 80 and Y-axis 82, while allowing the probe chip 310,16 to ride in the Z-Direction 84 as the Z-Block 342 expands due to thermal variation. The compliant membrane or mesh 326 keeps the bottom interposer 348 fully compressed at all times, by a controlled flex in the decal 326 caused by a Z plane delta between the bottom of the bottom interposer 348 and the mounted position of the decal ring 328, which provides compliance for TCE differences between the probe chip 310 and the decal ring 328.

Enhanced Probe Chip. Figure 44 is a cross sectional view 370 of a probe chip 310 prior to finger lift. Figure 45 is a cross sectional view 400 of a probe chip 310 after finger lift 404 and plate 406. Prior to finger lift, a portion of the probe surface 62a typically comprises a release layer 376, preferably comprising titanium, which is formed on the ceramic substrate 372. A composite layer 380 is then formed on the release layer 376, wherein the composite layer 380 typically comprises a plurality,, preferably but not restricted to 5 layers, of metallic film layers 17a-17n (FIG. 37), in which at least two adjoining layers have different inherent levels of stress before spring lift 404 (FIG. 45).

Such composite layers are formed preferably by techniques such as sputter deposition.

In some embodiments, such composite layers are formed by electrodeposition technique also. In the probe chip structure 310 shown in Figure 44, a conductive seed layer 384, preferably comprising gold, is preferably formed on the composite layer 380. Conductive traces 382 are then controllably formed over portions of the composite layer 380, typically through photolithographic patterning.

In some embodiments, upper layers, typically comprising a first polyimide (PMID) layer 384, a ground plane 388, a second PMID layer 390, and a hard mask layer 392, are then applied over the trace regions, typically providing shielding for signal traces.

Figure 44 and Figure 45 also show an exemplary formed connection structure 393 on the connector side of the substrate 372. The connection structure 393 provides contact pads 396, for connection to an interposer 348, e.g. such as seen in Figure 43, and also to a connection trace between a pad 445 (Fig. 49) and vias 374, as seen in Figure 44 and Figure 45.

The connection structure 393 is comprised of a stack of one or more metal layers 376,394, 396, which typically includes an outer layer 396 comprising gold, for reliable contact to interposer contacts or springs. In some embodiments of the probe chip 310, the connection structure 393 comprises a first titanium metal layer 376, a second nickel layer 394, and a third gold metal layer 396. In other embodiments of the probe chip 310, the connection structure 393 comprises a first Cr layer and Cu metal layer 376, a second nickel layer 394, and a third gold metal layer 396. The thickness of the layers 376,394, 396 is typically controlled to reduce electrical sheet resistance. In some embodiments of the connection structure, the first Cr/Cu layer 374 comprises a thickness 1-4um, and/or the outer gold layer comprises a thickness of 1-4um.

In some embodiments, a dielectric layer such as polyimide, photoresist, epoxy, or kapton, can be added on top of the stacked structure 393, such as to serve as a solder mask 398 for soldering components, e.g. bypass capacitors to the supper surface 396, or to serve as insulation against shorting to undesired locations in the interposer 348.

As seen in Figure 45, after finger lift 404, the probe spring 61 extends away from the probe surface 62a of the substrate 372 within the release region 386 (FIG. 44) as a result of inherent stress gradient in the plurality of film layers. In general, the free non- planar portion of the probe spring assumes substantially an arc shape upon release with the top deposited layer of the Mo-Cr film 380 and the gold seed layer 384 taking a concave shape. This results in the formation of a three-dimensional spring structure that is desirable for fabricating arrays of probe springs en masse with very small pitch using integrated circuit fabrication techniques including photolithography. A probe spring plating layer 402 is then preferably formed 406 on the extended, i.e. nonplanar, probe spring 61. In some embodiments of the probe springs, the plating layer 402 provides enhanced spring force, increased abrasion resistance, increased strength, and/or increased electrical conductivity. In some of the embodiments, a plurality of films comprising different materials, such as nickel, palladium alloy such as palladium colbalt, Rh, Au, is plated sequentially onto the lifted fingers.

In alternate embodiments of the probe spring 61,412, the spring tip may further comprise a secondary contact tip detail, i.e. a spring tip button contact, typically comprising rhodium, palladium or cobalt, as disclosed in Provisional Application Serial No. 60/365625, Filed 18 March 2002, Nanospring with Increased Resistance to Failure, the disclosure of which is incorporated herein by reference.

Probe Chip Component Functions. On the probe side 62a of the probe chip substrate 372, the probe chip 16,310 supports the probe springs, such as photo- lithographically defined springs 61, for contact to a wafer under test 92. The probe chip 16,310 provides signal and power fan-out to a grid achievable by the thick motherboard PWB 304, such as seen in Figure 40. As seen in Figure 44 and Figure 45, the probe chip 16,310 provides signal connections through the substrate 372, to a matrix of pads or solder balls 398 on the connection surface 62b, typically at a pitch matrix that matches the PTH matrix pitch on the motherboard 304.

The probe chip 16,310 also preferably provides controlled impedance to the fan-out traces 307 (FIG. 40), and preferably provides regions to mount bypass capacitors, such as on the connector surface 311 b of the probe chip 16,310. Some preferred embodiments of the probe chip 16,310 comprise two or more electrically conducting vias 374 per electrical path, so as to increase manufacturing yield through redundancy. Similarly, some preferred embodiments of the probe chip comprise two or more probe springs 61 per electrical path (FIG. 9).

Probe Card Assembly Sequence. As seen in Figure 43, the enhanced probe card assembly 310 is readily assembled to provide enhanced connections to one or more devices under test on a wafer 92.

In an exemplary assembly process, as seen in Figure 43, the motherboard PWB 304 is attached to the top stiffener plate 302, typically using fasteners, such as screws 324, forming a first sub-assembly 361 a. In some embodiments, the hole 309 could be a through hole, and the top stiffner plate with a larger dimension is attached to the mother board using a pin inserted into the through hole (not shown).

A second sub-assembly 361 b is typically formed by pushing alignment pins 330,350 into the Z-block PWB 342, and by attaching the Z-block 342 into the bottom stiffener ring 346, such as by alignment pins 350, Z-block retaining clip rings 352, and screws 353.

The second sub-assembly 361 b is then typically placed on a temporary support structure, such that the top interposer 344 is controllably placed onto the Z-block 342, and aligned with the interposer alignment pins 333. Small blind holes (not shown) are provided in the motherboard PWB 304, corresponding to the pins 333, to make room for the insertion of the exposed sections of the pins.

The first sub-assembly 361a is then placed over the second sub-assembly 361 b, while aligning the stiffener ring alignment pins 354 on the top surface 347b (FIG. 43) of the stiffener ring 346 (FIG. 43) to alignment holes 309 (FIG. 43) defined on the lower surface 305a (FIG. 43) of the motherboard PWB 304 (FIG. 43), and while gently compressing the top interposer springs on the top interposer 344. The first sub- assembly 361 a and the second sub-assembly 361 b are then secured together, such as by screws 834 (FIG. 75), forming a third sub-assembly 361 c.

The third sub-assembly 361 c is then typically turned over and placed onto a second temporary support structure, such that the bottom interposer 348 is controllably placed onto the lower surface 343a of the Z-block 342, and aligned with the interposer alignment pins 330. The decal subassembly 361 d is then positioned onto the bottom interposer 348, which in some embodiments is guided by the alignment pins 354. The use of a temporary assembly support fixture ensures alignment between the sub- assemblies 361c, 361d, and allows the decal assembly 361 d to be lowered gently onto the interposer 348, such that the outer ring 328 is then further compressed, to contact the bottom stiffener 346, while stretching the membrane 326. The assembly fixture holds the ring 328 in place, while fasteners 356 are tightened.

It should be appreciated that the descriptions for the assembly given above is for a typical embodiment. Some variations in the assembly fixture and process, which are based on the basic conception outlined above are within the scope of this invention. For example, the interposers 348 or 344 may not be required to have any holes as shown in Fig. 43. Instead, a number of additional pins such as 333 or 330 may be provided to hold the interposers in place. Similarly, the outer ring 328 in Fig. 43 may preferably be placed above the membrane 326.

Probe Chip Structure Details. Figure 46 is a plan view 410 of a spring side surface 311 a of a probe chip 16,310. Figure 47 is a detailed partial layout view 430 of a spring side surface 311 a of a probe chip 16,310. Probe chip probe springs 412, such as photlithographically formed probe springs 61a-61n, are generally arranged within probe spring groups 422, to provide a plurality of connections to one or more devices 44 on a wafer under test 92. The probe springs 412, such as photlithographically formed probe springs 61 a-61 n match contact pads 47 on a target wafer 92. The exemplary arrangement of probe spring groups 422 shown in Figure 46 is a 4x8 set of spring groups 422, for contacting thirty two die positions in parallel.

Vias 414 extend through the probe chip substrate 372, from the spring side surface 311 a to the connector, i.e. interposer, side surface 311b. The vias are preferably arranged in redundant via pairs 416, such as to increase the manufacturing yield of the probe chip 16,310, and/or to promote electrical conduction, particularly for power traces.

The via pairs 416 shown in Figure 46 and Figure 47 are arranged in a via grid array 417, comprising via rows 418 and via columns 420. The via grid array 417 preferably matches the plated through hole (PTH) grid on the motherboard 304. Where springs 412 coincide with vias 414,416, the vias 414,416 are moved to an adjacent row 418, and the traces 424 are typically routed to where the original vias 414,416 would have been located, on the opposite interposer side 311 b of the probe chip 16,310.

Figure 48 is a plan view 432 of an interposer side surface 311 b of a probe chip 16,310.

Figure 49 is a detailed partial layout view 444 of an interposer surface 311 b of a probe chip 16,310. Contact pads 434 are located on the interposer side surface 311 b, and are typically arranged in a pad grid array 436, which generally coincide with the vias 214 (FIG. 49), preferably arranged in redundant via pairs 416, such as to provide electrical contacts to an interposer 348. In some embodiments of the probe chip 16,310, the pad grid array 436 has a pitch of 0.056", to match a similar contact pitch on a connected interposer 348. As seen in Figure 48, a central region 438 is generally defined within the interposer side surface 311 b, wherein contact pads 434 and probe chip upper traces 442 are typically arranged in device regions 440.

An adhesive region 437 is located along the outer periphery of the interposer side surface 311b of the probe chip 16,310, whereby an adhesive 329 is attached or applied between the probe chip 16,310 and a compliant member or decal 326.

As seen in Figure 49, the interposer side surface 62b, 311 b of the probe chip 16,310 provides signal contacts 445, active ground contacts 446 (S) power contacts (P) 448, and ground contacts 450 (G). The traces 442 provide routing of the contacts 445,446, 448, and/or 450, to allow relocation of vias 414,416, such as due to any conflict of spring location with the vias 414,416. The probe chip 16,310 shown in Figure 48 and Figure 49 further comprises capacitors 452, such as between an active ground 446 and a power contact 448, between an active ground 446 and a ground contact 450, and/or between a power contact 448 and a ground contact 450.

Probe Chip Assembly Structure. Figure 50 is an expanded assembly view of a decal to probe chip assembly fixture 460. A fixture base 462 comprises a probe chip ledge 464 surrounding an inner probe spring relief region 466. One or more probe chip wafer alignment pins 468 are typically located on the probe chip ledge 464, to accurately receive a probe chip 16,310. The fixture base 462 similarly comprises a decal ring cavity 470 surrounding the probe chip ledge 464.

One or more lamination plate and stencil alignment pins 472 are typically located on the decal ring cavity 470, to accurately receive either a top lamination plate 474, having matching plate alignment holes 476, or a decal excise stencil 478, having stencil alignment holes 479. As described below, the top lamination plate 474 is used during lamination of the compliant member 326 to the probe chip 16,310, such as with an adhesive 329.

The decal excise stencil 478 further comprises a stencil template opening 480. When the decal excise stencil 478 is positioned on the fixture base 462 over an assembly comprising a probe chip 310 adhesive mounted to a compliant member or decal 326, the stencil template opening 480 is generally aligned about the outer adhesive region of the probe chip 310, such that cutting and removal of the inner region of the decal 326 provides access to the surface of the probe chip 16,310.

Decal Assembly Sequence. Figure 51 is a flowchart showing a decal probe chip assembly process 490. A completed probe chip 310 (lifted 402 and plated 404) is placed 492 face down onto the bottom plate 462 of the lamination fixture 460, which supports the probe chip 310 under the adhesive region only. A cavity 466 is provided to protect the springs 61. The probe chip 310 is aligned to alignment pins 468 on the fixture base 462.

An adhesive sheet 329 (FIG. 43) which is preferably pre-cut to the shape of the desired adhesive region on the probe chip 16,310 is then applied 494 to the perimeter of the probe chip 16,310. The adhesive sheet 329 is aligned to the features on the probe chip 16,310, and is then typically tacked with a hot iron. In one embodiment of the assembly 300, the adhesive sheet 329 is a B-stage adhesive sheet 329.

A compliant member or decal 326, which has been pre-strung to a specified tension and attached on the decal ring 328, is then placed 496 on the fixture 462 over the adhesive 329. In some embodiments of the lamination fixture 460, the ring side of the decal 326 is placed side up within the fixture 460. In alternate embodiments of the lamination fixture 460, the ring side of the decal 326 is placed side down within the fixture 460. The decal ring 328 is aligned to the fixture 462 using the alignment pins 472.

The fixture top plate 474 is then placed 498 over the decal 326, which compresses the decal 326 against the adhesive sheet 329, while held in place via the alignment pins 472 to the fixture base 462.

The assembly fixture 460 is then used to cure the adhesive 329, which typically comprises the steps of placing 500 the assembly 460 into an oven, placing 502 a weight or other compressive force on the fixture top plate 474, baking 504 to cure the adhesive 329, and removing 506 the weight and assembly 460 from the oven.

Removal of the inner portion of the compliant decal 326 comprises the placement 508 of the decal excise template 478 over decal 326. The decal excise template 478 is aligned to alignment pins 472 on the fixture bottom plate. The inner portion of the decal 326, i.e. inside the adhesive area 319, is then separated or excised 510 with a knife.

Interposer Structures. Figure 52 is a partial cross sectional view of an interposer structure 520, such as for a top interposer 344 or a bottom interposer 348.

Interposer springs 521, such as photolithographically formed probe springs 521, are generally arranged within an interposer grid array, to provide a plurality of standardized connections. For example, in the top interposer 344 shown in Figure 43, the interposer springs 521 provide connections between a motherboard 304 and a Z-block 342.

Similarly, in the bottom interposer 348shown in Figure 43, the interposer springs 521 provide connections between the Z-block 342 and the probe chip 16,310.

Interposer vias 524 extend through the interposer substrate 522, from the first surface 523a to the second surface 523b. The interposer vias 524 are preferably arranged in redundant via pairs, such as to increase the manufacturing yield of the interposer 520,344, 348, and/or to promote electrical conduction, particularly for power traces.

The opposing surfaces 523a, 523b are typically comprised of a release layer 526, such as comprising titanium, and a composite layer 530,532, typically comprising a plurality of conductive layers having different inherent levels of stress. Interposer vias 524, e.g. such as CuW or gold filled, extend through the central substrate 522, typically ceramic, and provide an electrically conductive connection between the release layers 526. The composite layer 530,532 typically comprises MoCr, in which the interposer probe springs 521 are patterned and subsequently to be later released within a release region 528.

A seed layer 534, such as a 0.5 to 1 um thick gold layer, is preferably formed over the composite layers 530,532. A tip coating 540, such as rhodium or palladium alloy, is controllably formed at least over the tips of spring fingers 532, such as to provide wear durability and/or contact reliability. Traces 536, typically comprising copper, are selectably formed by plating over the structure 520, as shown, such as to provide reduced resistance. As well polyimide PMID layers 538 are typically formed over the structure 520, as shown, to define the spring finger lift regions. A thick gold layer 534 remains on the lifted fingers 521, so as to reduce sheet resistance of the fingers 521.

Figure 53 is a plan layout view 550 of a interposer 520. A spring set array 552 is located within an inner contact region 554 of the interposer substrate 522, and comprises a plurality of multiple finger sets 560, which preferably provide redundancy of connections through a plurality of finger springs 521 (FIG. 52, FIG. 54), as well through a plurality of redundant interposer vias 524, as seen in Figure 52 and Figure 54. An outer support region 556 is located about the periphery of the interposer substrate 522. The inner contact region 554 of the interposer substrate 522 is located within the defined saw boundary 558.

Figure 54 is a partial detailed layout view 562 of an interposer 520. The multiple finger sets 560 are preferably arranged on the interposer substrate on a pitch 566, which preferably matches the corresponding connection pitch on the motherboard 304. Each of the multiple finger sets comprise a common conductive region 564, from which a plurality of redundant spring fingers 521 are formed, which extend from the substrate 522 over a lift region 528. A plurality of redundant interposer vias 524 are electrically connected to the common conductive region 564, and extend through the interposer substrate 522, such as from a first surface 523a to an opposing surface 523b (FIG. 52).

In the interposer 520 shown in Figure 54, each multiple finger set 560 comprises four redundant fingers 521 and four redundant interposer vias 524. In alternate interposers 520, any number of redundant fingers 521 and/or interposer vias 524 may be provided, e.g. such as two or three fingers 521 and/or interposer vias 524. While the multiple finger sets 560 shown in Figure 54 are generally arranged in a clover or cross configuration, a wide variety of geometries may be suitably used.

Some embodiments of the interposer 520 comprise integrated capacitors 576, which are typically smaller in thickness than the thickness of the interposer substrate 522.

Figure 55 is a partial plan view 570 of an interposer 520 having vacated springs within a capacitor cavity region 574. Figure 56 is a partial plan view 580 of an interposer 520 having embedded bypass capacitors 576. The multiple finger sets 560 are readily arranged or modified to provide vacated regions 572, within which thru hole cavities or cut-out regions 574 are located. As seen in Figure 56, vias 582, e.g. such as dog bone vias 582, provide electrical contacts between multiple finger sets 560 and the embedded capacitors 576. The material which comprises the dog bone vias 582 is not effected by laser during substrate cut. The exposed via surface is preferably treated by plating, to provide a suitable surface for a conductive epoxy or solder connection to the capacitor 576.

Before finger lift, an adhesive tape is applied to one side 523 of the interposer substrate 522. The capacitors 576 are picked and placed into the cavities 574 from the opposite side 523, e.g. 523b, of the substrate 522, and are temporarily held in place by the adhesive tape. Epoxy is then dispensed to either end, to complete contact between the capacitors 576 and the dog bone vias 582. The epoxy is then cured, after which the adhesive tape is removed. The fingers 521 are then lifted from the substrate 522.

Alternate Interposer Structures & Processes. Figure 57 is a partial cross sectional view 590 of an alternate decal interposer structure 520b before spring lift. Figure 58 is a partial cross sectional view 620 of an alternate decal interposer structure 520b after spring lift.

The alternate decal interposer structure 520b comprises pairs of oppositely pointing springs 521 that are joined at the base, and are attached to the edge of an opening in a membrane 602 which in turn is supported like a drum inside a support frame 610. The spring pair 521 extends away from the plane of the membrane 602 protruding through the hole in the membrane 602. The membrane 602 is preferably held under tension inside the frame 610, such that the locations of the spring pairs 521 are held constant.

This interposer structure 520b has the advantage that it does not require a substrate with vias and also that fingers 520 from both sides can be patterned with a singe photolithographic step, thus greatly lowering the fabrication cost. In addition, in embodiments where the alternate interposer 520 is used to interface between the connector side 62b of a probe chip 310 and a PWB board 304 where there are passive components, such as bypass capacitors, mounted on the connector side 62b of the probe chip 310, openings for the components to protrude through the interposer 520 can be easily provided in the photolithographic steps to pattern the membrane 602 and not requiring an expensive laser step to drill holes in the interposer substrate 602.

Being suspended on a flexible membrane 602, the alternate interposer 520b can also can be used in applications where the interfacing surfaces are not flat. Also, the thickness of an alternate interposer 520b as a connector can be as small as several microns. The alternate interposer 520b can also have a much smaller connector pitch, since the alternate interposer 520b is not confined by the via pitch of the probe chip substrate 310. The pitch can be as small as the photolithographic process can pattern the springs.

Unreleased portions of the metal can also be patterned on the membrane 602 to form interconnect as well. The support frame 610 can also be bent if it is made of a ductile material such as metal. This allows the membrane interposer 520b to be shaped into infinite topological shapes.

Figure 59 is a flow chart showing an exemplary interposer construction process 550. A first release layer 594, typically comprising titanium, having an approximate thickness of 2,000 angstroms, is formed 622 by sputtering on a substrate 592, which typically comprises silicon, aluminum, ceramic, or glass. A MoCr stressed sandwich layer 596, comprising a plurality of stress layers 17a-17n, such as seen in Figure 37 and/or Figure 38, having a resultant downward peeling stress, is sputter formed 634 on the first release layer 594. The plurality of stress layers 17a-17n in the first stressed sandwich layer 596 typically comprises a first tensile layer 17a, and a final compressive layer 17n, to provide an inherent stress gradient comprising a downward peeling stress.

A second layer 598, typically comprising titanium, is formed on top of the first stress sandwich, comprising a pattern which defines the region of the fingers that will lift away from each other. Note that there is a region 597 of the finger where this release layer is removed. This region 597 forms the contact region between the top and bottom finger 521. The second release layer 598, typically having an approximate thickness of 1,000 angstroms, is sputter formed 636 and patterned 638, with connection regions 597 that generally define the end of the finger lift regions 528. A second MoCr layer 600 is sputter formed 640 on the second release layer 598, comprising a plurality of stress layers 17a-17n, such as seen in Figure 37 and/or Figure 38, having a resultant upward peeling stress. The plurality of stress layers 17a-17n in the second stressed sandwich layer 600 typically comprises a first compressive layer 17a, and a final tensile layer 17n, to provide an inherent stress gradient comprising an upward, i.e. opposing, peeling stress.

Photoresist is then spun 642 to define the finger layout for the springs 521, and the second MoCr layer 600 is etched 644, such as by either a wet or dry etch process, to open regions where the springs 521 will lift, as well as any other regions where a hole is desired in the membrane, such as for mating components to protrude through. The second release layer 598 is then etched 646, and the first MoCr tensile stress layer 596 is etched 648, such as to allow the fingers 521 in the first MoCr layer 596 to be more undercut, while the first release layer 594 is left to overhang.

A polyimide layer 602, of an exemplary thickness of 18, um to 25 um, is spun and patterned 650 by a wet etch process, to open the lift window and the flat base region in the fingers 521. A seed layer 604, typically comprising gold, is then sputtered and patterned 652, to provide shorts between all fingers 521, but not to impede lift.

A plating mask 606, such as anodized aluminum, is then patterned 654 over the seed layer 604, so that only the fingers 521 are plated. The plating mask 606 is comprised of a material that can withstand being flexed after the substrate 592 has been removed and also that is easily removed, without attacking the fingers 521.

A support ring 610, typically comprising stainless steel, is attached 656 to the outside of the wafer substrate 592, such as by an epoxy adhesive 608, to serve as a decal support ring. The attached region of the substrate does not include either the seed layer 604 or the plating mask 606. Stainless steel is a preferred material for the support ring, where the probe springs are to contact a printed wiring board. In some embodiments requiring the pins to contact silicon devices, the support ring is comprised of a material, such as molybdenum, with a relatively low thermal coefficient of expansion that is relatively close to that of silicon.

The assembly is then finger lift etched 658, causing the entire membrane to delaminate from the substrate 592, as the fingers 521a, 521b defined layers 596,600 lift in opposing directions. The finger lift etching step 658 may require a longer time period in regions located under the support ring 610.

Altemately, the substrate 592 can be made of a material such as aluminum, which can b e removed by mechanical or chemical means.

As seen in Figure 58, one or more plating layers 622a, 622b are preferably formed on 660 the lifted fingers 521 a, 521 b as needed. In some interposer embodiments 520b, the plating layers 622a, 622b comprise a 1 urn to 10um nickel layer, as undercoat, and a 0.2 to 5um contact wear layer, such as rhodium, palladium palladium cobalt, or gold.

The plating mask 606 is then removed 662, followed by the removal 664 of the exposed seed layer 604.

In alternate embodiments of the decal interposer assembly 300, such as an <BR> <BR> ISOCONTM Connector, available through Circuit Components, Inc., of Tempe AZ. , is used as the upper interposer 344. While a ISOCON Connector typically requires a higher force to establish electrical contacts, ISOCON Connectors often provide a suitable cost-effective interconnection for the upper interposer 344, since the required force can be achieved between the Z-Block and the motherboard PWB 304.

Interposers which require relatively low contact forces are typically chosen for connection between the Z-Block 342 and the probe chip 16,310.

Z-Block Architecture. Figure 60 is a plan layout view 670 of a Z-block printed wiring board 342. Figure 61 is a partial detailed view 680 of a Z-block printed wiring board 342. The Z-block 342 comprises a plurality of electrical connections, such as an array of plated through holes 674, which extend between opposing surfaces 343a, 343b of the Z-block substrate 672. As seen in Figure 60, interposer alignment pins 330 and stiffener ring alignment pins 350 are located through the Z-block substrate 672, for alignment within the enhanced probe assemblies 300. The Z-block substrate 672 shown in Figure 60 also comprises retaining clip recesses 676, by which the Z-block is aligned to the retaining clips 352 (FIG. 43). As seen in Figure 61, the plurality of electrical connections typically comprise signal contact sites 682, including device Vcc and Vss connections, as well as system ground GND contact sites 684. In the exemplary Z-block 342 shown in Figure 43, every signal 682 plated through hole 674 is surrounded by four ground 684 plated through holes 674, to improve impedance and crosstalk performance, such as to provide an impedance matching structure for high frequency system environments.

Alternate Enhanced Probe Assemblies. Figure 62 is a partial cross sectional view 700 of a Z-block decal interposer assembly 300c having permanent interface 702 between the motherboard PWB 304 and the Z-block.

In contrast to the Z-block decal interposer assembly 300b, as seen in Figure 42 and Figure 43, in which the upper interface 308 (FIG. 40) comprises an interposer 344, the upper interface 308 in the Z-block decal interposer assembly 300c comprises a permanent interface 702, such as a solder ball array 756, an anisotropic conducting film, or electrically conductive pins to stake the Z-block 342 to the motherboard 304.

In some embodiments of the Z-block decal interposer assembly 300c, planarity adjustment is provided by shims 827 (FIG. 72) located between the motherboard 304 and the bottom stiffener ring 346. The probe force is supported through the upper interface connections 702, located between the Z-Block 342 and the motherboard 304, and not through the bottom stiffener ring 346. The Z-block decal interposer assembly 300c provides both Z-axis translation and planarity compliance, while providing a relatively inexpensive permanent upper interface 702, and eliminating the cost of an upper interposer 344.

Figure 63 is a partial cross sectional view of a Z-block decal interposer 300d having a probe chip 16,310 which comprises tester side springs 64a-64n. In contrast to the Z- block decal interposer assembly 300c, in which the lower interface 312 (FIG. 40) comprises an interposer 348, the lower interface 308 in the Z-block decal interposer assembly 300d comprises spring connections 64a-64n on the upper surface 311 b of the probe chip 16,310, which directly contact the Z-block 342.

In some embodiments of the Z-block decal interposer assembly 300d, the spring connections 64a-64n on the upper surface 311b of the probe chip 16,310 are redundant springs 64a-64n. The Z-block 342 shown in Figure 63 preferably comprises a cavities 722, typically formed by milling, which provide room for bypass capacitors 724 mounted on the probe chip 16,310.

In a similar manner to the Z-block decal interposer assembly 300c, the Z-block decal interposer assembly 300d provides planarity adjustment 324, either by shims 827 located between the motherboard 304 and the probe ring 346, or by other planarity adjustment mechanisms, such as differential screw assemblies 824 (FIG. 73). The Z- block decal interposer assembly 300d provides both Z-axis translation and planarity compliance, while eliminating the cost of a lower interposer 348. The Z-block decal interposer assembly 300d also comprises bypass capacitors 724 located within the Z- block cavities 722, which are typically lower in cost than interposer cutout regions 574 (FIG. 55, FIG. 56), which are commonly formed by laser cutting.

Figure 64 is a partial cross sectional view 740 of a probe card assembly 300e having a pin block 742 with a solder ball array 756 (FIG. 65). Figure 65 is a detailed schematic view 750 of a pin block having a solder ball array 756. In the probe card assembly 300e, the upper interface 308 (FIG. 40) comprises a pin grid array 745, the lower interface 312 (FIG. 40) comprises a solder ball array 756, and the probe chip 310 comprises one-sided springs 61a-61n located on the lower surface 311 a. A ZIF actuation template 743 is located between the metal pin block 742 and the motherboard 304. System planarity is preferably provided, such as by three differential screw assemblies 745.

The pin block 742 comprises a plurality of dielectric holes 746, through which the array 745 of pin connectors 744 extend. Pin templates 748,752, typically comprised of KAPTON, are located on opposing surfaces of the pin block 742. Ground contacts 758 within holes 746 provide ground connections between one or more pins 744 and the pin block 742. The pin grid array 745 electrically contacts a pin socket array 755 h the motherboard 304.

Figure 66 is a partial cross sectional view 760 of a probe card assembly 300f having a pin block 742, in which the probe chip card 16,310 comprises backside springs 64a- 64n. Figure 67 is a detailed schematic view 770 of a pin block 742, in which the probe chip card 16,310 comprises backside springs 64a-64n. The pin block 742 comprises a plurality of dielectric holes 746, through which the array 745 of pin connectors 744 extend. The pin grid array 745 extends through the motherboard 304, wherein solder joints 762 provide both mechanical and electrical connections between the motherboard 304 and the pin block 742. The probe chip 16,310 is flexibly suspended from the assembly by the compliant member or decal 326, while upper springs 64a-64n provide a compliant electrical interface 312 (FIG. 40). The probe chip 16,310 is therefore moveable with respect to the pin block 742, and provides compliance for the probe card assembly 300f.

As seen in Figure 67, Figure 69, and Figure 71, one or more capacitors 724, such as bypass capacitors 724, may be located on the pin block 742, based upon the intended probing environment.

Figure 68 is a partial cross sectional view 780 of a probe card assembly 300g having a pin block 742, in which the probe chip card 16,310 comprises backside springs 64a- 64n, and in which the assembly comprises press fit pin connections 782. Figure 69 is a detailed schematic view 790 of a pin block 742, in which the probe chip card 16,310 comprises backside springs 64a-64n, and in which the assembly comprises press fit pin connections 782. The pin grid array 745 extends through the motherboard 304, wherein press fit pin connections 782 provide both mechanical and electrical connections between the motherboard 304 and the pin block 742. The probe chip 16,310 is flexibly suspended from the assembly by the compliant member or decal 326, while upper springs 64a-64n provide a compliant electrical interface 312 (FIG. 40). The probe chip 16,310 is therefore moveable with respect to the pin block 742, and provides compliance for the probe card assembly 300g.

Figure 70 is a partial cross sectional view of a probe card assembly 300h having a pin block 742 with SMT solder and top interposer 344. Figure 71 is a detailed schematic view of a pin block 742 with SMT solder and top interposer.

Planarity Adjustment Mechanisms for Compliant Carrier Probe Card Assemblies. As described above, many embodiments of the decal interposer assemblies 300 comprise planarity adjustment mechanisms 324. Figure 72 is a partial cross sectional view of a Z-block decal interposer assembly 300i having a planarity adjustment mechanism 324, comprising one or more shims 827 affixed to the lower surface 305a of the motherboard PWB 304 by a shim plate 825 and attachment screws 823. The use of one or more shims 827 at a plurality of locations about the periphery of the motherboard PWB 304 allows the planarity entire interposer assembly 300i to be adjustable relative to the probe ring 284 (FIG. 39).

Figure 73 is an expanded side assembly view 830 of a Z-block decal interposer assembly 300 having planarity adjustment 324, comprising a plurality of differential screw assemblies 824 and a plurality of shoulder bolt assemblies 832. Figure 74 is an expanded perspective assembly view 850 of a Z-block decal interposer assembly 300 having planarity adjustment. Figure 75 is a partial cross sectional view of a Z-block decal interposer assembly 300 having a planarity adjustment 324, comprising a plurality of differential screw assemblies 824. The plurality of differential screw assemblies 824 typically comprises three assemblies 824, such that a plane is readily defined by the leading tips of the assembly, e.g. such as the plungers 826, as seen in Figure 75. As seen in Figure 74, the motherboard PWB 304 is fixedly attached to the upper stiffener plate 302 by a plurality of fasteners 322. The differential screw assemblies 824 extend from the bottom stiffener ring 346 to the top stiffener 302, such that adjustment of the differential screw assemblies 824 provides planarity to be adjusted between the lower stiffener ring 346 and the upper portion 871 (FIG. 75) of the assembly 300i.

As seen in Figure 73, each of the differential screw assemblies 824 comprise a plunger 826, a differential screw 828, and a jam nut 830. As seen in Figure 75, the leading edge 872 of the plunger 826 is rotationally fixed within a matching slot 827 defined within the motherboard PWB 304. The differential screw 828 is threadably engaged 874 to the plunger 826, and is also threadably engaged 876 to the upper stiffener 302. The threaded interfaces 874,876 shown in Figure 75 have different rotational pitches, i.e. thread gages, such that a rotation of differential screw 828 causes axial movement 878 of the plunger 826, thereby adjusting the separation 838 (FIG. 75) between the lower stiffener 346 and the motherboard PWB 304. The pitch differential 877 allows fine adjustment 878 of the plunger, such that the planarity of the probe chip relative to the motherboard PWB 304 can be finely adjusted.

As seen in Figure 73 and Figure 75, each of the shoulder bolt assemblies 832 comprise a shoulder bolt screw 834 and a compliant O-ring or spring washer 836. As described above, adjustment of the differential screw assemblies 824 moves the bottom stiffener 346 in relation to the motherboard 304. The planarity, i. e separation 838, of the bottom stiffener 346 is adjustable in relation to the motherboard 304, since the bottom stiffener 346 is relatively affixed in relation to the leading edge 872 of the plungers 826, while the shoulder bolt screws 834 and compliant 0-rings or spring washers 836 keep the stiffener 346 in contact with the plungers 826.

High Performance Spring Contact Packages. Figure 76 is a partial cross sectional view 900 of a high performance spring package 902a for integrated circuits 44. Figure 77 is a partial cross sectional view 920 of an alternate high performance spring package 902b for integrated circuits 44, further comprising multilayer routings 924. Figure 78 is a topside view 940 of a high performance spring package 902 for integrated circuits 44.

As seen in Figure 76, the package substrate 903 comprises a first substrate 904, having a first surface 906a and a second surface 906b, in which probe springs 61 are located on the first surface 906a, and extend to electrical connections 908, e.g. such as plated through holes, which extend from the first surface 906a to the second surface 906b. As seen in Figure 77, the package substrate 903 may further comprise additional routing layers 922 located on the first surface 906a of the first substrate 904, such that the probe springs 61 are located on the outer surface 923a of the routing layers 922, and are connected to the electrical connections 908 by multilayer routing 924.

The high performance spring package 902 comprises a structure for building a package utilizing springs 61 on a substrate 903, such as for a single IC or MCM package. The probe tips 61 are fabricated on the substrate 903 using thin film or IC or MEMS based processing methods to achieve low manufacturing cost and well-controlled uniformity, as well as to fabricate arrays of highly miniaturized probe springs with ultra-small pitch, e.g. 10-50 micron.

The probe springs 61 are fabricated on either the first surface 906a of the first substrate 904 (FIG. 76), or on the outer surface 923a of the multi-layer routing layers 924, using either thin-film or IC or MEMS based processing methods, as described above.

Signals from the probe springs 61 extend from connected integrated circuits 44, preferably using multilayer routings 924. The opposite side 906b of the package 902a, 902b comprises electrical contacts 910, such as either common micro-ball grid solder array pads 918 (FIG. 76), typically at an array pitch such as 1.0 mm or with braised on pins or package pins 928 which are typically solderable to holes 936 on the PCB 912a, 912b. The package 902a, 902b is therefore connectable to the printed circuit board 912a, 912b, such as for an end product 934, through the electrical contacts 910.

A single substrate wafer, which may contain multiple high performance spring packages 902 can be built on the first substrate wafer 904, providing cost-effective fabrication. For spring probe substrates 904 having a small surface area, several spring probe contactor packages may typically be fabricated from a single wafer 92 (FIG. 13). For example, as many as twenty-four sites, may be established on a standard four inch round starting wafer.

As seen in Figure 78, an array 942 of Micro BGA pads 944 located on the substrate assembly 902 are preferably arranged on a standard pitch, e.g. such as a pitch of 0.5mm, 1 mm, or 1.27mm. As seen in Figure 76 and Figure 77, the high performance spring package 902 may further comprise capacitors 932, typically to aid in high frequency power decoupling. The capacitors 932 are either mounted to either surface 906a, 906b of the substrate 904, or are formed, such as a parallel plate capacitor 932, within the substrate 904, typically between the reference plane and a plane formed on the unused trace areas of the substrate 904. For embodiments in which the first substrate 904 is comprised of silicon, an integral capacitor 932 may preferably be formed within the silicon substrate 904, typically comprising metal-dielectric-metal construction, metal-dielectric-heavily-doped semiconductor, or p-n junction using integrated circuit fabrication techniques. For embodiments in which the first substrate 904 is composed of a non-semiconductive material, metal-dielectric-metal capacitors may preferably be fabricated on or within the substrate 904, using integrated circuit fabrication techniques.

For embodiments in which the substrate 903 is composed of electrically conducting or semiconducting materials, such as doped silicon, the fabrication process is modified similar to the modification of the probe chip, as explained earlier with respect to Fig. 38.

As in the probe chip substrate fabrication, such modifications include deposition or formation of an electrically insulative film, e.g. oxide, on the substrate surfaces, as well as on the walls of the vias running through the substrate.

As stated above, the structure of the probe card packages 902 provides very short electrical distances between the probe tips 61 and the controlled impedance environment. This allows the high performance spring packages 902 to be used for high frequency applications. As shown in Figure 76, the high performance spring package 902 provides access for a shielded high frequency pin 916, by having grounds surrounding the pin 916. As seen in Figure 77, a ground is provided through a routing trace 938 under the spring 61 in layer 924. As well, the spring probe substrate 903 may preferably be modified for ultra high frequency applications.

For embodiments wherein the traces on one or both surfaces of the substrate 903 are required to be impedance controlled, one of more conductive reference planes may b e added within the substrate, either on top of the traces, below the traces, or both above and below the traces. The substrate 903 may also contain alternating ground reference traces, which are connected to the one or two reference planes, effectively providing a shielded coaxial transmission line environment. While the first substrate 904 typically comprises a ceramic material, the layered substrate 922 comprises conductive traces within a dielectric material, such as an organic or inorganic material. For some other embodiments controlled impedance environment in electrically conducting vias or through holes within the substrate 903 can be provided by constructing the via as shown in Fig. 37B and described earlier. Such vias provide a ground plane that is separated from the core electrical conductor by a dielectric film.

Advantages of High Performance Spring Packages. As described above, MEMS or IC processing methods can be used to fabricate the springs 61,412. The high performance spring package 902 has all of the benefits of"chips-first"and/or BBUL packages, while also allowing the replacement of any defective IC 44. The high performance spring package 902 can be tested prior to attaching the ICs 44, which significantly lowers the cost and risk of placing single or multiple ICs 44 in one package 902.

The substrate structure, with through holes 908, preferably comprises a plurality of electrical routing layers built on top of it to provide the functions of the HDI. These functions include : 1) much higher density of l/O connections than attainable in flip-chip or wire bonded packages, 2) higher interconnect performance can be achieved on an I C and 3) thinner packages.

This spring-based package 902 maintains low power and signal inductance by pushing the spring nearly flat to the multi-layer routing traces. The routing, directly under the spring, can be constructed to maintain controlled impedance as described in WO/09623.

The distances through the substrate can be kept very short and a ceramic substrate 904 supports RF frequency operations.

The high performance spring package 902 can be used as a"test package", such as to test expensive ICs 44 before committing them to the package 902. Some embodiments of the high performance spring package 902 comprise similar decoupling and impedance control features, as describe above for probe card embodiments. If permanent connection is desired between the probe springs 61 and the electrical contact pads on the IC 44, following the testing and identification of good chips, it can b e done using conventional joining techniques used in packaging, such as by reflowing the solder at the contact or by adhesive.

As seen in Figure 76, the high performance spring package 902 may further comprise means for holding integrated circuit devices in contact with the package, such as vacuum pull-down ports 905 defined through the substrate 903, and/or a temporary lid or pushdown plungers 909.

High performance spring packages 902 are an effective debugging tool for testing individual ICs 44 at RF frequencies, where the load of the package 902 affects the final packaged IC performance. The use of high performance spring packages 902 therefore helps to reduce the time-to-market characterization for new high-performance devices used in MCM packages, since integrated circuit devices 44 are not required to be committed to the final package 902 until the testing under load is verified. The test package also provides means to mix one or more RF signals with hundreds or thousands of digital signals. At high testing and/or manufacturing volumes, high performance spring packaging 902 significantly lowers package cost, as compared to a BBUL approach comprising similar frequency capabilities.

Quick-Turn Probe Assemblies. Figure 83 is a flowchart for a quick-tum probe assembly fabrication process 960. As describe above, many embodiments of probe card assemblies 60, enhanced probe card assemblies 300, and high performance spring packages 902 typically substantially comprise standardized componentry, which is readily reusable for connection to one or more devices 44 on a wafer 92.

The probe assembly 60,300 shown in Figure 83 comprises a master slice structure 962 comprising standardized portions, such as a motherboard substrate 304 (Fig. 40), at least one intermediate connector 306, and standardized portions of a probe chip 310, which typically includes the probe chip substrate and connector surface electrical connections, and may also typically include standardized electrical connections which extend from the connector surface 62b to the probe surface 62a of the probe chip substrate 310. It is to be understood that that the intermediate connector may include one or more components, such as an interposer and/or Z-block.

As seen in Figure 83, the quick-tum process 960, i.e. the method for developing a probe assembly for connection to at least one device on a wafer, typically comprises the establishment of the master slice 962, which comprises the steps of: providing a motherboard substrate 304, at step 964, having a bottom surface and a top surface, and a plurality of electrical conductors extending from the top surface to the bottom surface; providing at least one intermediate connector 306, and step 966, having an upper interface and a lower interface, the upper surface locatable proximate to the bottom surface of the motherboard substrate 304, the intermediate connector 306 comprising at least one electrically conductive connection between the upper interface and the lower interface corresponding to each of the electrical conductors on the bottom surface of the motherboard substrate 304; and providing a probe chip substrate design, at step 968, comprising a connector surface 62b, a probe surface 62a opposite the connector surface 62b, and a plurality of contacts on the connector surface arranged in a fixed layout, the connector surface 62b locatable proximate to the lower surface of the intermediate connector 306.

The quick turn process 960 shown in Figure 83 then includes the receiving of an interconnection specification 972 for the at least one device 44 on the wafer 92, in which the interconnection specification 972 comprises interconnection locations for the device 44.

The probe chip substrate 310 is then produced, at step 974, based on both the standardized information 968 and the received, i.e. customized, information 970, wherein the probe chip substrate comprises a plurality of spring probe contact tips 412, such as seen in Figure 40 on the probe surface 62a, corresponding to the interconnection locations 972 on the wafer 92, and wherein each of the spring probe contact tips is electrically connected to at least one contact on the connector surface 62a.

System Advantages. As described above, the probe card assemblies 60, enhanced probe card assemblies 300, interposer structures 520,520b, and high performance packages 902 provide several advantages over conventional probe and package technologies.

For example, many embodiments of the probe card assemblies 60, enhanced probe card assemblies 300, interposer structures 520,520b, and high performance packages 902 comprise photolithography-patterned springs 61,412, 521, and are typically formed by stress metal film batch processing, which is inherently lower in cost then either mechanical or MEMS based processing. The springs are formed through the use of two dimensional processing methods, through which three dimensional structures are formed.

In contrast, other conventional spring processes require additional processing steps to create three dimensional springs. In alternate processing embodiments, the springs are formed through the use of two dimensional plating processing methods to form a differential stress gradient between plating layers, through which three dimensional structures are formed.

The assembly techniques described in this document use components or substrates comprising stress metal probe elements, i.e. springs, which are all batch-fabricated on each of the substrates by IC processing techniques including photolithography. As a result the process allows fabrication of probe card assemblies and packages using arrays of very short, e.g. 100-200 micron long and 10-20 micron wide, probe elements with ultra-small pitch, e.g. 10-50 micron, that is not manufacturable by any currently used technology. It should also be understood that the application of the assembly and packaging techniques disclosed in this document are also applicable to substrates or components comprising arrays of probe elements with larger dimensions and greater pitch that can be fabricated using the technology presented in this document or by any other method available today.

Furthermore, the probe architecture typically comprises a substrate having through holes for direct connection to the other side, in which photolithographic methods define both the placement of the springs and the route connections from the holes to the springs, which enables the use of simpler starting material and shorter processing times.

As well, the disclosed probe and package architectures 60,300, 902 preferably comprise reusable components, i.e. standards, through the use of standardized master slices 962, such that the majority of a probe card 60,300 and/or or package assembly 902 can be implemented for a variety of connected devices 44 and wafers 92. The master slice 962 is readily matched to a particular interconnection specification 972, thereby reducing the"cost of ownership"needs for a particular customer.

As described above, the probe card assemblies 60, enhanced probe card assemblies 300, interposer structures 520,520b, and high performance packages 902 comprise one or more substrates, which can be built using a variety of starting materials, such as ceramic, glass or quartz, silicon, organic board, and/or multi-layer ceramic, depending on the requirements. The assembly substrate may further comprise multi-layer metal layers, such as to improve routing and performance.

Furthermore, the probe card assemblies 60, enhanced probe card assemblies 300, interposer structures 520,520b, and high performance packages 902 may further comprise stand offs on any of the substrates, such as to protect the probes 61,412, protect the devices 44 and/or wafer 92, or to control the amount of pad scrub in a clamped wafer level cartridge or cassette application. The stand offs can also be applied to connector side 62b of the Probe chip 16,310, such as to protect springs that interface from an interposer 520,520b.

Although the disclosed probe card assembly systems and improved non-planar spring probes and methods for production are described herein in connection with integrated circuit test probes, probe cards, and/or packages, the system and techniques can be implemented with other devices, such as interconnections between integrated circuits and substrates within electronic components or devices, bum-in devices and MEMS devices, or any combination thereof, as desired.

In addition, it is within the scope of this invention that assembly structures and methods disclosed in this document can be used to test, bum-in or package various miniaturized integrated solid state circuits, including both silicon and III-V semiconductor devices, as well as liquid crystal display panels, solid state sensor arrays, such as biosensors, environmental sensors and surface acoustic wave device sensors.

Accordingly, although the invention has been described in detail with reference to a particular preferred embodiment, persons possessing ordinary skill in the art to which this invention pertains will appreciate that various modifications and enhancements may be made without departing from the spirit and scope of the claims that follow.

「特表2006-507479およびWO2004001807より引用」

|
|

[Claims] 微小電子機械誘導性スイッチ

【特許請求の範囲】
【請求項1】
少なくとも1つの可動インダクタと、
固定された基板上の少なくとも1つの固定されたインダクタとを有し、前記少なくとも1つの固定されたインダクタに対する前記少なくとも1つの可動インダクタの相対的移動により、前記少なくとも1つの固定インダクタに対して前記少なくとも1つの可動インダクタを電気的に結合及び減結合する、
半導体微小電子機械システム(MEMS)スイッチ。
【請求項2】
前記少なくとも1つの可動インダクタに結合された駆動手段をさらに有する、請求項1に記載のMEMSスイッチ。
【請求項3】
前記少なくとも1つの可動インダクタおよび前記少なくとも1つの固定インダクタがそれぞれ第1および第2の平面または基板上に配置される、請求項1に記載のMEMSスイッチ。
【請求項4】
前記第1の平面が、前記第2の平面の上または下の可動プラットフォームである、請求項3に記載のMEMSスイッチ。
【請求項5】
前記駆動手段が前記可動プラットフォームの外部にあり、前記駆動手段が回転運動を提供する、請求項4に記載のMEMSスイッチ。
【請求項6】
そのそれぞれの端部で互いに相互接続された前記可動プラットフォーム上の第2および第3のインダクタをさらに有する、請求項4に記載のMEMSスイッチ。
【請求項7】
互いに接近してグループ化された複数のインダクタの配置と、前記複数のインダクタの配置とは別の少なくとも1つのインダクタと、
回転プラットフォーム上で互いに相互接続された2つのインダクタとを有し、前記回転プラットフォーム上の前記インダクタのうちの一方が、回転プラットフォームの回転により前記複数のインダクタの配置内の前記インダクタの1つと交互に結合及び減結合し、前記回転プラットフォーム上の他方のインダクタが固定基板上の前記複数のインダクタの配置とは別の前記少なくとも1つのインダクタと結合及び減結合する、
固定基板上のMEMSスイッチ。
【請求項8】
前記回転プラットフォームがその一方の端部でピボット・ピンに取付けられる、請求項7に記載のMEMSスイッチ。
【請求項9】
前記回転プラットフォームが支点に取付けられ、前記支点がその一方の端部で前記ピボット・ピンの周りを回転する、請求項7に記載のMEMSスイッチ。
【請求項10】
固定基板上の複数のインダクタより成る第1および第2の配置と、
2対のインダクタであって、前記2対がそれぞれ2つの可動プラットフォーム上に配置され、前記プラットフォームが支点のそれぞれの端部で互いに結合され、前記支点が前記可動プラットフォームの回転運動を可能にするピボット・ピンに接続される2対のインダクタとを有し、前記回転運動により、前記固定基板上の前記複数のインダクタの第1および第2の配置のそれぞれの対に対して前記可動プラットフォーム上の前記インダクタのそれぞれの対を結合及び減結合する、
MEMSスイッチ。
【請求項11】
固定基板上の複数のインダクタの配置と、
ピボット・ピンの周りを回転する可動プラットフォームであって、前記可動プラットフォームがその上に2つのインダクタを有し、前記可動プラットフォームにより、その上の前記インダクタのそれぞれが前記固定基板上の前記配置のインダクタと交互に結合及び減結合することができる、可動プラットフォームと、
を有する、MEMSスイッチ。
【請求項12】
固定基板上の複数のインダクタと、
可動プラットフォーム上の少なくとも1つのインダクタであって、前記可動プラットフォームが、前記固定基板に対する前記可動プラットフォームの横変位を可能にするよう駆動手段に結合され、前記可動プラットフォームが前記固定基板上の前記複数のインダクタに対して前記少なくとも1つのインダクタを結合及び減結合する、可動プラットフォーム上の少なくとも1つのインダクタと、
を有する、半導体多極誘導性MEMSスイッチ。
【請求項13】
前記インダクタが、単一閉ループまたは複数ループからなるらせんのうちの一方を形成する、請求項13に記載の半導体多極誘導性MEMSスイッチ。
【請求項14】
固定プラットフォーム上の少なくとも1つのインダクタと、
2対のインダクタであって、前記2対のそれぞれがそれぞれ1つの回転プラットフォーム上に配置され、前記回転プラットフォームのそれぞれがピボット・ピンに接続された端部を有する2対のインダクタとを有し、前記2つのプラットフォームの回転運動により、前記固定プラットフォーム上の前記少なくとも1つのインダクタに対して前記複数対のインダクタを交互に結合及び減結合する、
半導体誘導性MEMSスイッチ。
【請求項15】
固定基板上の複数のインダクタの配置と、
回転プラットフォーム上の複数の相互接続インダクタとを有し、前記回転プラットフォームが前記固定基板上の前記複数のインダクタの配置に対して前記複数の相互接続インダクタを結合及び減結合する、
半導体誘導性MEMSスイッチ。
【請求項16】
前記回転プラットフォームが前記回転プラットフォームの重心上またはその付近のピボット・ピンに接続され、前記ピボット・ピンが前記固定基板上に固定される、請求項16に記載の半導体誘導性MEMSスイッチ。
【請求項17】
固定基板上の複数のインダクタと、
その上に少なくとも1つのインダクタを有する回転プラットフォームであって、前記回転プラットフォームがその一方の端部に接続されたピボット・ピンの周りを回転し、前記ピボット・ピンが前記固定基板上に固定され、前記回転プラットフォームが前記固定基板上の前記インダクタに対して前記複数のインダクタを交互に結合及び減結合する回転プラットフォームと、
を有する、半導体MEMSスイッチ/スイッチの組合せ装置。

1. A semiconductor micro-electromechanical (MEM) switch comprising: at least one moveable inductor; and at least one inductor on a stationary substrate, wherein the relative movement of said at least one moveable inductor with respect to said at least one stationary inductor electrically couples and decouples said at least one moveable inductor to and from said at least one stationary inductor.

2. The MEM switch as recited in claim 1 further comprising driving means coupled to said at least one moveable inductor.

3. The MEM switch as recited in claim 1 wherein said at least one moveable inductor and said at least one stationary inductor are respectively placed on a first and second planes or substrates.

4. The MEM switch as recited in claim 3, wherein said first plane is a moveable platform above or below said second plane.

5. The MEM switch as recited in claim 4, wherein said driving means are external to said moveable platform, said driving means providing a rotational movement.

6. The MEM switch as recited in claim 4, further comprising a second and third inductors on said moveable platform interconnected to each other at respective ends thereof.

7. A MEM switch on a stationary substrate comprising: an arrangement of inductors grouped in close proximity of each other, and at least one inductor separate from said arrangement of inductors; and two inductors interconnected to each other on a rotating platform, wherein one of said inductors on said rotating platform alternatively couples and decouples one of said inductors in said arrangement of inductors by the rotation of rotating platform, while the second inductor on said rotating platform couples and decouples said at least one inductor separate from said arrangement of inductors on said stationary substrate.

8. The MEM switch as recited in claim 7 wherein said rotating platform is attached to a pivot pin at one end thereof.

9. The MEM switch as recited in claim 7 wherein said rotating platform is attached to a fulcrum, said fulcrum rotating about said pivot pin at one end thereof.

10. A MEM switch comprising: a first and second arrangement of inductors on a stationary substrate; and two pairs of inductors, said two pairs respectively placed on two moveable platforms, said moveable platforms being coupled to each other at respective ends of a fulcrum, said fulcrum being attached to a pivot pin allowing a rotating movement of said moveable platforms, wherein said rotational movement couples and decouples respective pairs of said inductors on said moveable platforms to and from respective pairs of said first and second arrangement of inductors on said stationary substrate.

11. A MEM switch comprising: an arrangement of inductors on a stationary substrate; and a moveable platform rotating about a pivot pin, said moveable platform having two inductors thereon, said moveable platform allowing each of said inductors thereon alternatively couple and decouple inductors of said arrangement on said stationary substrate.

12. A semiconductor multi-pole inductive MEM switch comprising: a plurality of inductors on a stationary substrate; at least one inductor on a moveable platform, said moving platform being coupled to driving means providing a lateral displacement of said moveable platform with respect to said stationary substrate, said moveable platform coupling and decoupling said at least one inductor to said plurality of inductors on said stationary substrate.

13. The semiconductor multi-pole inductive MEM switch as recited in claim 13, wherein said inductors form either a single closed loop or a spiral of consisting of multiple loops.

14. A semiconductor inductive MEM switch comprising: at least one inductor on a stationary platform; two pairs of inductors, each of said two pair being respectively placed on one rotating platform, each of said rotating platforms having an end thereof attached to a pivot pin, wherein the rotational movement of said two platforms alternatively couples and decoupled said pairs of inductors to and from said at least one inductor on a stationary platform.

15. A semiconductor inductive MEM switch comprising: an arrangement of inductors on a stationary substrate; a plurality of interconnected inductors on a rotating platform, wherein said rotating platform couples and decouples said plurality of interconnected inductors to and from said arrangement of inductors on said stationary substrate.

16. The semiconductor inductive MEM switch as recited in claim 16 wherein said rotating platform is attached to a pivot pin on or about the center of gravity of said rotating platform, said pivot pin being anchored on said stationary substrate.

17. A semiconductor MEM switch-switch combination comprising: a plurality of inductors on a stationary substrate; a rotating platform having at least one inductor thereon, said rotating platform rotating around a pivot pin attached at an end thereof, said pivot pin being anchored on said stationary substrate, said rotating platform alternatively coupling and decoupling said plurality of inductors to and from said inductor on said stationary substrate.

「特表2006-514414およびWO2004078638より引用」

|
|

微小電子機械誘導性スイッチ

【発明の詳細な説明】
【技術分野】
【0001】
本発明は一般に、微小電子機械システム(MEMS:micro-electromechanicalsystem)、特にMEMSスイッチに関し、より具体的には、誘導結合および減結合を使用し、標準的なCMOS製造材料およびプロセスと完全に互換性のある誘導性MEMSスイッチに関する。
【背景技術】
【0002】
スイッチング動作は、多くの電気的、機械的、および電子機械的応用例の基本部分である。MEMSスイッチは、この数年にわたり、相当な関心を引きつけてきた。MEMSスイッチ技術を使用する製品は、バイオメディカル・システム、航空宇宙システム、および通信システムで普及している。
【0003】
MEMSスイッチは様々な構成を使用して製造されてきたが、それらは、容量性スイッチを形成するために誘電体ストップを使用する金属同士の接触または同様の構造を作成する静電制御されたビーム(梁)である。このデバイスを特徴付ける一般的な特徴は、回路を完成するために他の可動エレメントに接触する少なくとも1つの可動エレメントが設けられていることである。
【0004】
本発明をより十分に理解するために、次に、変形可能なビーム5の両端が誘電体2上に固定されているMEMSスイッチの断面図を示す図1に関連して従来のMEMSスイッチについて説明する。最低レベルは、デバイスの様々な電気コンポーネントを接続または形成するために使用される導電エレメント3および4からなる誘電材料1から構成される。番号3および6によって参照される導体は、ビームを曲げさせる(または変形させる)動作電圧電位を提供するために使用される。導体4は、電気信号を伝導するものであり、MEMSスイッチが動作しているときに変形可能なビームに接触する。図2は、同じ従来のスイッチの平面図を示している。
【0005】
従来のMEMSスイッチの典型的な実現例では、接点ビームは、たとえば、SiO2で作られた誘電体の上にポリシリコンを付着させることによって形成される。周囲の材料はエッチングで除去され、シリコン・ビーム5に接続された隆起構造を残す。シリコン・ビーム5上の一方の端部に固定された接触エレメント6は、導体3および4の上のそのもう一方の端部で宙に浮いており、好ましくはポリシリコンで作られている。その後、デバイスは、通常、金の無電解めっきが施され、これはポリシリコンに付着して、導電エレメント3、4、および6の製作を完了する。
【0006】
このスイッチは、接点ビーム6と電極3との間に電位差を提供することによって操作される。この電圧は、ビーム6を電極4に接触させるような静電引力を発生し、その結果、スイッチを閉じる。固定されたビーム5に付与されたねじれは、制御電圧電位が降下したときに、接点ビーム6をその開位置に戻すために使用される。
【0007】
一般に、すべての従来のMEMSスイッチは、スイッチング動作を実行するために、物理的接触、特に金属同士の接触に依存する。このため、アークの発生、物質移動、微細溶接、静摩擦などに関連する多くの信頼性問題が発生する。これらのスイッチの大部分が高周波数で信頼性が低下することは当技術分野では周知のことである。金など、使用される冶金の一部は、これらの問題を緩和しようとして一般に使用されるものであるが、標準的なCMOS製作と互換性のあるものではない。以下に説明する本発明の誘導性MEMSスイッチは、任意の数の周知のMEMSアクチュエータによって操作するのに向いている。
【0008】
MEMSアクチュエータの例は、Sandia NationalLaboratoryのWebサイト(www.sandia.gov)か、または2001年12月11日に発行されたGeorge E.Vernon, Sr.による「Surface-Micromachined Chain for Use in Micro-MechanicalStructures」という米国特許第6328903号など、アクチュエータに関連するいくつかのMEMS特許で見つけることができる。以下に説明するくし形駆動システムを特に対象とするその他の特許は、たとえば、1999年12月7日に発行されたJerman他による「Electrostatic Microactuator and Method for UseThereof」という米国特許第5998906号で見つかるはずである。
【0009】
たとえば、2000年6月13日に発行されたYao他による「Method of Fabricating Suspended Single Crystal Silicon MEMS Devices」という米国特許第6074890号に記載され、2001年12月に発行されたIEEE Microwaveにさらに記載されている従来のMEMSスイッチでは、概して、スイッチング回路内の少なくとも1つの電極には、静電作動の一部としてDC電位が印加されている。したがって、静電作動を実行するために、少なくとも1つの接点にDC制御電圧がまったく印加されないように、スイッチング回路から駆動システムを分離する明確な必要性が存在する。
【特許文献1】米国特許第6328903号
【特許文献2】米国特許第5998906号
【特許文献3】米国特許第6074890号
【特許文献4】米国特許第6465929号
【特許文献5】米国特許第6404599号
【特許文献6】米国特許第6305779号
【非特許文献1】2001年12月に発行されたIEEE Microwave
【発明の開示】
【発明が解決しようとする課題】
【0010】
したがって、本発明の第一の目的は、電気信号の誘導結合および減結合に基づく誘導性MEMSスイッチを提供することにある。
【0011】
他の目的は、デバイスを操作するために使用される制御回路からスイッチされた信号の経路を分離することにより、スイッチされた信号から制御信号を隔離する、誘導性MEMSスイッチを提供することにある。
【0012】
さらに他の目的は、オフ状態で従来のスイッチをしのぎ、概して6GHzで約50dBの隔離を行うように制限される、オフ状態隔離を有する誘導性MEMSスイッチを提供することにある。
【0013】
他の一目的は、様々な多極多連配置(multi-pole, multi-throw arrangement)として構成可能であり、任意の数のMEMS直線または回転駆動システムによって制御される、誘導性MEMSスイッチを提供することにある。
【0014】
さらに他の目的は、「ホットスイッチング(hot-switching)」、すなわち、公称電力レベルで動作している間のスイッチングを確実に実行可能である、誘導性MEMSスイッチを提供することにある。スイッチングは非接触であり、したがって、接点のアーキングまたは溶接がまったく発生しないので、スイッチングは1ワット、5ワット或いは回路の残りの部分が対処できる値までで実施することができる。
【0015】
さらに他の目的は、潜在的にアーキング、溶接または物質移動、および劣化を発生する可能性のある信号経路内のDC電位または物理的接触点がまったくない状態で確実に動作する、誘導性MEMSスイッチを提供することにある。
【0016】
さらに他の一目的は、高周波数でその効率を増進し、信号の周波数が増加したときにコイルのサイズを低減することができ、効率の増進がスイッチ・コンポーネント間の磁界結合によって達成され、したがって、対応する隔離性能の低下なしに高周波数でより良好な挿入損特性を提供する、誘導性MEMSスイッチを提供することにある。
【0017】
本発明の他の目的は、インピーダンス整合を達成するためのスイッチ/変圧器の組合せを提供することにある。誘導性スイッチの各部分のインダクタンスを適切に選択することにより、スイッチの入出力インピーダンスを独立して調整することができる。この調整は、インピーダンス整合とスイッチングを同時に可能にするものである。変圧器の特殊構成を使用してシングルエンド/ダブルエンド変換器(single-ended to double-ended converter)またはバラン(balun)(平衡不平衡(BAlanced-UNbalanced)を作成し、スイッチングと信号変換の両方を単一デバイスで行うことができる。
【0018】
さらに他の一目的は、CMOS互換プロセスおよび材料を使用して製造可能な誘導性MEMSスイッチを提供することにある。
【課題を解決するための手段】
【0019】
本発明の一態様では、信号のスイッチングは固定コイルと可動コイルとの間の誘導結合および減結合によって実施される。可動コイルが固定コイルに対して位置合せされるかまたは位置合せされないときに、スイッチングが行われる。
【0020】
金属厚が4μm、巻き幅(turn width)が10μm、外径が150μmであり、スイッチの1つのエレメントととして構成された4回巻きらせんインダクタは、すぐ上またはすぐ下の他の同様のらせんに磁気結合され、約0.85の結合係数をもたらす。これらのらせんが上述の通りに構成されると、13GHzで6.6dBの閉スイッチ挿入損および65dBの開スイッチ隔離が達成される。これにより、スイッチの2つのポート間に外部同調コンデンサを追加することにより、13GHz以下の周波数に同調された優れたオン/オフ・スイッチ比が得られる。同様に、金属厚が4μm、巻き幅が10μm、外径が150μmであり、スイッチの1つのエレメントととして構成された1回半巻きらせんインダクタは、すぐ上またはすぐ下の他の同様のらせんに磁気結合され、約0.85の結合係数をもたらす。これらのらせんがこのように構成されると、25GHzで10dBの閉スイッチ挿入損および60dBの開スイッチ隔離が達成される。これにより、スイッチの2つのポート間に外部同調コンデンサを追加することにより、25GHz以下の周波数に同調された優れたオン/オフ・スイッチ比が得られる。
【0021】
本発明の他の態様では、このMEMSスイッチは、スイッチ接点の静摩擦、アーキング、および溶接として知られる問題を解決し、そのいずれもスイッチング・エレメント間の物理的接触がないために除去される。誘導結合により一方のコイルともう一方のコイルとの間で信号を転送できるように、コイル同士はきわめて接近して単純に位置合せされる。この特性を考慮して、MEMSスイッチは、全動力でのスイッチング(ホットスイッチング)と、明らかに従来のMEMSスイッチより多くの電力を容易に処理することができる。
【0022】
固定コイルまたは可動コイルの数を変化させることにより、あるいはコイルのコイル幾何構成および対応する可動エレメントの変位を変更することにより、またはその両方により、複数のスイッチ構成が実現される。さらに、駆動回路はスイッチング回路から完全に独立しているので、制御信号およびスイッチされた信号の経路からの完全隔離が可能である。
【0023】
本発明の上記その他の目的、態様、および利点は、添付図面に併せて読んだときに、本発明の好ましい実施形態に関する以下の詳細な説明から明らかになるであろう。
【発明を実施するための最良の形態】
【0024】
次に、好ましい諸実施形態が示されている図面に関連して、本発明についてより詳細に説明する。
【0025】
図3は、その最も単純な形で本発明を図示する、第1の実施形態の概略図である。可動コイル・アセンブリ10は、ピボット・ピン70の周りを回転する可動インダクタ20および30の基板、プラットフォーム、またはモジュール15から構成され、プラットフォーム15の上または下の第2の基板上に位置決めされた固定コイル40および50に対して可動コイル20および30を誘導結合または減結合する。他の駆動システムも同様に効果的に使用できるが、くし形駆動部8および9は、ある動作モードを示すためにアセンブリに駆動機能を提供する。固定コイルに対する可動コイルの相対位置をより十分に図示するために、後続の図では、両方の位置に位置決めされたデバイスを示す。
【0026】
2つのインダクタ20および30は、一方のコイルで誘導された電流がもう一方を流れるように、導体25および35により回路を閉鎖するように接続される。ピボット・ピン70は、ホール75(図6により詳細に図示する)を通る。インダクタ40および50は、基礎となる基板7(図6に図示する)上で位置決めされ、電力増幅器、受信機、アンテナなど、スイッチング・メカニズムを必要とする他の回路に接続される。
【0027】
それにより回転が達成されるメカニズムは、この実施形態の要素ではない。当業者であれば、必要な運動をデバイスに提供するために任意の数のMEMSスイッチに対処可能であることを容易に認識することになる。このようなデバイスの一例は、たとえば、単純なくし形駆動部を製作する方法を記載したYao他による「Method of Fabricating Suspended Single Crystal Silicon MEMS devices」という米国特許第6074890号、またはLevitan他による「Micro-electromechanical systemactuator for extended linearmotion」という米国特許第6465929号で見つかるが、どちらの特許も参照により本明細書に組み込まれる。
【0028】
図4は、可動コイル30が固定コイル50に結合される間に、コイル同士が誘導結合される、すなわち、可動インダクタ20が固定インダクタ40に誘導結合される位置まで回転された、図3のMEMSスイッチを図示している。このように、コイル40に注入された信号は、コイル20に誘導的に転送することができる。電気信号はコイル30に転送され、次にコイル30がそれをコイル50に誘導結合する。したがって、たとえば、送信機の電力増幅器からの電気信号はコイル40に印加され、その信号はデバイスを通ってコイル50まで流れ、次にコイル50は、たとえば、アンテナに接続することができる。追加の機能を達成するために、追加のコイルがアセンブリの一部になりうることに留意されたい。これについては、図9~12、図17、および図22~24に関連して以下に例証する。
【0029】
図5は、コイル20および30と、それらの内側および外側のコイル接続部35および25をそれぞれ図示する、アセンブリ10の平面図である。また、図6に関連して以下に使用することになる区切り文字(delimiter)A-Aも示されている。
【0030】
図6は、線A-Aによって取られた図5の断面図である。同図には、固定コイルと、おそらく他の関連回路(明瞭にするために同図には図示せず)とを含む、下部基板7が図示されている。ピボット・ピン70を取り囲む肩部80は基板7上に構築される。この肩部は、デバイス10が移動できるようにするために必要なクリアランスを提供する。肩部の高さは、このクリアランスと、ある程度まで、コイルを結合する効率のレベルを決定する。肩部は、特定の応用例に応じて、約1000Å~2μm以上の厚さである。ピボット・ピン70は肩部の領域内に含まれ、そのサイズはデバイス10が束縛または著しい揺れを発生せずにそこで自由に動ける能力によって決定される。ピボットの直径は、必要な機械的信頼性を提供するのに十分なものでなければならず、それに応じてスケーリングされる。また、その直径は、材料の選択およびプロセス・ケイパビリティにも左右される。同様に、デバイス10内のホール75は、束縛なしに所望の範囲の運動が発生するようにピボット・ピンを収容するように設計される。ピボットの高さは、デバイス10の厚さに適合するようにスケーリングされる。一例として、デバイス10が3μmの厚さである場合、ピボットは、デバイスを確実に保持するために、厚さの相当な部分を占めることになる。優先的に、デバイスを密封し保持するために上部肩部85(図18を参照)にピボットが接触するように、ピボットはデバイス10の厚さよりわずかに高くなるように作られる。
【0031】
図7は、図8でさらに使用するために、区切り文字B-Bによって定義された異なる位置から見たデバイス10の可動部分の平面図である。
【0032】
図8は、区切り文字B-Bから見た図7の断面透視図である。これは、真のらせんインダクタを形成するために必要なマルチレベル構造を製作するために使用されるコイルとスタッド60との間の上部接続部35を具体的に図示している。このように、コイル20および30の内側端部は相互接続することができる。外側接続部25は、コイルと同じ層に構築され、したがって、スタッドを必要としない。この配置は唯一可能な配置ではないことに留意されたい。以下に記載する図20および図21は、1つのレベルの配線のみを使用して構築されるコイルを図示することになる。
【0033】
図9は、ピボット70がデバイス10の可動部分の一方の端部に向かって移動し、追加の固定コイル42および45がコイル40および50のそばに組み込まれている、本発明の第2の実施形態を図示している。これは、多連配置を達成する1つの方法を図示している。図示の配置では、出力コイル30を固定コイル42、45、および50のいずれかに結合することができる。可動コイルへの入力は、固定コイル40をコイル20に結合することによって提供される。ピボット・ピン70はコイル20の中心に位置するので、くし形駆動部13がかみ合い、デバイスが反時計回りに回転し、コイル42および30が重なり合って、誘導結合状態になっている図10に示した通り、デバイス10が回転したときに40に誘導結合されたままになる。
【0034】
図11は、可動コイル20および30が支持部(fulcrum)11の一方の端部に構築される、本発明のさらに他の実施形態を示している。これらのコイルは、移動すると、固定コイル40、42、および50に対して2つ一組になって結合または減結合する。図11は、固定コイル42および40に結合するための位置にあるデバイス10を示している。図12は、コイル42が減結合され、したがって、コイル40および50が20および30とそれぞれ結合できるように、くし形駆動部13によって時計回りに回転された後のデバイス10を図示している。
【0035】
図13は、多極多連スイッチを形成するために、図11の実施形態が複数の可動コイル・セットを含むように拡張されている、本発明の他の実現例を図示している。図11のコイル配置を複製するために図11に示した構成に対して左右対称の構成を使用するが、その構成では、固定コイル42A、40A、および50Aがそれぞれ可動コイル20Aおよび30Aに結合され、その可動コイルが支持部11に接続され、ピボット・ピン70の周りを回転する。図13は、コイル20および30がそれぞれ42および40に誘導結合されるようなデバイスを図示している。同時に、コイル20Aおよび30Aはそれぞれコイル42Aおよび40Aに結合される。図14では、同じデバイスが、駆動部13によって時計回りに回転した後の相補的位置に示されている。この位置では、コイル20および30はそれぞれ40および50に結合された状態で示され、コイル20Aおよび30Aはそれぞれ40Aおよび50Aに結合されている。
【0036】
図15は、マルチモード・スイッチングを可能にするために可動コイル・デバイス10が回転駆動部(図示せず)によって作動される、本発明のさらに他の実施形態を示している。この応用例は、たとえば、携帯電話での帯域スイッチングの場合に特に有利である。駆動部は、当技術分野で知られており、たとえば、Veganによる「High Performance Integratedmicro-actuator」という米国特許第6404599号に完全に記載されている。図15は、コイル20および30がそれぞれ固定コイル42および42Aに結合されたデバイス10を示している。図16は、コイル20および30がそれぞれ40および40Aに結合されるように時計回りに回転したときのデバイス10を示している。この時計回りの回転により、20および30がそれぞれ50および50Aに結合されるように、デバイスはさらに移動する。回転は他の組合せのために時計回り方向にさらに継続するか、この時点で逆転してそれぞれの固定コイルとの可動コイル20および30の上記の結合および減結合を繰り返すこともできる。この回転運動は、当技術分野で現在知られている任意の数の手段により、デバイス10に付与することができる。
【0037】
図17は、デバイス10がレール16またはトレンチ・ガイドによって拘束され、直線駆動部を介して、または図示した通り、ラック17およびピニオン18を使用して、横方向に移動する、本発明のさらに他の実施形態を図示している。図示したデバイスは、唯一の固定コイル42と、2つの可動コイルとを含む。固定コイル42に隣接してデバイス10の下(または上)に他のコイル(図19の40Aおよび50Aと同様のもの)を組み込むこともできるが、明瞭にするためにそのいずれもこの図には示されていない。デバイス10は、前後に移動すると、デバイスの下(または上)に位置決めされた様々な固定コイルに対して結合及び減結合する。ラックアンドピニオン式のMEMSデバイスの一例は、Capurso他による「MEMS ink-jet nozzle cleaning and closing mechanism」という米国特許第6305779号に記載されているが、同特許は参照により本明細書に組み込まれる。
【0038】
図18は、信頼性のために完全にカプセル化された完成デバイスの断面図である。下部固定コイル40および50は、下部誘電体層7に構築される。さらに、下部肩部80も同じ層7上に構築され、同一平面上の誘電体層90と同じ材料で製作される。キャビティ12は誘電体層90内に形成され、デバイス10を移動するために必要な空間を提供する。上部誘電体層100は、デバイス10をすっぽり包み、ピボット・ピン70に接触する上部肩部85により示す通り、その構造に対して追加の機械的支持を提供する。
【0039】
図19は、同図が上部誘電体100に追加のコイル40Aおよび50Aを組み込むことにより追加の機能を達成する方法を図示することを除いて、図18と同様のものである。
【0040】
図20は、スタッド(図8の60など)の必要性を除去する、より単純な1レベルの配線方式のデバイス10の平面図である。この実施形態は、デバイス10の構造を単純化するが、らせんインダクタを可能にするものではなく、したがって、特定の応用例では用途が限られている可能性がある。区切り文字C-Cは、図21に示した断面図に関する参照を示す。
【0041】
図21は、図20の区切り文字C-Cにより示された構造の断面図を表す図である。この図は、らせんインダクタに必要な多層配線を図示する前述の図8に匹敵するものである。
【0042】
図22は、固定コイル30に対して複数の可動デバイス10および10Aを使用する、本発明のさらに他の実施形態である。これらのデバイスは、3つ以上のインダクタが同時に垂直結合する際に図23に示したように移動するように、構造の異なるレベルに形成することができる。図22は減結合状態の可動コイル10および10Aを示し、図23は結合状態の図22のデバイスを図示している。
【0043】
図24は、3つ以上のコイルと、ピボット・ピン70の周りを回転する三角形の基板とを含む、誘導性スイッチの構造を図示している。この配置では、コイル20、21、および30は、それぞれコイル56、41、および51に結合された状態で示されている。図示したミスアライメントは例示のみのためのものである。デバイスが時計回りに回転した場合、コイル20、21、および30はそれぞれコイル40、50、および55と位置合せされた状態になる。
【0044】
図25は、スイッチ/バランの組合せを提供するために単一可動コイルが使用される、減結合状態の誘導性MEMSスイッチを図示している。前述の通り、インピーダンス整合およびバラン機能ならびにスイッチングは、誘導性スイッチ内の適切な位置に異なるインダクタンス値を提供することによって使用可能になる。
【0045】
図26は、結合状態にある、図25と同じデバイスを図示している。可動コイル/バラン20Aは、インダクタ40および50の両方と同時に結合する。
【0046】
図27は、固定コイルおよび可動コイルが基板10の平面に対して垂直または直角に構築された、直線配置の平面図である。
【0047】
図28は、図27の可動コイル配置10の断面図を示し、可動コイルが可動エレメント10内にどのように構築されるかを図示している。図29は、垂直コイルの構造をより詳細に示している。この構造は、層Aがインダクタ・ループの下部部分を形成する、複数の層として製作される。このプロセスは、標準的なダマスク線(damascene line)レベルと同様のものである。層Bは、バイアおよび相互接続線を提供する二重ダマスク構造(dualdamascene structure)と同様に、ループ内の外側および内側垂直導体ならびに水平内側導体の一部分を含む。層Cは、この場合も二重ダマスク構造のように、コイル(複数も可)を完成する垂直ループの上部部分と上部水平導体とを含む。層Dは、金属をカプセル化し保護するという目的にもかなう、上部絶縁層である。
【0048】
図30は、図27に図示したものと同様の、より複雑な可動コイル配置のもう1つの平面図であり、図31は、前述の水平2次元コイルの代わりに垂直3次元構造のコイル20および30を備えた、図30に示した構造の断面図表現である。図32は、図30および図31に図示した3次元コイル配置の透視図を示している。この配置は、ダマスクまたは二重ダマスク処理で通常行われるようにコイルの様々なセグメントが層として構築される、図28に関して記載したものと同様に構築される。
【0049】
本発明の誘導性MEMSスイッチは、高周波数で効率の増進を示し、信号の周波数が増加したときにコイルのサイズを低減することができる。効率の増進はスイッチング・コンポーネント間の磁界結合によって達成される。この磁界結合は、対応する隔離性能の低下なしに高周波数でより良好な挿入損特性を提供する。この高周波数で効率の良い動作は、スイッチ開状態のときにその隔離性能が低下するので典型的な金属同士の接触および容量性スイッチの性能がますます低下することと対比される。典型的な金属接触スイッチは2~3GHzのみを確実に処理できるが、本発明は25GHz以上を容易に処理することができる。(高周波数で効率の良い動作は、スイッチ開状態のときに隔離性能が低下するので典型的な金属接触および容量性スイッチの性能がますます低下することと対比される。)
【0050】
本発明のもう1つの利点は、組込みインピーダンス整合のためにスイッチ/変圧器の組合せを構築できることにある。誘導性スイッチの各部分のインダクタンスを適切に選択することにより、スイッチの入出力インピーダンスを独立して調整することができる。この調整は、インピーダンス整合と同期スイッチングを可能にするものである。変圧器の特殊構成を使用してシングルエンド/ダブルエンド変換器またはバランを作成し、スイッチングと信号変換の両方を単一デバイスで行うことができる。
【0051】
好ましい実施形態に関連して本発明を説明してきたが、上記の説明を考慮して多くの代替例、変更例、および変形例が当業者には明らかになることを理解されたい。したがって、特許請求の範囲の精神および範囲に含まれるこのような代替例、変更例、および変形例をすべて包含することが意図されている。本明細書に示したものまたは添付図面に示したものはいずれも、例示的かつ非制限的な意味で解釈すべきものである。

MICRO-ELECTROMECHANICAL INDUCTIVE SWITCH

BACKGROUND OF THE INVENTION

This invention is generally related to micro-electromechanical systems (MEMS) especially MEMS switches, and more specifically, to an inductive MEMS switch utilizing inductive coupling and decoupling, and which is fully compatible with standard CMOS manufacturing materials and processes.

Switching operations are a fundamental part of many electrical, mechanical and electromechanical applications. MEMS switches have drawn considerable interest over the last few years. Products using MEMS switch technology are widespread in biomedical, aerospace and communications systems.

While MEMS switches have been manufactured using various configurations, they are electrostatically controlled beams that make metal-to-metal contact or a similar structure that uses a dielectric stop to form a capacitive switch. A common feature that characterizes the device is that it is provided with at least one moving element contacting another to complete the circuit.

In order to better understand the present invention, a conventional MEMS switch will now be described with reference to Fig. 1, showing a cross-section view of a MEMS switch having both ends of a deformable beam 5 anchored on dielectric 2. The lowest level consists of a dielectric material 1 consisting of conductive elements 3 and 4 which are used to connect or form the various electrical components of the device. Conductors referenced by numerals 3 and 6 are used to provide an operating voltage potential that causes the beam to bend (or deform). Conductor 4, which conducts an electrical signal, contacts the deformable beam when the MEM switch is in operation. Fig. 2 shows a top-down view of the same conventional switch.

In a typical implementation of a conventional MEM switch, the contact beam is formed by depositing polysilicon over a dielectric made of, e. g., Sitz. The surrounding material is etched away leaving a raised structure that is attached to silicon beam 5. The contact element 6, anchored at one end on silicon beam 5 is suspended at its other end above conductors 3 and 4, and is preferably made of polysilicon. Subsequently, the device is subjected to electroless plating, usually of gold, that adheres to the polysilicon to complete the fabrication of conductive elements 3,4 and 6.

The switch is operated by providing a potential difference between contact beam 6 and electrode 3. This voltage generates an electrostatic attraction that brings beam 6 in contact with electrode 4, thus closing the switch. The twist imparted to the anchored beam 5 is used to restore the contact 6 to its open position once the control voltage potential is dropped.

Generally, all conventional MEMS switches rely upon physical contact, especially metal- to-metal contact to perform the switching operation. This leads to many reliability problems related to arcing, material transfer, micro-welding, station, and the like. It is well known in the art that most of these switches become less reliable at higher frequencies. Some of the metallurgies used, such as gold, that are commonly used in an attempt to alleviate these problems are not compatible with standard CMOS fabrication. The inductive MEMS switch of the present invention which will be described hereinafter lends itself to be operated by any number of well known MEMS actuators.

Examples of MEMS actuators can be found at the Sandia National Laboratory web site (www. sandia. gov), or in several MEMS patents related to actuators such as U. S. Patent No.

6,328, 903, George E. Vernon, Sr., "Surface-Micromachined Chain for Use in Micro-Mechanical Structures", issued 12/11/2001. Other patents specifically directed to comb drive systems described hereinafter are to be found, for instance, in U. S. Patent No. 5,998, 906, Jerman et al., "Electrostatic Microactuator and Method for Use Thereof', issued 12/07/1999.

In conventional MEMS switches, as described, for instance, in U. S. Patent No. 6,074, 890, Yao et al., "Method of Fabricating Suspended Single Crystal Silicon MEMS Devices", issued 06/13/2000, and further described in IEEE Microwave issued December 2001, typically, at least one electrode in the switching circuit has a DC potential applied as part of the electrostatic actuation. Thus, a distinct need exists to separate the drive system from the switching circuit such that no DC control voltage is applied to at least one contact in order to perform electrostatic actuation.

SUMMARY OF THE INVENTION Accordingly, it is an object of the invention to provide an inductive MEMS switch based on inductive coupling and decoupling of electrical signals.

It is another object to provide an inductive MEMS switch that isolates the control signal from the switched signals by separating the path of the switched signal from the control circuit used to operate the device.

It is still another object to provide an inductive MEMS switch having an off-state isolation that surpasses a conventional switch in the off-state, and which are typically limited to provide isolation of about 50dB at 6GHz.

It is a further object to provide an inductive MEMS switch that may be configured in a variety of multi-pole, multi-throw arrangements and which is controlled by any number of MEMS linear or rotary drive systems.

It is yet another object to provide an inductive MEMS switch that can reliably perform 'hot-switching', namely, switching while operating under nominal power levels. Switching can be achieved at 1 watt, 5 watts up to whatever value the remaining part of the circuit can accommodate since the switching is non-contact and, thus, there is no arcing or welding of contacts.

It is still another object to provide an inductive MEMS switch that operates reliably with no DC potential or physical contact point in the signal path which can potentially lead to arcing, welding or material transfer and degradation.

It is still a further object to provide an inductive MEMS switch that increases its efficiency at higher frequencies, allowing the size of the coils to decrease when the frequency of the signal increases, the increase in efficiency being achieved by magnetic field coupling between the switch components, thus providing better insertion loss characteristics at higher frequencies without a corresponding decrease in isolation performance.

It is another object of the invention to provide a switch/transformer combination for achieve impedance matching. By selecting the inductance of each portion of the inductive switch appropriately, the input and output impedance of the switch can be adjusted independently.

This adjustment allows for impedance matching and switching at the same time. A special configuration of the transformer can be utilized to create a single-ended to double-ended converter or balun (BAlanced-UNbalanced), providing both switching and signal conversion in a single device.

It is still a further object to provide an inductive MEMS switch that can be manufactured using CMOS compatible processes and materials.

In one aspect of the invention, switching of the signal is accomplished by inductive coupling and decoupling between stationary coils and moveable coils. Switching occurs as the moveable coils are or not aligned with respect to the stationary coils.

A four turn spiral inductor, with a metal thickness of 4, um, a turn width of 10pm, and an outer diameter of 150pm, configured as one element of the switch, is magnetically coupled to another similar spiral, directly above or below, yielding a coupling coefficient of about 0.85. When these spirals are configured as described, a closed-switch insertion loss of 6.6 dB and a opened-switch isolation of 65 dB is achieved at 13 Ghz. This yields an excellent on-off switch ratio tuned to frequencies below 13 Ghz by adding an external tuning capacitor between the two ports of the switch. Similarly, a one and a half turn spiral inductor, with a metal thickness of 4pm, a turn width of 10pm, and an outer diameter of 150pm, configured as one element of the switch, is magnetically coupled to another similar spiral, directly above or below, yielding a coupling coefficient of about 0.85. When the spirals are configured as such, a closed-switch insertion loss of 10 dB and an opened-switch isolation of 60 dB is achieved at 25 Ghz. This yields an excellent on-off switch ratio tuned to frequencies below 25 Ghz by the addition of an external tuning capacitor between the two ports of the switch.

In another aspect of the invention, the present MEMS switch solves problems known as stiction, arcing and welding of the switch contacts, all of which are eliminated because of a lack of physical contact between the switching elements. The coils are simply aligned in close proximity such that inductive coupling can transfer the signal between one and the other. In view of this characteristic, the MEMS switch can easily handle switching at full power (hot switching) and, clearly more power than a conventional MEMS switch.

Multiple switch configurations are realized by varying the number of stationary or moveable coils, and/or by altering the coil geometric configuration of the coils and the corresponding displacement of the moveable elements. Additionally, total isolation from the control signal and the switched signal path is possible since the drive circuit is totally independent of the switching circuit.

BRIEF DESCRIPTION OF THE DRAWINGS These and other objects, aspects and advantages of the invention will become apparent from the following detailed description of preferred embodiments of the invention, when taken in conjunction with the accompanying drawings.

Figs. 1 and 2 are schematic diagrams respectfully showing a cross-section and a top-down view of a prior art MEMS switch.

Fig. 3 shows a first embodiment of the invention, wherein a module of movable inductors rotate about a pivot pin to inductively couple and decouple one set of coils from another. Also shown are comb drives illustrating one possible mode of operation. Fig. 3 illustrates the switch positioned in a decoupled mode to better view the stationary coils.

Fig. 4 is the same device shown of Fig. 3, but substantially rotated to illustrate the mode of coupling the moveable inducto