TWI276162B - Multi-variable analysis model forming method of processing apparatus, multi-variable analysis method for processing apparatus, control apparatus of processing apparatus, and control system of processing apparatus - Google Patents
Multi-variable analysis model forming method of processing apparatus, multi-variable analysis method for processing apparatus, control apparatus of processing apparatus, and control system of processing apparatus Download PDFInfo
- Publication number
- TWI276162B TWI276162B TW092115289A TW92115289A TWI276162B TW I276162 B TWI276162 B TW I276162B TW 092115289 A TW092115289 A TW 092115289A TW 92115289 A TW92115289 A TW 92115289A TW I276162 B TWI276162 B TW I276162B
- Authority
- TW
- Taiwan
- Prior art keywords
- processing device
- processing
- data
- setting data
- correlation
- Prior art date
Links
- 238000012545 processing Methods 0.000 title claims abstract description 527
- 238000000034 method Methods 0.000 title claims abstract description 122
- 238000010202 multivariate logistic regression analysis Methods 0.000 title claims abstract description 29
- 238000001514 detection method Methods 0.000 claims description 91
- 238000000491 multivariate analysis Methods 0.000 claims description 87
- 230000008569 process Effects 0.000 claims description 62
- 238000004458 analytical method Methods 0.000 claims description 12
- 239000000463 material Substances 0.000 claims description 6
- 230000005540 biological transmission Effects 0.000 claims description 5
- 238000004519 manufacturing process Methods 0.000 abstract description 7
- 210000002381 plasma Anatomy 0.000 description 225
- 235000012431 wafers Nutrition 0.000 description 69
- 239000007789 gas Substances 0.000 description 50
- 230000003287 optical effect Effects 0.000 description 21
- 238000012360 testing method Methods 0.000 description 18
- 238000004364 calculation method Methods 0.000 description 13
- 238000005259 measurement Methods 0.000 description 13
- 238000012549 training Methods 0.000 description 12
- 238000005530 etching Methods 0.000 description 11
- 230000005856 abnormality Effects 0.000 description 7
- 238000003745 diagnosis Methods 0.000 description 7
- 238000002474 experimental method Methods 0.000 description 7
- 239000003990 capacitor Substances 0.000 description 6
- 238000009826 distribution Methods 0.000 description 6
- 230000002159 abnormal effect Effects 0.000 description 5
- 239000003507 refrigerant Substances 0.000 description 5
- 239000013078 crystal Substances 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 239000002002 slurry Substances 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 239000004065 semiconductor Substances 0.000 description 3
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 2
- 229910052782 aluminium Inorganic materials 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 239000000835 fiber Substances 0.000 description 2
- 239000011810 insulating material Substances 0.000 description 2
- 238000004020 luminiscence type Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- LCFFREMLXLZNHE-GBOLQPHISA-N (e)-2-[(3r)-3-[4-amino-3-(2-fluoro-4-phenoxyphenyl)pyrazolo[3,4-d]pyrimidin-1-yl]piperidine-1-carbonyl]-4-methyl-4-[4-(oxetan-3-yl)piperazin-1-yl]pent-2-enenitrile Chemical compound C12=C(N)N=CN=C2N([C@@H]2CCCN(C2)C(=O)C(/C#N)=C/C(C)(C)N2CCN(CC2)C2COC2)N=C1C(C(=C1)F)=CC=C1OC1=CC=CC=C1 LCFFREMLXLZNHE-GBOLQPHISA-N 0.000 description 1
- 235000006732 Torreya nucifera Nutrition 0.000 description 1
- 244000111306 Torreya nucifera Species 0.000 description 1
- 229910052770 Uranium Inorganic materials 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 239000004020 conductor Substances 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000005684 electric field Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 229910052734 helium Inorganic materials 0.000 description 1
- 239000001307 helium Substances 0.000 description 1
- SWQJXJOGLNCZEY-UHFFFAOYSA-N helium atom Chemical compound [He] SWQJXJOGLNCZEY-UHFFFAOYSA-N 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000013101 initial test Methods 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000001020 plasma etching Methods 0.000 description 1
- 238000004886 process control Methods 0.000 description 1
- 239000000523 sample Substances 0.000 description 1
- 241000894007 species Species 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
- 230000007723 transport mechanism Effects 0.000 description 1
- JFALSRSLKYAFGM-UHFFFAOYSA-N uranium(0) Chemical compound [U] JFALSRSLKYAFGM-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
- H01L21/02—Manufacture or treatment of semiconductor devices or of parts thereof
- H01L21/04—Manufacture or treatment of semiconductor devices or of parts thereof the devices having potential barriers, e.g. a PN junction, depletion layer or carrier concentration layer
- H01L21/18—Manufacture or treatment of semiconductor devices or of parts thereof the devices having potential barriers, e.g. a PN junction, depletion layer or carrier concentration layer the devices having semiconductor bodies comprising elements of Group IV of the Periodic Table or AIIIBV compounds with or without impurities, e.g. doping materials
- H01L21/30—Treatment of semiconductor bodies using processes or apparatus not provided for in groups H01L21/20 - H01L21/26
- H01L21/302—Treatment of semiconductor bodies using processes or apparatus not provided for in groups H01L21/20 - H01L21/26 to change their surface-physical characteristics or shape, e.g. etching, polishing, cutting
- H01L21/306—Chemical or electrical treatment, e.g. electrolytic etching
- H01L21/3065—Plasma etching; Reactive-ion etching
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01J—ELECTRIC DISCHARGE TUBES OR DISCHARGE LAMPS
- H01J37/00—Discharge tubes with provision for introducing objects or material to be exposed to the discharge, e.g. for the purpose of examination or processing thereof
- H01J37/32—Gas-filled discharge tubes
- H01J37/32917—Plasma diagnostics
- H01J37/32935—Monitoring and controlling tubes by information coming from the object and/or discharge
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Plasma & Fusion (AREA)
- General Physics & Mathematics (AREA)
- Analytical Chemistry (AREA)
- Condensed Matter Physics & Semiconductors (AREA)
- Chemical & Material Sciences (AREA)
- Manufacturing & Machinery (AREA)
- Computer Hardware Design (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Power Engineering (AREA)
- Drying Of Semiconductors (AREA)
- Plasma Technology (AREA)
Abstract
Description
1276162 (1) 玖、發明說明 【發明所屬之技術領域】 本發明係關於處理裝置之多變量解析模型作 處理裝置用之多變量解析方法、處理裝置之控制 理裝置之控制系統。 【先前技術】 在半導體製造工程中，有使用種種的處理裝 導體晶圓或玻璃基板等之被處理體的成膜工程或 中，廣泛使用電漿處理裝置等之處理裝置。各處 別具有對於被處理體之固有的製程特性。因此， 裝置的製程特性，或者預測製程特性等，以進行 佳處理。 例如，在日本專利特開平6- 1 3 22 5 1號公報 有關於電漿蝕刻裝置的蝕刻監視器。在此情形下 究蝕刻的處理結果（均勻性、尺寸精度、形狀或 的選擇性等）和電漿的分光分析結果或製程條件 氣體流量、偏壓電壓等）的變動狀況等之關係， 些記憶爲資料庫，能夠不直接檢查晶圓而間接地 結果。在監視結果不符合檢查條件時，將該資訊 刻裝置，修正處理條件，或者中止處理，同時對 報該旨意。 另外，在日本專利特開平10-125660號公報 關於電漿處理裝置之製程監視方法。在此情形下 成方法、 裝置、處 置。在半 蝕刻工程 理裝置分 監視每一 晶圓的最 中，提案 ，事先硏 和基底膜 (壓力、 藉由將這 監視處理 傳送給蝕 管理者通 中，提案 ，在處理 -5- (2) 1276162 前’利用試用晶圓，製作使反映電漿狀態的電氣訊號和電 漿處理特性相關連的模型，將實際處理晶圓時所獲得的電 氣訊號的檢測値代入模型，與預測、診斷電漿狀態。 另外，在日本專利特開平1 1 - 8 7 3 2 3號公報中，提案 有關於利用半導體晶圓處理系統的多數參數，以監視製程 之方法及裝置。在此情形下，分析多數的製程參數，使這 些參數統計上相關，以檢測製程特性或者系統特性的變 化。多數的製程參數係使用：發光、環境參數（反應腔內 的壓力或溫度等）、RF功率參數（反射功率、調諧電壓 等）、系統參數（特定的系統構造或控制電壓）。 但是，在習知技術之情形下，藉由多變量解析等之統 計手法而解析種種的測量資料以作成模型，利用此模型以 掌握、監視處理裝置的狀態或製程特性，例如，在附設於 各處理裝置之感測器間的個體差等，每一處理裝置在製程 特性上有差異時，即使就一個之處理裝置作成模型，也無 法將此模型運用於同統之其他的處理裝置，必須每一處理 裝置取得種種的測量資料，而每次作成模型，存在有在模 型的作成上需要多的工夫和時間之課題。另外，在製程條 件改變時，也需要每一個製程條件取得種種的測量資料， 每次作成模型，存在有在模型的作成上需要多的工夫和時 間之課題。 本發明係爲了解決上述課題所完成者，目的在於提 供：每一處理裝置即使在製程特性或製程條件有差異，如 就一個處理裝置作成模型，也可以將該模型援用於同種的 -6 - (3) 1276162 其他處理裝置，能夠減輕每一處理裝置在作成模型時的工 夫或負擔，另外，即使不每一處理裝置新作成模型，也可 以評估各處理裝置之裝置狀態的處理裝置之多變量解析模 型作成方法及處理裝置用之多變量解析方法。 【發明內容】 爲了解決上述課題，如依據本發明之第1觀點，係提 供一種處理裝置之多變量解析模型作成方法，是針對作成 藉由多變量解析，以評估處理裝置之裝置狀態，或者預測 處理結果時的多變量解析模型之方法，其特徵爲具有：在 多數的處理裝置中，在個別依據第1設定資料而動作時， 藉由多變量解析而就上述每一個處理裝置求得由上述各處 理裝置的多數感測器所檢測的檢測資料和上述第1設定資 料的相關關係的第1工程；及如將在上述各處理裝置中的 1個當成基準處理裝置時，在此基準處理裝置中，於依據 新的第2設定資料而動作時，藉由多變量解析以求得由上 述基準處理裝置的多數感測器所檢測的檢測資料和上述第 2設定資料的相關關係的第2工程；及依據在上述第1工 程所求得的上述其他處理裝置的相關關係，和在上述第1 工程所求得的上述基準處理裝置的相關關係，和在上述第 2工程所求得的上述基準處理裝置的上述相關關係，以求 得上述基準處理裝置以外的其他處理裝置的上述第2設定 資料和檢測資料的相關關係，作成依據如此求得的相關關 係，以評估上述其他處理裝置的裝置狀態或者預測處理結 (4) 1276162 果的多變量解析模型的第3工程。 爲了解決上述課題，如依據本發明之第2觀點，係提 供一種處理裝置用之多變量解析模型作成方法，是針對作 成藉由多變量解析，以評估處理裝置之裝置狀態，或者預 測處理結果時的多變量解析方法，其特徵爲具有：在多數 的處理裝置中，在個別依據第1設定資料而動作時，藉由 多變量解析而就上述每一個處理裝置求得由上述各處理裝 置的多數感測器所檢測的檢測資料和上述第1設定資料的 相關關係的第1工程；及如將在上述各處理裝置中的！個 當成基準處理裝置時，在此基準處理裝置中，於依據新的 第2設定資料而動作時，藉由多變量解析以求得由上述基 準處理裝置的多數感測器所檢測的檢測資料和上述第2設 定資料的相關關係的第2工程；及依據在上述第1工程所 求得的上述其他處理裝置的相關關係，和在上述第1工程 所求得的上述基準處理裝置的相關關係，和在上述第2工 程所求得的上述基準處理裝置的上述相關關係，以求得上 述基準處理裝置以外的其他處理裝置的上述第2設定資料 和檢測資料的相關關係，作成依據如此求得的相關關係， 以評估上述其他處理裝置的裝置狀態或者預測處理結果的 多變量解析模型的第3工程。 另外，在依據上述第1觀點及第2觀點的發明中，上 述第3工程也可以依據對於在上述第丨工程所求得的上述 其他處理裝置的相關關係的上述其他處理裝置的上述第2 設定資料和檢測資料的相關關係，和對於在上述第1工程 -8- (5) 1276162 所求得的上述基準處理裝置的相關關係的在上述第2工程 所求得的上述基準處理裝置的上述相關關係之比例關係， 以求得上述其他處理裝置的上述第2設定資料和檢測資料 的相關關係。另外，上述多變量解析例如也可以藉由部份 最小平方法（PLS法）進行。 另外，在依據上述第1觀點及第2觀點的發明中，處 理裝置可以爲電漿處理裝置。在此情形下，上述設定資料 係使用可以控制電漿狀態的多數之控制參數，同時，上述 檢測資料可以使用由反映電漿狀態的多數之電漿反映參 數、與裝置狀態相關的多數之裝置狀態參數、反映製程完 成之參數群中所選擇的至少其中1種或者2種以上的參 數。 另外，在依據上述第2觀點的發明中，上述多變量解 析模型可以爲由在上述第3工程所求得的上述其他處理裝 置之相關關係和上述第2設定資料所算出的檢測資料和上 述第2設定資料的相關關係式。 爲了解決上述課題，如依據本發明之第3觀點，係提 供一種處理裝置之控制裝置，是針對設置在處理被處理體 的處理裝置，依據特定的設定資料，以進行上述處理裝置 的控制之處理裝置之控制裝置，其特徵爲：設置連接於上 述處理裝置及至少成爲基準的處理裝置及主機裝置相連接 的網路，可以進行資料之交換的發送接收手段，在基於第 1設定資料而動作時，藉由上述發送接收手段，將由上述 處理裝置的多數感測器所檢測出的檢測資料和上述第1設 -9- (6) 1276162 定資料透過上述網路而發送給上述主機裝置，將依據所發 送的資料，藉由上述主機裝置以多變量解析所求得之上述 第1設定資料和上述檢測資料的相關關係由上述主機裝置 錯由上述發送接收手段而透過網路予以接收，藉由上述發 送接收手段將新的第2設定資料透過上述網路發送給主機 裝置，將依據所發送的資料，藉由上述主機裝置所求得的 上述第2設定資料和基於此第2設定資料之檢測資料的相 關關係由上述主機裝置藉由上述發送接收手段透過上述網 路而予以接收，依據由上述主機裝置所接收的上述第2設 定資料的相關關係，作成多變量解析模型，依據此多變量 解析模型，以評估上述處理裝置之裝置狀態或者預測處理 結果，因應該結果以控制上述處理裝置。 另外，在依據上述第3觀點的發明中，可以使上述檢 測資料算出手段計算藉由上述發送接收手段透過上述網路 接收作成在以上述其他處理裝置進行特定製程處理時，評 估裝置狀態或者預測處理結果之多變量解析模型用的其他 處理裝置的設定資料，由接收的上述設定資料和上述處理 裝置的上述相關關係，以與上述其他的處理裝置的上述特 定製程處理相同的條件使上述處理裝置動作時的上述處理 裝置之檢測資料。 另外，在依據上述第3觀點的發明中，上述其他處理 裝置的設定資料可以利用：在上述特定製程處理前，藉由 多變量解析所求得之上述其他處理裝置的設定資料，和由 基於此設定資料而動作時的上述其他處理裝置之多數感測 -10- (7) 1276162 器所檢測的檢測資料的相關關係，以及由上述其他處理裝 置在進行上述特定製程處理時的上述其他處理裝置之多數 感測器所檢測的檢測資料而被算出。 另外，在依據上述第3觀點的發明中，關於上述處理 裝置的上述第2設定資料的相關關係也可以依據：由上述 主機裝置藉由多變量解析所求得之關於上述處理裝置的上 述第1設定資料的相關關係，和由上述主機裝置藉由多變 量解析所求得之上述基準處理裝置依據第1設定資料而動 作時，由上述基準處理裝置的多數感測器所檢測的檢測資 料與上述第1設定資料的相關關係，和由上述主機裝置藉 由多變量解析所求得之上述基準處理裝置依據新的第2設 定資料而動作時，由上述基準處理裝置的多數感測器所檢 測的檢測資料與上述第2設定資料的相關關係而藉由上述 主機裝置所算出。 另外，在基於上述第3觀點的發明中，處理裝置也可 以爲電漿處理裝置。在此情形下，上述設定資料係使用可 以控制電漿狀態的多數之控制參數，同時，上述檢測資料 可以使用由反映電漿狀態的多數之電漿反映參數、與裝置 狀態相關的多數之裝置狀態參數、反映製程完成之參數群 中所選擇的至少其中1種或者2種以上的參數。另外，上 述多變量解析可藉由部份最小平方法進行。另外，上述處 理裝置也可以爲電漿處理裝置。 爲了解決上述課題，如依據本發明之第4觀點，係提 供一種處理裝置之控制系統，是針對具備依據特定的設定 -11 - (8) 1276162 資料，以進行處理被處理體之處理裝置的控制之控制裝置 的處理裝置之控制系統，其特徵爲：具備透過發送接收手 段被連接於網路的多數之前述處理裝置，及連接在上述網 路的主機裝置，上述主機裝置在多數的處理裝置分別依據 第1設定資料動作時，一由上述多數的處理裝置透過上述 網路接收由上述各處理裝置的多數感測器所檢測的檢測資 料和上述第1設定資料時，便藉由多變量解析而每一上述 各處理裝置求得接收的上述第1設定資料和上述檢測資料 的相關關係，將求得之相關關係透過上述網路而發送給對 應的處理裝置，上述主機裝置在上述各處理裝置中的當成 基準之處理裝置依據新的第2設定資料動作時，一由上述 基準處理裝置透過上述網路接收由上述基準處理裝置的多 數感測器所檢測的檢測資料和上述第2設定資料時，便藉 由多變量解析以求得接收的上述第1設定資料和上述檢測 資料的相關關係，將求得的相關關係透過上述網路而發送 給上述基準處理裝置，上述主機裝置一透過上述網路由上 述基準處理裝置以外的其他處理裝置接收上述第2設定資 料時，便依據藉由多變量解析所求得之關於上述其他處理 裝置的上述第1設定資料的上述相關關係，和藉由上述多 變量解析所求得之關於上述基準處理裝置的上述第1設定 資料之上述相關關係，和藉由上述多變量解析所求得之關 於上述基準處理裝置的上述第2設定資料之上述相關關 係，以求得接收的上述第2設定資料和基於此第2設定資 料之檢測資料的相關關係，將所求得之相關關係透過上述 -12- 12761621276162 (1) Technical Field of the Invention The present invention relates to a multivariate analysis method for a processing device, a multivariate analysis method for a processing device, and a control system for a control device of the processing device. [Prior Art] In a semiconductor manufacturing process, a processing apparatus such as a plasma processing apparatus or the like is widely used for forming a film-forming project of a processed object such as a conductor wafer or a glass substrate. Each process has inherent process characteristics for the object being processed. Therefore, the process characteristics of the device, or the process characteristics are predicted, for better processing. For example, an etch monitor for a plasma etching apparatus is disclosed in Japanese Laid-Open Patent Publication No. Hei 6- 1 3 22 5 1 . In this case, the relationship between the processing result of the etching (uniformity, dimensional accuracy, shape, or selectivity) and the spectral analysis result of the plasma or the variation of the process condition gas flow rate, bias voltage, etc., etc., For the database, the results can be indirectly without directly inspecting the wafer. When the monitoring result does not meet the inspection condition, the information is engraved, the processing condition is corrected, or the processing is suspended, and the purpose is reported. In addition, Japanese Laid-Open Patent Publication No. Hei 10-125660 discloses a process monitoring method for a plasma processing apparatus. In this case, the method, device, and disposal. In the semi-etching engineering device to monitor the most of each wafer, the proposal, the prior 硏 and the base film (pressure, by transmitting this monitoring process to the eclipse manager, the proposal, in the processing -5 - (2) 1276162 Before the 'use of the test wafer, the model that correlates the electrical signal reflecting the plasma state with the plasma processing characteristics, and the detection of the electrical signal obtained when the wafer is actually processed is substituted into the model, and the prediction and diagnostic plasma In addition, a method and apparatus for monitoring a process using a plurality of parameters of a semiconductor wafer processing system are proposed in Japanese Patent Laid-Open Publication No. Hei No. Hei No. 1 - 8 7 3 2 3 . Process parameters that make these parameters statistically related to detect changes in process characteristics or system characteristics. Most process parameters are used: luminescence, environmental parameters (pressure or temperature in the reaction chamber, etc.), RF power parameters (reflected power, tuning) Voltage, etc.), system parameters (specific system configuration or control voltage). However, in the case of conventional techniques, multivariate analysis, etc. The statistical methods are used to analyze various kinds of measurement data to make a model, and the model is used to grasp and monitor the state of the processing device or the process characteristics, for example, individual differences between the sensors attached to the respective processing devices, etc., each processing device is When there are differences in process characteristics, even if a processing device is modeled, it cannot be applied to other processing devices of the same system. Each processing device must obtain various kinds of measurement data, and each time a model is created, there is In the creation of the model, it takes a lot of work and time. In addition, when the process conditions change, it is necessary to obtain various measurement data for each process condition. Every time a model is created, there is a need for more work on the creation of the model. The present invention has been made to solve the above problems, and an object of the present invention is to provide that each processing device can be applied to the same species even if there is a difference in process characteristics or process conditions, such as modeling a processing device. -6 - (3) 1276162 Other processing devices that can alleviate each processing device The multi-variable analysis model creation method of the processing device for evaluating the device state of each processing device and the multi-variable analysis method for the processing device can be performed without any new processing model for each processing device. In order to solve the above problems, according to a first aspect of the present invention, a multivariate analysis model creation method of a processing device is provided, which is configured to evaluate a device state of a processing device by multivariate analysis, or to predict a processing result. The method of multivariate analysis model of the case is characterized in that, in a plurality of processing apparatuses, when operating individually according to the first setting data, each of the processing apparatuses is obtained by the multivariate analysis In the first processing of the correlation between the detection data detected by the majority of the sensors of the device and the first setting data; and when one of the processing devices is used as the reference processing device, in the reference processing device When operating in accordance with the new second setting data, the multivariate analysis is used to obtain the above reference. a second project relating to the correlation between the detection data detected by the majority of the sensors of the processing device and the second setting data; and the correlation between the other processing devices obtained based on the first engineering, and the first The correlation between the reference processing device obtained by the project and the correlation relationship between the reference processing device obtained in the second project to obtain the second setting data of the processing device other than the reference processing device Based on the correlation between the detected data and the correlation relationship thus obtained, the third project of the multivariate analysis model of the device state of the other processing device or the prediction processing node (4) 1276162 is evaluated. In order to solve the above problems, according to a second aspect of the present invention, a multivariate analysis model creation method for a processing device is provided, which is configured to evaluate a device state of a processing device by multivariate analysis, or to predict a processing result. The multivariate analysis method is characterized in that, in a plurality of processing apparatuses, when operating individually according to the first setting data, the majority of the processing apparatuses are obtained for each of the processing apparatuses by multivariate analysis. The first item of the correlation between the detection data detected by the sensor and the first setting data; and, as in each of the above processing apparatuses! When the reference processing device is used as the reference processing device, when the reference processing device operates in accordance with the new second setting data, the detection data detected by the majority of the sensors of the reference processing device is obtained by multivariate analysis. a second relationship of the correlation between the second setting data; and a correlation between the other processing device obtained in the first engineering and the reference processing device obtained in the first engineering, And the correlation between the second setting data and the detection data of the processing device other than the reference processing device is obtained in accordance with the correlation relationship between the reference processing device obtained in the second engineering, and the correlation between the second setting data and the detection data is obtained. Correlation, a third project of a multivariate analysis model for evaluating the device state of the other processing device or the prediction processing result. Further, in the invention according to the first aspect and the second aspect, the third item may be based on the second setting of the other processing device relating to the other processing device obtained in the third item. The correlation between the data and the test data, and the correlation between the reference processing device obtained in the second project, which is related to the reference processing device obtained in the first project -8-(5) 1276162 The proportional relationship of the relationship is obtained in order to obtain the correlation between the second setting data and the detection data of the other processing device. Further, the multivariate analysis described above can be performed, for example, by a partial least squares method (PLS method). Further, in the invention according to the first aspect and the second aspect, the processing device may be a plasma processing device. In this case, the above setting data uses a plurality of control parameters that can control the state of the plasma, and at the same time, the detection data can use a majority of the plasma reflecting parameters reflecting the state of the plasma, and a plurality of device states related to the state of the device. The parameter and at least one or more of the parameters selected in the parameter group reflecting the completion of the process. Further, in the invention according to the second aspect, the multivariate analysis model may be a detection data calculated by the correlation between the other processing devices obtained in the third project and the second setting data, and the 2 Set the correlation of the data. In order to solve the above problems, according to a third aspect of the present invention, a control device for a processing device for processing a control of the processing device in accordance with a specific setting data is provided for a processing device provided in a processing target object. The device control device is characterized in that a network connected to the processing device and at least the reference processing device and the host device is provided, and the transmitting and receiving means for exchanging data can be operated when the first setting data is operated. And transmitting, by the transmitting and receiving means, the detection data detected by the plurality of sensors of the processing device and the first setting -9-(6) 1276162 data to the host device through the network, The transmitted data is obtained by the host device by the multi-variable analysis and the correlation between the first setting data and the detected data is received by the host device via the network by the transmitting and receiving means. The transmitting and receiving means transmits the new second setting data to the host device through the network, and According to the transmitted data, the correlation between the second setting data obtained by the host device and the detection data based on the second setting data is received by the host device through the network through the transmitting and receiving means. And generating a multivariate analysis model based on the correlation relationship between the second setting data received by the host device, and evaluating the device state or the prediction processing result of the processing device according to the multivariate analysis model, and controlling the result according to the result Processing device. Further, in the invention according to the third aspect, the detection data calculation means can calculate that the transmission/reception means is transmitted through the network, and when the specific processing is performed by the other processing means, the evaluation apparatus state or the prediction processing can be performed. As a result of the setting data of the other processing device for the multivariate analysis model, the processing device is operated under the same conditions as the specific processing of the other processing device by the correlation between the received setting data and the processing device. The detection data of the above processing device. Further, in the invention according to the third aspect, the setting data of the other processing device may be obtained by using the multi-variable analysis to determine the setting data of the other processing device before the specific process processing, and based on the The majority of the other processing devices that operate while setting the data sense the correlation between the detected data detected by the device 10-(7) 1276162, and the other processing device when the other processing device performs the specific process processing described above. The detection data detected by most sensors is calculated. Further, in the invention according to the third aspect, the correlation between the second setting data of the processing device may be based on the first aspect of the processing device obtained by the host device by multivariate analysis. The relationship between the setting data and the detection data detected by the majority of the sensors of the reference processing device when the reference processing device obtained by the multi-variable analysis by the host device operates according to the first setting data When the correlation between the first setting data and the reference processing device obtained by the multi-variable analysis by the host device operates in accordance with the new second setting data, the majority of the sensors of the reference processing device detect The correlation between the detection data and the second setting data is calculated by the host device. Further, in the invention according to the third aspect described above, the processing apparatus may be a plasma processing apparatus. In this case, the above setting data uses a plurality of control parameters that can control the state of the plasma, and at the same time, the detection data can use a majority of the plasma reflecting parameters reflecting the state of the plasma, and a plurality of device states related to the state of the device. The parameter and at least one or more of the parameters selected in the parameter group reflecting the completion of the process. In addition, the above multivariate analysis can be performed by a partial least squares method. Further, the above processing means may be a plasma processing apparatus. In order to solve the above problems, according to a fourth aspect of the present invention, a control system for a processing apparatus is provided, which is provided with a control device for processing a processed object according to a specific setting -11 - (8) 1276162 A control system for a processing device of a control device, comprising: a plurality of processing devices connected to a network via a transmitting and receiving means; and a host device connected to the network, wherein the host device is in a plurality of processing devices According to the first setting data operation, when the plurality of processing devices receive the detection data detected by the plurality of sensors of the processing devices and the first setting data through the network, the multi-variable analysis is performed. Each of the processing devices obtains a correlation between the received first setting data and the detection data, and transmits the obtained correlation relationship to the corresponding processing device through the network, wherein the host device is in each of the processing devices When the processing device that serves as the benchmark operates according to the new second setting data, When the device receives the detection data detected by the plurality of sensors of the reference processing device and the second setting data through the network, the multi-variable analysis is performed to obtain the received first setting data and the detection data. Correlation, the correlation relationship is transmitted to the reference processing device via the network, and the host device transmits the second setting data through the network processing other processing device other than the reference processing device The correlation relationship between the first setting data of the other processing device obtained by the multivariate analysis, and the correlation relationship between the first setting data of the reference processing device obtained by the multivariate analysis, And the correlation relationship between the second setting data of the reference processing device obtained by the multivariate analysis to obtain the correlation between the received second setting data and the detection data based on the second setting data , to obtain the relevant relationship through the above -12- 1276162
網路而發送給上述其他處理裝置，上述其他處理裝置依據 由上述主機裝置所接收的關於上述第2設定資料的相關關 係，以作成多變量解析模型，依據此多變量解析模型，以 評估上述處理裝置的裝置狀態或者預測處理結果，因應該 結果，以控制上述處理裝置。 另外’在依據上述第4觀點的發明中，處理裝置可以 爲電漿處理裝置。在此情形下，上述設定資料係使用可以 控制電漿狀態的多數之控制參數，同時，上述檢測資料可 以使用由反映電漿狀態的多數之電漿反映參數、與裝置狀 態相關的多數之裝置狀態參數、反映製程完成之參數群中 所選擇的至少其中1種或者2種以上的參數。另外，上述 多變量解析可藉由部份最小平方法進行。另外，上述處理 裝置也可以爲電漿處理裝置。 【實施方式】 以下一面參考所附圖面，一面詳細說明關於本發明之 裝置的合適實施形態。另外，在本說明書及圖面中，關於 實質上具有相同的機能構造之構成要素，藉由賦予相同的 圖號，省略重複說明。 首先，參考第1圖、第2圖以說明關於本發明之第1 實施形態的電漿處理裝置。本實施形態之電漿處理裝置 100係如第1圖所示般，具備：鋁製的處理室（處理腔） 101，及透過絕緣材ίο2A以支持配置在此處理室101內 的下部電極102而可以升降之鋁製支持體103，及配置在 -13- (10) 1276162 此支持體103的上方，且供給製程氣體，而且兼爲上部電 極的淋浴頭（以下，需要時，也稱爲「上部電極」） 104。 上述處理室1 0 1係形成爲上述爲小直徑的上室 101A，下部形成爲大直徑的下室101B。上室101A係由 偶極子環型磁鐵1 〇 5所包圍。此偶極子環型磁鐵1 〇 5係在 由環型的磁性體所形成的機殼內收容多數的非等向性弧形 柱狀磁鐵而形成，在上室101A內形成整體爲朝向一個方 向的一樣的水平磁場。在下室1 0 1 B之上部形成搬入搬出 晶圓W用之出入口，在此出入口安裝閘門閥1〇6。 在下部電極1〇2透過匹配器7A而連接高頻電源 107，由此高頻電源107對於下部電極102施加13.56 Μ 之高頻電力Ρ，在上室101Α內，與上部電極104之間形 成垂直方向的電場。此高頻電力Ρ係藉由連接在高頻電源 107和匹配器7Α間的瓦特計1〇7而檢測出。此高頻電力 Ρ爲可以控制的參數，在本實施形態中，將高頻電力Ρ與 後述的氣體流量、電極間距離等之可控制參數一同定義爲 控制參數。另外，控制參數係對於電漿處理裝置爲可以設 定之參數，所以也稱爲設定資料。 在上述匹配器7Α之下部電極1〇2側（高頻電壓的輸 出側）安裝電氣量測器（例如，VI探針）1 07C，藉由透 過此電氣量測器107C而施加在下部電極1〇2的高頻電力 Ρ，將基於發生在上室101Α內之電漿的基本波及高次諧 波的高頻電壓V、高頻電流I、電壓波形和電流波形間的 -14 - (11) 1276162 相位差P當成電氣資料予以檢測。這些電氣資料係與後述 的光學資料一齊地反映電漿狀態之可監視的參數，在本實 施形態中，定義爲電漿反映參數。另外，電漿反映參數係 藉由電氣量測器1 07C所檢測出之資料，所以也稱爲檢測 資料。 上述匹配器7A例如內藏2個可變電容器Cl、C2、 電容器C及線圈L，透過可變電容器C 1、C2可以取得阻 抗匹配。在匹配狀態的可變電容器C1、C2之電容、由上 述匹配器7A內之量測器（未圖示出）所測量之高頻電壓 Vpp 係與後述的 APC(Automatic pressure controller:自動 壓力控制器）開度等一同爲顯示處理時的裝置狀態之參 數，在本實施形態中，將顯示裝置狀態之可變電容器 C1、C2之電容、高頻電壓Vpp及APC之開度分別定義爲 裝置狀態參數。然後，裝置狀態參數係無法控制的參數， 爲可以檢測之資料，也稱爲檢測資料。 在上述下部電極102的上面配置靜電夾頭108，在此 靜電夾頭108之電極板108A連接直流電源109。因此， 在高真空下，藉由直流電源109對於電極板108A施加高 電壓，以靜電夾頭1 〇8靜電吸引晶圓 W。在此下部電極 102的外圍配置聚焦環110，將在上室101A內所產生的 電漿聚集在晶圓W。另外，在聚焦環1 1 0之下側配置安裝 於支持體1 03之上部的排氣環1 1 1。在此排氣環1 1 1涵蓋 全周在圓周方向等間隔形成多數孔，透過這些孔，將上室 101A內之氣體排往下室101B。 -15- (12) 1276162 上述支持體103係透過滾珠導螺桿機構1 12及波紋管 113而可在上室101A和下室101B間升降。因此’在將晶 圓W供應給下部電極1 0 2上時，下部電極1 0 2透過支持 體1 0 3而下降至下室1 0 1 B，開放閘門閥1 0 6透過未圖示 出的搬運機構而將晶圓W供應給下部電極1 02上。下部 電極1 0 2和上部電極1 0 4之間的電極間距離係可設定爲特 定値之參數，如上述般，構成爲控制參數。 在支持體103的內部形成連接於冷媒配管1 14之冷媒 流路103A，透過冷媒配管1 14在冷媒流路103A內使冷媒 循環，將晶圓W調整爲特定溫度。在支持體1 03、絕緣材 102A、下部電極102及靜電夾頭108分別形成氣體流路 103B，由氣體導入機構115透過氣體配管115A將He氣 體當成背面氣體以特定壓力而供應於靜電夾頭1 08和晶圓 W之間的間隙，透過He氣體，提高靜電夾頭1 08和晶圓 W間的熱傳導性。另外，1 1 6係波紋管蓋。 在上述淋浴頭104之上面形成氣體導入部104A，在 此氣體導入部104A透過配管117連接製程氣體供給系統 1 1 8。製程氣體供給系統1 1 8系具有 Ar氣體供給源 118A、COr氣體供給源118B、C4F6氣體供給源118C及 〇2氣體供給源 118D。這些氣體供給源 118A、118B、 118C、118D 分別透過閥門 118E、118F、118G、118H 及 質流控制器1 181、1 18J、1 18K、1 18L以特定之設定流量 將個別的氣體供應給淋浴頭1 04，在其內部調整爲具有特 定的混合比之混合氣體。各氣體流量爲藉由個別之質流控 -16- (13) 1276162 制器1 181、1 18J、1 18K、1 18L而可以控制，且可以檢測 之寥1數’如上述般，構成爲控制參數。 在上述淋浴頭104之下面涵蓋全面而均等配置多數的 孔104B，透過這些孔i〇4B由淋浴頭104將混合氣體當成 製程氣體供應給上室101A內。另外，在下室101B之下 部的排氣孔連接排氣管1 0 1 C，透過由連接於此排氣管 1 〇 1 C之真空泵等所形成的排氣系統1丨9而排氣處理室】〇 j 內以保持特定的氣體壓力。在排氣管101C設置APC閥門 101D’依循處理室1〇1內之氣體壓力而自動調節開度。 此開度爲顯示裝置狀態的裝置狀態參數，是無法控制的參 數。 在上述處理室1 0 1的側壁設置檢測窗1 2 1，在處理室 1 〇 1的側壁外側透上述檢測窗1 2 1將處理室1 0 1內之電漿 發光橫跨多波長予以檢測之分光器（以下，稱爲「光學檢 測器」）1 2 0。依據關於藉由此光學檢測器1 2 0所獲得之 特定波長的光學資料以監視電漿狀態，例如，檢測電漿處 S之終點。此光學資料係與基於藉由高頻電力p所發生的 電漿的電氣資料一齊地構成反映電漿狀態的電漿反映參 數。 接著’一面參考圖面一面說明設置在上述電漿處理裝 置1 0 0的多變量解析手段。電漿處理裝置i 〇 〇例如具備如 第2圖所示之多變量解析手段200。此多變量解析手段 2 0 0爲具備：記億多變量解析程式的多變量解析程式記憶 手段20 1 ;及間歇地取樣由控制參數量測器22 1、電漿反 -17- (14) 1276162 映參數量測器222及裝置狀態參數量測器223來之檢測訊 號的控制參數訊號取樣手段202、電漿反映參數訊號取樣 手段203及裝置狀態參數訊號取樣手段204。另外，具 備：記憶使多數的電漿反映參數（電氣資料及光學資 料）、與裝置狀態相關的多數的裝置狀態參數及多數的控 制參數相關連之模型等的解析結果或解析所必要的資料之 解析資料記憶手段205 ;及透過模型而依據目的以運算控 制參數、電漿反映參數及裝置狀態參數之運算手段206; 及依據來自運算手段206的運算訊號，依據目的而進行控 制參數、多數的電漿反映參數及裝置狀態參數之預測、診 斷、控制的預測·診斷·控制手段2 0 7。 另外，在多變量解析手段200分別連接依據控制參數 以控制電漿處理裝置1 00之處理裝置控制手段225、警報 器226及顯示手段224。處理裝置控制手段225例如系依 據來自預測·診斷·控制手段207之訊號以繼續或者中斷 晶圓W之處理。警報器226及顯示手段224係如後述 般，依據來自預測·診斷·控制手段207的訊號，基於目 的以通報控制參數、多數的電漿反映參數及裝置狀態參數 之何者之異常。另外，解析資料記憶手段205係記憶關於 上述各參數之資料或這些之加工資料（使用於多變量解析 之加工資料）。另外，控制參數量測器22 1、電漿反映參 數量測器222、裝置狀態參數量測器223係分別將流量檢 測器、光學量測器、高頻電壓Vpp量測器等之多數的控 制參數之量測器、多數的電漿反映參數的量測器、多數的 -18- (15) 1276162 裝置狀態參數的量測器彙整爲一個而顯示。 此處，I兌明本發明之原理。例如，作爲作成新模型時 的基準之處理裝置，係考慮電漿處理裝置100A，作爲此 基準處理裝置以外的處理裝置係考慮電漿處理裝置 10 0B。在電漿處理裝置100A、100B之間，由於製造上之 偏差等，存在稍微之個體差。另外，上述電氣量測器 1 07C、光學檢測器1 20等之感測器也分別由於製造誤差 等，而每一個電漿處理裝置存在個體差，所以即使在同一 種的電漿處理裝置使用同一種之感測器，也無法獲得相同 的檢測資料。因此，即使爲同一種之電漿處理裝置，需要 每一個電漿處理裝置作成多變量解析模型，無法將一個多 變量解析模型援用爲其他同種之電漿處理裝置的多變量解 析模型。 因此，在本實施形態中，例如，在電漿處理裝置 100A、100B間，即使有製造上的個體差，或者在個別的 多數感測器間有個體差，也可以將關於電漿處理裝置 1 00A所作成的多變量解析模型源用於其他的電漿處理裝 置1 0 0B。在本實施形態中，多變量解析之一手法係利用 部份最小平方法（以下，稱爲「PLS(Partial Least Squares)法」）以作成電漿處理裝置100A、100B個別的 多變量解析模型，找出裝置間的個體差，作成吸收此個體 差的模型。PLS法之詳細例如揭露在 JOURNAL OF CHEMOMETRICS，VOL. 2 (PP.21 卜228)( 1 998)。 例如，電漿處理裝置1 00A、1 00B都設多數的控制參 «19- (16) 1276162 數（設定資料）爲目的變數’設多數的電漿反映參數（含 電氣資料及光學資料的檢測資料）爲說明變數，作成使以 目的變數爲成分之行列x和以說明變數爲成分之行列Y 相關連的下述（1 )所示之回歸式（以下，單單稱爲「模 型」）（第1工程）。 在電漿處理裝置 100A、100B個別的運算手段 206 中，依據利用多變量解析之一手法的P L S法，在個別的 實驗所獲得之說明變數和目的變數，分別算出模型之回歸 行列K、Kb，如上述般，將這些模型記憶在解析資料記憶 手段205。另外，在下述（1 ) 、（ 2 )的模型中，Ka、Kb 爲個別之模型的回歸行列，a係表示電漿處理裝置1 〇〇a、 b係表示電漿處理裝置100B。The network transmits to the other processing device, and the other processing device creates a multivariate analysis model based on the correlation relationship between the second setting data received by the host device, and evaluates the processing according to the multivariate analysis model. The device state of the device or the predicted processing result is controlled to control the above-described processing device. Further, in the invention according to the fourth aspect, the processing device may be a plasma processing device. In this case, the above setting data uses a plurality of control parameters that can control the state of the plasma, and at the same time, the detection data can use a majority of the plasma reflecting parameters reflecting the state of the plasma, and a plurality of device states related to the state of the device. The parameter and at least one or more of the parameters selected in the parameter group reflecting the completion of the process. In addition, the above multivariate analysis can be performed by a partial least squares method. Further, the above processing means may be a plasma processing apparatus. [Embodiment] Hereinafter, a preferred embodiment of the device of the present invention will be described in detail with reference to the accompanying drawings. In the specification and the drawings, constituent elements that have substantially the same functional structure are denoted by the same reference numerals, and the description thereof will not be repeated. First, a plasma processing apparatus according to a first embodiment of the present invention will be described with reference to Figs. 1 and 2 . As shown in Fig. 1, the plasma processing apparatus 100 of the present embodiment includes an aluminum processing chamber (processing chamber) 101 and a transparent insulating material ίο2A to support the lower electrode 102 disposed in the processing chamber 101. The aluminum support body 103 that can be lifted and lowered, and the shower head that is disposed above the support body 103 of -13-(10) 1276162 and that supplies the process gas and also serves as the upper electrode (hereinafter, if necessary, also referred to as "upper part" Electrode") 104. The processing chamber 101 is formed as the upper chamber 101A having a small diameter as described above, and the lower portion is formed as a lower chamber 101B having a large diameter. The upper chamber 101A is surrounded by a dipole ring magnet 1 〇 5 . The dipole ring magnet 1 〇5 is formed by accommodating a plurality of anisotropic arcuate columnar magnets in a casing formed of a ring-shaped magnetic body, and is formed in one direction in the upper chamber 101A as a whole. The same horizontal magnetic field. An inlet and outlet for loading and unloading the wafer W are formed in the upper portion of the lower chamber 1 0 1 B, and the gate valve 1〇6 is attached to the inlet and outlet. The lower electrode 1〇2 is connected to the high-frequency power source 107 through the matching unit 7A, whereby the high-frequency power source 107 applies a high-frequency power 13 of 13.56 对于 to the lower electrode 102, and forms a vertical line with the upper electrode 104 in the upper chamber 101Α. The electric field in the direction. This high-frequency power system is detected by a wattmeter 1〇7 connected between the high-frequency power source 107 and the matching unit 7Α. This high-frequency power Ρ is a controllable parameter. In the present embodiment, the high-frequency power Ρ is defined as a control parameter together with controllable parameters such as a gas flow rate and an electrode distance which will be described later. In addition, the control parameters are parameters that can be set for the plasma processing apparatus, and are therefore also referred to as setting data. An electrical measuring device (for example, a VI probe) 107C is mounted on the lower electrode 1〇2 side (the output side of the high-frequency voltage) of the matching device 7Α, and is applied to the lower electrode 1 through the electrical measuring device 107C. The high-frequency power 〇 of 〇2 will be based on the high-frequency voltage V, the high-frequency current I, the voltage waveform, and the current waveform of the fundamental wave and higher harmonics of the plasma generated in the upper chamber 101Α-14 - (11) 1276162 Phase difference P is detected as electrical data. These electrical data are monitorable parameters reflecting the state of the plasma together with the optical data described later, and in the present embodiment, are defined as plasma reflection parameters. In addition, the plasma reflection parameter is the data detected by the electrical measuring device 107C, so it is also called the detection data. The matching unit 7A includes, for example, two variable capacitors C1, C2, a capacitor C, and a coil L, and impedance matching can be obtained by the variable capacitors C1 and C2. The capacitance of the variable capacitors C1 and C2 in the matching state, the high frequency voltage Vpp measured by the measuring device (not shown) in the matching unit 7A, and the APC (Automatic Pressure Controller) described later The opening degree and the like are parameters of the state of the device at the time of the display processing. In the present embodiment, the capacitances of the variable capacitors C1 and C2 of the display device state, the high-frequency voltages Vpp, and the opening degree of the APC are respectively defined as device state parameters. . Then, the device status parameter is an uncontrollable parameter, which is a detectable data, also called detection data. An electrostatic chuck 108 is disposed on the upper surface of the lower electrode 102, and the electrode plate 108A of the electrostatic chuck 108 is connected to the DC power source 109. Therefore, under high vacuum, a high voltage is applied to the electrode plate 108A by the DC power source 109, and the wafer W is electrostatically attracted by the electrostatic chuck 1 〇8. A focus ring 110 is disposed on the periphery of the lower electrode 102 to collect the plasma generated in the upper chamber 101A on the wafer W. Further, an exhaust ring 1 1 1 attached to the upper portion of the support body 103 is disposed below the focus ring 1 10 0. Here, the exhaust ring 1 1 1 includes a plurality of holes formed at equal intervals in the circumferential direction throughout the circumference, and the gas in the upper chamber 101A is discharged to the lower chamber 101B through the holes. -15- (12) 1276162 The support body 103 is vertically movable between the upper chamber 101A and the lower chamber 101B by the ball screw mechanism 1 12 and the bellows 113. Therefore, when the wafer W is supplied to the lower electrode 10 2 , the lower electrode 1 0 2 passes through the support 1 0 3 and descends to the lower chamber 1 0 1 B, and the open gate valve 1 0 6 transmits through the unillustrated The wafer W is supplied to the lower electrode 102 by the transport mechanism. The distance between the electrodes between the lower electrode 110 and the upper electrode 104 can be set to a specific parameter, and as described above, it is configured as a control parameter. The refrigerant flow path 103A connected to the refrigerant pipe 1 14 is formed inside the support body 103, and the refrigerant is circulated through the refrigerant pipe 1 14 in the refrigerant flow path 103A to adjust the wafer W to a specific temperature. A gas flow path 103B is formed in each of the support member 103, the insulating material 102A, the lower electrode 102, and the electrostatic chuck 108, and the gas introduction mechanism 115 transmits the He gas as a back gas through the gas pipe 115A at a specific pressure to the electrostatic chuck 1 The gap between the 08 and the wafer W is transmitted through the He gas to improve the thermal conductivity between the electrostatic chuck 108 and the wafer W. In addition, 1 16 is a bellows cover. A gas introduction portion 104A is formed on the upper surface of the shower head 104, and the gas introduction portion 104A is connected to the process gas supply system 1 18 through a pipe 117. The process gas supply system 1 18 includes an Ar gas supply source 118A, a COr gas supply source 118B, a C4F6 gas supply source 118C, and a 〇2 gas supply source 118D. These gas supply sources 118A, 118B, 118C, and 118D supply individual gases to the shower head through valves 118E, 118F, 118G, and 118H and mass flow controllers 1 181, 1 18J, 1 18K, and 18 18, respectively, at a specific set flow rate. 1 04, adjusted internally to have a specific mixing ratio of the mixed gas. Each gas flow rate can be controlled by individual mass flow control-16-(13) 1276162 controllers 1 181, 1 18J, 1 18K, 1 18L, and the number of 可以1 that can be detected is configured as described above. parameter. A plurality of well-arranged holes 104B are provided under the shower head 104, through which the mixed gas is supplied as a process gas into the upper chamber 101A by the shower head 104. Further, the exhaust pipe in the lower portion of the lower chamber 101B is connected to the exhaust pipe 1 0 1 C, and is exhausted through the exhaust system 1丨9 formed by a vacuum pump or the like connected to the exhaust pipe 1 〇1 C. 〇j to maintain a specific gas pressure. The APC valve 101D' is provided in the exhaust pipe 101C to automatically adjust the opening degree in accordance with the gas pressure in the processing chamber 1〇1. This opening is a device status parameter indicating the state of the device and is an uncontrollable parameter. A detection window 1 2 1 is disposed on the sidewall of the processing chamber 110, and the plasma light in the processing chamber 1 0 1 is detected across the multi-wavelength through the detection window 1 2 1 outside the sidewall of the processing chamber 1 〇1. A spectroscope (hereinafter referred to as "optical detector") 1 2 0. The state of the plasma is monitored based on the optical data of the particular wavelength obtained by the optical detector 120, for example, the end of the plasma S is detected. This optical data constitutes a plasma reflection parameter reflecting the state of the plasma together with the electrical data based on the plasma generated by the high-frequency power p. Next, the multivariate analysis means provided in the above-described plasma processing apparatus 100 will be described with reference to the drawing. The plasma processing apparatus i 〇 具备 has, for example, a multivariate analysis means 200 as shown in Fig. 2 . The multivariate analysis means 2000 is a multivariate analysis program memory means 20 1 having a multivariate analysis program; and intermittent sampling by the control parameter measurer 22 1 , plasma counter -17- (14) 1276162 The control parameter signal sampling means 202, the plasma reflection parameter signal sampling means 203 and the device status parameter signal sampling means 204 of the detection signal from the parameter parameter measuring device 222 and the device state parameter measuring device 223. In addition, it includes information necessary for analyzing the analysis results of a model such as a plurality of plasma reflection parameters (electrical data and optical data), a plurality of device state parameters related to the device state, and a plurality of control parameters. The data memory means 205 is analyzed; and the calculation means 206 for calculating the control parameter, the plasma reflection parameter and the device state parameter according to the purpose through the model; and the control signal from the calculation means 206, the control parameter and the majority of the electricity are performed according to the purpose. Prediction, diagnosis, control prediction, diagnosis and control means of slurry reflection parameters and device state parameters. Further, the multivariable analysis means 200 is connected to the processing means control means 225, the alarm 226 and the display means 224 of the plasma processing apparatus 100 in accordance with the control parameters. The processing device control means 225 continues or interrupts the processing of the wafer W based on, for example, the signal from the prediction/diagnosis/control means 207. The alarm 226 and the display means 224 are based on the signal from the prediction/diagnosis/control means 207, based on the purpose of reporting the control parameter, the majority of the plasma reflection parameter, and the device state parameter. Further, the analytical data memory means 205 stores data on the above parameters or processing data (processing data used for multivariate analysis). In addition, the control parameter measuring device 22, the plasma reflection parameter measuring device 222, and the device state parameter measuring device 223 control the majority of the flow rate detector, the optical measuring device, the high-frequency voltage Vpp measuring device, and the like, respectively. The parameter measurer, the majority of the plasma reflectance parameter measurer, and the majority of the -18-(15) 1276162 device status parameters are measured and displayed as one. Here, I understand the principles of the present invention. For example, the plasma processing apparatus 100A is considered as a processing device for designing a new model, and the plasma processing apparatus 100B is considered as a processing device other than the reference processing device. Between the plasma processing apparatuses 100A and 100B, there is a slight individual difference due to variations in manufacturing or the like. In addition, the sensors of the electrical measuring device 100C, the optical detector 120, and the like are also individually poor in each of the plasma processing devices due to manufacturing errors, etc., so even if they are used in the same plasma processing device. A sensor does not have the same test data. Therefore, even for the same plasma processing apparatus, each of the plasma processing apparatuses is required to be a multivariate analysis model, and a multivariate analysis model cannot be used as a multivariate analysis model of other plasma processing apparatuses of the same kind. Therefore, in the present embodiment, for example, between the plasma processing apparatuses 100A and 100B, even if there is an individual difference in manufacturing or there is an individual difference between the individual sensors, the plasma processing apparatus 1 can be used. The multivariate analysis model source made by 00A is used for other plasma processing devices 100B. In the present embodiment, one of the multivariate analysis methods uses a partial least squares method (hereinafter referred to as "PLS (Partial Least Squares) method) to create individual multivariate analysis models of the plasma processing apparatuses 100A and 100B. Find the individual differences between the devices and create a model that absorbs this individual difference. Details of the PLS method are disclosed, for example, in JOURNAL OF CHEMOMETRICS, VOL. 2 (PP. 21 228) (1 998). For example, the plasma processing devices 100A, 100B all have a large number of control parameters «19- (16) 1276162 (setting data) for the purpose of the variable 'multiple plasma reflection parameters (including electrical and optical data detection data) In order to explain the variables, a regression equation (hereinafter, simply referred to as "model") in which the row of the target variable is a component and the rank of the variable Y is associated with the rank Y (the following is simply referred to as "model") is created. engineering). In the individual calculation means 206 of the plasma processing apparatuses 100A and 100B, the regression ranks K and Kb of the model are respectively calculated based on the explanatory variables and the target variables obtained in the individual experiments by the PLS method using one of the multivariate analysis methods. These models are stored in the analytical data memory means 205 as described above. Further, in the models of the following (1) and (2), Ka and Kb are regression rows of individual models, and a indicates that the plasma processing apparatus 1a and b represent the plasma processing apparatus 100B.
Xa = KaYa··· (1) Xb = KbYb··· (2) P L S法係於行列X、γ中，即使個別有多數的說明變 數及目的變數，只要具有個別之少數的實測値，便可以求 得行列X和行列Υ之關係式。而且，即使爲以少數的實 測値所獲得之關係式，其穩定性及可靠性高，此爲PLS 法之特徵。在實測成爲說明變數及目的變數之各資料時， 分配控制參數以檢測控制參數，以多數感測器分節檢測電 漿反映參數。 在此情形下，在分配控制參數（高頻電力、處理室內 -20- (17) 1276162 壓力、製程氣體流量等）之範圍窄時，如下述式（3 )所 示般，可以對於控制參數以線性形式予以近似之，在分配 參數之範圍大時，如下述式（4 )所示般，對於控制參數 可以置入平方、立方及1次和2次交叉項之非線性形式予 以近似之。 此種控制參數在電漿處理裝置100A和電漿處理裝霞 1 0 0B係使用相同範圍、相同値之控制參數。在求得回歸 行列Ka、Kb時，可以與本案申請人在日本專利特_ 2001-398608號|兌明書中所提案的pls法相同的運算步驟 而求得。此處，省略該運算步驟之說明。電漿處理裝寶 100A和電漿處理裝置100B之間的個體差及個別之感測器 間的個體差’成爲上述式（1 )、式（2 )之回歸行列 Ka、Kb之差而表現。 X = [ X 1 , X 2 5 …，X η ] ··· (3) X = [ X 1，X 2， … ? X η ? (xl)2,（χ2)2，…，（χη)2， (X 1 )3，（χ2)3（χη)3， X 1 χ2, X 1 χ3，…，xn-1χη， (xl)2x2? (χ1)2χ3··· (xn-l)2xn] …（4) 然後，在Ιθ由PLS法以求得上述模型時，事先藉由 利用晶圓之訓練裝置（training set )的實驗，以測量多數 的說明變數和多數的目的變數。爲此，例如作爲訓練裝 -21 - (18) 1276162 置，準備18片之晶圓（丁11-0乂3丨）。另外，1^-0乂5丨係 形成熱氧化膜之晶圓。在此情形下’利用實驗計畫法’有 效率設定控制參數（設定資料），可以最小限度的實驗完 成。 在電漿處理裝置100A中，例如在以標準値爲中心而 於特定範圍中，每一個訓練晶圓分配成爲目的變數之控制 參數，以蝕刻處理訓練晶圓。然後，在蝕刻處理時，關於 各訓練晶圓各多數次測量製程氣體之各氣體的流量、處理 室內的壓力等之控制參數、電氣資料及光學資料等的電漿 反映參數，透過運算手段20 6算出這些控制參數、電漿反 映參數的平均値。然後，將控制參數的平均値當成設定資 料使用，將電漿反映參數當成檢測資料使用。 分配控制參數之範圍係假定在進行蝕刻處理時，控制 參數最大限度變動之範圍，在此假定的範圍內，分配控制 參數。在本實施形態中，將高頻電力、處理室內壓力、上 下兩電極102、104間的間隙尺寸及各製程氣體（Ar氣 體、CO氣體、C4F6氣體及02氣體）的流量當成控制參數 (設定資料）使用。各控制參數之標準値因蝕刻對象而不 同。在電漿處理裝置100B也以與電漿處理裝置100A相 同之要領，以同一控制參數（設定資料）進行實驗，獲得 控制參數（設定資料）及電漿反映參數（檢測資料）。 具體爲，以標準値爲中心而在下述表1所示之等級1 及等級2之範圍內，每一訓練晶圓分配控制參數予以設 定’進行各訓練晶圓之蝕刻處理。然後，在處理各訓練晶 -22- (19) 1276162 圓間，透過電氣量測器107C將基於電漿之高頻電壓（由 基本波至4倍波爲止）V、高頻電流（由基本波至4倍波 爲止）I、相位差4等之電氣資料當成檢測資料予以測量 的同時，透過光學檢測器120例如將200〜95 Onm之波長 範®的發光光譜強度（光學資料）當成檢測資料予以測 量，將這些檢測資料（電氣資料及光學資料）當成電漿反 映參數使用。另外，同時利用個別之控制參數量測器22 1 以測量下述表1所示之各控制參數。 (表1 ) 電力 壓力 間隙 Ar CO C4F6 02 W m Τ ο rr mm seem seem seem seem 等級1 1400 3 8 25 170 36 9.5 3.5 標準値 1500 40 27 200 50 10 4 等級2 1540 42 29 230 64 10.5 4.5 2.67% 5.0% 7.41% 15.00% 2 8.00% 5.0 0 % 1 2.5 0 % 然後，在處理訓練晶圓時，將上述各控制參數設定爲 熱氧化膜之標準値，以標準値事先處理5片之僞晶圓’以 謀求電漿處理裝置100A、100B之穩定化。接著，在電漿 處理裝置100A、100B中，進行18片之訓練晶圓的蝕刻 處理。此時，在電漿處理裝置10 0A中，如下述表2所示 般，在上述等級1及上述等級2之範圍內，每一訓練晶圓 分配上述各控制參數，即製程氣體（Ar、CO、Cd6、 -23- (20) 1276162 〇2 )流量、處理室內之壓力、高頻電力，以處理各訓練晶 圓。 接著，關於各訓練晶圓，由個別之量測器獲得多數的 電氣資料及多數的光學資料。這些例如當作實測値而記億 在解析資料記憶手段205。然後，在運算手段206中，算 出多數之控制參數個別的實測値之平均値、多數的電漿反 映參數（電氣資料、光學資料）個別之實測値的平均値， 將這些平均値當成目的變數及說明變數，記憶在解析資料 記憶手段20 5中。接著，在運算手段206中，利用PLS 法，依據這些運算資料，求得上述（1 )之模型的回歸行 列Ka(第1工程）。 另外，在電漿處理裝置100B中也與電漿處理裝置 10〇A相同，如下述表2所示般，分配控制參數，算出各 參數的實測値之平均値，將這些平均値當成目的變數及說 明變數使用，求得上述（2 )之模型的回歸行列Kb(第1 工程）。另外，在下述表2中，L1〜L8係表示訓練晶圓之 號碼。 -24- (21) 1276162 (表2 )Xa = KaYa··· (1) Xb = KbYb··· (2) The PLS method is used in the ranks X and γ. Even if there are a large number of explanatory variables and target variables, as long as there are a few actual measured flaws, Find the relationship between the rank X and the rank Υ. Moreover, even if it is a relationship obtained by a small number of actual measurements, its stability and reliability are high, which is characteristic of the PLS method. When the actual measurement becomes the data for the variables and the purpose variables, the control parameters are assigned to detect the control parameters, and the plasma reflection parameters are detected by the majority of the sensors. In this case, when the range of the distribution control parameters (high-frequency power, processing chamber -20-(17) 1276162 pressure, process gas flow rate, etc.) is narrow, as shown in the following formula (3), The linear form is approximated. When the range of the distribution parameters is large, as shown in the following formula (4), the control parameters can be approximated by a nonlinear form in which square, cubic, and one-time and two-time cross terms are placed. Such control parameters use the same range and the same control parameters in the plasma processing apparatus 100A and the plasma processing apparatus. When obtaining the regression ranks Ka and Kb, it can be obtained by the same calculation procedure as the pls method proposed by the applicant in Japanese Patent Application No. 2001-398608. Here, the description of the arithmetic steps is omitted. The individual difference between the plasma processing package 100A and the plasma processing apparatus 100B and the individual difference between the individual sensors are expressed as the difference between the regression lines Ka and Kb of the above equations (1) and (2). X = [ X 1 , X 2 5 ..., X η ] ··· (3) X = [ X 1,X 2, ... ? X η ? (xl) 2, (χ2) 2,...,(χη)2 , (X 1 )3, (χ2)3(χη)3, X 1 χ2, X 1 χ3,...,xn-1χη, (xl)2x2? (χ1)2χ3··· (xn-l)2xn] ... (4) Then, when the above model is obtained by the PLS method at Ιθ, a majority of the explanatory variables and a plurality of purpose variables are measured in advance by experiments using a wafer training set. To do this, for example, as a training pack -21 - (18) 1276162, prepare 18 wafers (Ding 11-0乂3丨). In addition, 1^-0乂5丨 is a wafer in which a thermal oxide film is formed. In this case, the 'experimental plan' is used to set the control parameters (setting data) efficiently, and it can be done with a minimum of experiments. In the plasma processing apparatus 100A, for example, in a specific range centered on the standard enthalpy, each training wafer is assigned a control parameter of the target variable, and the wafer is trained by etching. Then, at the time of the etching process, the flow rate of each gas of the process gas, the control parameters such as the pressure in the processing chamber, the plasma reflection parameters of the electrical data and the optical data, and the like are measured by the calculation means. Calculate the average enthalpy of these control parameters and plasma reflection parameters. Then, the average of the control parameters is used as the set data, and the plasma reflection parameter is used as the test data. The range of the distribution control parameters is assumed to be the range in which the control parameters are maximally changed when the etching process is performed, and the control parameters are assigned within the assumed range. In the present embodiment, the high-frequency power, the pressure in the processing chamber, the gap size between the upper and lower electrodes 102 and 104, and the flow rates of the respective process gases (Ar gas, CO gas, C4F6 gas, and 02 gas) are regarded as control parameters (setting data) )use. The standard of each control parameter varies depending on the object to be etched. The plasma processing apparatus 100B also performs experiments with the same control parameters (setting data) in the same manner as the plasma processing apparatus 100A, and obtains control parameters (setting data) and plasma reflection parameters (detecting data). Specifically, each training wafer distribution control parameter is set in the range of level 1 and level 2 shown in Table 1 below, centered on the standard ’, and the etching process of each training wafer is performed. Then, between the processing of each training crystal-22-(19) 1276162 circle, through the electrical measuring device 107C, the high-frequency voltage based on the plasma (from the fundamental wave to 4 times the wave) V, the high-frequency current (from the fundamental wave) The electrical data of I, the phase difference 4, etc. are measured as the detection data, and the luminescence intensity (optical data) of the wavelength range of 200 to 95 Onm is transmitted as the detection data by the optical detector 120, for example. For measurement, these test data (electrical data and optical data) are used as plasma reflection parameters. In addition, the individual control parameter estimators 22 1 are simultaneously utilized to measure the respective control parameters shown in Table 1 below. (Table 1) Power Pressure Clearance Ar CO C4F6 02 W m Τ ο rr mm seem I seem to rank 1 1400 3 8 25 170 36 9.5 3.5 Standard 値 1500 40 27 200 50 10 4 Level 2 1540 42 29 230 64 10.5 4.5 2.67 % 5.0% 7.41% 15.00% 2 8.00% 5.0 0 % 1 2.5 0 % Then, when processing the training wafer, the above control parameters are set as the standard 热 of the thermal oxide film, and the pseudo-crystals of 5 pieces are processed in advance by standard 値The circle ' is stabilized by the plasma processing apparatuses 100A and 100B. Next, in the plasma processing apparatuses 100A and 100B, etching processing of 18 training wafers is performed. At this time, in the plasma processing apparatus 100A, as shown in the following Table 2, in the range of the above-mentioned level 1 and the above-mentioned level 2, each of the control parameters, that is, the process gas (Ar, CO) is distributed for each training wafer. , Cd6, -23- (20) 1276162 〇 2 ) Flow, processing chamber pressure, high frequency power to process each training wafer. Next, for each training wafer, a majority of the electrical data and most of the optical data are obtained by individual measuring instruments. These are, for example, regarded as actual measurements and are recorded in the data memory means 205. Then, in the calculation means 206, the average 値 of the actual measured 个别 of the plurality of control parameters and the average 値 of the actual measured 参数 of the plurality of plasma reflection parameters (electrical data, optical data) are calculated, and the average 値 is regarded as the target variable and Explain the variables, and the memory is in the analytical data memory means. Next, the calculation means 206 obtains the regression row Ka (first project) of the model of the above (1) based on these calculation data by the PLS method. Further, in the plasma processing apparatus 100B, similarly to the plasma processing apparatus 10A, as shown in the following Table 2, the control parameters are distributed, and the average value of the measured parameters of each parameter is calculated, and these average values are regarded as the target variables and Explain the use of the variable and find the regression rank Kb (1st project) of the model of (2 above). Further, in Table 2 below, L1 to L8 indicate the number of the training wafer. -24- (21) 1276162 (Table 2)
No. 壓力 Ar CO C 4 F 6 〇2 間隙 電力 [m T 〇 r r ] [seem] [seem] [seem] [seem] [mm] [W] L1 42 170 64 10 4.5 25 1500 L2 3 8 200 36 9.5 4.5 29 1500 L3 40 230 64 9.5 3.5 27 1500 L4 42 1 70 50 9.5 4.5 27 1540 L5 3 8 1 70 36 9.5 3.5 25 1460 L6 3 8 200 50 10 4 27 1500 L7 38 230 50 10 3.5 25 1540 L8 36 230 64 10.5 4.5 29 1540 L9 42 200 64 10 3.5 29 1460 L10 40 1 70 50 10.5 3.5 29 1500 L1 1 40 200 54 9.5 4 25 1540 L12 42 200 36 10.5 3.5 27 1540 L1 3 42 230 36 10.5 4 25 1500 L14 40 230 36 10 1 .5 27 1460 L 1 5 40 200 50 10.5 4.5 25 1460 L16 42 230 50 9.5 3.5 29 1460 L1 7 40 170 36 10 3.5 29 1540 L1 8 38 170 54 10.5 3.5 27 1460 求得回歸行列 Ka、Kb後，利用電漿處理裝置 100A，在下述表3所示之新的製程條件下，如下述表3 -25- (22) 1276162 所示般，由標準値分配製程氣體流量等之控制參數，處理 20片之測試晶圓（ΤΗ-OX Si )，藉由個別之感測器檢測 此時的電漿反映參數及裝置狀態參數。此時，如下述表3 所示般，將多數的控制參數設定爲製程條件的標準値，使 電漿處理裝置運轉，將5片之裸矽晶圓當成僞晶圓流通於 處理室1 〇 1內，使電漿處理裝置穩定化。No. Pressure Ar CO C 4 F 6 〇2 Clearance power [m T 〇rr ] [seem] [seem] [seem] [seem] [mm] [W] L1 42 170 64 10 4.5 25 1500 L2 3 8 200 36 9.5 4.5 29 1500 L3 40 230 64 9.5 3.5 27 1500 L4 42 1 70 50 9.5 4.5 27 1540 L5 3 8 1 70 36 9.5 3.5 25 1460 L6 3 8 200 50 10 4 27 1500 L7 38 230 50 10 3.5 25 1540 L8 36 230 。 。 。 。 。 。 。 。 。 。 。 40 230 36 10 1 .5 27 1460 L 1 5 40 200 50 10.5 4.5 25 1460 L16 42 230 50 9.5 3.5 29 1460 L1 7 40 170 36 10 3.5 29 1540 L1 8 38 170 54 10.5 3.5 27 1460 Regression rank Ka After Kb, using the plasma processing apparatus 100A, under the new process conditions shown in Table 3 below, the control parameters of the process gas flow rate and the like are distributed from the standard enthalpy as shown in Table 3-25-(22) 1276162 below. The 20 test wafers (ΤΗ-OX Si ) were processed, and the plasma reflection parameters and device state parameters at this time were detected by individual sensors. At this time, as shown in Table 3 below, a plurality of control parameters are set as the standard of the process conditions, the plasma processing apparatus is operated, and five bare wafers are circulated as dummy wafers in the processing chamber 1 〇1 Internally, the plasma processing apparatus is stabilized.
-26- (23) 1276162 (表3 ) 電力 壓力 間隙 Ar CO C 4 F 6 〇2 NO. W m T o rr mm seem seem seem seem Bare Si 1 2000 100 35 300 50 10 8 Bare Si 2 2000 1 00 35 300 50 10 8 Bare Si 3 2000 100 35 300 50 10 8 Bare Si 4 2000 100 35 300 50 10 8 Bare Si 5 2000 100 35 300 50 10 8 TH-OX Si 6 2000 100 35 300 50 10 8 TH-OX Si 7 1980 100 35 300 50 10 8 TH-OX Si 8 1900 100 35 300 50 10 8 TH-OX Si 9 1980 100 35 280 50 10 8 TH-OX Si 10 2000 95 3 5 300 50 10 8 TH-OX Si 11 2 00 0 100 33 300 50 10 8 TH-OX Si 1 2 2000 100 37 300 50 10 8 TH-OX Si 13 2000 100 35 270 50 10 8 TH-OX Si 14 2000 98 35 300 50 10 8 TH-OX Si 15 2000 100 35 300 50 10 8 TH-OX Si 16 2000 100 35 300 70 10 8 TH-OX Si 17 2000 100 35 300 50 8 8 TH-OX Si 18 2000 100 35 300 50 12 8 TH-OX Si 19 1900 95 35 300 50 10 6 TH-OX Si 20 1980 102 35 300 50 10 10 TH-OX Si 2 1 1900 98 33 300 50 10 10 TH-OX Si 22 1980 98 33 300 50 10 8 TH-OX Si 23 1900 100 35 270 50 10 8 TH-OX Si 24 1980 100 35 350 50 10 8 TH-OX Si 25 2000 100 35 300 50 10 8 -27- (24) 1276162 即在將處理室101內之上下電極102、104之間隙設 定爲3 5mm後，一開始電漿處理裝置的運轉時，支持體 103透過滾珠導螺桿機構112下降至處理室101之下室 1 0 1 B的同時’由閘門閥1 0 6開放之出入口搬入僞晶圓， 載置在下部電極1〇2上。晶圓W搬入後，閘門閥106關 閉的同時，排氣系統1 1 9動作，將處理室1 〇 1內維持在特 定的真空度。藉由此排氣，APC閥門101D之開度依據排 氣量而自動進行調整。此時，由氣體導入機構115將He 氣體當成背景氣體供給，提高晶圓W和下部電極1 〇 2，具 體爲靜電夾頭1 08和晶圓W間的熱傳導性，以提高晶圓 W的冷卻效率。 然後，由製程氣體供給系統118分別以3 00sccm、 50sccm、lOsccm以及8sccm之流量供給Ar氣體、CO氣 體、<：4Ηό氣體以及Ο:氣體。此時，將處理室1〇1內之製 程氣體的壓力設定爲lOOmTorr之故，APC閥門101D之 開度便依據製程氣體供給量和排氣量而自動調整。在此狀 態下’如由高頻電源107施加2000W之高頻電力時，與 偶極子環型磁鐵105之作用相輔，發生磁控管放電，而產 生製程氣體的電漿。開始爲裸矽晶圓之故，不進行蝕刻處 理。在特定時間（例如，1分鐘）處理裸晶源後，以與搬 入時相反的操作，將處理後之晶圓W由處理室1 0 1內搬 出’以相同條件處理至後述的第5片之僞晶圓爲止。 藉由僞晶圓之處理，電漿處理裝置穩定後，處理測試 晶圓。關於最初的測試晶圓（即爲第6片晶圓），將控制 -28· (25) 1276162 參數維持爲標準値之原樣，進行蝕刻處理 之間，透過電氣量測器107C以及光學檢 資料以及光學資料分別當成檢測資料多數 未圖示出之記憶手段記憶這些量測値。然 測値，利用運算手段206算出平均値。 在處理第2片之測試晶圓時，將高; 改變爲設定値1 980W，其他的控制參數係 來進行鈾刻處理。在其間，與最初的測試 氣資料以及光學資料當成檢測資料予以量 之平均値。 在處理第3片以後的測試晶圓時，每 般分配設定各控制參數，蝕刻處理各測試 試晶圓，將電漿反映參數（電氣資料、光 測資料予以量測，算出個別之平均値。 與上述（1 )之模型相同，由此種控 之行列Xa’和電漿反映參數的平均値之行 (5 )所示之新的模型（第2工程）。-26- (23) 1276162 (Table 3) Power Pressure Clearance Ar CO C 4 F 6 〇2 NO. W m T o rr mm seem seem seem seem Bare Si 1 2000 100 35 300 50 10 8 Bare Si 2 2000 1 00 35 300 50 10 8 Bare Si 3 2000 100 35 300 50 10 8 Bare Si 4 2000 100 35 300 50 10 8 Bare Si 5 2000 100 35 300 50 10 8 TH-OX Si 6 2000 100 35 300 50 10 8 TH-OX Si 7 1980 100 35 300 50 10 8 TH-OX Si 8 1900 100 35 300 50 10 8 TH-OX Si 9 1980 100 35 280 50 10 8 TH-OX Si 10 2000 95 3 5 300 50 10 8 TH-OX Si 11 2 00 0 100 33 300 50 10 8 TH-OX Si 1 2 2000 100 37 300 50 10 8 TH-OX Si 13 2000 100 35 270 50 10 8 TH-OX Si 14 2000 98 35 300 50 10 8 TH-OX Si 15 2000 100 35 300 50 10 8 TH-OX Si 16 2000 100 35 300 70 10 8 TH-OX Si 17 2000 100 35 300 50 8 8 TH-OX Si 18 2000 100 35 300 50 12 8 TH-OX Si 19 1900 95 35 300 50 10 6 TH-OX Si 20 1980 102 35 300 50 10 10 TH-OX Si 2 1 1900 98 33 300 50 10 10 TH-OX Si 22 1980 98 33 300 50 10 8 TH-OX Si 23 1900 100 35 270 50 10 8 TH-OX Si 24 1980 100 35 350 50 10 8 TH-OX Si 25 2000 1 00 35 300 50 10 8 -27- (24) 1276162 After the gap between the upper and lower electrodes 102, 104 in the processing chamber 101 is set to 35 mm, the support body 103 is guided through the ball guide when the plasma processing apparatus is started. When the screw mechanism 112 is lowered to the lower chamber 1 0 1 B of the processing chamber 101, the dummy wafer is carried into the inlet and outlet opened by the gate valve 106, and placed on the lower electrode 1〇2. After the wafer W is carried in, the gate valve 106 is closed, and the exhaust system 1 1 9 is operated to maintain the inside of the processing chamber 1 〇 1 at a specific degree of vacuum. By this exhausting, the opening degree of the APC valve 101D is automatically adjusted in accordance with the amount of exhaust gas. At this time, He gas is supplied as a background gas by the gas introduction mechanism 115, and the wafer W and the lower electrode 1 〇2 are specifically increased, specifically, the thermal conductivity between the electrostatic chuck 108 and the wafer W to improve the cooling of the wafer W. effectiveness. Then, the process gas supply system 118 supplies Ar gas, CO gas, <:4 gas and helium: gas at a flow rate of 300 sccm, 50 sccm, 10 sccm, and 8 sccm, respectively. At this time, the pressure of the process gas in the processing chamber 1〇1 is set to 100 mTorr, and the opening degree of the APC valve 101D is automatically adjusted in accordance with the process gas supply amount and the exhaust amount. In this state, when high-frequency power of 2000 W is applied from the high-frequency power source 107, in addition to the action of the dipole ring-shaped magnet 105, magnetron discharge occurs, and plasma of the process gas is generated. It started as a bare wafer and was not etched. After processing the bare crystal source at a specific time (for example, 1 minute), the processed wafer W is carried out from the processing chamber 1 0 1 by the operation opposite to the loading, and is processed under the same conditions to the fifth sheet to be described later. Pseudo wafers up to now. After the pseudo wafer is processed, the plasma processing apparatus is stabilized and the test wafer is processed. Regarding the initial test wafer (ie, the sixth wafer), the control -28·(25) 1276162 parameters are maintained as standard ,, and between the etching process, through the electrical measuring device 107C and the optical inspection data and The optical data is used as the detection means, and most of the memory means not shown indicate these measurements. However, the average 値 is calculated by the calculation means 206. When processing the second test wafer, it will be high; change to 値1 980W, and other control parameters will be used for uranium engraving. In the meantime, the average test data and the optical data are used as the average amount of the test data. When processing the test wafer after the third chip, each control parameter is allocated and set, and each test wafer is etched, and the plasma reflection parameters (electrical data and optical measurement data are measured, and the average average enthalpy is calculated. In the same manner as the above model (1), the new model (second project) shown by the row (5) of the average of the parameters Xa' and the plasma reflection parameter is reflected.
Xa，= Ka，Ya，…（5) 接著，在將電漿處理裝置 100B以 1 00 A相同條件分配控制參數時，關於電！ 係可不用如電漿處理裝置100A般進行實 。在進行此處理 測器1 2 0將電氣 次予以測量，以 後，依據這些量 頻電力由 1 5 00W 以上述的標準値 晶圓相同，將電 測後，算出個別 次都如表3所示 晶圓，關於各測 學資料）當成檢 制參數的平均値 列Ya’作成下述 與電漿處理裝置 资處理裝置100B 驗而援用電漿處 -29- (26) 1276162 理裝置100A之上述（5 )所示的模型。即在電漿 置100B中，以與電漿處理裝置100A相同的條件 制參數之故，下述（6)式在電漿處理裝置100B之 數的行列Xb’成立。因此，電漿處理裝置100B之 數在行列Xb’時，上述（2)之模型係成爲下述 模型。 然後，由上述（1)式以及下述（5)式所示的 理裝置100A之模型和上述（2)式以及下述（7) 的電漿處理裝置1 00B之模型的關係’可以獲得下ί 式所示之模型。即在電漿處理裝置10〇Α之回歸行: 新的回歸行列Ka’，和電漿處理裝置100Β之回 Kb ， 新的回歸行列 Kb’之間，比例 (Kb，/Ka，=Kb/Ka )成立，所以 Kb ’ =Ka，Kb/Ka。如 (7 )式之Kb，中適用此關係時，則可以獲得下述（Xa, = Ka, Ya, (5) Next, when the plasma processing apparatus 100B is assigned the control parameters under the same conditions of 100 A, the electricity is concerned! It is not necessary to perform as in the plasma processing apparatus 100A. In this process, the measuring device 1 2 0 measures the electrical quantity. Thereafter, according to these frequency-frequency powers, the wafer is the same as the above-mentioned standard 値 wafer, and after the electrical measurement, the crystals are calculated as shown in Table 3. Round, regarding each test data), the average queue Ya' of the inspection parameters is prepared as follows with the plasma processing apparatus processing apparatus 100B, and the plasma is used -29-(26) 1276162. ) The model shown. That is, in the plasma set 100B, the following equation (6) is established in the rank Xb' of the number of the plasma processing apparatus 100B in the same condition as the plasma processing apparatus 100A. Therefore, when the number of the plasma processing apparatus 100B is in the rank Xb', the model of the above (2) is the following model. Then, the relationship between the model of the device 100A shown in the above formula (1) and the following formula (5) and the model of the plasma processing apparatus 100B of the above formula (2) and (7) below can be obtained. The model shown by ί. That is, in the regression line of the plasma processing apparatus 10: the new regression rank Ka', and the plasma processing apparatus 100 back Kb, the new regression rank Kb', the ratio (Kb, /Ka, = Kb / Ka ) is established, so Kb ' = Ka, Kb / Ka. When the relationship is applied to Kb of the formula (7), the following can be obtained (
Xb，=Xa， …（6)Xb,=Xa, ...(6)
Xb，=Kb，Yb， …（7)Xb,=Kb,Yb, ...(7)
Xb’ = (Ka’Kb/Ka) Yb’ ·_· (8) 因此，在以行列Xb ’分配控制參數的新製程條 關於電漿處理裝置100A，如求得模型（5 ) ’由事 得之電漿處理裝置100A的模型（1)和電獎處 1 0 ο B之模型（2 )和上述模型（7 ) ’則如（8 ) 般，可以作成關於電漿處理裝置100B之新的模型 處理裝 分配控 目的變 目的變 )式之 電漿處 式所示 1(8) 可Ka， 歸行列 關係 在下述 件中， 先所求 理裝置 式所示 (第3 -30- (27) 1276162 •X 程）。 即藉由求得使關於以新的製程條件所檢測的電-_ @ 裝置100A的控制參數之平均値（設定資料）的行列 Xa’，和電漿反映參數的平均値（檢測資料）之行列 Ya’ 相關連的回歸行列Ka，，可以作成電漿處理裝置1 〇OB的 新的模型（8)，藉由此新的模型（8)可以評估電漿處理 裝置1 00B之裝置狀態。此係意味著如依據實驗作成關於 電漿處理裝置100A之上述模型（5 )，則關於電漿處理 裝置100B，即使不重新實驗，也可以將上述（8 )式當成 電漿處理裝置100B的新的模型予以作成。 如此作成的新模型（8 )也可以記憶在電漿處理裝置 100B的解析資料記憶手段205中。藉此，在藉由電漿處 理裝置100B之平常運轉以做晶圓處理時，在由多數的電 漿反映參數之個別平均値（檢測資料）預測計算多數的控 制參數値時，可以使用解析資料記憶手段20 5之新模型 (8 ) 〇 在此情形下，藉由預測·診斷·控制手段207比較所 預測的控制參數値（所預測的設定資料）和實際設定在電 漿處理裝置100B之設定資料的變動容許範圍，在判斷爲 異常時，例如藉由處理裝置控制手段22 5以停止電漿處理 裝置100B的同時，以顯示手段224、警報器226通知異 常。 如以上說明過的，在本實施形態中，具有：例如在電 漿處理裝置100A、100B中，分別基於第i設定資料（例 -31 - (28) 1276162 如’控制梦數）而動作時，藉由多變量解析而每一電黎處 理裝置100A、100B求得由各電漿處理裝置100A、100B 之多數感測器所檢測的檢測資料（例如，電漿反映參數） 和第1設定資料的相關關係（（1 )式之Ka’、（ 2 )式之 Kb)的第1工程；以及在各電漿處理裝置100A、100B中 之當成基準處理裝置的電漿處理裝置10 0A中，依據新的 第2設定資料（例如，分配控制參數之範圍與第1設定資 料不同的新的設定資料）動作時，藉由多變量解析以求得 由電漿處理裝置1 00A的多數感測器所檢測的檢測資料和 第2設定資料的相關關係（（5 )式之Ka’）的第2工 程；以及基於在第1工程所求得之電漿處理裝置100B的 相關關係Kb，和在第1工程所求得之電漿處理裝置1 00A 的相關關係Ka，和在第2工程所求得知電漿處理裝置 100A的相關關係Ka’以求得基準處理裝置以外的其他處 理裝置之電漿處理裝置1 00B的第2設定資料和檢測資料 的相關關係（（8 )式之Kb’），作成依據如此求得之相 關關係Kb’，以評估電漿處理裝置100B的裝置狀態或者 預測處理結果之多變量解析模型（（8 )式）的第3工 程。 因此，關於藉由新製程條件的新的設定資料，在當成 基準之電漿處理裝置100A中，如對於晶圓進行電漿處理 之實驗而作成模型（5 )時，援用電漿處理裝置100A之 模型（5)，可以作成當成基準之處理裝置以外的處理裝 置，例如電漿處理裝置100B之新的模型（8)。因此，關 -32- (29) 1276162 於電漿處理裝置100B，即使不爲了作成新的模型而 的設定資料進行實驗’也可以作成新模型（8 )。藉 可以大幅減輕關於電漿處理裝置1 0 0 B之模型作成 擔。 另外’在本實施形態中，第3工程係依據對於在 工程所求得之電漿處理裝置100B的相關關係Kb的 設定資料和檢測資料的相關關係Kb ’，和對於在第1 所求得之電漿處理裝置100A的相關關係Ka的第2 所求得之相關關係Ka’的比例關係，以求得電漿處理 100B之第2設定資料和檢測資料的相關關係Kb’ 此，電漿處理裝置100B之相關關係Kb’可以不藉由 量解析而簡單算出。 另外，在本實施形態中，使可以控制電漿狀態之 的控制參數和反映電漿狀態的多數的電漿反映參數相 作成多變量解析模型。具體爲，利用電漿處理 100A、100B，分別以設定資料（控制參數等）爲目 數，同時以檢測資料（電漿反映參數等）爲說明變數 成多變量解析模型（1 ) 、（ 2 )。然後，在新的設定 中，關於電漿處理裝置100A而作成多變量解析 (5 )時，利用相關關係Kb ’和設定資料Xb ’，算出電 理裝置1 00B之檢測資料（電漿反映參數、裝置狀態 等）’可以作成關於電漿處理裝置100B之藉由新的 資料的多變量解析模型（8 )。 另外，爲了利用PLS法以作成多變量解析模型 以新 此， 的負 第1 第2 工程 工程 裝置 。藉 多變 多數 關以 裝置 的變 ，作 資料 模型 漿處 參數 設定 ，即 -33- (30) 1276162 使實驗數少，也可以高精度預測、評估上述各參數。另 外’藉由以電漿處理裝置100B之預測値爲主成分分析， 可以綜合地評估電漿處理裝置1 00B之運轉狀態。 另外，進行利用P L S法以求得設定資料和檢測資料 之相關關係的多變量解析，可以少的資料進行高精度的多 變量解析以作成多變量解析模型。 接著’ 一面參考圖面一面說明本發明之第2實施形 態。第3圖係顯示關於本實施形態之控制系統整體的槪略 構造方塊圖。此控制系統3 0 0係由藉由網路3 2 0以連接主 機裝置3 1 0和多數的電漿處理裝置1 00A.....100N而構 成。電漿處理裝置100A.....100N係分別與第1圖所示 者同樣的構造故，省略其之詳細說明。另外，電漿處理裝 置1〇〇 A.....100N係分別具備如第2圖所示之多變量解 析手段200。另外，在本實施形態中，例如第2圖所示之 多變量解析手段200、處理裝置控制手段225、第3圖所 示之發送接收裝置1 5 0係擔負當成處理裝置的控制裝置之 任務。 主機裝置3 1 0係至少具備：進行各種運算的運算手段 3 1 2、記億上述之PLS法等之多變量解析程式的多變量解 析程式記憶手段3 1 4、記憶解析結果或解析所必要的增要 的解析資料記憶手段3 1 6、透過上述網路320與各電漿處 理裝置100A.....10ON進行資料之交換的發送接收手段 318。另外，上述主機裝置310例如可以半導體製造工廠 的主電腦構成，或以連接於此主電腦之個人電腦構成。 -34- (31) 1276162 電漿處理裝置100A.....100N分別具備：進行在各 電漿處理裝置100A.....100N和主機裝置310之間或考 各電漿處理裝置100A.....1 00N間的各種資料的發送接 收的發送接收裝置150A.....150N、輸入控制參數（設 定資料）等之各種資料用的輸入手段152A.....152N。 上述發送接收裝置1 5 0 A.....1 5 ON係分別與第2圖所示 之多變量解析手段200連接，可以與各電漿處理裝置 100A.....100N之多變量解析手段200進行資料之交 換。 上述網路3 20係可以雙向通訊以連接主機裝置310、 各電漿處理裝置100A.....100N之網路，典型上可舉網 際網路等之公眾線路網。另外，網路3 20在上述公眾線路 網之外，也可以爲 WAN(Wide Area Network :廣域網 路）、LAN(Local Area Network :區域網路）、1卩-VPN(Internet Protocol-Virtual Private Network ··網際網 路通訊協定-虛擬私人網路）等之封閉線路網。另外，對 於網路 3 20 之連接媒體，可以爲藉由 FDDI(Fiber Distributed Data Interface:光纖分配數據介面）等之光纖 電纜、藉由Ethernet之同軸電纜或者對絞電纜、或者藉由 IEEE8 02 1 lb等之無線等，不管有線無線，也可以爲衛星 網路等。 各電漿處理裝置1 00在以所期望的製程條件進行蝕刻 處理時，藉由將作成評估裝置狀態用之新模型所必要的資 料，由主機裝置3 1 0透過發送接收裝置1 5 0而發送給所期 -35- 00 (32) 1276162 望之電漿處理裝置100，可以減輕以該電漿處理裝置1 的多變量解析手段200作成模型時的負擔。而且，由藉 電漿處理裝置100之實際的晶圓處理時，利用新的模型 評估裝置狀態，依據因應其結果而由預測·診斷·控制 段2 07所輸出的指示，藉由處理裝置控制手段225得以 制電漿處理裝置100。 接著，參考圖面說明此種控制系統300之處理。控 系統3 00之處理例如係如在第1實施形態說明過的，可 將在電漿處理裝置100A所作成的新模型援用於電漿處 裝置100B，以作成電漿處理裝置10 0B的新模型之例子 第4圖〜第6圖係顯示作成電漿處理裝置100B之 模型時的處理之動作流程。更詳細爲第4圖〜第6圖係 示以電漿處理裝置100A爲基準處理裝置，以電漿處理 置100B.....100N爲基準處理裝置以外的處理裝置時 主機裝置、基準處理裝置、基準處理裝置以外的處理裝 的動作流程。另外，在第4圖〜第6圖中，基準處理裝 以外的處理裝置係以電漿處理裝置100B之處理爲代表 記載。在關於其他的電漿處理裝置 1 00 C.....100N要 成新模型時，也進行與電漿處理裝置100B同樣的動作^ 首先，如第 4圖所示般，求得各電漿處理裝 !〇〇A.....10 ON之回歸行列Ka.....Kn。以下說明具 之處理。Xb' = (Ka'Kb/Ka) Yb' ·_· (8) Therefore, in the new process bar for assigning control parameters in the row Xb ' with respect to the plasma processing apparatus 100A, as in the case of obtaining the model (5) The model (1) of the plasma processing apparatus 100A and the model (2) of the electric prize office 10B and the above model (7)' can be made into a new model regarding the plasma processing apparatus 100B as in (8). The treatment of the distribution control purpose of the change of the type of plasma type 1 (8) can be Ka, the relationship between the ranks in the following parts, as shown in the first device (3-30- (27) 1276162 • X procedure). That is, by judging the rank Xa' of the average 値 (setting data) of the control parameters of the electric-_@ device 100A detected by the new process conditions, and the average 値 (detection data) of the plasma reflection parameter Ya's associated regression rank Ka, can be made into a new model (8) of the plasma processing apparatus 1 〇OB, by which the device state of the plasma processing apparatus 100B can be evaluated. This means that if the above model (5) regarding the plasma processing apparatus 100A is made experimentally, with respect to the plasma processing apparatus 100B, the above formula (8) can be regarded as a new type of the plasma processing apparatus 100B even without re-experimentation. The model is made. The new model (8) thus created can also be stored in the analytical data memory means 205 of the plasma processing apparatus 100B. Thereby, when the wafer processing is performed by the normal operation of the plasma processing apparatus 100B, the analysis data can be used when the majority of the control parameters are predicted by the average average 値 (detection data) of the majority of the plasma reflection parameters. The new model (8) of the memory means 20 5 In this case, the predicted control parameter 値 (predicted setting data) and the setting actually set in the plasma processing apparatus 100B are compared by the prediction/diagnosis/control means 207. When it is determined that the data is abnormal, for example, the processing device control means 22 stops the plasma processing apparatus 100B, and the display means 224 and the alarm 226 notify the abnormality. As described above, in the present embodiment, for example, in the plasma processing apparatuses 100A and 100B, when operating based on the ith setting data (example -31 - (28) 1276162 such as "control dream number"), Each of the electric power processing apparatuses 100A and 100B determines the detection data (for example, the plasma reflection parameter) detected by the majority of the sensors of the respective plasma processing apparatuses 100A and 100B by the multivariate analysis, and the first setting data. The first relationship of the correlation (K' of the formula (1), and the Kb of the formula (2); and the plasma processing apparatus 100A which is the standard processing apparatus in each of the plasma processing apparatuses 100A and 100B, according to the new When the second setting data (for example, a new setting data whose distribution control parameter range is different from the first setting data) is operated, multivariate analysis is performed to obtain detection by a plurality of sensors of the plasma processing apparatus 100A. The second relationship between the detection data and the second setting data (Ka' of the formula (5)); and the correlation Kb based on the plasma processing apparatus 100B obtained in the first project, and the first project Plasma processing device 100 00A The correlation relationship Ka and the correlation relationship Ka' of the plasma processing apparatus 100A are obtained by the second engineering unit to obtain the second setting data and the detection data of the plasma processing apparatus 100B of the processing apparatus other than the reference processing apparatus. The correlation (Kb' of the formula (8)) is based on the correlation Kb' thus obtained to evaluate the device state of the plasma processing apparatus 100B or the multivariate analysis model ((8)) The third project. Therefore, regarding the new setting data by the new process conditions, in the plasma processing apparatus 100A which is the standard, when the model (5) is performed on the experiment of the plasma processing on the wafer, the plasma processing apparatus 100A is used. The model (5) can be made as a processing device other than the processing device as a reference, for example, a new model (8) of the plasma processing device 100B. Therefore, the -32-(29) 1276162 can be made into a new model (8) in the plasma processing apparatus 100B even if the experiment is not performed for the setting data for creating a new model. By borrowing, the model for the plasma processing apparatus 100B can be greatly reduced. In addition, in the present embodiment, the third engineering is based on the correlation Kb' between the setting data and the detection data of the correlation Kb of the plasma processing apparatus 100B obtained in the engineering, and the first correlation is obtained. The proportional relationship of the correlation relationship Ka' obtained by the second correlation of the correlation relationship Ka of the plasma processing apparatus 100A, in order to obtain the correlation relationship Kb' between the second setting data of the plasma processing 100B and the detection data, the plasma processing apparatus The correlation Kb' of 100B can be easily calculated without the amount analysis. Further, in the present embodiment, a control parameter that can control the plasma state and a plurality of plasma reflection parameters that reflect the plasma state are made into a multivariate analysis model. Specifically, the plasma processing 100A, 100B is used to set the data (control parameters, etc.) as the mesh number, and the detection data (plasma reflection parameters, etc.) are used as explanatory variables to form a multivariate analysis model (1), (2) . Then, in the new setting, when the multi-variable analysis (5) is performed on the plasma processing apparatus 100A, the correlation data Kb' and the setting data Xb' are used to calculate the detection data of the electrical processing device 100B (plasma reflection parameter, The device state, etc. ' can be made into a multivariate analysis model (8) on the new material of the plasma processing apparatus 100B. In addition, in order to use the PLS method to create a multivariate analysis model, the new first, second engineering equipment. By changing the number of devices, the parameters of the model are set, that is, -33- (30) 1276162. The number of experiments is small, and the above parameters can be predicted and evaluated with high precision. Further, by using the predicted enthalpy of the plasma processing apparatus 100B as the main component analysis, the operational state of the plasma processing apparatus 100B can be comprehensively evaluated. In addition, multivariate analysis using the P L S method to obtain the correlation between the set data and the detected data is performed, and highly accurate multivariate analysis can be performed with a small amount of data to create a multivariate analysis model. Next, the second embodiment of the present invention will be described with reference to the drawings. Fig. 3 is a block diagram showing the overall configuration of the control system of the present embodiment. The control system 300 is constructed by connecting the host device 310 and the plurality of plasma processing devices 100A.....100N via the network 3 2 0. The plasma processing apparatuses 100A, ..., 100N are the same as those shown in Fig. 1, and detailed description thereof will be omitted. Further, the plasma processing apparatus 1A.....100N each has a multivariate analysis means 200 as shown in Fig. 2 . Further, in the present embodiment, for example, the multivariate analysis means 200, the processing means control means 225 shown in Fig. 2, and the transmission/reception apparatus 150 shown in Fig. 3 are responsible for the task of the control means of the processing means. The host device 3 1 0 includes at least a calculation means for performing various calculations. 3 1 2. A multivariate analysis program memory means for the multivariate analysis program such as the PLS method described above, and a memory analysis result or analysis necessary for the analysis. The analysis data storage means 316 of the additional information is a transmission/reception means 318 for exchanging data with each of the plasma processing apparatuses 100A.....10ON via the network 320. Further, the host device 310 may be constituted by, for example, a host computer of a semiconductor manufacturing factory or a personal computer connected to the host computer. -34- (31) 1276162 The plasma processing apparatuses 100A.....100N are respectively provided between each of the plasma processing apparatuses 100A.....100N and the host apparatus 310 or each of the plasma processing apparatuses 100A. ....1. The input/receiving means 152A.....152N for various data such as the transmission/reception device 150A.....150N for transmitting and receiving various data between 00N and input control parameters (setting data). The transmitting/receiving apparatus 1 500 A.....1 5 ON is connected to the multivariate analyzing means 200 shown in Fig. 2, and can be analyzed with the multivariate of each of the plasma processing apparatuses 100A.....100N. Means 200 exchanges data. The network 3 20 can be bidirectionally connected to connect to the host device 310, the plasma processing devices 100A.....100N, typically a public line network such as the Internet. In addition, the network 3 20 may be a WAN (Wide Area Network), a LAN (Local Area Network), or a VPN (Internet Protocol-Virtual Private Network) in addition to the above public line network. · Closed line network such as Internet Protocol - Virtual Private Network. In addition, the connection medium of the network 3 20 may be a fiber cable such as FDDI (Fiber Distributed Data Interface), a coaxial cable or a twisted pair cable by Ethernet, or IEEE 8 02 1 lb. Waiting for wireless, etc., regardless of wired or wireless, it can also be a satellite network. When the plasma processing apparatus 100 performs the etching process under the desired process conditions, the data necessary for creating a new model for evaluating the state of the device is transmitted from the host device 310 through the transmitting and receiving device 150. The desired plasma processing apparatus 100 of the -35-00 (32) 1276162 can reduce the burden when the multi-variable analysis means 200 of the plasma processing apparatus 1 is used as a model. Further, when the actual wafer processing by the plasma processing apparatus 100 is performed, the state of the apparatus is evaluated by a new model, and the instruction outputted by the prediction/diagnosis/control section 207 according to the result is controlled by the processing apparatus. The 225 is capable of making a plasma processing apparatus 100. Next, the processing of such a control system 300 will be described with reference to the drawings. The processing of the control system 300 is, for example, as explained in the first embodiment, and a new model created by the plasma processing apparatus 100A can be applied to the plasma apparatus 100B to create a new model of the plasma processing apparatus 10B. Examples 4 to 6 show an operational flow of processing when the model of the plasma processing apparatus 100B is created. More specifically, FIGS. 4 to 6 show a host device and a reference processing device when the plasma processing device 100A is used as a reference processing device and the plasma processing device 100B.....100N is used as a processing device other than the reference processing device. The flow of the processing of the processing equipment other than the reference processing device. Further, in the fourth to sixth figures, the processing device other than the reference processing device is described by the processing of the plasma processing device 100B. When the other plasma processing apparatus 100 C.....100N is to be a new model, the same operation as the plasma processing apparatus 100B is performed. First, as shown in Fig. 4, each plasma is obtained. Processing loading! 〇〇A.....10 ON returning ranks Ka.....Kn. The following instructions are handled.
基準處理裝置之電漿處理裝置100Α在求得回歸行 Ka用之設定資料（例如，控制參數）由輸入手段1 5 2 A 由 以 手 控 制 舉 理 〇 新 顯 裝 之 置 置 而 作 置 體 列 被 -36- (33) 1276162 輸入而設定時，在步驟S 11 0中，依據此設定資料處理晶 圓W，取得檢測資料（例如，電漿反映參數），將這些設 定資料、檢測資料透過網路3 20而發送給主機裝置3 1 0。 另一方面，基準處理裝置以外的處理裝置之例如電漿 處理裝置100Β在求得回歸行列Kb用之設定資料（例 如，控制參數）由輸入手段152 A被輸入而設定時，在步 驟S 5 1 0中，依據此設定資料處理晶圓W，取得檢測資料 (例如，電漿反映參數），將這些設定資料、檢測資料透 過網路320而發送給主機裝置310。 主機裝置310在步驟 S210中，由電漿處理裝置 1 〇〇A.....1 00N接收設定資料、檢測資料，記憶在解析 資料記億手段316。接著，在步驟S220中，藉由運算手 段3 1 2求的所接收之設定資料的每一晶圓之平均値，以這 些爲目的變數Xa.....Xn，記憶在解析資料記憶手段 3 1 6，同時，藉由運算手段3 1 2求得所接收之檢測資料的 每一晶圓之平均値，以這些爲說明變數Ya.....Yn，記 憶在解析資料記憶手段3 1 6。 接著，主機裝置310在步驟S230中，依據由多變量 解析程式記億手段3 1 4之藉由P L S法的程式，與上述第1 實施形態相同地’藉由運算手段3 1 2由設定資料（目的變 數）Xa.....Xn、檢測資料（說明變數）Ya.....Υη求 得各電漿處理裝置 100A..... 100N 的回歸行列The plasma processing apparatus 100 of the reference processing device determines the setting data (for example, the control parameter) for the regression line Ka by the input means 1 5 2 A by the hand control and the new display. -36- (33) 1276162 When input is set, in step S11 0, the wafer W is processed according to the setting data, and the detection data (for example, the plasma reflection parameter) is obtained, and the setting data and the detection data are transmitted through the network. 3 20 is sent to the host device 3 1 0. On the other hand, for example, the plasma processing apparatus 100 of the processing device other than the reference processing device is configured to input the setting data (for example, the control parameter) for the regression row Kb by the input means 152 A, in step S51. In the case of 0, the wafer W is processed according to the setting data, and the detection data (for example, the plasma reflection parameter) is acquired, and the setting data and the detection data are transmitted to the host device 310 via the network 320. In step S210, the host device 310 receives the setting data and the detection data from the plasma processing device 1 〇〇A....1 00N, and stores it in the analysis data recording means 316. Next, in step S220, the average 値 of each wafer of the received setting data obtained by the computing means 312 is used for the analytic data memory means 3 with the purpose variables Xa.....Xn 16. At the same time, the average 値 of each wafer of the received detection data is obtained by the operation means 3 1 2, and these are used as explanatory variables Ya.....Yn, and are stored in the analytical data memory means 3 1 6 . Next, in step S230, the host device 310 performs the setting data by the arithmetic means 3 1 2 in the same manner as in the first embodiment, in accordance with the program of the PLS method by the multivariate analysis program. The target variable) Xa.....Xn, the detection data (description variable) Ya.....Υ, the regression rank of each of the plasma processing apparatuses 100A.....100N is obtained.
Ka.....Kn，記憶在解析資料記憶手段3 1 6。接著，在步 驟 S240中，將這些設定資料Xa.....Xn、檢測資料 -37- (34) 1276162Ka.....Kn, memory in the analytical data memory means 3 1 6. Next, in step S240, these setting data Xa.....Xn, detection data -37-(34) 1276162
Ya.....Υη、回歸行列Ka、…、Κη透過網路3 20而 給各電漿處理裝置.........100Ν。 基準處理裝置之電漿處理裝置iOOA在步驟 中，由主機裝置3 1 0接收設定資料X &、檢測資料γ a 歸行列Ka ’當成如上述（1 )所示之模型予以記憶 外，基準處理裝置以外的處理裝置之例如電漿處理 100B在步驟S 520中，由主機裝置31〇接收設定 Xb、檢測資料Yb、回歸行列Kb，當成如上述（2 ) 之模型予以記憶。 接著，如第5圖所示般，作成基準處理裝置之電 理裝置100A的新模型。以下說明具體之處理。 電漿處理裝置100A在求得回歸行列Ka，用之新 定資料（例如，控制參數）由輸入手段1 5 2 A被輸入 定時，在步驟S 1 3 0中，依據此設定資料處理晶圓w 得新的檢測資料（例如，電漿反映參數），將這些新 定資料、新的檢測資料透過網路3 20而發送給主機 3 10° 主機裝置310在步驟S310中，由基準處理裝置 漿處理裝置1 00 A接收新的設定資料、新的檢測資料 憶在解析資料記憶手段3 1 6。接著，在步驟S3 20中 由運算手段3 1 2求得接收之新的設定資料的每一晶圓 均値，將這些當成說明變數Xa’ .....Xn’ ’記憶在 資料記億手段3 1 6，同時，藉由運算手段3 1 2求得接 新的檢測資料的每一晶圓的平均値，將這些當成目的 發送 S 1 20 、回 。另 裝置 資料 所示 漿處 的設 而設 ，取 的設 裝置 之電 ，記 ，藉 之平 解析 收之 變數 -38 - (35) 1276162Ya.....Υη, the return ranks Ka,...,Κη are given to each plasma processing apparatus through the network 3 20...100Ν. The plasma processing apparatus iOOA of the reference processing device receives, in the step, the setting data X & the detection data γ a is returned to the row Ka ' as the model shown in the above (1), and the reference processing is performed. For example, the plasma processing 100B of the processing device other than the device receives the setting Xb, the detected data Yb, and the regression rank Kb from the host device 31A in step S520, and is memorized as the model of the above (2). Next, as shown in Fig. 5, a new model of the power device 100A of the reference processing device is created. The specific processing will be described below. The plasma processing apparatus 100A obtains the regression matrix Ka, and uses the new data (for example, control parameters) to be input by the input means 1 2 2 A. In step S130, the wafer is processed according to the setting data. New test data (for example, plasma reflection parameter) is obtained, and the new data and the new test data are sent to the host through the network 3 20. The 10 10 host device 310 is processed by the reference processing device in step S310. The device 100 A receives the new setting data, and the new detection data is recalled in the analytical data memory means 3 16 . Next, in step S320, each wafer of the received new setting data is obtained by the computing means 3 1 2, and these are used as the explanatory variables Xa' ..... Xn' 'memory in the data recording means At the same time, the average value of each wafer of the new detection data is obtained by the operation means 3 1 2, and these are transmitted as S 1 20 and back. According to the design of the slurry, the device is set up, and the electrical, recording, and borrowing of the device are taken to analyze the variable -38 - (35) 1276162
Ya ’ .....Yn，，記憶在解析資料記憶手段3 1 6。 接著，主機裝置310在步驟S330中，依據多變量解 析程式記憶手段314之藉由PLS法的程式，與上述第！ 實施形態相同地，由新的設定資料（目的變數）Xa·、新 的設定資料（說明變數）Ya’藉由運算手段312以求得電 漿處理裝置1 00A的回歸行列Ka’，記憶在解析資料記憶 手段316。接著，在步驟S 340中，將這些新的設定資料 Xa’、新的檢測資料Ya’、新的回歸行歹U Ka’透過網路320 而發送給電漿處理裝置100A。 基準處理裝置之電漿處理裝置100A在步驟 S140 中，由主機裝置310接收設定資料Xa’、檢測資料Ya’、 回歸行列Ka’，當成新模型予以記憶。 接著，如第6圖所示般，求得基準處理裝置以外的處 理裝置之例如電漿處理裝置的模型。基準處理裝置 以外的處理裝置之新模型係依據基準處理裝置的新模型而 求得之故，在基準處理裝置以外的處理裝置中，不需要重 新進行對於晶圓之電漿處理。以下說明具體之處理。 電漿處理裝置100B在步驟S 5 3 0中，在求得回歸行 列Kb ’用之設定資料（與求得回歸行列Kb ’用之設定資料 相同的設定資料）由輸入手段1 5 2 B被輸入時’將此設定 資料透過網路320而發送給主機裝置310。 主機裝置310在步驟S4 10中，由基準處理裝置以外 的處理裝置之電漿處理裝置1 〇 〇 B接收新的設定資料，記 憶在解析資料記憶手段3 1 6，藉由運算手段3 1 2求得接收 -39- (36) 1276162 的新設定資料的每一晶圓之平均値，將這些當成設定資料 (說明變數）Xb， .....Xn，，記億在解析資料記憶手段 3 16° 接著，主機裝置310在步驟S420中’由基準處理裝 置以外的處理裝置之回歸行列（Kb.....Kn )、新的回 歸行列（Kb， .....Kn，），和基準處理裝置的回歸行列 (Ka )、基準處理裝置的新的回歸行列（Ka’）之比例關 係（例如（Kb，/Ka，=Kb/Ka))由運算手段312分別求得 新的回歸行列（K b ’ .....Κ η ’）。例如，電漿處理裝置 100Β之（7 )式所示的新的回歸行列 Kb’係藉由 Kb，= Ka’Kb/Ka所求得。藉此，在求得基準處理裝置以外 的處理裝置之新的回歸行列時’不需要重新進行PLS法 等之多變量解析處理，可以簡單求得。 接著，主機裝置310在步驟S43 0中，依據上述（7) 式所示之模型，由新的設定資料（Xb’ 、…、Xn，）、新 的回歸行列（Kb’ .....Kn’）算出新的檢測資料 (Yb ’ .....Yn ’），記憶在解析資料記憶手段3 1 6，將 這些新的設定資料（X b ’ .....X η ’）、新的回歸行列 (Kb’ .....Kn’）算出新的檢測資料（Yb’ ......... 透過網路 320分別發送給對應之電漿處理裝置 1 00Β.....1 00Ν。 例如在電漿處理裝置iOOB中’在步驟S540中，由 主機裝置3 10接收新的設定資料（xb’ .....χη，）、新 的回歸行列（Kb， .....Kn’）算出新的檢測資料 -40- (37) 1276162 (Yb’ .....Yn’），當成如上述（8 )式所示之新 予以記憶。如此，在基準裝置以外的處理裝置中，得 成適合個別之處理裝置的新模型。 接著，參考圖面說明依據如此獲得之新模型，評 置狀態時的控制系統之處理。第7圖係顯示依據各電 理裝置所分別作成的新模型以評估裝置狀態時的主機 之動作流程和各電漿處理裝置的動作流程。 首先，在某電漿處理裝置100中，在步驟S610 對於設定資料的標準條件之容許變動範圍一被輸入時 記憶此容許變動範圍。此容許變動範圍係使用於判定 狀態爲正常或異常用之臨界値，例如，設爲對於分配 新模型用之設定資料，例如控制參數時的各控制參數 準値的最大値和最小値之範圍。 接著，上述電漿處理裝置100在步驟S620中， 處理晶圓用的設定資料（標準條件，例如表1所示之 値）由輸入手段1 52被輸入時，依據此設定資料，電 理晶圓W，取得每一晶圓所量測之設定資料和檢測資 將這些設定資料、檢測資料透過網路320發送給主機 3 10° 主機裝置310在步驟S710中，由上述電漿處理 1 00每一晶圓的接收設定資料和檢測資料而記憶在解 料記憶手段3 1 6。然後，求得個別的平均値，當成設 料（目的變數）X，、檢測資料（說明變數）Y’，記憶 析資料記憶手段3 1 6。接著，主機裝置3 1 0在步驟 模型 以作 估裝 漿處 裝置 中， ，便 裝置 作成 的標 實際 標準 漿處 料， 裝置 裝置 析資 定資 在解 S720 -41 - (38) 1276162 中，將設定資料X，、檢測資料Y ’發送給上述電漿處理裝 置 1 00。 電漿處理裝置100在步驟S63 0中，接收設定資料 X ’、檢測資料Υ，，將這些當成實際的設定資料X 〇 b s ’、實 際的檢測資料Y 〇 b s ’而記憶在解析資料憶手段2 0 5。接 著，在步驟S640中，在上述（8)式所示之新模型適用實 際的檢測資料Yobs，，算出預測設定資料xPre’，記憶在 解析資料記憶手段205。 接著，上述電漿處理裝置100在步驟S650中，藉由 預測設定資料Xpre，對於實際的設定資料X〇bs’是否在容 許變動範圍內，以判定爲正常或異常。例如，預測設定資 料Xpre’對於實際的設定資料Xobs’如在容許變動範圍 內，則判斷爲正常，如超過容許變動範圍，則判斷爲異 常。在判斷爲異常時，在步驟S 6 6 0中，例如藉由處理裝 置控制手段225，停止上述電漿處理裝置1〇〇，同時，以 顯示手段224、警報器226通知異常。 如此，主機裝置310依據來自各電漿處理裝置的資料 以求得平均値，而進行多變量解析處理’所以可以大幅減 輕各電漿處理裝置的計算處理負擔。另外，在各電漿處理 裝置中，不需要暫時記憶在進行電漿處理時所獲得的大量 設定資料或檢測資料等，也不需要多變量解析程式，所以 不需要爲此之記憶手段。藉此，可以使各電漿處理裝置之 構造變得簡單，另外可以抑制製造成本。 另外，在第2實施形態中，雖就以各電漿處理裝置側 -42- (39) 1276162 的新模型來判斷裝置狀態時做說明，但是並不一定 此，新模型也記憶在主機裝置310之故，也可在主 3 1〇側判定各電漿處理裝置 100A.....100N之 態。在此情形時，於判定爲異常時，可將異常判定 送給各電漿處理裝置100A.....1〇0N。各電漿處 1〇〇 A.....100N可以依據異常判定資訊，例如藉 裝置控制手段225以停止處理裝置，藉由顯示手段 警報器226以通知異常。如依據此，可以主機裝置 中監視各電漿處理裝置之裝置狀態。 以上，雖一面參考所附圖面一面說明關於本發 適的實施形態，但是不用說，本發明並不限定於 子。只要是該行業者，在申請專利範圍所記載的範 很淸楚可以想到各種之變更例或者修正例，關於那 也應理解係屬於本發明之技術範圍。 例如，作爲上述第1及第2實施形態的設定資 如利用第2實施形態的新模型以判定裝置狀態時般 漿處理晶圓時，可以使用藉由控制參數量測器22 1 之設定資料，另外，也可以使用由輸入手段1 52所 設定資料本身。在此情形下，設定資料之全部都可 制參數量測器22 1測量時，雖可以使用藉由控制參 器22 1所測量的設定資料，但是在設定資料中，含 藉由控制參數量測器22 1所測量者時，以利用所輸 定資料本身爲有效。 另外，在上述實施形態的多變量解析中，雖然 限制於 機裝置 裝置狀 資訊發 理裝置 由處理 224、 310集 明之合 此種例 疇中， 些當然 料，係 ，在電 所測量 輸入的 藉由控 數量測 有無法 入的設 不使用 -43- (40) 1276162 裝置狀態參數，但是也可以將裝置狀態參數當成目的變數 或者說明變數使用。另外，在上述實施形態中，於構築模 型時，作爲目的變數之控制參數，雖使用製程氣體流量、 電極間的間隙及處理室內的壓力，但是只要是可以控制的 參數，並不限定於這些參數。 另外，作爲裝置狀態參數，雖利用可變電容器電容、 高頻電壓、APC開度，但是只要是顯示裝置狀態參數之可 測量的參數，並不限定於這些。另外，作爲反映電漿狀態 之電漿反映參數，雖使用基於電漿之電氣資料及光學資 料，但是只要是反映電漿狀態之參數，並不限定於這些。 另外，作爲電氣資料，雖使用基本波及高次諧波（至4倍 波爲止）之高頻電壓、高頻電流，但是並不限定於這些。 另外，也可將來自組裝在電漿處理裝置內的測量晶圓 完成結果之手段（例如，掃描量測儀）的輸出資料當成檢 測資料使用。具體爲作爲檢測資料可以使用形成在晶圓上 之膜的膜厚、蝕刻晶圓上的被處理膜時的蝕刻量或其面內 均勻性等之特徵値。另外，在本實施形態中，雖係每一晶 圓求得電漿反映參數之資料的平均値，使用此平均値而每 一晶圓預測控制參數及裝置狀態參數，但是也可以利用一 片之晶圓處理中的即時的電漿反映參數，即時地預測控制 參數及裝置狀態參數。 另外，在上述實施形態中，雖利用有磁場平行平板型 電漿處理裝置，但是只要是具有控制參數和電漿反映參數 及/或者裝置狀態參數之裝置，都可以適用本發明。 -44- (41) 1276162 如以上詳細說明般，如依據本發明，即時每一處理裝 置其製程特性或者製程條件有差異，如就一個處理裝置作 成模型，可將該模型援用於同種類的其他處理裝置，不需 要每一處理裝置重新作成模型，能夠提供可以減輕模型作 成之負擔的處理裝置之多變量解析模型作成方法及處理裝 置用之多變量解析方法。 產業上之利用可能性 本發明例如可以適用於電漿處理裝置等之處理裝置之 多變量解析模型作成方法、處理裝置用之多變量解析方 法、處理裝置之控制裝置、處理裝置之控制系統。 【圖式簡單說明】 第1圖係顯示關於本發明之第1實施形態的電漿處理 裝置之槪略構造的剖面圖。 第2圖係顯示第1圖所示之電漿處理裝置的多變量解 析手段之一例的方塊圖。 ' 第3圖係顯示關於本發明之第2實施形態的處理裝置 控制系統的構造方塊圖。 第4圖係說明關於本實施形態之處理控制系統的模型 作成時的動作流程圖。 第5圖係說明關於本實施形態之處理裝置控制系統的 模型作成時之動作流程圖，爲第4圖之延續。 第6圖係說明關於本實施形態之處理裝置控制系統的 模型作成時之動作流程圖，爲第5圖之延續。 -45- (42) 1276162 第7圖係說明藉由關於本實施形態5 統的模型以進行控制時的動作流程圖。 主要元件對照表 100 :電漿處理裝置，101 :處理室 極，104 :上部電極（淋浴頭），105 : il 107:高頻電源，107Α:匹配器，107Β: 電氣量測器，1 1 8 :製程氣體供給系統 器，200:多變量解析手段，201:多變量 段，205 :解析資料記憶手段，206 :運： 測·診斷·控制手段，22 1 :控制參數量 反映參數量測器，223 :裝置狀態參數量 裝置控制手段，3 00 :控制系統，3 20 :網 處理裝置控制系 |，1 0 2 :下部電 !極子環型磁鐵， 瓦特計，107C : ，120 :光學檢測 t解析程式記憶手 障手段，207 :預 測器，222 :電漿 測器，225 :處理 路’ W :晶圓 -46-Ya ’ .....Yn,, memory is in the analytical data memory means 3 16 . Next, in step S330, the host device 310 analyzes the program by the PLS method according to the multivariate analysis program memory means 314, and the above-mentioned! In the same manner, the new setting data (destination variable) Xa· and the new setting data (description variable) Ya' are obtained by the arithmetic means 312 to obtain the regression row Ka' of the plasma processing apparatus 100A, and the memory is analyzed. Data memory means 316. Next, in step S340, the new setting data Xa', the new detection data Ya', and the new return line U Ka' are transmitted to the plasma processing apparatus 100A through the network 320. In step S140, the plasma processing apparatus 100A of the reference processing device receives the setting data Xa', the detection data Ya', and the regression matrix Ka' from the host device 310, and memorizes it as a new model. Next, as shown in Fig. 6, a model of a plasma processing apparatus such as a plasma processing apparatus other than the reference processing apparatus is obtained. The new model of the processing device other than the reference processing device is obtained based on the new model of the reference processing device, and it is not necessary to re-process the plasma for the wafer in the processing device other than the reference processing device. The specific processing will be described below. In step S530, the plasma processing apparatus 100B inputs the setting data for the regression row Kb' (the same setting data as the setting data for obtaining the regression row Kb') by the input means 1 5 2 B. This setting data is transmitted to the host device 310 via the network 320. In step S410, the host device 310 receives the new setting data from the plasma processing device 1B of the processing device other than the reference processing device, and stores it in the analytical data memory means 3 1 6 by means of the arithmetic means 3 1 2 You must receive the average 每一 of each wafer of the new setting data of -39- (36) 1276162, and treat these as the setting data (description variable) Xb, .....Xn, and remember the data memory means 3 16 ° Next, the host device 310 'regresses the ranks (Kb...Kn), the new regression ranks (Kb, .....Kn)), and the reference by the processing devices other than the reference processing device in step S420. The proportional relationship between the regression row (Ka) of the processing device and the new regression row (Ka') of the reference processing device (for example, (Kb, /Ka, = Kb/Ka)) is determined by the arithmetic means 312 to obtain a new regression row ( K b ' .....Κ η '). For example, the new regression rank Kb' shown by the equation (7) of the plasma processing apparatus 100 is obtained by Kb, = Ka'Kb/Ka. As a result, when a new regression sequence of the processing device other than the reference processing device is obtained, it is not necessary to perform the multivariate analysis processing such as the PLS method again, and it is possible to easily obtain it. Next, the host device 310, in step S430, according to the model shown in the above formula (7), from the new setting data (Xb', ..., Xn,), the new regression row (Kb' ..... Kn ') Calculate new test data (Yb ' .....Yn '), memorize the data memory means 3 1 6 , and add these new settings (X b ' .....X η '), new The regression sequence (Kb' .....Kn') calculates new detection data (Yb' ... ... is transmitted to the corresponding plasma processing device through the network 320, respectively. .1 00 Ν. For example, in the plasma processing apparatus iOOB 'in step S540, the new setting data (xb' ..... χη,) is received by the host device 3 10, a new regression rank (Kb, ... ..Kn') Calculate the new test data -40 - (37) 1276162 (Yb' .....Yn'), as a new memory as shown in the above formula (8). Thus, outside the reference device In the processing device, a new model suitable for the individual processing device is obtained. Next, the processing of the control system when the state is evaluated based on the new model thus obtained will be described with reference to the drawings. A new model created by the electrical device to evaluate the operation flow of the host and the operation flow of each plasma processing device in the state of the device. First, in a certain plasma processing device 100, the standard conditions for setting data in step S610 are The permissible variation range is memorized when the allowable variation range is input. This permissible variation range is used as a threshold for determining whether the state is normal or abnormal, for example, for setting data for assigning a new model, for example, each of control parameters. The range of the maximum 値 and the minimum 控制 of the control parameter is controlled. Next, the plasma processing apparatus 100 processes the setting data for the wafer (standard conditions, for example, 値 shown in Table 1) by the input means 1 52 in step S620. When input, according to the setting data, the wafer W is obtained, and the setting data and the detection resources measured by each wafer are obtained, and the setting data and the detection data are transmitted to the host through the network 320. The 10° host device 310 is In step S710, the receiving setting data and the detection data of each wafer are processed by the plasma processing and stored in the de-storing memory means 3 16 . Then Find the average average 値, as the material (destination variable) X, the test data (describe the variable) Y', and memorize the data memory means 3 16 6. Then, the host device 3 10 is used in the step model to estimate the slurry In the device, the actual standard slurry material is prepared by the device, and the device device is funded and capitalized in the solution S720-41 - (38) 1276162, and the setting data X and the detection data Y ' are sent to the plasma processing. Device 100. In step S63 0, the plasma processing apparatus 100 receives the setting data X′ and the detection data Υ, and treats these as the actual setting data X 〇bs ', the actual detection data Y 〇bs ' and is stored in the analytical data recall means 2 0 5. Then, in step S640, the actual detection data Yobs is applied to the new model shown in the above formula (8), and the prediction setting data xPre' is calculated and stored in the analysis data storage means 205. Next, in step S650, the plasma processing apparatus 100 determines whether the actual setting data X〇bs' is within the allowable fluctuation range by predicting the setting data Xpre to determine whether it is normal or abnormal. For example, the prediction setting data Xpre' is judged to be normal if the actual setting data Xobs' is within the allowable variation range, and if it exceeds the allowable variation range, it is judged to be abnormal. When it is determined that the abnormality has occurred, in step S660, the plasma processing apparatus 1 is stopped by the processing device control means 225, and the abnormality is notified by the display means 224 and the alarm 226. As described above, the host device 310 performs the multivariate analysis processing based on the data from the respective plasma processing devices to obtain the average 値, so that the calculation processing load of each of the plasma processing devices can be greatly reduced. Further, in each of the plasma processing apparatuses, it is not necessary to temporarily memorize a large amount of setting data or detection data obtained when performing plasma processing, and a multivariate analysis program is not required, so that there is no need for a memory means for this. Thereby, the structure of each plasma processing apparatus can be simplified, and the manufacturing cost can be suppressed. Further, in the second embodiment, the state of the device is determined by a new model of each of the plasma processing apparatus sides -42-(39) 1276162, but this is not necessarily the case, and the new model is also stored in the host device 310. Therefore, it is also possible to determine the state of each of the plasma processing apparatuses 100A.....100N on the main side. In this case, when it is determined that the abnormality is abnormal, the abnormality determination can be sent to each of the plasma processing apparatuses 100A.....1〇0N. Each of the plasmas 1 〇〇 A.....100N can be based on the abnormality determination information, for example, by the device control means 225 to stop the processing means, and by the display means alarm 226 to notify the abnormality. According to this, the state of the devices of the respective plasma processing devices can be monitored in the host device. Although the embodiments of the present invention have been described above with reference to the drawings, it is needless to say that the present invention is not limited thereto. As long as it is the industry, it is obvious that various modifications and alterations are conceivable in the scope of the patent application, and it should be understood that it is within the technical scope of the present invention. For example, when the setting of the first and second embodiments is performed by using the new model of the second embodiment to determine the apparatus state, the setting data by the control parameter measuring unit 22 1 can be used. Alternatively, the material itself set by the input means 1 52 can be used. In this case, all of the setting data can be measured by the parameter measuring device 22 1 , although the setting data measured by the control parameter 22 1 can be used, but in the setting data, the measurement parameter is measured by the control parameter. When the person measured by the device 22 1 is used, it is effective to use the data to be transferred. Further, in the multivariate analysis of the above-described embodiment, it is limited to the case where the device device-like information processing device is composed of the processes 224 and 310, and of course, the input is measured by the electric device. The -43- (40) 1276162 device status parameter is not used by the control quantity, but the device status parameter can also be used as the destination variable or explanatory variable. Further, in the above-described embodiment, when the model is constructed, the process gas flow rate, the gap between the electrodes, and the pressure in the processing chamber are used as the control parameters of the target variable. However, the parameters are not limited to these parameters as long as they are controllable parameters. . Further, although the variable capacitor capacitance, the high-frequency voltage, and the APC opening degree are used as the device state parameters, the parameters are not limited to these as long as they are measurable parameters of the display device state parameters. Further, as the plasma reflection parameter reflecting the state of the plasma, although electrical data based on plasma and optical materials are used, it is not limited to these as long as it is a parameter reflecting the state of the plasma. In addition, as the electrical data, a high-frequency voltage or a high-frequency current that fundamentally affects harmonics (up to 4 times) is used, but the present invention is not limited thereto. Alternatively, the output data from the means for measuring the completion of the measurement wafer (e.g., the scanning gauge) assembled in the plasma processing apparatus may be used as the detection data. Specifically, as the detection data, the film thickness of the film formed on the wafer, the etching amount when etching the film to be processed on the wafer, or the in-plane uniformity thereof can be used. Further, in the present embodiment, although the average 资料 of the data of the plasma reflection parameter is obtained for each wafer, the average 値 is used to predict the control parameter and the device state parameter for each wafer, but it is also possible to use a piece of crystal. The instantaneous plasma reflection parameters in the circle processing predict the control parameters and device status parameters in real time. Further, in the above embodiment, the magnetic field parallel flat type plasma processing apparatus is used, but the present invention can be applied to any apparatus having control parameters, plasma reflection parameters, and/or device state parameters. -44- (41) 1276162 As described in detail above, according to the present invention, the processing characteristics or process conditions of each processing device may be different immediately. For example, if a processing device is modeled, the model may be applied to other types of the same type. The processing device does not require each processing device to recreate the model, and can provide a multivariate analysis model creation method and a multivariate analysis method for the processing device that can reduce the burden of model creation. Industrial Applicability The present invention can be applied to, for example, a multivariate analysis model creation method of a processing device such as a plasma processing apparatus, a multivariate analysis method for a processing device, a control device for a processing device, and a control system for a processing device. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a cross-sectional view showing a schematic structure of a plasma processing apparatus according to a first embodiment of the present invention. Fig. 2 is a block diagram showing an example of a multivariate analysis means of the plasma processing apparatus shown in Fig. 1. Fig. 3 is a block diagram showing the configuration of a processing device control system according to a second embodiment of the present invention. Fig. 4 is a flow chart showing the operation at the time of model creation of the process control system of the present embodiment. Fig. 5 is a flow chart showing the operation at the time of model creation of the processing device control system of the present embodiment, and is a continuation of Fig. 4. Fig. 6 is a flow chart showing the operation at the time of model creation of the processing device control system of the present embodiment, and is a continuation of Fig. 5. -45- (42) 1276162 Fig. 7 is a flowchart showing an operation when control is performed by the model of the fifth embodiment. Main component comparison table 100: plasma processing device, 101: processing chamber pole, 104: upper electrode (shower head), 105: il 107: high frequency power supply, 107 Α: matching device, 107 Β: electrical measuring device, 1 1 8 : Process gas supply system, 200: multivariate analysis means, 201: multivariate segment, 205: analytical data memory means, 206: transport: measurement, diagnosis, control means, 22 1 : control parameter quantity reflects parameter measurer, 223: Device state parameter amount device control means, 300: Control system, 3 20: Network processing device control system |, 1 0 2 : Lower electric charge! Pole ring magnet, Watt meter, 107C: , 120: Optical detection t analysis Program Memory Handicap Means, 207: Predictor, 222: Plasma Detector, 225: Processing Road 'W: Wafer-46-
Claims (1)
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2002163869 | 2002-06-05 | ||
| JP2002168653 | 2002-06-10 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| TW200404333A TW200404333A (en) | 2004-03-16 |
| TWI276162B true TWI276162B (en) | 2007-03-11 |
Family
ID=29738319
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW092115289A TWI276162B (en) | 2002-06-05 | 2003-06-05 | Multi-variable analysis model forming method of processing apparatus, multi-variable analysis method for processing apparatus, control apparatus of processing apparatus, and control system of processing apparatus |
Country Status (5)
| Country | Link |
|---|---|
| JP (1) | JP4224454B2 (en) |
| KR (1) | KR100628392B1 (en) |
| CN (1) | CN100426471C (en) |
| TW (1) | TWI276162B (en) |
| WO (1) | WO2003105210A1 (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006001416A1 (en) * | 2004-06-29 | 2006-01-05 | Nikon Corporation | Management method, management system, and program |
| JP4643392B2 (en) * | 2005-08-24 | 2011-03-02 | 東京エレクトロン株式会社 | Operating state determination method, operating state determination apparatus, program, and storage medium for plasma processing apparatus |
| JP4164835B2 (en) * | 2005-12-20 | 2008-10-15 | 新東工業株式会社 | Projection state information estimation method, projection state information estimation device, and projection state information estimation program using a blast device |
| US7286948B1 (en) * | 2006-06-16 | 2007-10-23 | Applied Materials, Inc. | Method for determining plasma characteristics |
| US20100332010A1 (en) * | 2009-06-30 | 2010-12-30 | Brian Choi | Seasoning plasma processing systems |
| US10176279B2 (en) * | 2015-06-05 | 2019-01-08 | Uptake Technologies, Inc. | Dynamic execution of predictive models and workflows |
| US9934351B2 (en) * | 2015-11-09 | 2018-04-03 | Applied Materials, Inc. | Wafer point by point analysis and data presentation |
| JP6546867B2 (en) * | 2016-03-10 | 2019-07-17 | 東京エレクトロン株式会社 | How to adjust the processing process |
| JP7034646B2 (en) * | 2017-09-25 | 2022-03-14 | 株式会社Ｓｃｒｅｅｎホールディングス | Anomaly detection device and anomaly detection method |
| US10896833B2 (en) * | 2018-05-09 | 2021-01-19 | Applied Materials, Inc. | Methods and apparatus for detecting an endpoint of a seasoning process |
| JP6990634B2 (en) * | 2018-08-21 | 2022-02-03 | 株式会社日立ハイテク | State prediction device and semiconductor manufacturing device |
| JP6841980B2 (en) * | 2019-07-30 | 2021-03-10 | 株式会社日立ハイテク | Equipment diagnostic equipment, plasma processing equipment and equipment diagnostic method |
| WO2023162856A1 (en) * | 2022-02-22 | 2023-08-31 | 株式会社Ｓｃｒｅｅｎホールディングス | Substrate processing apparatus management system, assistance apparatus, substrate processing apparatus, inter-chamber performance comparison method, and inter-chamber performance comparison program |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH1194735A (en) * | 1997-09-22 | 1999-04-09 | Horiba Ltd | Quantitative analyzer for sample characteristic using spectrophotometry and multivariate analysis and analysis method using it |
| US6153115A (en) * | 1997-10-23 | 2000-11-28 | Massachusetts Institute Of Technology | Monitor of plasma processes with multivariate statistical analysis of plasma emission spectra |
| US6582618B1 (en) * | 1999-09-08 | 2003-06-24 | Advanced Micro Devices, Inc. | Method of determining etch endpoint using principal components analysis of optical emission spectra |
| AU4733601A (en) * | 2000-03-10 | 2001-09-24 | Cyrano Sciences Inc | Control for an industrial process using one or more multidimensional variables |
| TW499702B (en) * | 2000-07-04 | 2002-08-21 | Tokyo Electron Ltd | Method for monitoring operation of processing apparatus |
-
2003
- 2003-06-05 TW TW092115289A patent/TWI276162B/en not_active IP Right Cessation
- 2003-06-05 WO PCT/JP2003/007132 patent/WO2003105210A1/en active Application Filing
- 2003-06-05 JP JP2004512183A patent/JP4224454B2/en not_active Expired - Fee Related
- 2003-06-05 KR KR1020047019644A patent/KR100628392B1/en active IP Right Grant
- 2003-06-05 CN CNB038130823A patent/CN100426471C/en not_active Expired - Lifetime
Also Published As
| Publication number | Publication date |
|---|---|
| JP4224454B2 (en) | 2009-02-12 |
| TW200404333A (en) | 2004-03-16 |
| JPWO2003105210A1 (en) | 2005-10-13 |
| WO2003105210A1 (en) | 2003-12-18 |
| CN1659690A (en) | 2005-08-24 |
| KR100628392B1 (en) | 2006-09-26 |
| CN100426471C (en) | 2008-10-15 |
| KR20050010021A (en) | 2005-01-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7505879B2 (en) | Method for generating multivariate analysis model expression for processing apparatus, method for executing multivariate analysis of processing apparatus, control device of processing apparatus and control system for processing apparatus | |
| JP4464276B2 (en) | Plasma processing method and plasma processing apparatus | |
| US7289866B2 (en) | Plasma processing method and apparatus | |
| TWI276162B (en) | Multi-variable analysis model forming method of processing apparatus, multi-variable analysis method for processing apparatus, control apparatus of processing apparatus, and control system of processing apparatus | |
| JP2018117116A (en) | Characteristic constraint method from time sequence of spectrum for controlling end point in processing | |
| JP2018026558A (en) | Methods and systems for monitoring plasma processing systems and advanced process and tool control | |
| WO2003056618A1 (en) | Plasma treatment apparatus | |
| JP2004349419A (en) | Method and device for judging cause of abnormality in plasma processor | |
| JP2004335841A (en) | Prediction system and prediction method for plasma treatment apparatus | |
| TW200401179A (en) | Method of predicting processing device condition or processed result | |
| JP2004039952A (en) | Plasma treatment apparatus and monitoring method thereof | |
| EP1700333A1 (en) | Method and apparatus for seasoning semiconductor apparatus of sensing plasma equipment | |
| JP5105399B2 (en) | Data collection method, substrate processing apparatus, substrate processing system | |
| US20240096713A1 (en) | Machine-learning in multi-step semiconductor fabrication processes | |
| JPH07307301A (en) | Semiconductor device and its manufacture | |
| JP4220378B2 (en) | Processing result prediction method and processing apparatus | |
| TWI297904B (en) | ||
| JP2007250902A (en) | Estimation method and estimation device for substrate processor | |
| KR100446926B1 (en) | Method of monitoring and/or controlling a semiconductor manufacturing apparatus and a system therefor | |
| JP2009010370A (en) | Semiconductor processing apparatus | |
| KR20230055609A (en) | Real time monitoring method and system for plasma appatus | |
| JPH11238723A (en) | Method and apparatus for plasma processing | |
| CN116583938A (en) | Machine learning in a multi-step semiconductor manufacturing process |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| MM4A | Annulment or lapse of patent due to non-payment of fees |