| 研究生: |
高銘聰 Kao, Ming-Tsung |
|---|---|
| 論文名稱: |
AlNx/Al雙層複合材料應用於TFT之可行性分析與鋁薄膜產生鋁尖凸特性研究 AlNx/Al Composite Film Prepared for Use in Thin-Film Transistors and Effects of Deposition Conditions of Al film and Annealing Conditions on Internal Stresses and Hillock Formations |
| 指導教授: |
林仁輝
Lin, Jen-Fin |
| 學位類別: |
博士 Doctor |
| 系所名稱: |
工學院 - 機械工程學系 Department of Mechanical Engineering |
| 論文出版年: | 2012 |
| 畢業學年度: | 100 |
| 語文別: | 中文 |
| 論文頁數: | 101 |
| 中文關鍵詞: | AlNx/Al雙層複合材料 、鋁尖凸 、閘極層電阻 、內應力 、微孔隙 、楔形角 |
| 外文關鍵詞: | AlNx/Al composite film, hillocks, electrical resistance, internal stress, nanovoids, wedge angle |
| 相關次數: | 點閱:85 下載:3 |
| 分享至: |
| 查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
本研究主要分兩大內容,內容一探討的主題是使用氮化鋁(AlNx)當作緩衝層,利用接近玻璃熱膨脹係數的氮化鋁鋪在玻璃基板與純鋁金屬中間來抑制鋁尖凸的產生,也是本論文所提出一個新的思考方向。AlNx/Al雙層複合材料用來取代傳統TFT製程中閘極金屬所使用高單價的AlNd,選擇不會產生鋁尖凸的AlNx/Al膜厚配比的實驗條件,來驗證此AlNx/Al雙層複合材料用作TFT閘極導線材料時,其薄膜導線電阻、薄膜導線與ITO間接觸電阻、經過蝕刻後金屬薄膜的profile呈現、薄膜導線經過顯影後與蝕刻後金屬線寬差異(critical dimension loss)及TFT電壓電流特性曲線等重要的製程特性均在用作TFT閘極導線上允許的變動範圍區間內,因此可以得到在不用加入任何稀有材料的情況下,只需在TFT鍍製鋁薄膜製程時通入氮氣,變更此製程,即可達到成本降低與抑制鋁尖凸產生的雙重效應。
本論文另一個探討的主題是濺鍍在玻璃基板上的鋁薄膜產生鋁尖凸(hillock)或微孔隙(nanovoid)的行為模式,設計六個主要影響參數包含鋁薄膜厚度、鍍膜功率、鍍膜壓力、鍍膜溫度、退火溫度及退火時間等,利用田口中六因子五水準直交表施作25組實驗,來求得退火前後的壓力值(σ0與σf),而退火過程中的熱應力σan則是利用Flinn所建立的方程式[1]來求得。爾後再利用反應曲面法(response surface methodology)來求得三種應力(σ0、σf、σan)與六個設計參數間的關係。實驗結果得知退火過程中的熱應力σan須大於28.7 MPa以上(壓應力),鋁尖凸才會產生。鋁薄膜厚度、退火溫度及退火時間是三個主要影響σf與σan大小的關鍵因素。鋁薄膜厚度越厚、退火溫度越高及退火時間越長,其σf與σan應力值也會越大,則鋁尖凸的現象越容易產生。而在微孔隙方面可以發現25個樣本均有微孔隙的存在,我們針對在1000 μm2內的微孔隙數量及其平均粒徑作比較分析與探討,退火後應力值(σf)越大可以發現微孔隙的數量有較多的趨勢,且微孔隙的平均粒徑也有越大的情形,微孔隙的數量多與粒徑大會微幅增加鋁薄膜的電阻。由實驗數據來看,此增加的幅度,尚不至於影響TFT 閘極層電子傳導的速率,但若閘極層電阻過高,則應用於TFT 閘極層的鋁薄膜會因為電阻過高導致電子傳導的速率不夠快,使得螢幕畫面左右亮度不均及灰階錯誤等情形產生,因此對於閘極金屬層,需求的電阻值是越低越好。文中我們也分析出退火後與退火過程中熱應力的應力變化,兩者間壓力差值(σf-σan)大於130MPa,鋁尖凸才會產生,無論鋁尖凸或微孔隙的產生,都跟退火後與退火過程中熱應力的變化息息相關,文中也針對蝕刻後楔形角(wedge angle)的大小影響鋁尖凸的關係,提出了相關的說明,wedge angle與tap length 成反比關係,當wedge angle越小則tap length會變長,而tap length長代表Mo未覆蓋鋁膜的面積增加,Mo覆蓋對於抑制鋁尖凸,具有一定的成效,因此對於會產生鋁尖凸的製程條件來說,tap length長則單位面積內鋁尖凸數量會變多,文中我們也作出wedge angle與退火後應力值相對應的關係,wedge angle小退火後應力值會較大。後續的實驗結果,也呼應文中的論述。
AlNx/Al/Mo composite films with various thicknesses of AlNx and Al layers were prepared to replace commercial AlNd/Mo composite film as the gate metal of the two metal layers (namely the gate metal and the source-drain metal) in thin-film transistor (TFT) specimens. The prerequisite for the TFT device is that no hillock is formed. The electrical properties of the AlNx/Al/Mo TFT device rival those of the AlNd/Mo TFT device. One of eight kinds of AlNx/Al/Mo composite films (0.05μm/0.2μm/0.07μm) without hillocks was compared with the AlNd/Mo (0.25μm /0.07μm) composite film. The line width after development and strip inspections, the Ig (gate leakage current)-Vg (gate voltage) curve, the coating film resistance to electricity, the contact resistance between the indium tin oxide (ITO) film and the metal film, the Id-Vg curve, and the critical dimension loss (CD loss) were compared. The experimental results indicate that the metal line widths for these two composite films are similar. The coating film resistance, the contact resistance between the ITO film and the metal film, and the Id-Vg curve for the AlNx/Al/Mo TFT device were similar to those for the AlNd/Mo TFT device. The CD loss shown in the AlNx/Al/Mo TFT device was lower than that for the AlNd/Mo TFT device.
In the present study, 25 kinds of specimen with five Al-film thicknesses were prepared to investigate the relation between the internal stress formed during the annealing process and hillocks. In the preparation of specimens, the governing factors including deposition conditions, annealing temperature, and annealing time, were arranged following the orthogonal table of five-level and six-factorial (L25(56) ) design. Stoney’s formula is applied to describe the internal stresses before and after annealing (σ0 and σf), respectively. The internal stress arising during the annealing process (σan) is evaluated using the model developed by Flinn et al. [1]. Then, the response surface methodology (RSM) is used to express the three stress parameters in terms of influential factors. The incipient σan value for hillocks appearing in the specimens was found to be between -28.7 MPa and -32 MPa in a compressive form. The annealing temperature and time and the Al-film thickness are the three major factors, affecting internal stress σan. The density of hillocks increased linearly with (σf-σan) when the value of the stress change parameter was beyond the critical value (130MPa). Nanovoids were also produced in specimens without hillocks. The wedge angle formed in the specimen after wet etching linearly increases with (σf-σan). The electrical resistance of the gate layer linearly increases with increasing the product value (R*) of the mean size (area) and the density of nanovoids, it nonlinearly increases with increasing (σf-σan). An increase in the annealing time reduces the tensile stress or increases the compressive stress, or both. The tensile stress decreases and the compressive stress increases during the annealing process with increasing Al film thickness and annealing temperature. The number of hillocks formed in a unit of area is linearly proportional to both σan and (σf-σan).
1.P. A. Flinn, D. S. Gardner, and W. D. Nix, IEEE Trans. Electron Devices, ED-34 (1987) 689.
2.薛英家,液晶顯示器專題,科儀新知第二十三卷第四期 91.2C.
3.P. Chaudhari, J. Appl. Phys. 45 (1974) 4339.
4.G. Dehm, B. J. Inkson, and T. Wagner, Acta Mater. 50 (2002) 5021.
5.D. S. Gardner, and P. A. Flinn, IEEE Trans. Electron Devices, ED-35 (1988) 2160.
6.P. Chaudhari, J. Vac. Sci. Technol. 9 (1972) 520.
7.D. S. Gardner, and P. A. Flinn, J. Appl. Phys. 67 (1990) 1831.
8.S. Dutta, H. M. Chan, and R. P. Vinci, J. Am. Ceram. Soc. 90 (2007) 2571.
9.E. Iwamura, T. Ohnishi, and K. Yoshikawa, Thin Solid Films 270 (1995) 450.
10.S. J. Hwang. Y. D. Lee, Y. B. Park, J. H. Lee, C. O. Jeong, and Y. C. Joo, Scrip. Mater. 54 (2006) 1841.
11.S. J. Hwang. J. H. Lee, C. O. Jeong, and Y. C. Joo, Scrip. Mater. 56 (2007) 17.
12.S. J. Hwang, W. D. Nix, and Y. C. Joo, Acta Mater. 55 (2007) 5297.
13.E. Iwamura, T. Ohnishi, K. Yoskikawa, and K. Itayama, J. Vac. Sci. Technol. A12(5) (1994) 2922.
14.R. C. Cammarata, T. M. Trimble and D.J. Srolovitz, Journal of Materials Research, 15 (2000) 2468.
15.P. Chaudhari, Journal of Applied Physics 45 (1974). 4339.
16.H. C. W. Huang, P. Chaudhari, C. J. Kircher, and M. Murakami, Philos. Mag. A 54 (1986) 583.
17.M. Zaborowski, and P. Dumania, Microelec. Eng. 50 (2000) 301.
18.M. Ronay, and G. F. Aliotta, Philos. Mag. A 42 (1986) 161.
19.R. A. Schwarzer, and D. Gerth, J. Electronic Mater. 22 (1993) 607.
20.C. Y. Chang, and R. W. Vook, J. Vac. Sci. Technol., A9 (3) (1991) 559.
21.U. Smith, N. Kristensen, F. Erison, and J.-A. Schweitz, J. Vac. Sci. Technol., A9 (1991) 2527.
22.S. Takayama, and N. Tsutsui, J. Vac. Sci. Technol., 14 (1996) 2499.
23.T. Onishi, E. Iwamura, K. Takagi, and T. Watanabe, J. Vac. Sci. Technol., 15 (1997) 2339.
24.H. Takatsuji, H. Iiyori, S. Tsuji, K. Tsujimoto, K. Kuroda, and H. Saka, Mater. Res. Soc. Symp. Proc. 471 (1997) 99.
25.T. Arai, H. Iiyori, Y. itiromasu, M. Atsumi, S. Ioku, and K. Furuta, IBM J. Res. Dev. 42 (1998) 491.
26.C. Y. Chang, and R. W. Vook, J. Mater. Res. 4 (1989) 1172.
27.F. Ericson, N. Kristensen, J. Schweitz, and U. Smith, J. Vac. Sci. Technol., B 9 (1991) 58.
28.D. Gerth, D. Katzer, and M. Krohn, Thin Solid Films 208 (1992) 67.
29.T. Onishi, E. Iwamura, K. Takagi, and K. Yoshikawa, KOBE steel engineering reports, 48 (1998).
30.T. Arai, A. Makita, Y. Hiromasu, and H. Takatsuji, Thin Solid Films 383 (2001) 287.
31.A. T. Voutsas, Y. Hibino, R. Pethe, and E. Demaray, J. Vac. Sci. Technol., A16(4) (1998).
32.T. Onishi, E. Iwamura, and K. Takagi, Thin Solid Films, 340 (1999) 306.
33.H. Takatsuija, K. Tsujimoto, K. Kuroda, and H. Saka, Thin solid Films, 343-344 (1999) 461.
34.H. Takatsuji, K Haruta, S. Tsuji, K. Kuroda, and H. Saka, Surf. Coat. Technol. 125 (2000) 167.
35.B. Cao Martin, C. J. Tracy, J. W. Mayer, and L. E. Hendrickson, Thin Solid Films 271 (1995) 64
36.C. Kylner, and L. Mattsson, Thin Solid Films 307 (1997) 169.
37.A. Nathan, R. V. R. Murthy, B. Park, and S. G. Chamberlain, Microelectron. Reliab. 40 (2000) 947.
38.D. K. Kim, B. Heiland, W. D. NiX, E. Arzt, M. D. Deal, and J. D. Plummer, Thin Solid Films 371 (2000) 278.
39.李輝煌, 田口方法品質設計的原理與實務,高立圖書有限公司 (2008).
40.G. E. P. Box, and N. R. Draper, “Empirical Model Building and Response Surfaces”, Wiley, New York, 1987.
41.白木 靖寬,吉田 貞史編著,王建義編譯,"薄膜工程學",全華科技圖書股份有限公司 (2006).
42.Y. Huang, and A. J. Rosakis, J. Mech. Phys. Solids 53 (2005) 2483.
43.S.J. Bull, D.S. Rickerby, J.C. Knight and T.F. Page, Surface Engineering. 8 (1992) 193.
44.G. G. Stoney, Mathematical and Physical Sciences, 82 (1909) 172.
45.J. F. Lin, Z. C. Wan, P. J. Wei, H. Y. Chu, and C. F. Ai, Thin Solid Films, 466 (2004) 137.
46.A. K. Sinha and T.T. Sheng, Thin Solid Films, 48 (1978) 117.
47.M. F. Ashby and H. J. Frost, “The kinetics of inelastic deformation above 0 K”, in Constitutive Equatins in Plasticiity, A. S. Argon, Ed. Cambridge, MA:MIT press, (1975) 117.
48.郭侃誠,電漿輔助化學氣相沉積系統淺論,半導體設備技術專輯,機械工業雜誌, p. 198 (90年9月).
49.G. E. Dieter, “Mechanical Metallurgy”, New York, McGraw-Hill (1988).
50.C. C. Chou, and J. F. Lin, ASME J. of Tribology, 121 (1999) 185.
51.C. F. Yeh, J. Y. Cheng, and J. H. Lu, Journal of Applied Physics 32 (1993) 2803.
52.紀國鐘,鄭晃忠,液晶顯示器技術手冊(2001).
53.顧鴻壽,光電液晶平面顯示器技術基礎及應用(2001).