在這一篇論文當中，我們設計出一套方法對氮化鎵磊晶片在還沒製程前，就先去預測晶片的相關參數，如此，對於預測出來元件特性差的晶片，我們將可避免再對其做製程，如此將可大量減少生產成本。在預測的參數中包含了導通電壓(Vf)、主波長(λd)、峰波長(λp)、半高寬(Δλ)、崩潰電壓(Vr)、漏電流(Ir)和發光亮度(Iv)。

　　其中在Vf、λd、λp和Δλ的預測，我們希望能直接預測到當它們操作在20mA電流時的值，但由於p-GaN電阻率過大問題，使我們必須還要再加一層電流傳輸層在p-GaN上才得以使電流驅動到20mA。如此之後，對於光特性λd、λp和Δλ，其預測結果可說是非常準確；而對於Vf的預測，其預測值與結果值都能固定在一個等差值，這樣的一個結果就可以被我們拿來應用

　　在崩潰電壓(Vr)、漏電流(Ir)和發光亮度(Iv)的預測中，我們取三片磊晶片並量測其dead current，由於dead current相對於主動層有絕對關係，dead current越大主動層品質越差，所以我們將可藉由一開始量dead current與最後量Vr、Ir和Iv而作成一比較表。如此，當我們取第四片晶片時，就可藉由一開始量得第四片晶片之dead current，而相對地比較預測出其Vr、Ir和Iv值。

　　最後，我們發現當探針在擊穿磊晶片後，將會在磊晶片表面燒出一個孔洞，我們知悉此預測實驗之所以能夠執行，必定與這個孔洞有極大關聯。所以我們利用AFM去量測此孔洞的深度與電流分布，然後發現會有一極大電流分布在孔洞之周圍與底部，這大電流遠比一般表面電流大了數百倍，而也因為此電流之劇增，才得以使我們在未做製程之前就能驅動電流達數毫安培，而預測參數成功。

　In this dissertation, we have designed a set of methods to predict the relative parameters of GaN LED epi-wafer before the process. Accordingly, we can avoid processing the epi-wafers that we have predicted that they will have poor device quality . By means of this, we can reduce a great number of costs. The prediction parameters include forward voltage(Vf), dominant wavelength(λd), peak wavelength(λp), Full Width at Half Maximum(Δλ), break down voltage(Vr), leakage current(Ir)and electroluminescence intensity(Iv).

　In the prediction of Vf, λd, λp and Δλ, because we have the problem of p-GaN whose resistivity is too high, we have to add a current spreading layer on p-GaN to drive the inject current to 20mA so as to directly predict the value of Vf, λd, λp and Δλ. As to the prediction of optical properties, such as λd, λp and Δλ, the prediction is quite accurate. In terms of the prediction of Vf, the difference between the prediction and the result always keeps in a constant. Therefore this prediction of Vf is also applicable.

　In the prediction of Vr, Ir and Iv, we take three epi-wafers to measure their dead current. This is because dead current absolutely has relation with the active layer of LED. The higher the dead current is, the poor quality the active layer will has. Thus, we can make the values of dead current, Vr, Ir and Iv a graph. Afterward, we can relatively predict the values of Vr, Ir and Iv of the fourth epi-wafer through the dead current of the fourth epi-wafer that we have measured first.

　Finally, we find that after the probes damage the epi-wafer, there will be a burned hole in the surface of the epi-wafer. We know that the hole has great relation with the practice of the prediction. So we make use of AFM to measure the depth and current distribution of the hole. We eventually find there is a great value of current spreading in the ramp and the bottom of the hole. The current is hundreds of times higher than usual one. Because the sudden increase of the current, we can drive the inject current upto 20mA before the process and successfully predict the parameters of the epi-wafers.

[1] S. Nakamura, T. Mukai, and M. Senoh,
“Candela-class High-brightness
InGaN/AlGaN Double- heterostructure Blue-
light-emitting Diodes”, Appl.
Phys. Lett., 64, 1687, 1994.

[2] S. Nakamura, M. Senoh, N. Iwasa, and S.
Nagahama, “High-power InGaN
Single-quantum-well-structure Blue and
Violet Light-emitting Diodes”, Appl.
Phys. Lett. 67, 1868, 1995.

[3] S. Chichibu, T. Sota, K. Wada, and S.
Nakamura, “Exciton Localization in
InGaN Quantum Well Devices”, J. Vac. Sci.
Technol., B16, 2204, 1998.

[4] H. Kim, S-J Park, and H. Hwang, “Effects of
Current Spreading on the Performance of GaN-
based Light-emitting Diode”, IEEE
Transactions on Electron Devices, 48, 1065,
2001.

[5] S. Nakamura, M. Senoh, S. Nagahama, N.
Iwasa, T. Yamada, T. Matsushita,
H. Kiyoku, and Y. Sugimoto, “InGaN-Based
Multi-Quantum-Well-Structure Laser Diodes”,
Jpn. J. Appl. Phys., 35, L74, 1996.

[6] K. Itaya, M.Onomura, J. Nishino, L. Sugiura,
S. Saito, M. Suzuki, J. Rennie,
S. Nunoue, M. Yamamoto, H. Fujimoto, Y.
Kokubum, Y. Ohba, G.Hatakoshi, and M.
Ishikawa, “Room Temperature Pulsed
Operation of Nitride Based Multi-Quantum-
Well Laser Diodes with Cleaved Facets on
Conventional C-Face Sapphire Substrates”,
Jpn. J. Appl. Phys., 35, L1315,1996.

[7] S. Nakamura, M. Senoh, S. Nagahama, T.
Matsushita, H. Kiyoku, Y.Sugimoto, T.
Kozaki, H. Umemoto, M. Sana, and T. Mukai,
“Violet InGaN/GaN/AlGaN-Based Laser Diodes
Operable at 50oC with a Fundamental
Transverse Mode “, Jpn. J. Appl. Phys., 38,
L226, 1999.

[8] V. W. L. Chin, T. L. Tansley, and T.
Osotchan, “Electron Mobilities in Gallium,
Indium, and Aluminum Nitrides”, J. Appl.
Phys., 75, 7365, 1994.

[9] M. Shur, B. Gelmont, and M. A. Khan,
“Electron Mobility in Two-Dimensional
Electron Gas in AlGaN/GaN Heterostructures
and in Bulk GaN”, Journal of Electronic
Materials, 25, 777, 1996.

[10] K. S. Kim, C. S. Oh, W. H. Lee, K. J. Lee,
G. M. Yang, C. H. Hong, E. K. Suh, K. Y.
Lim, H. J. Lee, and D. J. Byun“Comparative
Analysis of Characteristics of Si, Mg, and
Undoped GaN“, Journal of Crystal Growth,
210, 505, 2000.

[11] B. E. Foutz, S. K. O’Leary, M. S. Shur,
and L. F. Eastman, “Transient Electron
Transport in Wurtzite GaN, InN, and AlN“,
J. Appl. Phys., 85, 7727, 1999.

[12] W. J. Fan, M. F. Li, T. C. Chong, and J. B.
Xia, “Band Structure Parameters of Zinc-
blende GaN, AlN and Their Alloys Ga1-
xAlxN”, Solid State Communications, 97,
381, 1996.

[13] Hari Singh Nalwa, “Handbook of thin film
materials”, ISBN: 0-12-512912-6. Vol.4,
117-125, 2002.

[14] S. Nakamura, T. Mukai, and M. Senoh, Jpn.
J. Appl. Phys., Part 131, 2883 ,1992.

[15] W. R. Runyan and T. J. Shaffer,
Semiconductor Measurements and
Instrumentation, The McGraw-Hill
Companies,Inc, pp. 386~389.

[16] Y. S. Zhao, C.L. Jensen, Member, IEEE, R.
W. Chuang, Member, IEEE, H.P. Lee, Z. J.
Dong, and R. Shih, “A Simple and Reliable
Wafer-Level Electrical Probing Technique
for III-Nitride Light-Emitting Epitaxial
Structures”, IEEE ELECTRON DEVICE LETTERS,
VOL. 21, NO. 5, 2000.