In this thesis, a quick turn-around, experimental methodology based on nanoindentation has been successfully developed to accelerate the evaluation of various BOAC layout designs without a full array of reliability tests. Nanoindentation,
which simulated the impact force of wi N), was
adopted to characterize mechanical resistance of BOAC structures, which consisted of oxide/multi-layered copper/low-k structures below Al bond pad, as a function of indenter depth by mean
structures could be quantif g King suming layered
Al/substrate system with their bon esults.
Four parameters affecting the mechanical strength of BOAC such as bond pad types, copper density, line width/pitch (technology nodes) and low-k materials were invest thesis to identify general design rules for BOAC layouts and structures. In the bond pad design, normal pa ssessed much stronger mechanical suppo ad beca normal pads had full array of dum ed trench/via inforcement in M1-6, while BOAC structures had only sparse vias (< 6 %) ra
M1-6 layer were higher by increased aspect ratio of copper in 65 nm node.
re bonding (down force of 5~25 m
s of substrate effect. The mechanical stiffness of BOAC ied usin ’s model as a uniform
and correlated dability r
igated in this
d po
rt than BOAC p use mifi
re
ndomly layout in M1-6.
For BOAC structures with different copper density and layout (block copper vs.
ring copper) in the top metal layer, metal density had little influence on the mechanical strength of BOAC structures. This indicated that the top Al pad (1.2 μm) and oxide layer (1.2 μm) absorbed the majority of impact force; thus provided good protection for the structures underneath the top Al pad and oxide layer.
When the metal line widths and pitches scaled from 90 nm to 65 nm process node, the modulus of composite substrate, Ecs in 65 nm parts was larger because the Cu fractions in
The cal strength of BOAC structures. E
2
type of low-k materials was found to have great influence on the mechani
cs of ultra low-k parts was much lower than Ecs for BOAC with low-k due to low modulus of ultra low-k and its weak interfacial adhesion. Care and consideration of weak modulus, and more importantly the interfacial adhesion shall be taken to avoid mechanical reliability or device failure when ultra low-k materials are chosen as the ILD materials. In summary, BOAC structures can be strengthened by (1) adding a stronger buffer layer such as oxide layer on the top of BOAC structures, (2) increasing aspect ratio (AR) in Cu/low-k layer, and (3) using low-k materials with better modulus and interface adhesion.
Different BOAC stacks were also investigated in this thesis. The first BOAC structure was Al/composite substrate system which could be categorized as a soft film/hard substrate system. The other BOAC structure was oxide/composite substrate system which could be categorized as a hard film/soft substrate system. The objective was to cross-check the fitting results based on King’s model and to understand any difference in the values obtained by nanoindentation method. The results showed that the Ecs of oxide film/composite substrate were lower than those of Al film/composite substrates for the same BOAC design because the affected region within the composite substrates changed with increasing indenter depth during nanoindentation.
Moreover, the results also indicated that the hardness would affect the modulus in the nanoindentation measurement. For Al film/composite substrate system, the Ecs was overestimated because the contact area (A) was underestimated. For the oxide film/composite substrate, the Ecs was underestimated because the contact area (A) was overestimated.
Then P/S term was then used to eliminate the contact area effect from nanoindentation in order to study the true modulus behavior during indentation. The
depth. This method could be used to analyze the changing mechanical strength of multilayer in BOAC structures as a function of indenter depth. A 3-layered model based on Al/oxide/soft composite substrate was used to estimate the mechanical strength of BOAC structures under oxide layer by using (1) experimental observations, which used P/S2 to exclude hardness effect and (2) theoretical calculations, which used the equation of composite materials. From the results, we found the values from the equation of composite material Er were lower than those observed in experiments.
The lower values in theoretical values may be due to the underestimation of contribution from top harder layer in soft composite substrate such as Cu/FSG and the strain-hardening induced by the Cu in copper/low-k layers. Moreover, the theoretical calculation cannot take weak adhesion due to porosity into account. The P/S2 (experimental observations) results showed not only the same tendency of mechanical strength as that of theoretical calculation using the equation of composite materials, but also smaller deviation from theoretical calculation, as compared to values obtained from King’s model fitting. A modification of King’s model will be needed if the exact modulus of the multi-layered substrates is sought after.
verall, a novel methodology based on nanoindentation has been successfully established to distinguish the mechanical strength of various BOAC structures through a composite modulus values using a simplified film/uniform substrate model or P/S2 model. Unfortunately the BOAC structures in this thesis were all passing the bondability test. Nevertheless, a modulus of 70 GPa in composite substrate of bond pad structure based on King’s model fitting results, can be considered as a sufficient condition for new BOAC layouts meeting the bondability tests.
O
Reference
[1]
of 55th ECTC, 861
re Bonding in Microelectronics Materials, Processes,
stig, P. Roper, International Technology Roadmap for Semiconductors, executive summary (2005).
[2] G. Heinen, R. J. Stierman, D. Edwards, and L. W. Nye, IEEE Proc. of 44th ECTC, 922 (1994).
[3] K. M. Chen B. C. Wu, K. H. Tang, F. Y. Cheng, N. H. Kao, and J. Y. Lai, Microelectronics Reliability, 46, 335 (2006).
[4] B. Halg, IEEE Micro Electro Mechanical Systems Workshop, 172 (Feb 1990).
[5] L. Shen and K. Zeng, Microelectronic Engineering, 71, 221 (2004).
[6] Y. Liu, D. Desbiens, S. Irving, and T. Luk, IEEE Proc.
(2005).
[7] Y. Liu, S. Irving, and T. Luk, IEEE Proc. of 54th ECTC, 383 (2004).
[8] 呂宗興, 電子構裝技術的發展歷程, 工業材料 (1997).
[9] 鐘文仁, 陳佑任, IC 封裝製程與 CAE 應用 (2005).
[10] International Technology Roadmap for Semiconductors, executive summary (2006).
[11] G. G. Harman, Wi
Reliability, and Yield (second ed.), McGraw-Hill (1997).
[12] R. Singh, and R. K. Ulrich, The Electrochemical Society Interface, 8 (2), 26 (1999).
[13] M. T. Bohr, IEEE International Electronic Device Meeting, 241 (1995).
[14] D. Edelstein, J. Heidenreich, R. Goldblatt, W. Cote, C. Uzoh, N. Lu
T. Mcdevittt, W. Motsifft, A. Simon, J. Dukovic, R. Wachnik, H. Rathore, R.
Schulz , L. Su, S. Lucet, and J. Slatteryt, IEEE International Electronic Device
[15] 楊正杰, 張鼎張, 鄭晃忠, 銅金屬與低介電常數材料與製程, 毫微米通訊 7, 40.
[16] J. Leu, Low-k Materials and Processing Thchnologies Class (2006).
[17] K. Mikagi, N. Ishikawa, T. Usami, M. Suzuki, K. Inoue, N. Oda, S. Chikaki, I.
Sakai, and T. Kikkawa, IEEE International Electronic Device Meeting, 365 (1996).
[18] S. J. Martin, J. P. Godschalx, M. E. Mills, E. O. Shaffer, and P.H. Townsend, Adv.
Mater., 12, 1769 (2000).
[19] H. J. Lee, E. K. Lin, H. Wang, W. L. Wu, W. Chen, and E. S. Moyer, Chem.
ENTS AND PACKAGING TECHNOLOGIES, 27, 4 (2004).
[2
Mater., 14, 1845 (2002).
[20] J. H. Zhao, I. Malik, T. Ryan, E. T. Ogawa, and P. S. Ho, APPLIED PHYSICS LETTERS, 74, 944 (1999).
[21] S. Allada, “Low K Adhesion Issues in Cu/Low K Integration”, IEEE (1999).
[22] J. B. VELLA, I. S. ADHIHETTY, K. JUNKER, and A. A. VOLINSKY, International Journal of Fracture, 119/120, 487 (2003).
[23] K. Y. Chou, M. J. Chen, and C. W. Liu, IEEE Trans. Electron Dev., 49, 2279 (2002).
[24] K. J. Hess, S. H. Downey, G. B. Hall, T. Lee, L. L. Mercado, J. W. Miller, W. C.
Ng, and D. G. Wontor, IEEE Proc. Of 53th ECTC., 1344, 2003.
[25] B. Chandran, R. Mahajan, M. Bohr, C. H. Jan, Q. T. Vu, “The Mechanical Side of Ultra-Low k: Can it Take the Strain?”, Future Fab Intl., 17 (2004).
[26] D. Degryse, B. Vandevelde, and E. Beyne, IEEE TRANSACTIONS ON COMPON
7] O. van der Sluis, R. A. B. Engelen, W. D. van Driel, M. A. J. van GiIs, and R. B.
R. van Silfhout, 7th. Int. Conf: on Thermal, Mechanical and Multiphysics Simulation and Experiments in Micro-Electronics and Micro-Systems,
EuroSimE (2006).
[28] D. Tabor, The hardness of metals., Oxford: Oxford Univ. Press (1951).
[29] I. N. Sneddon, The relaxation between load and penetration in the axisymmetric Bussinesq problem for a punch of arbitrary profile.
[30] G. M. Pharr, and W. C. Oliver, J. Mater. Res., 4, 94 (1989).
[31] W. C. Oliver and G. M. Pharr, J. Mater. Res., 7, 1564 (1992).
[32] A. A. Volinsky, J. B. Vella, and W. W. Gerberich, Thin Solid Films, 429, 201 (2003).
[33] A. K. Bhattacharya and W. D. Nix, International Journal of Solids and Structures, 24, 1287 (1988).
[34] R. B. King, Int. J. Solids Struct., 23, 1657 (1987).
[35] B. A. Walmsley, et al., Journal Applied Physics, 98, 044904 (2005).
[36] R. Saha and W. D. Nix, Acta. Mater., 50, 23 (2002).
[37] J. Vitiello, A. Fuchsmann, L. L. Chapelon, V. Arnal, D. Barbier, and J. Torres, Microelectronic Engineering, 82, 422 (2005).
[38] S. Chen, L. Liu, and T. Wang, Surf. Coat. Technol., 191, 25 (2005).