Objective The preparation of a high consistency, high aspect ratio indium bump arrays represent a pivotal technology for fabricating high density focal plane array (FPA). Indium is one of the most commonly utilized materials for interconnecting bumps in FPAs, particularly cooled FPAs, due to its exceptional ductility, electrical conductivity, thermal conductivity, and other properties. The current research on indium bump growth is primarily concerned with the macroscopic effects of process parameters on the growth of indium bump. However, there is a paucity of in-depth reports on the growth mechanism of indium bump. When considered in conjunction with the theory of thin-film nucleation growth, the impact of pivotal process parameters, such as substrate temperature and deposition rate, on the growth of indium bumps can be elucidated from a microscopic perspective. This offers a theoretical and technical foundation for developing and preparing indium bump arrays with smaller pixel pitch.
Methods According to the indium bump growth model (Fig.1), indium bump arrays with an image element spacing of 10 μm and a photolithography pattern diameter of 6 μm (abbreviated as
P10-
Φ6) were studied, and indium bump arrays were obtained by thermal evaporation method under different growth parameters. The morphological differences (Fig.2(a)-(f)), high uniformity (Fig.3(a)-(b)), top diameter uniformity (Fig.4(a)-(b)), and indium bump growth rate differences (Fig.5(a)-(b)) of the indium bump arrays were comparatively analyzed with the theory of thin film micro growth for the indium bump arrays with three different substrate temperatures (
T0 K,
T0-30 K, and
T0-45 K). Comparative analysis of indium bump arrays at three different deposition rates (
VV0+2 Å/s,
VV0+15 Å/s, and
VV0+23 Å/s) in terms of morphology difference (Fig.6(a)-(f)), height uniformity (Fig.7(a)-(b)) and top diameter uniformity (Fig.8(a)-(b)), indium bump growth rate differences (Fig.9(a)-(b)).
Results and Discussions The standard deviation of the height distribution of indium bumps decreased from 0.22 to 0.14 and 0.09(Fig.3(a)-(b)), the height nonuniformity from 15.4% to 11.8% and 6.6%(Fig.3(a)-(b)), and the standard deviation of the top diameters from 0.64 to 0.43 and 0.37(Fig.4(a)-(b)), respectively, when the substrate temperature was decreased from
T0 K to
T0-30 K and
T0-45 K. This resulted in an improvement in the homogeneity of the indium bump arrays. Nevertheless, a reduction in the substrate temperature from
T0-30 K to
T0-45 K has been observed to increase the lateral growth rate of the indium layer from 3.4 Å/s to 4.1 Å/s(Fig.5(a)-(b)). This is an unfavorable phenomenon for the growth of indium bumps with high aspect ratio. The increase in deposition rate from
VV0+2 Å/s to
VV0+15 Å/s and
VV0+23 Å/s resulted in a reduction in the standard deviation of the indium bump height distribution from 0.14 to 0.09(Fig.7(a)-(b)), the height nonuniformity from 11.8% to 8.1%(Fig.7(a)-(b)), and the standard deviation of the top diameter of the indium bumps from 0.49 to 0.31(Fig.8(a)-(b)). The standard deviation of the top diameter of the indium bumps decreased from 0.49 to 0.31 and 0.34(Fig.8(a)-(b)), respectively. The increase in deposition rate resulted in enhanced indium bump uniformity. Nevertheless, when the deposition rate is elevated from
VV0+15 Å/s to
VV0+23 Å/s, the indium film roughness undergoes a mere 0.001 μm alteration(Fig.9(a)-(b)). Additionally, the lateral growth rate of the indium layer demonstrates a linear correlation with the deposition rate, thereby limiting the potential for further enhancement of indium bumps through the continued pursuit of elevated deposition rates.
Conclusions On the one hand, high deposition rates and low substrate temperatures result in the refinement of indium film grains and facilitate the deposition of highly uniform indium bump arrays. Following a decrease in substrate temperature of 45K, the nonuniformity of the indium bump height was observed to decrease from 15.4% to 6.6%. In contrast, the indium bumps' top diameter range difference decreased by 0.9 μm. Following an increase in deposition rate of 21 Å/s, the indium bump height nonuniformity decreased from 11.8% to 6.6%, and the top diameter range difference of indium bumps decreased by 0.4 μm. Conversely, the impact of kinematic roughening in the high deposition rate specific gravity increases, while indium film roughness moderates with the deposition rate. Concurrently, the capacity of indium diffusion from the photolithography pattern edge to the surrounding area is diminished at a high deposition rate and low substrate temperature.Furthermore, the lateral growth rate of the indium layer increases from 3.4 Å/s to 4.1 Å/s following sustained temperature reduction. However, accelerated closure of the photolithography pattern is not conducive to the growth and preparation of high density, high aspect ratio and high uniformity indium bumps. The effects of the deposition rate and substrate temperature on indium bump growth are twofold. In order to obtain indium bump arrays with flat tops, high uniformity, and full morphology, it is necessary to select a suitable combination of substrate temperature and deposition rate parameters for small pitch focal planes.