Carbon Source and Buffer Layer Effects on 4H-SiC Homoepitaxial Growth

Carbon Source and Buffer Layer Effects on 4H-SiC Homoepitaxial Growth –

1. Abstract

This study used a horizontal hot wall chemical vapor deposition reactor to achieve homogeneous epitaxial growth of 4H-SiC on a 150 mm 4 ° angle substrate. The experiment compared the effects of using propane (C3H8) and ethylene (C2H4) as carbon sources: the epitaxial layer grown using C2H4 showed better thickness uniformity and surface roughness, although the growth rate was slower and the doping concentration was lower, indicating its greater potential in commercial epitaxial wafer production. The study also found that an increase in growth pressure can lead to a decrease in growth rate, deterioration of thickness and doping uniformity, and a nonlinear effect of first improving and then deteriorating surface roughness. Under relatively low pressure conditions of 40 Torr, the overall quality of the epitaxial layer is optimal. In addition, the effect of buffer layer technology on epitaxial quality is regulated by the C/Si ratio, indicating that the reasonable selection of growth parameters is crucial for the performance of subsequent devices.

2. Sample Preparation

This study conducted epitaxial growth experiments on a 4 ° off-axis 4H-SiC substrate with a 150mm in a horizontal hot wall chemical vapor deposition system. The growth process is carried out under the process conditions of 1570 ° C and 40 Torr, using trichlorosilane (TCS) as the silicon source at a flow rate of 50 sccm, and H2 as the dilution, carrier, and etching gas at a flow rate of 100 slm.

The experiment first selected C3H8 and C2H4 as carbon sources, and systematically evaluated the influence of carbon source types on the quality of the epitaxial layer under the condition of a fixed C/Si ratio of 1. Subsequently, pressure parameter comparison experiments were conducted under the conditions of a C/Si ratio of 0.72 and pressures of 40 Torr, 50 Torr, 60 Torr, and 70 Torr, respectively. In addition, buffer layer growth (4 minutes) and subsequent 30 minutes of standard epitaxial process were carried out under the conditions of C/Si=1 and N2 flow rate of 50 sccm.

Fourier transform infrared spectrometer was used to measure the thickness of all epitaxial wafers, and 29 detection points on the wafer surface were selected (including 8 points in the central area and 4 points 5 mm away from the edge), with a focus on analyzing the data from points 1 to 15. The doping concentration is characterized by capacitance voltage testing of mercury Schottky contacts, avoiding the 5mm edge area during measurement. Surface roughness was evaluated using atomic force microscopy within a scanning range of 10μ m × 10μ m.

Fig. 1 Epitaxial layers with different carbon sources: (a) Changes of growth rate and uniformity; (b) Changes of doping concentration and uniformity; (c) Roughness variation

Fig. 2 Influence of different carbon sources on the surface roughness of 4H-SiC Epi: (a), (b) C3H8; (c),(d) C2H4

Fig. 3 Under different pressures (a) Changes in epitaxial growth rate and thickness uniformity; (b) Changes in epitaxial doping concentration and uniformity; (c) Epitaxial roughness variation

Fig. 4 Effects of buffer layer: (a) C/Si=1, growth rate and uniformity; （b）C/Si=0.72, growth rate and thickness uniformity of the epitaxial layer; (c) C/Si=1, the roughness of the epitaxial layer; (d) C/Si=0.72, the roughness of the epitaxial layer

Fig. 5 AFM of SiC epitaxial wafers with and without buffer layers at different C/Si ratios: (a) C/Si=1, without buffer layer; ( b )C/Si = 1, with a buffer layer; ( c )C/Si = 0.72, No buffer layer; ( d )C/Si = 0.72, with a buffer layer

3. Conclusion

This study systematically investigated the effects of carbon source types, growth pressure, and buffer layer conditions on the homogeneous epitaxial growth of 4H-SiC. The results indicate that when C2H4 is used as the carbon source, although the growth rate is slightly lower than C3H8, it has better thickness uniformity and doping distribution stability, making it more suitable for high-quality epitaxial growth. C3H8 helps to achieve lower surface roughness, but its process fluctuates greatly. Overall, C2H4 is the preferred carbon source for 4H SiC epitaxial growth.

Increasing growth pressure can lead to a decrease in growth rate and degradation of thickness and doping uniformity. The surface roughness shows a trend of first decreasing and then increasing with pressure, and the main reason for these changes is the influence of growth pressure on material transport, reaction kinetics, and substrate surface conditions, indicating the need for fine pressure regulation to balance various indicators.

In the buffer layer experiment, the C/Si ratio significantly affects the epitaxial behavior: growing a buffer layer at C/Si=1 inhibits growth rate and uniformity, but can improve roughness; When C/Si=0.72, it is beneficial to improve the quality and growth efficiency of the epitaxial layer. The introduction of the buffer layer needs to be considered in conjunction with the C/Si ratio, and its optimization should serve specific application needs.

Whether you need 4H-SiC homoepitaxial wafer for research or for industrial applications, please contact us email at [email protected] and [email protected].

Reference：

Yuan, W., Pei, Y., Li, Y., Guo, N., Zhang, X., & Liu, X. (2024). Research on the Influence of Carbon Sources and Buffer Layers on the Homogeneous Epitaxial Growth of 4H-SiC. Micromachines, 15(5), 600.