How Defect Density Impacts Mechanics of 4H-SiC Substrate?

  • Home
  • News
  • How Defect Density Impacts Mechanics of 4H-SiC Substrate?

How Defect Density Impacts Mechanics of 4H-SiC Substrate?

PAM-XIAMEN can supply SiC substrates with various specifications, please get more info from: https://www.powerwaywafer.com/sic-wafer/sic-wafer-substrate.html

1. Significance of Research on Correlation Between Defect Rate and Mechanical Strength

It is well known that SiC defects have great negative impact to the electrical reliability and performance of chips; Their impact on the mechanical yield of devices has not been properly studied. Therefore, it is useful to study the correlation between the defect rate (whether intrinsic or process induced) of 4H-SiC substrates and its mechanical strength. Bending strength, referring to the stress that the mold can withstand under the load applied at the time of fracture, has been used as a reference parameter. The three-point bending (3PB) test has been performed on identified substrate areas and collected data, which is correlated with the corresponding defect density.

2. Study of Defect Density on Mechanical Characteristics of 4H-SiC

This study conducted a control experiment using a 150 mm 4° off-angle 4H-SiC substrate:

One group of SiC substrates characterized by different dislocation densities (wafer # 1 and wafer # 2) was used to study the effect of inherent SiC defects on mechanical strength. Perform a 5 min KOH wet etching process at 500℃ to highlight SiC crystal defects. The optical inspection of the wafer surface after KOH etching has been carried out using an automatic optical microscope (nSPEC by Nanotronics) to monitor the density and distribution of defects. The entire chip was inspected using an x10 objective lens.

And another group of substrates underwent thinning treatment, followed by high-temperature bulk heat treatment and high-throughput laser treatment, to study the effect of process defects on mechanical strength.

A failure strength investigation was conducted on all wafers through a 3-point bending test (Instron 5566), and the correct selection was made based on the defect density on the wafers on various 5×4 mm molds. The tool accuracy is 0.5%, and the supporting span length is 3mm. The load is applied at the center of the span and increases at a constant rate of 0.5 mm/min.

Fig. 1 Distribution of defect density at the wafer scale after KOH         

Fig. 1 Distribution of defect density at the wafer scale after KOH using n-spec analysis, with high and low density areas marked by black and red circles respectively. Etch wafer # 1 a) and wafer # 2 b).

In the experiment, it was found that BPD was mainly caused by stress caused by lattice distortion strain in both high-density and low-density areas investigated. In high-density areas, BPD array half ring array (HLA) was found.

Conduct a three-point bending test on the selected mold to calculate the stress that the mold can withstand when it fractures σ (Bending strength), according to the theory of suspension beams with known geometric shapes supported at both ends, can be expressed as: σ= 3FL/2Wd2.

In the formula, F is the maximum load, which causes the mold to fracture, and L, W, and d are the length, width, and thickness of the mold, respectively. In this experiment, the mold size is 5 × 4 millimeters and the mold thickness is 350 um. In areas with defect densities as high as 2000/mold size, no significant changes in mold strength were observed. It is worth noting that molds with low bending strength corresponding to low defect density account for 7% -10% of the total analyzed low-density molds.

Fig. 2 Defect density as a function of each wafer # 1 (red) and wafer # 2 (blue) in the wafer region

Fig. 2 Defect density as a function of each wafer # 1 (red) and wafer # 2 (blue) in the wafer region

Fig. 3 Variation of bending strength of inherent defects with defect density

Fig. 3 Variation of bending strength of inherent defects with defect density

Fig. 4 Variation of bending strength as a function of defect density induces defects

Fig. 4 Variation of bending strength as a function of defect density induces defects

The results indicate that within the range of high defect density in SiC crystals, the bending strength of the mold remains basically unchanged. However, the defect density generated during the thinning process on the SiC surface reduces the mold strength by 1.4 times, and this value remains almost unchanged under subsequent high-temperature treatment.

For more information, please contact us email at victorchan@powerwaywafer.com and powerwaymaterial@gmail.com.

2024-04-01

Share this post

Related Posts

PMN-PT CRYSTALS

PAM XIAMEN offers PMN-PT Crystals. 10 x 10 x 0.5 mm PMN-PT Piezoelectric Crystal (111) Pb(Mg1/3Nb2/3)O3-PbTiO3 SSP 10 x 10 x 0.5 mm Pyroelectric Crystal PMN-PT (111) Pb(Mg1/3Nb2/3)O3-PbTiO3 SSP 5 x 10 x 0.5 mm PMNT Electro-Optic Crystal (001) Pack SSP 5 x 10 x 0.5 mm PMNT [...]

2019-03-14meta-author
GaSb (Gallium Antimonide) Substrate

GaSb (Gallium Antimonide) crystal substrate is available with P type or N type in various sizes. The minimum value of GaSb substrate mobility can reach 200 cm2/V.s, and the maximum can reach 3500 cm2/V.s. PAM-XIAMEN’s gallium antimonide semiconductor material is a single crystal grown by a special [...]

2019-03-12meta-author
4″ Silicon EPI Wafer-2

PAM XIAMEN offers 4″ Silicon EPI Wafers. Substrate EPI Comment Size Type Res Ωcm Surf. Thick μm Type Res Ωcm 4″Øx525μm n- Si:As[111] 0.0010-0.0035 P/E 37 p- Si:B 35±10% P/N/N+ 4″Øx525μm n- Si:As[111] 0.0010-0.0035 P/E 16.5 n- Si:P 12.5±10% P/N/N+ 4″Øx525μm n- Si:As[111] 0.0010-0.0035 P/E 45 p- Si:B 13±10% P/N/N+ 4″Øx525μm n- Si:As[111] 0.0010-0.0035 P/E 7±1 n- Si:P 12±10% P/N/N+ 4″Øx525μm n- Si:As[111] 0.0010-0.0035 P/E 45 p- Si:B 14.5±10% P/N/N+ 4″Øx525μm n- Si:As[111] 0.0010-0.0035 P/E 7 n- Si:P 12±10% P/N/N+ 4″Øx525μm n- Si:As[111] 0.002-0.005 P/E 88 p- Si:B 80.5±10% P/N/N+ 4″Øx525μm n- Si:As[111] 0.002-0.005 P/E 88 n- Si:P 27±10% P/N/N+ 4″Øx380μm n- Si:As[111] 0.002-0.005 P/E 105 p- Si:B 0.0035±10% P/N/N+ 4″Øx380μm n- Si:As[111] 0.002-0.005 P/E 26 n- Si:P 5±10% P/N/N+ 4″Øx525μm n- Si:As[111] 0.0010-0.0035 P/E 10.15 n- Si:P 3.8±0.5 N/N/N+ 4″Øx525μm n- Si:As[111] 0.0010-0.0035 P/E 6.8±0.8 n- Si:P 0.55±0.15 N/N/N+ 4″Øx380μm n- Si:As[111] 0.004-0.008 P/EOx 16.5 n- Si:P 35 ±10% N/N+ 4″Øx508μm n- Si:As[111] 0.002-0.005 P/E 19±1.3 n- Si:P 25±5 N/N/N+ 4″Øx508μm n- Si:As[111] 0.002-0.005 P/E 54.5±3.6 n- Si:P 4.4 N/N/N+ 4″Øx380μm n- Si:As[111] 0.001-0.005 P/EOx 20 n- Si:P 270 ±10% N/N+ 4″Øx400μm n- Si:As[111] 0.001-0.005 P/E 20 n- Si:P 0.09 ±10% N/N+ 4″Øx400μm n- [...]

2019-03-08meta-author
X-ray response of polycrystalline – CdZnTe

X-ray response of polycrystalline – CdZnTe X-ray response of polycrystalline-CdZnTe deposited by thermal evaporation were measured by signal to noise(S/N) analysis. The CdZnTe material has optimal property adquem in solid state X-ray detector and many research presented on single crystal CdZnTe with small sized silicon [...]

2014-03-21meta-author
Lapped Wafers

PAM-XIAMEN offers lapped silicon carbide wafers, As-cut wafers are lapped and cleaned to be qualified as lapped wafers. What is wafer lapping? Wafer lapping is a global flattening process that improves wafer flatness by removing surface damage (usually from back grinding). It is most commonly found [...]

2018-08-07meta-author
Getting to 450 mm Wafer Soone

Getting to 450 mm Wafer Sooner   As device circuits continue to get smaller, the wafers that they are made from are getting larger — again. A few years ago device manufacturers — both vertically integrated companies and independent foundries — pushed maximum wafer diameters from [...]

2013-03-14meta-author