Simulation of infrared spectra of trace impurities in silicon wafers based on the multiple transmission–reflection infrared method

The content of trace impurities, such as interstitial oxygen and substitutional carbon, in silicon is crucial in determining the mechanical and physical characteristics of silicon wafers. The traditional infrared (IR) method is adopted as a normal means to analyse their concentration at home and abroad, but there are two problems. The first problem is the poor representativeness of a single local sampling point because the impurity distribution in a solid sample is not as uniform as that in a liquid sample. The second problem is that interference fringes appear in the infrared spectra of the sample due to the thin wafer (≤ 300 μm thick). Based on this, controversial issues existed regarding the measured trace impurity concentrations between wafer manufacturers and solar cell assembly businessmen who used silicon sheets made by the former. Therefore, multiple transmission-reflection (MTR) infrared (IR) spectroscopy was proposed to solve the problems mentioned above. In the MTR setup, because light passes through different parts of the silicon chip several times, multiple sampling points make the final result more representative. Moreover, the optical path is lengthened, and the corresponding absorbance is enhanced. In addition to amplification of weak signals, the MTR-IR method can eliminate interference fringes via the integrating sphere effect of its special configuration. The signal-to-noise ratio of the corresponding spectrum is considerably improved due to the aforementioned dual effects. Thus, the accuracy and sensitivity of the detection method for trace impurities in silicon chips are greatly increased. In this study, silicon wafers were placed in the MTR setup, and then, their relative properties at room temperature were investigated. The corresponding theoretical calculation and simulation of infrared spectra of silicon chips were provided. This affords an optional method for the semiconductor material industry to analyse trace impurities in their chips.