Quasi-steady-state photoconductance bulk lifetime measurements on silicon ingots with deeper photogeneration

Mohsen Goodarzi,Ronald Sinton,Daniel Macdonald

doi:10.1063/1.5086378

Abstract

Quasi-steady-state photoconductance measurements on silicon ingots and blocks with different photo-generation profiles are simulated in this work. The results show that deeper generation profiles, achieved by using a long-pass optical filter with a longer cut-off wavelength, can reduce the impact of the high surface recombination velocity of the ingot surface. This results in higher measured effective lifetimes and reduces the reliance on transfer functions to convert the measured lifetimes into bulk lifetimes. However, there exists a trade-off between generating carriers further from the surface to reduce surface recombination and ensuring that the generated carriers are within the sensitivity range of the photoconductance sensing coil. The simulations are compared with experimental results measured on a monocrystalline silicon block using both quasi-steady-state and transient photoconductance decay, as the transient method is relatively less prone to the impact of surface recombination, and provides a lower bound on the bulk lifetime. The results confirm an increased accuracy in bulk lifetimes extracted from quasi-steady-state measurements when measured using deeper photo-generation profiles.

Highlights

As a very well established method for the measurement of carrier lifetimes on silicon wafers, the quasi-steady-state photoconductance (QSSPC) method1 has been recently applied to ingot and block lifetime measurements, and the characterization of crystal defects and impurities within them.2,3 This application can provide more immediate feedback on crystal growth quality through specific knowledge about the impurities and crystal defects in the ingots and blocks
QSSPC lifetime measurements were performed at one fixed point (10 mm from the bottom of the block) with different filters to compare the lifetime as a function of injection level for the two filters
This paper demonstrates improved QSSPC effective lifetime measurements, in the lifetime range above 50 μs, on silicon blocks by using deeper photo-generation profiles via filters with longer cut-off wavelengths

Summary

Introduction

As a very well established method for the measurement of carrier lifetimes on silicon wafers, the quasi-steady-state photoconductance (QSSPC) method has been recently applied to ingot and block lifetime measurements, and the characterization of crystal defects and impurities within them. This application can provide more immediate feedback on crystal growth quality through specific knowledge about the impurities and crystal defects in the ingots and blocks. As a very well established method for the measurement of carrier lifetimes on silicon wafers, the quasi-steady-state photoconductance (QSSPC) method has been recently applied to ingot and block lifetime measurements, and the characterization of crystal defects and impurities within them.. As a very well established method for the measurement of carrier lifetimes on silicon wafers, the quasi-steady-state photoconductance (QSSPC) method has been recently applied to ingot and block lifetime measurements, and the characterization of crystal defects and impurities within them.2,3 This application can provide more immediate feedback on crystal growth quality through specific knowledge about the impurities and crystal defects in the ingots and blocks. This may lead to making more reliable decisions for cropping ingots and blocks for wafering purposes. It may be possible to sort the wafers at the start of processing and tune the cell process for different quality wafers coming from the different sections of the block. Microwave-detected photoconductance decay mapping has been the most commonly used method for monitoring carrier lifetimes on silicon ingots. More recently, other characterization methods capable of providing such information at the ingot level have been developed, such as the QSSPC method, which is used in this work, and Photoluminescence (PL)-based methods for ingot characterization.

Objectives

Results

Conclusion