A Comparison of Surface Passivation Techniques for Measurement of Minority Carrier Lifetime in Thin Si Wafers: Toward a
- PDF / 3,058,426 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 28 Downloads / 167 Views
A Comparison of Surface Passivation Techniques for Measurement of Minority Carrier Lifetime in Thin Si Wafers: Toward a Stable and Uniform Passivation Bhushan Sopori1, Srinivas Devayajanam1, 2, Prakash Basnyat1, Vishal Mehta1, Helio Moutinho1, Bill Nemeth1, Vincenzo LaSalvia1, Steve Johnston1, N.M. Ravindra2, Jeff Binns3, and Jesse Appel4 1
National Renewable Energy Laboratory, Golden, CO, USA New Jersey Institute of Technology, Newark, NJ, USA 3 SunEdison, St. Peters, MO, USA 4 SunEdison, Portland, OR, USA 2
ABSTRACT We describe appropriate wafer cleaning procedure and surface passivation characteristics of various passivants used for making measurement of minority carrier lifetime (τB ) of very high quality Si wafers. These passivants include: iodine ethanol (I-E), quinhydrone methanol (QHM), SiO2, and Al2O3. The issues related to the passivation stability and the spatial uniformity for mapping τB are also discussed. INTRODUCTION Measurement of the bulk minority carrier lifetime (τB) is crucial in all semiconductor and solar cell technologies. In the Si technology, the most common applications of τB measurement are to evaluate quality of wafers/ingots, determine cleanliness of furnaces, and establish effects of various process steps during device processing. However, it is well known that a reliable and reproducible measurement of τB is not easy and involves very careful preparation of the wafer and an understanding of the conditions that are valid for appropriate measurement. One of the critical requirements for the measurement of τB is that the wafer surfaces and edges be well passivated (i.e must have low surface recombination velocity, SRV) and it becomes even more stringent for wafers that have higher lifetime. As clearly seen from fig.1, it is very difficult to obtain measured/effective minority carrier lifetime (τeff) close to τb on a τb = 10ms wafer than on a τb = 1ms wafer.
Figure. 1. Calculated effective lifetime as a function of SRV for two cases of bulk lifetime—1 ms and 10 ms.
Because, current wafers used in industry are of very high quality (FZ wafers can exceed 20 ms) it is essential to find ways to attain SRV 10 ms) with wide resistivity ranges, 1-2 Ω-cm, 8-12 Ω-cm, 24-34 Ω-cm, and 150-250 Ω-cm. Several wafer cleaning procedures were evaluated but we will only describe the procedures that gave best results (ref. wafer preparation section). Cleaned wafers were tested with Sinton tester (WTC 100) and Semilab μ-PCD tool for mapping the lifetime. Because of long lifetimes involved, Sinton testing was done in transient mode. WAFER PREPARATION AND PASSIVATION METHODS Wafer preparation is a critical step in getting the right lifetime value with all of these passivation methods. To obtain very low SRV, the wafer surfaces should be free of contamination over the entire passivant-Si interface. We have evaluated two major wafercleaning processes of RCA clean and Piranha clean. We found that, in most cases, like when using a used wafer (previously passivated) a gentle etch is needed to remove a thin layer
Data Loading...
