Electrical prosperties of reactive ion etached black silicon- Shaozhou Wang

Faculty of Engineering School of Photovoltaic and Renewable Energy Engineering Electrical Properties of Reactive Ion Etched Black Silicon Shaozhou Wang (PhD candidate) Research team: Malcolm D. Abbott1,2, Bram Hoex1, David N. R. Payne1,3, Giuseppe Scardera1， Tsun H. Fung1, Muhammad Umair Khan1,Yu Zhang1, Fajun Ma1 1 University of New South Wales, Sydney, NSW, Australia 2 PV Lighthouse, Coledale, NSW, Australia 3 Macquarie University, Sydney, NSW, Australia 2 Many Different Textures Studied • Studying both black silicon (MCCE and RIE) and conventional textures • Surface features and optical properties thoroughly studied and compared 3 Holistic Approach • Structural & Optical Characterization • Opto-electrical Simulations • Process Integration & Optimization • Energy Yield Analysis • Why study RIE b-Si? • Nano-texturing is industry trend. Near-zero-reflectance b-Si can potentially improve solar cell efficiency to a new level. • Symmetric RIE structure is ideal for the fundamental study of near-zero reflectance b-Si • Challenges with b-Si for solar cell integration: inferior electrical properties Characteristic Electrical Properties of RIE b-Si Guide the improvement on industrial (MACE) b-Si Fung, TH et al., Solar Energy Materials and Solar Cells, 2019 (submitted) RIE b-Si and random pyramid samples Emitter optimization Surface passivation Supported experiments: Importance of field- effect passivation Simulation: Minority carrier distribution in nano-texturing Collection efficiency Characterization: SEM-DCI to determine junction depth 4 Samples for Electrical Property Study • Extreme RIE b-Si (area factor 4.2) was compared by random pyramid (area factor 1.4) • 6 different diffusion recipes (different oxidation conditions) were used • Surface passivated by ALD SiO2/Al2O3 stack and enhanced by corona charge deposition • Characterization: collection efficiency (QE) and J0e Fung, TH et al., Solar Energy Materials and Solar Cells, 2019 (submitted) 5 Fixed charges (Qf) repel surface minority carriers 6 • Seff: effective surface recombination velocity Surface Passivation Mechanisms -10 -5 0 5 10 10 0 10 1 10 2 10 3 10 4 1/ Q f 2 No rmal ize d S e ff Q f (10 12 cm -2 ) 1/ Q f 2 Reduction of recombination centres (Dit) 10 8 10 9 10 10 10 11 10 12 10 13 10 14 10 15 10 -1 10 0 10 1 10 2 10 3 10 4 10 5 10 6 No pa s s i v ati on No rmal ize d S e ff D it ( eV -1 cm -2 ) Be s t p oss i bl e Seff scales with 1/Qf2 and is proportional with Dit, i.e. Qfis a more powerful “knob” to turn Surface Passivation Improvement by Increased Charge • Passivation by ALD SiO2/Al2O3 stack • Strong field-effect passivation can improve J0e effectively Fung, TH et al., Solar Energy Materials and Solar Cells, 2019 (submitted) 7 Challenges from Enhanced Surface Area: Higher J0e • We got the lowest J0e among the reported results in the literature • Our low J0e can be attributed to improved profile and excellent surface passivation • However, J0e of b-Si is still inferior to random pyramids and polished surface, i.e. there still room for optimisation Fung, TH et al., Solar Energy Materials and Solar Cells, 2019 (submitted) 8 Why Field-effect Passivation is Significant? 1E 14 1E 15 1E 16 1E 17 1E -04 1E -03 1E -02 Al2O3 o n Plan ar Al2O3 o n 2 min RIE b- Si Al2O3 o n 6 min RIE b- Si Mino rity Ca rr ier Life time (s) Mino rity Ca rr ier Density (cm -3 ) 1E 14 1E 15 1E 16 1E 17 1E -04 1E -03 1E -02 HfO2 on Plana r HfO2 on 2 min RIE b-Si HfO2 on 6 min RIE b-Si Mino rity Ca rr ier Life time (s) Mino rity Ca rr ier Density (cm -3 ) 6 min RIE2 min RIE 500 nm Area factors are 4.32 and 4.16 determined by AFM • Strong field-effect passivation can passivate the large surface area effectivelyALD films with different interfacial properties 9 3.5 0E +01 1 7.0 0E +01 1 1.0 5E +01 2 1.4 0E +01 2 0.0 0E +00 0 5.0 0E +01 0 1.0 0E +01 1 1.5 0E +01 1 2.0 0E +01 1 Al2O3 o n HfO2 on Midg ap Dit (e V -1 cm -2 ) Nega tiv e Q f / q ( cm -2 ) Characteristic Carrier Distribution in Nano-texturing Valley Peak Peak Valley • Simulated by Sentaurus TCAD • Undiffused RIE b-Si with area factor 5.1 • Low Sn0 and Sp0 • Doping = Injection = 1e15 cm-3 10 200 nm 500 nm • 2D/3D nano-scale simulations to verify hypothesises of field-effect passivation on b-Si Dopant Profile for Black Silicon • Advanced doping characterization method: Scanning Electron Microscope Dopant Contrast Image (SEM-DCI) • Valley of the feature has the same junction depth as RP • Peak of the feature has a much deeper junction depth SEM-DCI images of random pyramid and RIE b-Si with the same diffusion recipe Fung, TH et al., Solar Energy Materials and Solar Cells, 2019 (submitted) 11 Challenges from Enhanced Surface Area: Lower QE • Advanced QE measurement: photoluminescence- based spectro-response (PL-SR) method • Our optimized diffusion recipe can improve QE of b- Si, but it is still much lower than the referenced random pyramid at short wavelength Paduthol, A et al., IEEE J. Photovoltaics, 2018 Fung, TH et al., Solar Energy Materials and Solar Cells, 2019 (submitted) 12 • Increasing the oxygen concentration can improve the diffusion for b-Si, but the J0e and QE are still not comparable to conventional texturing wafers • b-Si J0e improved via emitter optimization and enhanced charge • For non-diffused b-Si, the film with strong Qf are necessary as Seff has a quadratic dependence on Qf • 2D/3D nano-scale simulations used to monitor impact of charge on minority carrier distributions • Contactless QE technique used to characterise collection efficiency of extreme b-Si • Dopant contrast imaging used to characterize doping in extreme b-Si features Conclusion 13 Questions? Acknowledgement Shaozhou Wang shaozhou.wang@unsw.edu.au 14