Abstract

We investigate the passivation of crystalline Si (c-Si) surfaces by phosphorus oxide (POx) thin films deposited in an atomic layer deposition (ALD) reactor and capped in-situ by ALD Al2O3. Passivation is demonstrated on both n- and p-type (100) Si surfaces, and for POx/Al2O3 stacks deposited at both 25 °C and 100 °C. In contrast to Al2O3 alone, POx/Al2O3 passivation is activated already by annealing at temperatures as low as 250 °C in N2 in all cases. Best results were obtained after annealing at 350 °C and 450 °C for films deposited at 25 °C and 100 °C respectively, with similar implied open-circuit voltages of 723 and 724 mV on n-type (100) Si. In the latter case an outstandingly low surface recombination velocity of 1.7 cm/s and saturation current density of 3.3 fA/cm2 were obtained on 1.35 Ω cm material. Passivation of p-type Si appeared somewhat poorer, with surface recombination velocity of 13 cm/s on 2.54 Ω cm substrates. Passivation was found to be independent of POx film thickness for films of 4 nm and above, and was observed to be stable during prolonged annealing up to 500 °C. This excellent passivation performance on n-type Si is attributed partly to an unusually large positive fixed charge in the range of 3–5 × 1012 cm−2 (determined from capacitance–voltage measurements) for stacks deposited at both temperatures, which is significantly larger than that exhibited by existing positively charged passivation materials such as SiNx. Indeed, passivation performance on n-type silicon is shown to compare favourably to state-of-the-art results reported for PECVD SiNx. POx/Al2O3 stacks thus represent a highly effective positively charged passivation scheme for c-Si, with potential for n-type surface passivation and selective doping applications.

Highlights

  • Effective surface passivation has long been understood to be essential to realizing high-efficiency crystalline silicon (c-Si) solar cells, and has been an important focus of research for many years

  • We investigate the passivation of crystalline Si (c-Si) surfaces by phosphorus oxide (POx) thin films deposited in an atomic layer deposition (ALD) reactor and capped in-situ by ALD Al2O3

  • Most of these new materials are metal oxides, and most feature a negative fixed charge at their interface with Si which makes them well-suited to passivating p-type Si surfaces, but less well-suited to passivating n-type surfaces. For the latter task SiNx deposited by plasma-enhanced chemical vapour deposition (PECVD) [10], which has been known since the 1980s, remains the preferred solution, due partly to its relatively large positive charge, which has remained unmatched by newer passivation materials

Read more

Summary

Introduction

Effective surface passivation has long been understood to be essential to realizing high-efficiency crystalline silicon (c-Si) solar cells, and has been an important focus of research for many years. In the last decade the list of materials known to be capable of passivating c-Si has expanded from the well-established Si-based compounds (SiO2, SiNx, and a-Si:H) to embrace a wide variety of newer materials [1], including Al2O3 [2], AlN [3], Ga2O3 [4], TiO2 [5], Ta2O5 [6], HfO2 [7], Nb2O5 [8], and ZnO [9] Most of these new materials are metal oxides, and most feature a negative fixed charge at their interface with Si which makes them well-suited to passivating p-type Si surfaces, but less well-suited to passivating n-type surfaces. We have recently demonstrated unprecedentedly effective passivation of InP surfaces using a non-metal oxide, namely phosphorus oxide (POx), capped by Al2O3 [11] In this structure, the Al2O3 capping layer acts as a moisture barrier to provide chemical stability to the POx, which is known to be highly hygroscopic, and unstable when exposed to atmospheric moisture (uncapped POx films were observed to undergo rapid visible degradation on exposure to atmosphere). We show that POx/Al2O3 occupies a unique position among existing passivation schemes due to its unmatched positive charge, and compare POx/Al2O3 passivation performance to state-of-the-art results for PECVD SiNx

Experimental details
Results and discussion
Conclusions
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call