A passivating contact for silicon solar cells formed during a single firing thermal annealing

Abstract

As highlighted by recent conversion efficiency records, passivating contacts are keys to fully exploit the potential of crystalline silicon as a light absorbing semiconductor. Prime passivating contact technologies include a-Si/c-Si silicon heterojunctions and high temperature tunnel oxide/polysilicon-based contacts. The first has the advantage of a simple fabrication process, but it is incompatible with standard metallization processes and bulk semiconductor defect treatments which take place at temperature > 800°C. The second relies on a buried junction or dopant profile near the tunnel oxide, and requires process times of several minutes at high temperature. In this paper, we solve the scientific question to know whether such a dopant profiles, with the possible the presence of nano-holes, is required to make an efficient contact when using a tunnel oxide. We show that, by leveraging the versatility of plasma deposition processes, it is possible to realize Si-based thin-film doped layers that withstand a short annealing at high temperature (> 800 for typ 10 s, called "firing"), passivate the c-Si interface and foster collection of photo-generated charge carriers by inducing a strong electric field at the Si-surface near the interface with SiOx. The contact has a high-compatibility with existing industrial process: a plasma deposition of a thin-film layer at the rear side followed by a rapid thermal treatment ("firing"), an essential process for metallization formation of industrial cells. With the developed technology, we fabricated proof-of-concept p-type solar cells with conversion efficiency up to 21.9%.

Analysis of hydrogen distribution and migration in fired passivating contacts (FPC)

“…A second observation is that all iVOC values are ~10 mV higher than in the previous experiments, thus exceeding 700 mV even for the samples processed with a chemical oxide (as previously published in [17]). The J0 value decreases from 22 to 8.3 fA/cm 2 for the samples featuring a chemical oxide on a DSP and SE wafer, respectively, and increases from 1.4 to 2.3 fA/cm 2 for the DSP and SE samples with a thermal oxide.…”

“…The first processing step consisted of a wet chemical cleaning, ending with a hot HNO3 treatment (69 %, 80 °C, 10 min) growing a ~1.3 nm thin wet chemical SiOx layer on the wafer surfaces [22,23]. Next, a ~25 nm thick hydrogenated amorphous a-SiCx(p):H (~2.5 at.% of carbon [17]) layer was deposited by Plasma Enhanced Chemical Vapor Deposition (PECVD) at 200 °C. Subsequently, the samples were fired for 3 s at ~800 °C.…”

“…The C content was tuned in order to avert blistering while fostering layer crystallization, which was found to be beneficial for surface passivation and charge carrier extraction. The integration of the FPC as rear hole selective contact, co-fired with a screen printed Ag grid contacting a POCl3 diffused front emitter, resulted in a conversion efficiency of 21.9 % [17].…”

Transparent silicon carbide/tunnel SiO2 passivation for c‐Si solar cell front side: Enabling Jsc > 42 mA/cm2 and iVoc of 742 mV

“…A comparison of both passivation concepts shows that record cells with a‐Si:H(i) passivation developed for the silicon heterojunction (SHJ) technology currently give rise to higher open‐circuit voltages ( V oc ) than using SiO 2 passivation from polycrystalline silicon on oxide (POLO) technology or tunnel oxide passivating contact (TOPCon) technology. Further, for a‐Si:H(i), the excellent interface passivation is directly achieved after the deposition of the a‐Si:H(i) film, whereas for SiO 2 , it requires several additional process steps to provide excellent interface passivation . Reducing the number of process steps is under current investigation within several research groups in the field of c‐Si solar cells.…”

“…Further, for a-Si:H(i), the excellent interface passivation is directly achieved after the deposition of the a-Si:H(i) film, whereas for SiO 2 , it requires several additional process steps to provide excellent interface passivation. 4 Reducing the number of process steps is under current investigation within several research groups in the field of c-Si solar cells. Although a-Si: H(i) is an excellent passivation material, it possesses an optical bandgap 5 of approximately 1.7 eV that leads to significant parasitic absorption of the incident sunlight.…”