Abstract
A strong injection level dependence of the effective minority carrier lifetime ($\tau_{eff}$) is typically measured at low injection levels for undiffused crystalline silicon (c-Si) wafers symmetrically passivated by a highly charged dielectric film. However, this phenomenon is not yet well understood. In this work, we concentrate on two of those possible physical mechanisms to reproduce measured $\tau_{eff}$ data of c-Si wafers symmetrically passivated by atomic layer deposited $Al_2O_3$. The first assumes the existence of a defective region close to the c-Si surface. The second assumes asymmetric electron and hole lifetimes in the bulk. Both explanations result in an adequate reproduction of the injection dependent $\tau_{eff}$ found for both n- and p-type c-Si wafers. However, modeling also predicts a distinctly different injection dependence of $\tau_{eff}$ for the two suggested mechanisms if the polarity of the effective surface charge is inverted. We test this prediction by experimentally inverting the polarity of the effective surface charge using corona charges. From the experiments and simulations, it is concluded that surface damage is the most likely cause of the significant reduction of $\tau_{eff}$ at low injection levels.
Introduction
In the development of high-efficiency crystalline silicon (c-Si) solar cells, surface passivation is critical. Dielectric films like Aluminum Oxide ($Al_2O_3$) are widely used to reduce carrier recombination at the silicon surface. However, researchers have consistently observed an unexpected drop in the effective minority carrier lifetime ($\tau_{eff}$) at low injection levels (low light intensities). Understanding this drop is vital for optimizing cells that operate under varying light conditions.
Historically, two primary theories have competed to explain this behavior:
- The Surface Damage Model: Proposes a thin, defective region near the silicon surface, likely introduced during processing, which acts as a site for high recombination.
- The Asymmetric Bulk Model: Suggests that the bulk silicon itself contains defects with significantly different capture cross-sections for electrons and holes.
Standard experimental data often fits both models equally well, making it difficult to distinguish the true physical cause. This study aims to resolve this ambiguity through advanced numerical modeling and targeted experimental validation.
Methodology
The research utilized a combination of high-fidelity simulations and a novel "charge-flipping" experimental technique.
Numerical Modeling
Carrier dynamics were simulated using Sentaurus TCAD. The models incorporated Fermi-Dirac statistics and comprehensive recombination mechanisms, including Shockley-Read-Hall (SRH), Auger, and radiative recombination. Two distinct simulation frameworks were established to represent the Surface Damage and Asymmetric Bulk hypotheses.
Experimental Design
The study used high-quality Float-Zone (FZ) c-Si wafers (both n-type and p-type) passivated with 15 nm of Plasma-Assisted Atomic Layer Deposition (PA-ALD) $Al_2O_3$. To differentiate between the two models, the researchers employed corona charging. Since $Al_2O_3$ naturally possesses a negative fixed charge, spraying the surface with ions allowed the researchers to invert the surface charge polarity to positive.
Characterization
The effective lifetime was measured using Quasi-Steady-State Photo-Conductance (QSSPC), tracking $\tau_{eff}$ across a wide range of injection levels to compare against the simulated predictions.
Results and Discussion
The core insight of this work lies in the "tie-breaker" logic provided by the polarity inversion. Simulations predicted that if the Asymmetric Bulk Model were correct, flipping the surface charge would have a negligible impact on the lifetime's injection dependence. Conversely, the Surface Damage Model predicted a drastic change in behavior upon inversion, as the charge carriers would be pushed into or pulled out of the damaged zone.
Model Validation
The experimental results closely matched the predictions of the Surface Damage Model. When the $Al_2O_3$ charge was flipped from negative to positive using corona charging, the resulting lifetime curves aligned almost perfectly with the surface damage simulations.
Characteristics of the Damaged Layer
The best-fit simulations suggest that the damaged region extends approximately 700 nm into the wafer. While thin relative to the total wafer thickness, this region has carrier lifetimes ($\tau_{n0,it}, \tau_{p0,it}$) in the microsecond range—orders of magnitude lower than the millisecond lifetimes found in the high-quality bulk silicon.
The study also found that this surface damage is most detrimental when the surface is in an "inverted" state (where the surface charge polarity is opposite to the bulk doping). When the surface is in "accumulation," the high carrier concentration effectively masks the impact of the damaged layer.
Conclusion
This investigation concludes that a thin layer of surface damage is the primary cause of the minority carrier lifetime drop at low injection levels in $Al_2O_3$-passivated silicon wafers. By successfully distinguishing between surface and bulk effects through polarity inversion, the study provides a clearer path for improving passivation processes. Future research is needed to identify the specific processing step—whether sawing, cleaning, or deposition—that introduces this damage, and to explore ultra-low injection regimes using more sensitive techniques like photoluminescence imaging.
Summary updated