Optimum Hydrogen Injection in Phosphorus-Doped Polysilicon Passivating Contacts

Abstract

It has previously been shown that ex situ phosphorus-doped polycrystalline silicon on silicon oxide (poly-Si/SiO x ) passivating contacts can suffer a pronounced surface passivation degradation when subjected to a firing treatment at 800 °C or above. The degradation behavior depends strongly on the processing conditions, such as the dielectric coating layers and the firing temperature. The current work further studies the firing stability of poly-Si contacts and proposes a mechanism for the observed behavior based on the role of hydrogen. Secondary ion mass spectrometry is applied to measure the hydrogen concentration in the poly-Si/SiO x structures after firing at different temperatures and after removing hydrogen by an anneal in nitrogen. While it is known that a certain amount of hydrogen around the interfacial SiO x can be beneficial for passivation, surprisingly, we found that the excess amount of hydrogen can deteriorate the poly-Si passivation and increase the recombination current density parameter J 0. The presence of excess hydrogen is evident in selected poly-Si samples fired with silicon nitride (SiN x ), where the injection of additional hydrogen to the SiO x interlayer leads to further degradation in the J 0, while removing hydrogen fully recovers the surface passivation. In addition, the proposed model explains the dependence of firing stability on the crystallite properties and the doping profile, which determine the effective diffusivity of hydrogen upon firing and hence the amount of hydrogen around the interfacial SiO x after firing.

Comparison of firing stability between p‐ and n‐type polysilicon passivating contacts

“…The p-type poly-Si had a lower doping concentration in the poly-Si film and a larger in-diffusion depth in the c-Si wafer, compared to the n-type poly-Si. In our previous study on the firing stability of n-type poly-Si, 20 the doping density within poly-Si films strongly affected the firing stability, whereas we did not observe a clear correlation between the penetration depth and level of the firing impact. It was found that samples with a lower doping concentration in poly-Si appeared to show stronger firing impact than the more heavily doped n-type poly-Si film, possibly related to the different hydrogen diffusivity.…”

“…It was found that samples with a lower doping concentration in poly-Si appeared to show stronger firing impact than the more heavily doped n-type poly-Si film, possibly related to the different hydrogen diffusivity. 20 Interestingly, this is not detected in the current study, suggesting that the difference in doping concentration in p-and n-type poly-Si films does not contribute to their distinct firing stability. Moreover, we observe different firing response in p-and n-type poly-Si with a similar in-diffusion depth (n-type poly-Si in Figure 8 and ptype poly-Si in Figure 2B).…”

“…Based on the hydrogen profiles measured by SIMS for industrial poly-Si fired at different temperatures or fired with various dielectric coating layers, samples with more hydrogen content around SiO x also exhibited a higher J 0 value. 20 Our previous results indicate that the passivation quality after firing could be determined by the amount of hydrogen released from the dielectric films to the SiO x interlayer, and excess hydrogen around the SiO x can be detrimental to the surface passivation, explaining the pronounced firing impact on industrially processed n-type poly-Si. This speculation was also supported by the increased degradation extent after injecting more hydrogen, as well as the recovered J 0 after reducing the hydrogen content around the SiO x thin layer via a 300 C N 2 anneal after the removal of fired SiN x , 20 similar to what can be found in this work (Figures 6 and 7), which suggests that the hypothesis proposed in our previous work for industrial samples could be also applied to the n-type poly-Si investigated in the current work.…”

“…Note that the hydrogen profiles for n-type poly-Si contacts demonstrated in Figure 4A are taken from our recent work, 20 whereas the SIMS data for p-type poly-Si given in Figure 4B are the actual measurements performed on p-type poly-Si samples shown in Figure 3.…”

“…[11][12][13][14][15][16][17][18] In our recent work, 19 a pronounced surface degradation was observed on n-type ex-situ phosphorus-doped poly-Si contacts fired either with or without the presence of SiN x capping layers at above 800 C, with the degradation extent depending on various processing parameters, such as the growth of interfacial SiO x , the poly-Si deposition temperature, the phosphorus diffusion conditions, and the dielectric capping layer. In other work, 20 we proposed a hypothesis that an excessive amount of hydrogen released from the dielectric capping layer to the poly-Si/ SiO x structure, particularly surrounding the interfacial SiO x , causes the degradation after firing.…”