Hydrogen Movement from Passivating Dielectrics Measured by Mass Spectrometry and Vibrational Spectroscopy

We relate the firing temperature of hydrogen-containing dielectric films on polycrystalline silicon on silicon oxide (<i>poly-</i>Si/SiO<i>_x</i>) passivating contacts to the final passivation of solar cell test structures. By subjecting passivating contact structures with different passivating dielectric films and film stacks to different thermal treatments, we can develop improved processes for firing of high-efficiency Si solar cells. Through isotopic studies using quadrupole mass spectrometry (QMS), Fourier transform infrared spectroscopy (FTIR), and Raman spectroscopy, we determine how hydrogen content and stability within each type of film relates to final passivation quality of solar cell test structures. Because H can be a very difficult element to observe and deuterium is chemically identical to hydrogen within these systems, we will test films containing D in place of H for such studies. D gives different signals from H in FTIR and Raman spectroscopy as well as in QMS, so isotopic substitution can be used as an excellent tool to probe the H behavior within films. Thermal treatments will determine at what temperatures H evolves from the dielectric films and serve as a guide to determine the optimal firing conditions. Following firing of the passivating films, lifetime and saturation current density (<i>J₀</i>) values of <i>poly-</i>Si/SiO<i>_x</i> passivated contact test structures and bare Si are measured using quasi-steady-state photoconductance decay on a Sinton lifetime tester to evaluate overall performance. Si solar cells using passivating contacts are at the forefront of Si solar cell research and emerging as top performers within industrial production. Performance of passivating contact Si solar cell test structures is largely determined by a parameter known as the implied open-circuit voltage <i>iV</i>_oc, which directly relates to material quality within the bulk of the device and at surfaces. High <i>iV</i>_oc is achieved when defects within the bulk crystalline silicon (<i>c-</i>Si) and at interfaces are passivated, preventing them from acting as charge carrier recombination centers. One of the most common ways of passivating defects within Si solar cells is hydrogenation, injecting large amounts of H to satisfy dangling bonds in the bulk and at interfaces. At elevated temperatures, the hydrogen becomes mobile enough to find and satisfy dangling bonds. Industrial silicon solar cells are often metallized through co-firing of silver pastes at high temperatures through SiN<i>_x</i>. This firing results in release of H out of the SiN<i>_x</i> film and into the rest of the device. Though some amount of H is beneficial to passivate defects in the bulk Si and at interfaces, too much H can lead to H-induced degradation or light and elevated temperature induced degradation (LeTID) [1]. It has been shown that SiN<i>_x</i>, Al₂O₃, and <i>a-</i>Si:H passivate the interfaces of <i>poly-</i>Si passivating contacts through different mechanisms, and that combinations of these films can lead to differing performance [2]. Though Al₂O₃ is a stoichiometric dielectric material, SiN_x can have many different values of <i>x</i> depending on deposition conditions. We observe FTIR and Raman spectra over several <i>x</i> values to determine the bonding within them and correlate the relative Si, N, and H concentrations to the stability of H within SiN<i>_x</i> and the passivation of each film. Investigations into the performance of these passivating films and film stacks will lead to greater understanding of dielectrics in semiconductor devices, further improvements in passivated contact design, and greater proliferation of solar energy worldwide.<br/><br/>[1] H. C. Sio<i> et al.</i>, "Light and elevated temperature induced degradation in p-type and n-type cast-grown multicrystalline and mono-like silicon," <i>Solar Energy Materials and Solar Cells, </i>vol. 182, pp. 98-104, 2018.<br/>[2] T. N. Truong<i> et al.</i>, "Hydrogenation mechanisms of poly-Si/SiOx passivating contacts by different capping layers," <i>Solar RRL, </i>vol. 4, no. 3, p. 1900476, 2020.