Novel Technique to Characterize Minority and Majority Carrier Contact Resistivity

In this work we report a novel method of measuring both majority and minority carrier contact resistivities. In the pursuit of highly efficient silicon solar cells, a great deal of research has been done on passivating and selective contacts [1-3]. The contacts of these cell must be well passivating and conductive, and the accurate characterization of contacts is of great importance to silicon PV. The transfer length method (TLM) is widely used to characterize the contact resistivity of PV devices. However, TLM can only be used to measure the majority carrier contact resistivity. To measure the minority carrier contact resistivity with TLM, a contact must be created on the same type of wafer as the minority carrier contact to avoid blocking diode effects. It is not clear that this contact will have the exact same properties as a contact on the original wafer. Our technique can be used to measure the resistance of electrons and holes separately on the same type of wafer used in the final cell, allowing for the accurate characterization of both contacts.<br/><br/>In our method, a thin tunneling oxide is grown on n-Cz silicon. Then heavily doped polysilicon of the same type, either n or p-type, is deposited on both sides of the device to create carrier selective contacts (either electron or hole selecting). The devices have a metal grid on the front, and a full back metal contact (the typical metallization scheme of a bipolar solar cell device). By increasing illumination, the apparent resistance (measured between the front and back metal) of the bulk silicon substrate will approach zero due to increasing wafer conductivity from the high injection of photocarriers, and only the resistance of the contacts will be detected. Under illumination, the effect of the blocking diode is decreased due to increased minority carrier density and a weaker field in the diode. Since the contacts are carrier selective, practically only electrons or holes will conduct through the contact, depending on the contact used, and allows for the measurement of both types of contacts as they would behave in practice. We can then plot the total resistance measured against the inverse of illumination. This gives a very distinct plot that approaches the total contact resistance at the y-intercept. The value of contact resistivity measured in this way has been verified with TLM measurements for n-type contacts on an n-type wafer. We have also characterized the resistivity of p-type contacts on an n-Cz wafer. Unlike the TLM method, we can observe the effects of spreading resistance caused by contact geometry without any additional calculations. This is seen as the total resistance levels out as the inverse of illumination approaches infinity. While this value should ideally correspond to the bulk resistance of the underlying wafer, we observe it to be higher than expected, likely due to spreading resistance between the two contacts.<br/><br/>This technique can give more accurate insight into the causes behind problematic series resistance than TLM and can be used to optimize contact design as well as contact geometry. In addition to the work mentioned above, we will present extensive simulations of this measurement done in Quokka 3, show how this can be used to characterize full devices, and demonstrate how this technique can be used in calculating contact selectivity [1]. We suspect that this method can be used to expand the definition of contact selectivity and will present our investigation of material property effects on selectivity.<br/><br/>References<br/>[1] Brendel et al., 2016 <i>IEEE J. Photovoltaics</i>, vol. 6, no. 6, pp. 1413–1420<br/>[2] Yan <i>et al.</i>, 2022 <i>Prog. Photovoltaics Res. Appl., </i>pp. 1- 17<br/>[3] Feldmann et al., 2014 <i>Sol. Energy Mater. Sol. Cells</i>, vol. 120, pp. 270–274