We present our own Interdigitated Back Contact (IBC) technology, which was developed at ISC Konstanz and implemented in mass production with and at SPIC Solar in Xining, China, with production efficiencies of over 24%. To our knowledge, this is the highest efficiency achieved in the mass production of crystalline silicon solar cells without the use of charge-carrier-selective contacts. With an adapted screen-printing sequence, it is possible to achieve open-circuit voltages of over 700 mV. Advanced module technology has been developed for the IBC interconnection, which is ultimately simpler than for conventional double-sided contacted solar cells. In the next step, we will realize low-cost charge-carrier-selective contacts for both polarities in a simple sequence using processes developed and patented at ISC Konstanz. With the industrialisation of this process, it will be possible to achieve efficiencies well above 25% at low cost. We will show that with the replacement of silver screen-printed contacts by copper or aluminium metallisation, future IBC technology will be the end product for the PV market, as it is the best performing c-Si technology, leading to the lowest cost of electricity, even in utility-scale applications.
An in situ oxidation during the boron tribromide diffusion process to form $\text{p}^{+}$-doped junctions on crystalline Si solar cells leads to the formation of a layer stack system consisting of a borosilicate glass (BSG) (a binary B$_{2}$O$_{3}$–SiO$_{2}$ system) and a SiO$_{2}$ layer at the Si interface. We present a method to passivate the $\text{p}^{+}$-doped regions by using this in situ grown SiO$_{2}$ in combination with a plasma-enhanced-chemical-vapor-deposited SiN$_{x}$ layer. We show that the etching rate of the BSG layer, in a hydrofluoric acid solution, varies over the wafer. The etching rate depends on its local B$_{2}$O$_{3}$ content in the BSG and is markedly higher than that of the SiO$_{2}$ layer. This difference in the etching rates can be used to controllably etch back the BSG layer in order to obtain a thin and uniform passivating oxide layer for solar cell application. Using this oxide/SiN$_{x}$ stack, we obtained implied $V_{\text{oc}}$ of 705 mV and $J_{0e}$ as low as 14 fA/cm$^{2}$ on symmetrically diffused boron emitters on $\text{n}$-type Czochralski wafers. These passivation results are comparable, on similar structure and boron emitters, with state-of-the-art Al$_{2}$O$_{3}$-based passivation methods. Moreover, we have successfully implemented this passivation method in the fabrication of n-type passivated emitter and rear totally diffused and interdigitated back contact solar cells, avoiding the need of adopting additional process steps and costs for boron emitter passivation.
Uniform and low contact resistances between Ag grids and Si emitters are very important for reducing series resistance losses in screen-printed silicon solar cells. In this paper, we studied systematically the contact resistance of firing-through Ag pastes on samples that were processed in a fast firing furnace with and without the presence of busbars. We fabricated p-type full-area aluminum back surface field and n-type passivated emitter rear totally diffused solar cells with various doping concentrations and surface morphologies. For all tested conditions, lower contact resistances were obtained when firing without busbars. For instance, by omitting the busbars during firing, on a 138 Ω/sq phosphorus emitter, the median contact resistance dropped from 7.5 to 0.7 mΩ·cm 2 , and on an 85 Ω/sq boron emitter, the median contact resistance decreased from 80 to 1 mΩ·cm 2 . Furthermore, we found that the pure Ag paste outperformed the AgAl paste in contacting boron emitters. An explanation of these results might be the short-circuit effect, which is a fundamental mechanism that strongly influences contact formation. In addition, in spite of the significant impact of the presence of busbars during firing on contact resistances, its impact on the 10,met was found to be only marginal.
8 August 2026 in Physica Status Solidi (RRL) – Rapid Research Letters