Gallium Nitride as Transparent Electron-Selective Contact in Silicon Heterojunction Solar Cells

Abstract

Carrier-selective, passivating contacts have allowed silicon heterojunction (SHJ) cells to reach recordbreaking efficiencies particularly in all-back-contacted designs. However, two-side-contacted SHJ cell efficiency has been limited due in part to parasitic absorption losses up to 3 mA/cm 2 in the a-Si:H layers. More transparent materials could reduce this current loss while minimizing process complexity. Gallium nitride (GaN), with a bandgap of 3.4 eV and an advantageous band alignment with silicon, could be applied as a transparent electron-selective layer. Here, we report on SHJ solar cells utilizing PECVD GaN layers grown at 200°C as electron-selective contact. First devices exhibited open-circuit voltages of ~575 mV due to poor passivation, and low conductivity of the as-yet undoped GaN layers induced high series resistance (Rs). However, this Voc suggests the potential for electron selectivity if appropriate passivation and doping strategies are implemented.

Dopant‐free passivating contacts for crystalline silicon solar cells: Progress and prospects

“…Reliable doping strategies or chemical potential tuning of relevant elements (e.g., oxygen) could be promising solutions and deserve efforts for further development 150,169 . Besides, other than the common metal oxides, materials such as III‐V compounds, 109 ternary semiconductors, 108 or hybrid semiconductors 225–227 could be further explored as potential CTLs due to their proper band matching with c‐Si and the high electrical conductivities 228,229 The instability of CTLs and its incompatibility with adjacent layers due to interfacial reactions or processing constrains need to be properly addressed.…”

“…In addition to oxides, other types of metal compounds such as nitrides, 59,108–111 sulfides, 112–114 halides, 115,116 fluorides, 63,117 phosphides, 118 and alkali/alkaline‐earth metal carbonates have also been exploited as dopant‐free passivating contact for c ‐Si solar cells 119,120 . Great attention has been paid to developing ETLs such as TiN x, 59 TaN x, 111 MgF x, 63 LiF x, 121 and so forth 122 .…”

“…Contact structure τ (μs) J 0c (fA/cm 5C) and a 4-nm MoO x layer contributes to the best cell performance showing a certified efficiency of 23.5% (Figure 5D). 48 In addition to oxides, other types of metal compounds such as nitrides, 59,[108][109][110][111] sulfides, [112][113][114] halides, 115,116 fluorides, 63,117 phosphides, 118 and alkali/alkaline-earth metal carbonates have also been exploited as dopant-free passivating contact for c-Si solar cells. 119,120 Great attention has been paid to developing ETLs such as TiN x,…”

Carrier‐selective contacts using metal compounds for crystalline silicon solar cells

“…363 GaN shows similar attractiveness, or even higher due to its wider bandgap, yet also faces similar implementation challenges. [367][368][369] There are even fewer reports about this approach, in spite of the prospect of epitaxial GaN growth on (111) Si, which is attractive when using textured Si. One common difficulty in the growth of III-V on Si is that traditionally, relatively thick inactive 'buffer' layers are used by the III-V community when growing on foreign substrates (be it sapphire or silicon), to guarantee a high-quality active layer despite the lattice mismatch.…”

“…Nevertheless, this remains an attractive material, and low‐deposition‐temperature approaches are being trialled 363 . GaN shows similar attractiveness, or even higher due to its wider bandgap, yet also faces similar implementation challenges 367–369 . There are even fewer reports about this approach, in spite of the prospect of epitaxial GaN growth on (111) Si, which is attractive when using textured Si.…”