Access Official Technical PDF

Share Summary:LinkedInEmail

Pyrite (FeS2) has attracted interest as a photovoltaic material because it combines a suitable band gap with strong light absorption and consists of abundant elements. Yet pyrite solar cells have persistently delivered low open-circuit voltages (VOC), limiting reported conversion efficiencies. Electrical transport measurements on pyrite single crystals now provide evidence for a previously unaccounted-for feature: a junction between a p-type surface layer and an n-type interior. Its relatively low barrier may help explain why pyrite-based devices have struggled to generate higher voltage.

Why an internal junction matters

A photovoltaic cell needs to separate light-generated charge carriers and maintain a voltage between its terminals. In pyrite, researchers have faced a longstanding materials challenge: controllable doping has not enabled practical p-n homojunction cells, while heterojunction devices have also shown low VOC. Understanding the electrical behavior of the pyrite surface and its interface with the crystal interior is therefore important to explaining the voltage shortfall.

Earlier work established that n-type pyrite can have a thin, conductive surface layer. The layer may conduct as either p-type or n-type in as-grown crystals, but polishing has generally produced p-type surfaces. The crystal interior, meanwhile, is n-type. This combination implies an internal p-n junction, with a depletion region between the surface and bulk. However, transport models commonly treated the surface and interior as two parallel conduction paths, without including resistance at the junction between them.

That omission matters because current traveling through the interior must cross the interface between the n-type bulk and p-type surface. If this junction presents substantial resistance, it can restrict charge flow even when the interior itself is conductive.

Transport measurements expose the junction

To make the junction’s contribution more visible, the researchers measured horizontal electrical transport in pyrite single crystals with different levels of sulfur-vacancy doping. Sulfur vacancies act as deep donors, increasing the electron concentration and changing the interior’s resistance. This variation helped distinguish the electrical response of the bulk from that of the surface and the intervening junction.

In heavily sulfur-vacancy-doped crystals, resistance rose sharply as the samples cooled below approximately 200 K. The behavior was particularly evident when the interior became more conductive with cooling. If current could continue to flow freely through the bulk, this trend would not produce the observed steep increase in sheet resistance. Instead, the rise indicates that an additional resistance is restricting transport into the interior.

At lower temperatures, transport shifted toward the conductive surface layer. The researchers interpreted the intervening resistance increase as the signature of the internal junction: as its resistance grows during cooling, the bulk contributes less to measured transport, and surface conduction becomes dominant. The effect was not attributable to the external electrical contacts; similar behavior appeared with different contact materials and methods.

A three-part circuit model

The measurements also showed why a two-resistor model is insufficient for these crystals. That model represents parallel conduction through the surface and interior, but cannot reproduce the pronounced resistance increase seen in heavily doped samples. Adding a junction resistance between those two paths allows the model to capture the behavior across the measured temperature range.

In the expanded model, three contributions shape the measured sheet resistance:

  • Interior resistance: transport through the n-type bulk, based on the temperature-dependent electron concentration and mobility.
  • Surface resistance: transport through the conductive surface layer, described using a variable-range hopping model at low temperatures.
  • Junction resistance: resistance associated with the interface between the p-type surface and n-type interior.

The relative influence of these components changes with temperature. The interior can dominate at higher temperatures, the junction can govern the response at intermediate temperatures, and the surface can carry most of the current at low temperatures. Including the junction resistance allowed the model to reproduce the measured behavior more closely than the two-channel approach.

Barrier heights and the voltage problem

From the transport data, the researchers extracted Schottky barrier heights for 19 crystals. Most values fell between 170 and 400 meV, with an average of about 320 meV; the full range was 130 to 560 meV. These values are substantially below the barrier heights expected from band-bending calculations for heavily doped pyrite, which predicted values above 750 meV.

The measured barriers are also similar in magnitude to open-circuit voltages reported for pyrite heterojunction solar cells. This does not establish that the internal junction is the only factor limiting device voltage. It does, however, directly implicate the junction as a plausible source of voltage loss: a relatively low barrier may make the interface leaky, constraining the voltage a device can sustain.

The barrier heights varied between crystals, including samples with similar nominal doping and surface preparation. The study does not establish the precise cause of this variation, but it points to the junction region’s defects and their spatial distribution as factors worth investigating.

Doping points to a possible control strategy

A further test used cobalt, a shallow donor, to increase electron concentration beyond the levels achieved with sulfur vacancies alone. In moderately cobalt-doped crystals, the junction’s resistance signature remained visible during cooling. At higher doping, with room-temperature electron concentration above 1018 cm-3, the sharp rise disappeared and transport through the interior remained accessible across the measured temperatures.

This result is consistent with the idea that sufficiently high doping can narrow the depletion region and facilitate charge transport across the interface. The researchers describe the heavily doped surface-to-interior contact as Ohmic-like. It is evidence that junction behavior can be altered through doping, not proof that the same approach will resolve voltage losses in a finished solar cell.

Practical implications for pyrite research

For photovoltaic researchers and materials engineers, the central implication is that pyrite’s surface and bulk should not be treated as independent conduction channels without considering the interface between them. The internal p-n junction adds a potential bottleneck that can influence transport and may contribute to low device voltage. Characterizing how doping and defects affect its barrier is therefore an important step toward understanding and potentially mitigating the limitation.

More broadly, these findings show the value of measuring transport across a range of temperatures and doping conditions. In pyrite, the junction becomes detectable when bulk carrier freeze-out no longer masks its contribution. The resulting three-part electrical model offers a more complete description of the crystal than a surface-and-interior model alone. Whether junction control can translate into improved solar-cell performance remains an open engineering question, but the internal interface is now a concrete target for further investigation.

Access Official Technical PDF

About The Author