GaAs Outshines Silicon in Polymer Solar Cells, Study Finds

In the crowded race to build cheaper, lighter solar cells, one of the most consequential decisions turns out to be one of the least visible to the naked eye: which crystal face of a semiconductor wafer a polymer film is laid upon. A new study from researchers at the University of Zakho in the Kurdistan Region of Iraq shows that the type of substrate and, crucially, its crystallographic orientation can completely reorder the performance ranking of organic/inorganic hybrid solar cells. The work, published in the Journal of Materials Science: Polymers, examined devices made by spinning thin films of the conducting polymer polyaniline (PANI) onto three different n-type semiconductor surfaces: gallium arsenide cut along the (111)B plane, silicon cut along the (100) plane, and silicon cut along the (111) plane. When the lights came on, the GaAs-based device won decisively as a solar cell. When the lights went off, the silicon-based device on the (111) plane emerged as the superior diode. The lesson, the authors argue, is that neither material nor orientation alone tells the whole story; together they govern how these hybrid junctions harvest light and move charge.

Hybrid solar cells of this kind sit at the intersection of two very different material worlds. Organic photovoltaics, built from polymers and small molecules, are inexpensive, lightweight and mechanically flexible, but they suffer from narrow band gaps that encourage rapid charge recombination and limit photocurrent. Inorganic semiconductors, by contrast, offer excellent charge transport and robust light absorption, though at higher fabrication cost. Combining the two promises the best of both, provided the interface between the polymer and the semiconductor can be engineered to separate and transfer photoexcited charges efficiently. PANI has long been a favorite for this role. In its doped, conductive emeraldine salt form it behaves as a p-type conductor, and acid doping shrinks its optical band gap from roughly 3.06 electron volts in the undoped emeraldine base to around 2.35 to 2.54 electron volts, making it far more useful as a light-absorbing, charge-transfer layer. Paired against n-type substrates, it forms the p-n style heterojunction at the heart of the devices in this study.

The choice of inorganic partner matters enormously. Gallium arsenide is a III-V direct-band-gap semiconductor with a gap of 1.42 electron volts, a property that allows it to absorb photons efficiently and to boast high electron mobility. Silicon, the workhorse of photovoltaics, has an indirect band gap of about 1.1 electron volts, which makes absorption less efficient per unit thickness. But the Zakho team’s earlier work had already hinted that something subtler was also at play. Previous studies by co-author Dler A. Jameel and colleagues found that PANI devices on (311)A and (311)B n-GaAs substrates outperformed those on (100) GaAs, with differences in open-circuit voltage, fill factor and rectification ratio attributable to orientation-dependent interface quality, defect densities and charge transport. The new study extends that logic into a direct GaAs-versus-silicon comparison, a combination the authors describe as previously unexamined.

The fabrication itself was deliberately simple, underscoring the practical relevance of the findings. Chemically synthesized PANI was spin-coated at 2000 revolutions per minute onto each of the three substrates, which had been cleaned with deionized water and acetone and dried under nitrogen gas. Current-voltage measurements were then performed at room temperature with a Keithley 2450 source meter, both in darkness and under illumination of 60 milliwatts per square centimeter. From the illuminated current density-voltage curves the team extracted the standard photovoltaic figures of merit: open-circuit voltage, short-circuit current density, fill factor and power conversion efficiency. From the dark measurements, analyzed with the thermionic emission model, they derived the diode parameters, including saturation current, barrier height, ideality factor, series resistance and rectification ratio.

Under illumination, the PANI/(111)B n-GaAs device led on nearly every measure. Its open-circuit voltage of 129 millivolts topped the 110 millivolts of the PANI/(100) n-Si device and the 95 millivolts of the PANI/(111) n-Si device, and its short-circuit current density of 10.8 milliamperes per square centimeter edged out 10.1 for (111) silicon and well above 7.93 for (100) silicon. Fill factors clustered tightly, at 36 percent for the GaAs device against 35 and 37 percent for the two silicon orientations. The decisive margin came in conversion efficiency: 0.83 x 10^-3 percent for the GaAs heterostructure, versus 0.59 x 10^-3 percent for PANI/(111) n-Si and 0.51 x 10^-3 percent for PANI/(100) n-Si. The team attributes the GaAs advantage primarily to its direct band gap, which generates photocarriers more efficiently than indirect-gap silicon, combined with favorable charge separation and transport at the (111)B interface.

The numbers, the authors are careful to note, remain modest compared with the best hybrid devices reported elsewhere. Open-circuit voltages of 342 millivolts on PANI/(311)B n-GaAs and 400 millivolts on PANI/n-Si have appeared in prior literature, and the low values here likely reflect higher interface recombination or differences in film quality and polymer-semiconductor interaction. But the relative rankings are the point. When illuminated performance was set against dark-rectification behavior, a striking inversion emerged: the material that made the better solar cell was not the material that made the better diode. Photovoltaic output, the study concludes, is governed chiefly by the bulk optical properties of the semiconductor, while diode quality is governed chiefly by the quality of the interface, and those two qualities do not necessarily reside in the same substrate.

Dark current-voltage analysis told the second half of the story. The PANI/(111) n-Si device posted the best diode parameters of the trio: a rectification ratio of 50.20 at plus and minus one volt, dwarfing 3.98 for PANI/(100) n-Si and 1.43 for PANI/(111)B n-GaAs, alongside the highest barrier height at 0.69 electron volts and the lowest ideality factor at 3.07. Its saturation current, 2.11 x 10^-7 amperes, was also the smallest of the three. Because an ideal diode has an ideality factor of one, all three devices deviate substantially from ideal behavior, a hallmark of organic-inorganic Schottky junctions where interface states and recombination within the polymer dominate charge transport. Even so, the silicon (111) device comes closest to the ideal, which the authors attribute to a junction with fewer recombination pathways. The (100) silicon device, by contrast, showed a saturation current roughly an order of magnitude higher than its rivals, suggesting greater carrier injection and recombination at that orientation.

Turn-on voltages added further texture. The GaAs device began conducting at the lowest bias, 0.52 volts, indicating easier carrier injection at the junction, while (100) and (111) silicon required 0.58 and 0.63 volts respectively. Series resistance followed the same pattern: 0.35 kilo-ohms for the GaAs device, helped by GaAs’s higher carrier mobility, against 0.81 and 0.37 kilo-ohms for the (100) and (111) silicon devices. Larger barrier heights, the authors note, generally correspond to fewer interface defects and reduced recombination, which is precisely what the silicon (111) interface appears to deliver under dark conditions. The comparison with earlier GaAs studies is instructive: the 0.59 electron-volt barrier on (111)B GaAs sits above the 0.45 electron volts reported for (110) GaAs but below the 0.75 and 0.79 electron volts seen on (100) and (311)B orientations, reinforcing that orientation tunes the barrier landscape in ways that are neither trivial nor predictable from band gaps alone.

The broader implication is that substrate selection in hybrid photovoltaics is a two-dimensional optimization problem. If the goal is light harvesting, the bulk optical properties of the semiconductor, here GaAs’s direct gap, dominate the outcome, and the (111)B GaAs orientation additionally promotes efficient charge separation. If the goal is a rectifying junction, say for photodetectors or diode applications, interface quality takes precedence, and silicon’s (111) plane delivers the cleanest, most rectifying PANI contact measured here. For a field searching for efficiency gains through interface engineering, molecular doping and nanostructuring, the message is that the crystallographic fingerprint of the substrate is not a detail to be controlled for; it is a design variable in its own right. As flexible, solution-processed hybrid devices edge toward real applications, the authors’ systematic comparison suggests that choosing the right crystal face could matter as much as choosing the right material.

Subject of Research: Effects of substrate type and crystallographic orientation on the photovoltaic and diode performance of PANI-based organic/inorganic hybrid solar cell devices.

Article Title: Effects of substrate type and orientation on the photovoltaic and diode performance of PANI-Based organic/inorganic hybrid devices

Article References: Tatar, H. H., & Jameel, D. A. (2026). Effects of substrate type and orientation on the photovoltaic and diode performance of PANI-Based organic/inorganic hybrid devices. Journal of Materials Science: Polymers, 1(1), Article 12. https://doi.org/10.1007/s44493-026-00012-7

Image Credits: AI Generated

DOI: 10.1007/s44493-026-00012-7

Keywords: hybrid solar cells, polyaniline, PANI, GaAs, silicon, substrate orientation, heterojunction, photovoltaics, open-circuit voltage, fill factor, barrier height, ideality factor