Metal sulfides represent a potentially disruptive innovation for solar energy conversion, but interfacial defects, poor energetic barrier heights, anodic decomposition, and other fundamental surface chemistry issues remain. Elucidating and controlling surface structure and reactivity will overcome these hurdles.
A detailed understanding of metal sulfide interfaces will result in materials with improved solar energy conversion characteristics. With bandgaps generally smaller than comparable oxides resulting in better spectral overlap, compounds such as stibnite, Sb2S3, show promise, but performance issues remain. Success in this program will be measured against our ability to attack the following knowledge gaps regarding an improved understanding of Sb2S3 and metal sulfides in general.
As tandem absorber structures become more popular, efforts must focus on materials with band gaps that will provide the best overlap with the solar spectrum. For a tandem structure with silicon (Eg = 1.12 eV) as the small Eg absorber, maximum efficiency occurs when the larger gap is 1.6-1.7 eV. Band gap engineering of oxides has produced materials such as BiVO4, but its 2.5 eV band gap remains well outside the range to maximize tandem light absorption.
Beyond well-researched MS2 absorbers such as MoS2 and WS2, recent experiments highlight the viability of < 2 eV band gap MS2 and M2S3 materials. Stibnite, Sb2S3, is particularly interesting with a direct Eg = 1.74 eV, > 7% photovoltaic efficiencies, and carrier collection up to 12.6 mA cm–2, (ideally 21 mA cm–2 under AM 1.5 illumination).
What are the stibnite knowledge gaps?
Despite promising results, reported Sb2S3 solar conversion efficiencies vary based on preparation method, surface passivation, and device structure. Interfacial MgO improves Sb2S3 photovoltaic performance, and WO3 species from silicotungstic acid increase PV and photoelectrochemical performance, however the exact mechanism of improvement remains unclear. Although researchers achieve improved voltages and current, interfacial stability significantly still affects Sb2S3 photoperformance.
The variation in reported performance, and the discrepancy between the 7% experimental and the 25% theoretically achievable efficiency indicate that interfacial properties must be stabilized and controlled. How do additives in the chemical bath synthesis affect the surface? What interfacial chemical states yield improved or decreased performance? Would organic functionalization or atomic layer deposition (ALD) similarly effect long-term passivation? Do adlayers surface states and passivating films promote hole conduction? Can we understand and manipulate surface passivation for stable water oxidation without anodic decomposition?
The central thesis of the Grimmgroup’s sulfide efforts is that we can characterize the chemical species that exist at Sb2S3 surfaces, and that the knowledge gained will lead to improved Sb2S3-based photovoltaics. Will it work? Stay tuned?
How do we synthesize metal sulfides?
Students in the Grimmgroup get to synthesize Sb2S3 from elemental Sb and S in a process called chemical vapor transport. (Have you put together the reason why antimony’s atomic symbol is “Sb”?)
The first step is to create a sealed quartz tube that will serve as the reaction vessel. The Sb, S, any dopant elements, and a small amount of iodine are loaded into a quartz tube. That tube is mounted into a heavily modified Rotovap and evacuated of all gas by a vacuum pump. (Remember, we’re making Sb2S3 at very high temperatures, and if there’s oxygen or water in the tube we won’t make antimony sulfide, we’ll make antimony oxide!) Rotating the tube and applying a natural gas-oxygen flame will seal the tube and create our reaction “vessel” of interest.
The second step is to load the tube into a two-zone tube furnace where the chemical vapor transport process takes place. Two zone control is ideal for vapor transport syntheses because the temperature gradient from the source Sb, and S zone to the slightly cooler Sb2S3 deposition zone drives the synthesis. Because we’re condensing the Sb2S3 out of gas-phase species, we’re basically throwing atoms at our crystal one-at-a-time. This is a very slow process that requires days of reaction time to produce millimeter-sized crystals. The upshot of this atom-by-atom growth is the ability to grow very high-quality crystals. The third step is to verify that your compound adjectives are hyphenated. That’s how we roll.
Sb2S3 and MQP’s
As an MQP student in the Grimmgroup, you would get to grow stibnite and vary the recipe somewhat to characterize the effect of stoichiometry, dopants, and temperature on the crystals. Cool, huh? Did we mention fire?