Non-oxide semiconductors have emerged as promising candidates for photocatalytic hydrogen generation due to their narrow bandgaps and visible-light absorption capabilities. Unlike traditional oxide-based photocatalysts such as TiO2, which are limited by wide bandgaps and UV-light dependency, materials like cadmium sulfide (CdS), graphitic carbon nitride (g-C3N4), and metal sulfides/nitrides exhibit tunable electronic structures that enable efficient solar energy utilization. However, these materials also face significant stability and charge recombination challenges, necessitating careful design and surface modifications to optimize their performance.

Cadmium sulfide (CdS) is a well-studied non-oxide semiconductor with a bandgap of approximately 2.4 eV, allowing it to absorb visible light up to 520 nm. The conduction band position of CdS is sufficiently negative to facilitate proton reduction, making it thermodynamically favorable for hydrogen evolution. However, CdS suffers from severe photocorrosion under prolonged irradiation, where photoexcited holes oxidize sulfur ions in the lattice, leading to material degradation. To mitigate this, surface passivation strategies such as the deposition of thin protective layers or the creation of sulfur-rich surfaces have been explored. These modifications reduce the exposure of CdS to oxidative environments while maintaining its photocatalytic activity. Another critical issue is the rapid recombination of photogenerated electron-hole pairs. Engineering defect sites or introducing electron traps can prolong charge carrier lifetimes, enhancing hydrogen production efficiency.

Graphitic carbon nitride (g-C3N4) is a metal-free semiconductor with a moderate bandgap of around 2.7 eV, corresponding to visible-light absorption up to 460 nm. Its layered structure, composed of tri-s-triazine units, provides abundant active sites for photocatalytic reactions. The material’s high thermal and chemical stability makes it attractive for long-term applications. However, the photocatalytic efficiency of pristine g-C3N4 is often limited by low surface area, poor electrical conductivity, and fast charge recombination. Surface modifications such as exfoliation into ultrathin nanosheets or the introduction of nitrogen vacancies can significantly enhance charge separation and light absorption. Exfoliation reduces the diffusion length for photogenerated carriers, while nitrogen vacancies create mid-gap states that act as electron traps, suppressing recombination. Additionally, protonation of g-C3N4 surfaces can improve hydrophilicity and interfacial charge transfer, further boosting hydrogen evolution rates.

Metal sulfides and nitrides, including MoS2, WS2, and Ta3N5, represent another class of narrow-bandgap semiconductors suitable for visible-light-driven hydrogen production. MoS2, for instance, has a bandgap ranging from 1.2 to 1.9 eV depending on its phase and layer thickness, enabling broad visible-light absorption. The edges of MoS2 nanosheets are particularly active for proton reduction due to their metallic character and favorable hydrogen adsorption energetics. However, the basal planes of MoS2 are catalytically inert, necessitating morphological control to maximize edge site exposure. Nitrides like Ta3N5 exhibit even narrower bandgaps (~2.1 eV) and suitable band positions for overall water splitting, but they are prone to self-oxidation by photogenerated holes. Surface oxidation layers can form, passivating the material and reducing its activity over time. To address this, in-situ generation of protective overlayers or the use of hole scavengers can help stabilize the nitride surfaces during photocatalysis.

Charge transfer mechanisms in these non-oxide semiconductors are strongly influenced by their electronic and crystallographic properties. In CdS, photogenerated electrons are typically localized on cadmium sites, while holes reside on sulfur sites. The spatial separation of charges reduces direct recombination but also necessitates efficient hole transfer to prevent sulfur oxidation. In g-C3N4, the delocalized π-conjugated system facilitates electron mobility along the planes, but interlayer transport is hindered without proper structural modifications. Metal sulfides like MoS2 exhibit anisotropic charge transport, with electrons moving more freely along the layers than across them. Engineering defects or heteroatom doping can alter these charge transfer pathways, improving overall photocatalytic efficiency.

Surface modification techniques play a crucial role in enhancing the performance of non-oxide semiconductors without relying on co-catalysts or composite formation. For CdS, controlled sulfur vacancy engineering can create electron-rich surfaces that promote proton adsorption and reduction. In g-C3N4, covalent functionalization with electron-withdrawing groups can shift the band positions, improving the driving force for hydrogen evolution. For metal sulfides and nitrides, phase engineering—such as converting 2H-MoS2 to the more metallic 1T phase—can dramatically increase charge carrier mobility and active site density. Additionally, surface plasmon resonance effects, when combined with certain sulfides, can enhance light absorption and hot electron injection, further boosting photocatalytic activity.

Despite their advantages, non-oxide semiconductors still face challenges in achieving large-scale photocatalytic hydrogen production. Stability under operational conditions remains a critical issue, particularly for sulfides and nitrides that are susceptible to oxidation or dissolution. Charge recombination, although mitigated through surface modifications, still limits the quantum efficiency of these materials. Future research should focus on developing in-situ characterization techniques to monitor surface transformations during photocatalysis and further refine defect engineering strategies to optimize charge separation.

In summary, non-oxide semiconductors like CdS, g-C3N4, and metal sulfides/nitrides offer compelling advantages for visible-light-driven hydrogen generation due to their narrow bandgaps and tunable electronic structures. However, their practical application requires overcoming stability limitations and charge recombination through targeted surface modifications and a deeper understanding of charge transfer dynamics. Advances in material design and interfacial engineering will be essential to unlock their full potential in sustainable hydrogen production.