Abstract
Traditional immersed photocatalysis is constrained by the “water layer dilemma,” where bulk water causes parasitic optical losses and impedes mass transfer. To overcome this, interfacial water state engineering via nonimmersed architectures offers a transformative paradigm shift by decoupling the reaction interface from the bulk liquid. Establishing a novel framework centered on water supply modalities and interfacial states, this review provides a fresh perspective on nonimmersed systems, tracing the transition from externally supplied to autonomous self‐supplied architectures. We dissect how engineering the physical form and phase of water reconfigures the local microenvironment to maximize photon utilization and accelerate gas evolution kinetics. Nevertheless, scaling these systems for widespread application remains challenged by key hurdles in water‐film thickness regulation and heat–mass coupling stability. Addressing these challenges through mechanistic understanding of interfacial water dynamics is essential to unlocking the full potential of nonimmersed architectures, which offer a transformative pathway for geographically versatile distributed hydrogen production and global carbon neutrality.
Keywords: interfacial water management, nonimmersed systems, photocatalytic water splitting
Conventional photocatalysis suffers from the “water layer dilemma.” This review highlights interfacial water state engineering via nonimmersed architectures as a transformative solution. We evaluate these systems across multiple dimensions, analyzing their advantages alongside existing challenges. This paradigm shift represents a crucial direction for advancing solar hydrogen production toward practical applications.
1. Introduction
Since the adoption of the Paris Agreement in 2015, the international community has continued to pursue coordinated mitigation and adaptation policies aimed at “holding the increase in the global average temperature to well below 2 °C above preindustrial levels and pursuing efforts to limit the temperature increase to 1.5 °C” [1]. In the decade since the agreement, parties have undertaken a range of mitigation and adaptation efforts; nevertheless, the latest climate science and policy assessments concur that substantially faster and more ambitious emission reductions, together with an accelerated transformation of energy systems, are still required to meet long‐term climate targets [2]. At the same time, although renewable power capacity and generation have grown markedly in recent years and are driving the decarbonization of electricity supply, the expansion of renewables and electrification alone will not fully address all hard‐to‐decarbonize demands or provide long‐duration energy storage at the necessary scale [3]. Consequently, the development of diverse, flexible, and low‐carbon energy carriers is essential for resilient and scalable decarbonization pathways. Hydrogen—by virtue of its high gravimetric energy density, long‐term storability, and cross‐sector applicability—has therefore emerged as a key complementary medium to renewable electricity, and the advancement of efficient, economical, and widely deployable hydrogen production technologies is of critical importance for achieving the objectives set out in the Paris Agreement [1].
Modern decarbonization strategies increasingly emphasize green hydrogen—hydrogen produced from renewable resources—as a central, low‐carbon energy carrier that can provide long‐duration storage and decarbonize hard‐to‐electrify sectors [4]. Contemporary technical pathways for producing green hydrogen include (i) water electrolysis driven by renewable electricity (alkaline, porton exchange membrane (PEM), and emerging anion exchange membrane (AEM) electrolyzers), (ii) photovoltaic (PV)–electrolysis hybrids and integrated PV–electrolyzer systems, (iii) photoelectrochemical (PEC) devices that combine light absorption and electrolysis within photoelectrodes, (iv) heterogeneous photocatalytic particulate systems that directly catalyze overall water splitting under illumination, and (v) various solar–thermochemical and concentrated solar processes as well as biomass‐derived and biophoto routes [5, 6].
Among various renewable hydrogen production routes, photocatalytic overall water splitting has attracted sustained interest due to its conceptual simplicity—directly converting sunlight into hydrogen within a single semiconductor‐based system through integrated light absorption, charge separation, and surface redox reactions [7, 8, 9, 10, 11, 12]. This inherently integrated approach eliminates the need for external electricity or high‐pressure electrolysis hardware, offering significant potential for low‐cost, modular, and decentralized hydrogen production, especially in off‐grid or resource‐limited settings. Recent advances in optical management, interfacial engineering, photothermal coupling, and water‐management strategies have opened new avenues to address long‐standing challenges related to quantum efficiency and operational stability [13, 14]. As catalyst materials continue to mature, system‐level architecture and reactor design have become increasingly critical, positioning photocatalytic water splitting as one of the most promising pathways in the broader landscape of renewable hydrogen technologies [15, 16, 17, 18].
Fujishima and Honda’s first demonstration of TiO2‐based PEC water splitting not only revealed the fundamental concept of semiconductor‐driven solar water splitting but also established experimental paradigms that have guided materials and reactor research for decades [19, 20, 21]. In subsequent laboratory studies, testing configurations for overall photocatalytic water splitting have been diverse, but they can be broadly classified into two categories. One category comprises dispersed particulate (slurry) systems evaluated in aqueous suspensions [15, 22]; the other comprises plate‐like testing configurations in which catalyst materials are immobilized on substrates and tested as plates or modules [23, 24].
In the process of translating laboratory discoveries into practical applications, several demonstration and engineering studies have adopted those experimental paradigms [25]. For production‐scale devices, however, immobilized plate‐ or fixed‐bed configurations are more commonly employed to facilitate scalable deployment and catalyst management [26, 27, 28], while some pilot‐scale experiments continue to use slurry systems in pursuit of higher instantaneous production rates [29, 30].
Although substantial progress has been made at the material level, most practical photocatalytic reactor designs still operate with the catalyst submerged in the aqueous phase. However, an increasing body of experimental and theoretical work indicates that the interfacial water layer itself acts as a primary bottleneck, introducing significant optical and mass‐transfer resistances that can collectively degrade device‐level performance and stability [31, 32, 33, 34]. To address these limitations, recent research has explored a class of nonimmersed photocatalytic systems founded on the principle of reconfiguring the catalyst–water–light interface [17, 35, 36].
While several reviews have summarized specific aspects such as photothermal coupling, floating platforms, and interfacial engineering [37, 38, 39], the essence of these technological routes converges on the precise modulation of the reaction microenvironment (particularly the liquid environment). However, with the rapid emergence of novel device geometries and material‐integration strategies, the field still lacks a systematic treatment from the perspective of “water supply modes and interfacial environment regulation.” Specifically, a comprehensive understanding of how to precisely tailor the physical state of interfacial water to synergistically couple with photocatalytic processes remains elusive.
While substantial progress continues to be made in photocatalyst material design, the primary aim of this article is to complement these material advances by synthesizing system‐level and interfacial‐control strategies that enable practical deployment. We structure the discussion around two principal technological pathways: (i) externally supplied systems, encompassing liquid‐supply modes that utilize capillary forces or hydrogel networks for directed transport and vapor‐fed modes that couple photothermal effects to construct “quasi‐gas‐phase” environments, and (ii) self‐supplied systems, leveraging hygroscopic materials to realize in situ moisture capture from ambient air for self‐driven operation (Figure 1). For each pathway, we dissect the operating principles, representative device architectures, and performance bottlenecks, with a particular emphasis on assessing the engineering challenges associated with scale‐up and long‐term durability. By adopting this system‐design‐centered review framework, we aim to delineate the critical technological milestones and future development directions for translating laboratory‐scale explorations into practical applications.
FIGURE 1.
Schematic illustration of immersed and nonimmersed photocatalytic water‐splitting systems [Reprinted with permission by Ref. [15]. Copyright 2023, Springer Nature and Ref. [40]. Copyright 2013, Royal Society of Chemistry].
2. Mechanism of Photocatalytic Water Splitting
2.1. Thermodynamic and Kinetic Fundamentals
The basic principle of using semiconductor photocatalysts to split water is that light excites electrons in the semiconductor, generating charge carriers that subsequently participate in the water‐splitting reactions. However, for a semiconductor to act as an effective catalyst, the photogenerated carriers must also meet the relevant thermodynamic requirements. For example, under standard conditions (pH = 0, 25 °C), the overall water‐splitting reaction requires a Gibbs free energy change of ΔG° = +237.13 kJ mol−1, which is the minimum energy input needed to produce H2 and O2 [41].
| (1) |
When decomposed into two half‐reactions, the hydrogen‐evolution reaction (HER) and the oxygen‐evolution reaction (OER) demand standard potentials of E°(HER) = 0.00 V and E°(OER) = +1.23 V versus the normal hydrogen electrode (NHE), respectively [42]. To drive both half‐reactions spontaneously, a photocatalyst’s conduction band minimum must lie more negatively than 0.00 V versus NHE and its valence band maximum more positively than +1.23 V versus NHE [43].
| (2) |
| (3) |
Consequently, an ideal single‐semiconductor absorber requires a bandgap of E g ≥ 1.23 eV. However, to surmount these kinetic barriers requires additional driving force beyond the 1.23 eV thermodynamic minimum. The oxygen evolution half‐reaction typically incurs an overpotential of 0.3–0.7 V to drive the sluggish four‐electron process on particulate photocatalysts [41]. The hydrogen evolution half reaction likewise demands an overpotential of approximately 0.05–0.2 V to overcome proton–electron recombination and sustain appreciable H2 production rates [42]. Consequently, the combined kinetic overpotentials raise the effective photovoltage requirement above 1.6 eV, meaning that a single‐absorber semiconductor must possess a practical bandgap of at least 1.6 eV to achieve efficient overall water splitting under solar illumination [7]. Beyond thermodynamic requirements, the overall efficiency of photocatalytic systems is fundamentally governed by intrinsic kinetic constraints. The process is conceptually divided into three sequential steps: (i) photon absorption, which generates electron–hole pairs when the incident energy exceeds the semiconductor bandgap (E ≥ E g); (ii) charge separation and migration, where photogenerated carriers diffuse to the catalyst surface while competing against bulk and surface recombination; and (iii) surface redox reactions, where protons are reduced to hydrogen at conduction band sites and water is oxidized to oxygen at valence band sites, typically facilitated by cocatalysts (Figure 2).
FIGURE 2.
Mechanism of photocatalytic water splitting on a semiconductor‐based photocatalyst.
Since the overall efficiency is strictly constrained by these three interdependent steps, extensive research has traditionally focused on intrinsic material modifications to optimize the microscale kinetics, utilizing strategies including bandgap engineering [44], heterojunction construction [10, 45, 46, 47, 48], and surface modification [49]. While these methods have historically yielded significant performance gains, relying solely on catalyst‐level improvements has recently encountered an efficiency bottleneck. Therefore, contemporary research is increasingly shifting toward manipulating broader systemic and macroscopic factors to further elevate overall performance.
2.2. The Challenge of Spatio–Temporal Mismatch
With the maturation of photocatalytic research, the field has shifted from purely material‐centric studies toward application‐driven system‐level engineering. Increasingly, investigators recognize that photocatalytic water splitting entails processes spanning more than 15 orders of magnitude in timescale [13]: ultrafast photon transmission through the reactor assembly (10−14–10−9 s) [13] and carrier excitation/separation within catalyst particles (10−15–10−6 s) [50, 51, 52, 53]; intermediate‐speed interfacial steps—adsorption, surface reaction, and desorption—occurring over 10−6–10−3 s [54, 55, 56]; and slow macroscopic bubble dynamics from nucleation to detachment (10−3–1 s) [57, 58, 59, 60] (Figure 3). At the system level, translating intrinsic material activity into practical device efficiency is primarily hindered by a fundamental spatiotemporal mismatch between ultrafast photophysical events and sluggish macroscopic bubble formation. This vast multiscale disparity causes severe product accumulation and carrier recombination, drastically degrading the excellent metrics observed in short‐duration laboratory tests when scaled to actual device operation.
FIGURE 3.
Schematic diagram of energy flow and mass flow within whole‐process photocatalytic water splitting [Reprinted with permission by Ref. [13]. Copyright 2019, Elsevier].
2.3. The “Water Layer” Dilemma: Optical and Kinetic Constraints
In traditional photocatalytic systems, the catalyst operates in a deep‐water immersion environment. Recent mechanistic insights reveal that this bulk interfacial water layer itself constitutes a critical, yet often overlooked, kinetic and optical barrier. This barrier manifests in three interconnected dimensions.
1. Energy Input: Spectrally Dependent Attenuation and Photothermal Screening
Although liquid water is relatively transparent in the visible region, it is not a lossless medium. The absorption spectrum of water exhibits increasing attenuation coefficients in both the ultraviolet (<300 nm) and near‐infrared (>700 nm) regions (Figure 4a) [61]. For wide‐bandgap photocatalysts relying on UV excitation or narrow‐bandgap materials utilizing visible light, even this seemingly minor parasitic absorption by a macroscopic water layer can reduce the effective photon flux reaching the surface. More critically, recent advances have highlighted the pivotal role of the photothermal effect in boosting catalytic kinetics (Figure 4b–e) [23]. However, water is a strong absorber of infrared radiation. A bulk liquid layer effectively acts as a “heat sink,” intercepting incident IR photons and dissipating their energy into the bulk fluid before it can heat the catalyst surface (Figure 4f). This “infrared screening effect” greatly restricts the potential of photothermal‐assisted photocatalysis in immersed systems [63, 64]. Therefore, minimizing the liquid‐phase optical path is a prerequisite not only for maximizing photon penetration across the full spectrum but also for enabling efficient localized thermal management.
FIGURE 4.
(a) Absorption coefficient of pure water [Reprinted with permission by Ref. [61]. Copyright 2018, The Optical Society]. (b) Temperature‐dependent STH efficiency of the Rh/Cr2O3/Co3O4–InGaN/GaN nanowires (NWs). (c) Stability test of the Rh/Cr2O3/Co3O4‐loaded InGaN/GaN NWs in the self‐heated photocatalytic overall water splitting (OWS) system [Reprinted with permission by Ref. [23]. Copyright 2023, Springer Nature]. (d) Temperature change, (e) hydrogen production, (f) schematic diagram, and thermographic image of C‐MF/CIZS@Ru in solution illuminated at different positions for 1 h [Reprinted with permission by Ref. [62]. Copyright 2013, Royal Society of Chemistry]. (g) Evolution of the hydrogen‐bond network on the TiO2 surface [Reprinted with permission by Ref. [33]. Copyright 2022, American Chemical Society]. (h) Schematic of photocatalytic gas product evolution and its effects on the reaction process.
2. Reaction Process: Nonmonotonic Influence of Interfacial H‐Bond Networks
A growing body of research indicates that the thickness of the interfacial water layer exerts a “nonmonotonic” effect on catalytic activity [32]. When the interface is covered only by a “thin water layer” of several nanometers or micrometers, a requisite hydrogen bond (H‐bond) network is established (Figure 4g). This network not only facilitates proton transfer via the Grotthuss mechanism and lowers the energy barrier for water dissociation but also assists in the timely detachment of product molecules. However, as the water layer thickens, the interfacial water grid becomes increasingly “rigid.” This overstructured network can impede the escape of intermediates and products into the gas phase and may kinetically inhibit catalytic turnover [31, 33, 65]. While the reported “optimal water layer thickness” varies across different material systems, the overarching trend suggests that an appropriately thin water film is superior to deep immersion.
3. Product Output: Mass Transfer Limitations and the Optical Feedback Loop
In deep immersion environments, the release of evolved H2 and O2 is kinetically hindered by solubility limits and the high energy barrier for bubble nucleation. This sluggish mass transfer frequently leads to the accumulation of microbubbles at active sites (the shielding effect), which not only blocks reactant access but also thermodynamically triggers unwanted reverse reactions (2H2 + O2 → 2H2O) [66]. Furthermore, this mass transfer bottleneck creates a deleterious feedback loop with optical input: The swarm of bubbles trapped in the thick water layer acts as a strong light‐scattering medium (Figure 4h). These scattering centers reflect incident light away from the catalyst, further exacerbating the optical losses discussed in the energy input section [67, 68, 69]. The construction of a thin water film or a gas–liquid–solid triphase interface compresses the reaction zone to the micro–nanoscale. This configuration allows product gases to be released directly into the gas phase via an extremely short diffusion path, largely circumventing the macroscopic bubble dynamics lag and the associated parasitic light scattering [70, 71, 72, 73, 74].
These fundamental constraints reveal that achieving high solar‐to‐hydrogen (STH) efficiency is not merely a materials challenge but a system‐level problem of interface management. Traditional reactor designs that immerse catalysts in deep water are inherently limited by these conflicting optical and kinetic requirements, necessitating a paradigm shift toward new reactor architectures that can decouple the catalytic interface from the bulk liquid environment.
3. Photocatalytic System Configurations: From Immersed to Nonimmersed
Although catalyst‐level modifications often deliver substantial performance improvements under laboratory conditions, translating these material advances into large‐area, long‐term, device‐level hydrogen production remains challenging. Early demonstration‐scale systems were primarily aimed at validating catalyst stability under realistic operating conditions, and their instantaneous activities measured in the field or in pilot trials frequently fall short of the short‐duration, stirred‐lab metrics. As understanding of reactor‐ and system‐level issues has deepened, an increasing number of studies have focused on compensating for the limitations of material‐only approaches via reactor architecture and interfacial design, with the goal of improving overall system efficiency under practical operating conditions.
Recent works have highlighted the important role of interfacial water layers on catalyst surfaces: The thickness and structure of these layers affect proton/ion transport and interfacial electric fields while also influencing light transmission and bubble detachment dynamics—factors that can materially constrain hydrogen evolution rates. This insight has prompted a shift toward integrated material–interface–device design strategies. Based on differences in how water is supplied and how the interface is engineered, photocatalytic systems can be broadly classified into two categories: immersed systems, in which the catalyst remains in direct contact with bulk liquid and which represent the prevailing laboratory and many demonstration configurations, and nonimmersed systems, which decouple the catalytic layer from the bulk liquid by means of interface reconfiguration (e.g., controlled thin liquid films, vapor/condensate feeding, or material‐level self‐supply), thereby directly addressing optical and mass‐transfer bottlenecks. Nonimmersed architectures have advanced rapidly in recent years and show strong potential for improving light utilization, accelerating gas release and enabling deployment in water‐limited environments [75].
Accordingly, this section first reviews prototypical immersed reactor configurations and their mechanism‐linked constraints, then focuses on the principal classes, implementation strategies, and engineering potential of non‐immersed systems, and concludes with a comparative synthesis that offers design guidance and evaluation metrics for scale‐up and long‐term operation. The aim is to provide a systems‐level roadmap to help translate promising laboratory catalyst developments into field‐deployable solar hydrogen technologies.
3.1. Immersed Systems: Current Standards and Limitations
Laboratory‐scale activity assessment for particulate photocatalysts is conventionally performed in a slurry system, with the powder dispersed in water (Figure 5a,b). In this arrangement, the liquid phase is in intimate contact with catalyst particles; to counteract the detrimental effects of stagnant water layers and particle agglomeration, experiments typically rely on mechanical stirring to maintain dispersion and enhance interfacial mass transfer. Inspired by the high activities often observed in these bench‐scale tests, several scale‐up strategies have been developed to retain the fundamental dispersed‐particle configuration (Figure 5d). For example, concentrator‐coupled tubular flow reactors use fluid flow to keep particles suspended, transparent tubing to promote uniform illumination, and compound parabolic concentrators to boost light collection (Figure 5c) [29].
FIGURE 5.
(a) Schematic of Type 1 reactor and Type 2 reactor [Reprinted with permission by Ref. [40]. Copyright 2013, Royal Society of Chemistry]. (b) Typical reaction cells for suspension [Reprinted with permission by Ref. [15]. Copyright 2023, Springer Nature]. (c) Marginal light and major parameters of the compound parabolic concentrator (CPC) [Reprinted with permission by Ref. [29]. Copyright 2018, Elsevier]. (d) Photo of the CPC demonstration plant [Reprinted with permission by Ref. [76]. Copyright 2018, Elsevier]. (e) Photographic image of a panel reactor unit (625 cm2). (f) The structure of the panel reactor unit viewed from the side. (g) An overhead view of the 100 m2 solar hydrogen production system consisting of 1600 panel reactor units and a hut housing a gas separation facility [Reprinted with permission by Ref. [77]. Copyright 2021, Springer Nature].
However, translating this dispersed configuration into practical, large‐scale deployment faces formidable engineering hurdles. Maintaining a uniform particle suspension over extended periods requires continuous mechanical agitation, which incurs significant parasitic energy costs. Furthermore, the random scattering of light by suspended particles often leads to inefficient photon utilization in deeper reactor volumes. Crucially, the challenges associated with the continuous separation of gaseous products from the liquid slurry, alongside the difficulty in catalyst recovery and recycling, complicate the translation of slurry reactors into scalable engineering solutions.
Accordingly, many pilot and demonstration projects have moved toward immobilized architectures in which catalysts are fixed on panels or modules to facilitate handling and module‐level scale‐up (Figure 5e–g). A range of fabrication methods has been explored—for example, drop coating [26], screen printing [78], and brush coating [77, 79] (Table 1). The shared goal is to deposit catalysts uniformly and stably over a substrate area so that active sites remain accessible while light harvesting, product separation, and catalyst retrieval become more tractable for modular deployment.
TABLE 1.
Comparison of various immersed photocatalytic systems for hydrogen production.
| Reactor type | Photocatalyst | Immobilization method | Reactor area, m 2 | STH | Reference |
|---|---|---|---|---|---|
| Slurry system | CdS | \ | 103.7 | 0.087% | [30] |
| Immobilized system | BiVO4 | Brush coating | 1 | 1.8% | [79] |
| Immobilized system | Pt@mp‐CN | Drop coating | 0.756 | 0.12% | [26] |
| Immobilized system | SrTiO3:Al | Brush coating | 100 | 0.76% | [77] |
| Immobilized system | InGaN/GaN | In situ growth | 0.0016 | 6.2% | [23] |
It is also worth noting that recent studies have pursued a more integrated approach by in situ growth of sheet‐like or film catalysts directly on substrates; such substrate‐grown layers have shown promising performance in outdoor tests. Although this route can improve interfacial contact and tune charge‐transport and optical coupling, it entails greater fabrication complexity and presents additional challenges for large‐area scale‐up [23].
Compared to slurry configurations, immobilizing photocatalysts onto substrates offers a more viable pathway for scalable engineering, simplifying both product separation and catalyst handling. However, this immobilized architecture introduces its own set of limitations. The dense packing or stacking of catalyst particles often results in the burial of active sites, leading to insufficient illumination and restricted reactant accessibility in the lower layers of the film. Furthermore, while panel reactors significantly reduce the liquid volume compared to slurry systems, the catalyst surface remains submerged under a macroscopic water layer. This “residual liquid film”—which is often orders of magnitude thicker than the catalytic interface—continues to impose substantial optical losses due to absorption and scattering, as well as mass‐transfer resistances that impede gas bubble detachment, thereby constraining the overall STH conversion efficiency.
3.2. Nonimmersed Systems: Principles and Classification
As the quantum efficiency of photocatalytic materials approaches intrinsic limits, the research focus is fundamentally transitioning from isolated catalyst modification toward integrated “material–interface–device” systems engineering [25, 27]. To circumvent the inherent “water layer dilemma” of immersed reactors, nonimmersed architectures have emerged. The fundamental premise is to decouple the photocatalytic interface from the bulk liquid environment. By confining the reaction to a micro/nanoscale wetted interface to overcome traditional bottlenecks, these systems simultaneously deliver three distinct advantages: (1) minimized light attenuation, by avoiding the parasitic absorption and scattering typically caused by a bulk liquid phase; (2) mass transfer enhancement, by enabling direct gas desorption that bypasses the slow bubble cycle; and (3) optimized triple‐phase boundaries, by maintaining a precisely controlled microwetted state that drives the reaction kinetics to their optimal state for maximized performance.
However, this conceptual advantage also introduces a series of complex engineering and scientific challenges centered on the precise regulation of the interfacial microenvironment. A primary issue is how to achieve and regulate the “effective water film thickness” and “water supply flux” to strike a delicate balance between preventing site dry‐out and maintaining rapid interfacial kinetics. In photothermal‐coupled systems, managing the temperature of the reaction surface to keep it within the optimal range for photocatalytic activity is also crucial. Furthermore, ensuring the operational stability of the system is a major obstacle to practical deployment. Based on the problems that nonimmersed systems can solve and the challenges they face, these designs are primarily classified into two main categories according to their water supply mechanism: externally supplied systems and self‐supply systems.
3.2.1. Externally Supplied Systems
Externally supplied systems, which rely on a continuous feed from an external water source, represent the most direct evolution from immersed designs. Based on the physical state of the water feed, these strategies can be further divided into two distinct paradigms. Liquid‐fed systems utilize microstructures such as capillary channels or hydrogel networks to transport liquid water as a controlled thin film. The primary advantage of this approach lies in the intuitive nature of the water source and mature flux regulation methods, but challenges remain regarding the risk of clogging and maintaining uniform film thickness over large areas. In contrast, vapor‐fed systems deliver water as vapor, often generated locally via photothermal conversion. Their appeal stems from the ability to utilize long‐wavelength solar flux and sustain ultrathin “effective water films,” though it introduces new constraints related to thermal management and vapor distribution uniformity.
To date, researchers have primarily engineered this controlled liquid supply by leveraging either the capillary action within porous substrates or the water‐retention properties of hydrogel networks. These strategies aim to maintain sufficient reactant supply while significantly compressing the macroscopic liquid volume. Their primary advantages lie in mature water‐supply management and facile integration with conventional photocatalytic reaction systems.
Capillary action refers to the spontaneous flow of liquid within fine pores or fibrous gaps, driven by surface tension and interfacial wettability. Its transport flux is described by the Lucas–Washburn equation and is governed by the pore size distribution and contact angle of the material [80]. Within photocatalytic systems, the capillary effect enables a continuous, passive supply of trace water without external energy consumption, making it an ideal mechanism for eliminating bulk water layers and constructing efficient triphase interfaces.
Early exploration of nonimmersed systems primarily hinged on the capillary characteristics of fibrous substrates to achieve passive water supply. Some studies utilized polypropylene nonwoven fabrics (PP NWF) as carriers, involving the in situ growth of iron oxyhydroxide photocatalysts. Driven by the capillary pressure differential generated within the loose, porous fiber interstices, water is supplied passively, while the low density of polypropylene confers ideal self‐floating capabilities to the device [81]. Given that in situ growth involves specific chemical synthesis directly on the substrate, its scope is generally defined by material applicability. Conversely, physical deposition methods such as vacuum filtration offer broader adaptability, accommodating almost any conventional powder‐based photocatalyst. For instance, using hydrophilic quartz fiber membranes as substrates (Figure 6a), various semiconductor catalysts can be constructed into uniform nonimmersed interfaces via simple filtration. This approach accommodates a broader range of catalysts and utilizes the capillary action of quartz fibers to manage the microscale water film distribution. Additionally, the reported configuration allows for standard device assembly (Figure 6b) and demonstrates targeted product selectivity under the specified reaction conditions (Figure 6c) [35].
FIGURE 6.
(a) The energy diagram of STO/Cu@Ni/SiO2/TiN, (b) schematic diagram of the reactor, and (c) the different products’ selectivity of STO/Cu@Ni/SiO2/TiN [Reprinted with permission by Ref. [35]. Copyright 2023, Royal Society of Chemistry]. (d) The interfacial redox reaction scheme of the PTPC film. (e) Schematic of the hierarchical design of the PTPC. (f) The comparison of the average H2 production rate of PTPC film and TiO2/Au/CuS PTC mixed PDMS film under UV light and full spectrum irradiation. (g) H2 production stability comparison between PTPC film and TiO2/Au film from seawater and plastic waste [Reprinted with permission by Ref. [82]. Copyright 2025, Wiley].
Although fiber‐membrane‐based strategies addressed the fundamental issues of water supply and catalyst loading, directly exposed catalysts still face severe risks of deactivation when deployed in complex or hostile aqueous environments (e.g., high‐salinity seawater or organic wastewater). Furthermore, when introducing photothermal effects to assist catalysis, direct contact often leads to photothermal material‐induced carrier recombination, disrupting charge transport pathways. Addressing these challenges, recent research has proposed a more sophisticated polymeric tandem photothermal catalytic film (PTPC) design (Figure 6d,e). This work constructs a multilayer structure featuring spatial decoupling: The TiO2/Au photocatalytic layer and the CuS photothermal layer are physically separated and encapsulated by specific polymeric layers. This design rationally integrates a triple function. First, the polymer layer acts as an “isolator,” blocking the interference of the photothermal layer with photogenerated charge transport, thereby achieving spatial decoupling and synergy among photothermal generation, charge separation, and redox reactions. Second, the encapsulation structure serves as a corrosion‐resistant redox medium, enabling the catalyst to maintain long‐term stability in harsh media characterized by high alkalinity, high salinity, or even plastic waste (Figure 6f,g). Finally, while passively floating, the film structure achieves efficient interfacial hydrogen production in hostile environments through optimized light absorption and interfacial mass transfer. This approach demonstrates the transition of nonimmersed membrane systems from simple “physical supports” toward configurations that offer multifunctional spatial decoupling and environmental adaptability [82].
Despite the significant progress in preparation simplicity and functional integration achieved by the aforementioned 2D planar designs—ranging from fabrics and filter membranes to multilayer polymeric films—such configurations still exhibit inherent physical limitations when pursuing ultimate interfacial stability and photothermal utilization efficiency. First, precisely because these two‐dimensional designs are in direct contact with the water source, they are susceptible to environmental disturbances, making it challenging to regulate the surface water layer thickness and often leading to an “over‐wetted” state. This significantly compromises the anticipated advantages of the thin liquid film in nonimmersed systems. Second, concerning thermal management, 2D films typically adhere closely to the bulk water surface, lacking an effective thermal insulating buffer layer. Consequently, the high temperatures generated by photothermal effects are prone to dissipating into the bulk water heat sink, making it difficult to sustain the high‐temperature microenvironment required at the reaction interface. To overcome these bottlenecks, researchers have begun to explore more complex hierarchical composite structures. This strategy aims to physically decouple the bulk water source from the photocatalytic reaction interface at a macroscopic level through 3D spatial design, thereby constructing a nonimmersed reaction microenvironment that combines high stability with high energy efficiency while simultaneously achieving more precise throttling of water flux and heat localization.
To balance the trade‐off between sufficient water supply and surface over‐wetting, researchers have turned to hierarchical composite structures. Melamine foam, characterized by its abundant and interconnected 3D skeletal structure, has been widely adopted as a directional bottom‐layer “hydraulic pump.” In an advanced “foam–composite membrane” architecture, the underlying melamine foam transports water from the reservoir to a top reaction layer consisting of a spin‐coated photocatalyst/MXene/PVDF composite (Figure 7a). This configuration achieves a precise division of functions: The foam provides a mild, continuous capillary supply to prevent overwetting, while the top composite membrane utilizes PVDF to establish an aerophobic/hydrophilic interface. Simultaneously, the embedded MXene generates in situ photothermal effects to enhance reaction kinetics, successfully realizing the physical coupling of mass flow and energy flow within an integrated space. This design has demonstrated significant activity enhancements when applied to various photocatalysts (Figure 7b) [83].
FIGURE 7.
(a) The interfacial photocatalytic mode. (b) Bar graph of hydrogen production efficiencies of the membrane samples with different photocatalysts in the bulk water mode and interfacial mode [Reprinted with permission by Ref. [83]. Copyright 2021, American Chemical Society]. (c) Schematic illustration of the macroscopic structural assembly and microscopic mapping of the interfacial photocatalytic reaction unit. (d) Cycling test of interfacial photocatalytic hydrogen generation of one sample with TiO2 [Reprinted with permission by Ref. [84]. Copyright 2022, American Chemical Society].
Similarly, biological structures in nature have inspired the construction of efficient 3D water supply networks. During their growth, biological organisms spontaneously evolve sophisticated and directional vascular systems—exemplified by the xylem in vascular plants—to achieve long‐distance water transport against gravity. These naturally formed microfluidic channels, characterized by high connectivity and low tortuosity, offer an ideal blueprint for engineering water management systems [85, 86]. Drawing inspiration from such biological architectures, Sun et al. identified similar hydraulic capabilities in biomass fungi. They demonstrated that the natural vascular bundles formed during mushroom growth are preserved after carbonization, enabling the stipe to efficiently extract water from a bottom reservoir like a hydraulic pump (Figure 7c). By dispersing and immobilizing the photocatalyst on the broad pileus (cap) surface, the team successfully constructed a natural, integrated, nonimmersed system. In this configuration, the carbonized stipe facilitates directional water supply, while the pileus provides a vast, wettable interface. Crucially, the carbonized biomass matrix inherently possesses photothermal absorption properties. This biomimetic design not only resolves the challenge of uniform wetting across large reaction areas but also realizes the in situ coupling of photothermal evaporation and photocatalytic reaction through the material’s intrinsic features, illustrating a design trend evolving from single‐function water supply toward multifunctional integration [84]. This design also exhibited excellent catalytic stability (Figure 7d).
Distinct from capillary‐driven strategies, hydrogel‐based systems leverage the swelling and water‐retention capacities of 3D polymer networks to confine water within a stable gel framework [87, 88]. In this configuration, the hydrogel transcends its role as a passive transport medium, acting instead as a critical regulatory unit for the interfacial microenvironment. From a material physics perspective, water confined within the nanometer‐scale pores of hydrogels interacts extensively with polymer chains via hydrogen bonding and electrostatic forces. This confinement restrains the freedom of water molecules, significantly altering their diffusion behavior and hydrogen‐bonding configurations compared to bulk water [89, 90]. Crucially, this restructuring of the hydrogen‐bond network—manifested as changes in bond strength and lifetime—has been correlated with enhanced proton transfer efficiency and interfacial reaction kinetics [91], providing a robust physical basis for the catalytic advantages of hydrogel systems.
In practical implementation, the inherent “swelling‐retention” mechanism of hydrogels enables the creation of a uniform and enduring liquid film that is less dependent on ambient humidity. This paradigm offers distinct engineering advantages: (1) high‐water‐uptake capacity allows for sustained reactions with a minimal footprint; (2) superior retention ensures stability even in arid conditions; and (3) tunable polymer chemistry enables precise modulation of water content and film thickness. Consequently, hydrogels have emerged as versatile platforms for constructing nonimmersed interfaces. They not only facilitate the uniform immobilization of powder photocatalysts via in situ growth or encapsulation but also actively modulate the intermediate hydrogen‐bonding states of internal water to boost activity [92]. This underscores that hydrogels are not merely physical supports but functional components that synergistically enhance reaction kinetics.
Building upon these design principles, Lee and co‐workers introduced a floatable elastic–hydrogel nanocomposite platform [36]. In their study, Pt/TiO2 aerogels or single‐atom Cu/TiO2 nanoparticles were embedded within a polyurethane–poly(propylene glycol) (HPU–PPG) hydrogel matrix, and a hydrophilic/hydrophobic Janus silica–aerogel layer was applied to the surface to provide buoyancy and a pathway for rapid gas desorption (Figure 8a). This architecture leverages the highly porous hydrogel matrix for passive microwater supply, while the Janus layer ensures efficient light transmission and facilitates the rapid escape of evolved gases from the upper surface. Under simulated solar illumination, the system achieved HER of 163 mmol H2 h−1 m−2 in the Pt/TiO2 regions (Figure 8b). Upon scaling the device to 1 m2 using more economical Cu/TiO2 catalysts, an average daily hydrogen production of 79.2 mL was recorded in field tests. A hallmark of this work was the successful 1 m2 outdoor demonstration without reliance on forced convection, directly validating the operational stability and viability of hydrogel systems in real‐world environments. However, the study also identified engineering challenges such as catalyst leaching, suggesting that the interfacial binding strength between the gel and the catalyst, as well as particle encapsulation strategies, requires further optimization.
FIGURE 8.
(a) Schematic of the nanocomposites and their advantages for photocatalytic HER: efficient light delivery (i), facile gas separation (ii), enhanced surface tension (floatability) (iii), stable catalyst immobilization (iv), suppressed back‐oxidation (v), and facile reactant supply (mass transfer) (vi). (b) H2 evolution rate of the floatable and sunken nanocomposites [Reprinted with permission by Ref. [36]. Copyright 2023, Springer Nature]. (c) A schematic illustration of photocatalytic waste upcycling using a floatable nanocomposite system and light concentrator. (d) H2 evolution rates in the solubilized solutions of PET bottle and PLA under one‐sun irradiation [Reprinted with permission by Ref. [93]. Copyright 2025, Springer Nature]. (e) Schematic illustration of H2O molecule types in the polymer matrix: bonded water (BW blue), intermediate water (IW yellow), and free water (FW light blue). (f) Raman spectra with the O–H stretching region and IW/FW ratio. (g) Water‐splitting activity of SrTiO3 embedded in PAM/PVA nanocomposite hydrogels with varying crosslinker concentrations. Photocatalytic reaction images and water‐splitting activity [Reprinted with permission by Ref. [92]. Copyright 2025, American Chemical Society].
In their subsequent research, the team further enhanced the stability and economic feasibility of the system through material and systemic refinements. They introduced waste polyethylene terephthalate (PET) and polylactic acid (PLA) as oxidizable reactants—effectively utilizing plastic waste as substrates for photoreforming to produce hydrogen (Figure 8c). By employing dynamically stabilized atomic Pt/TiO2 catalysts and concentrated sunlight, they reported a remarkable plastic reforming activity of 271 mmol H2 h−1 m−2. Scaled to a 1 m2 area under natural sunlight, the system yielded 0.906 L of hydrogen per day from PET waste. This result not only demonstrates the high efficiency of the hydrogel platform but also integrates hydrogen production, waste valorization, and economic viability into a pathway suitable for industrial consideration [93]. Collectively, these two studies provide compelling evidence for the engineering feasibility of hydrogel‐mediated water supply while highlighting the need for further research into stable catalyst immobilization, antileaching mechanisms, and long‐term fouling mitigation (Figure 8d).
Beyond microplastic degradation, utilizing abundant seawater represents another highly promising avenue for sustainable hydrogen production. To circumvent the severe catalyst corrosion typically caused by complex ions in raw seawater, recent strategies have coupled interfacial solar desalination with in situ water vapor splitting. For instance, a photothermal–photocatalytic layered aerogel was engineered to completely isolate the catalytic sites from the liquid electrolyte. The bottom photothermal layer generates water vapor that diffuses directly into the upper photocatalytic layer, enabling a stable hydrogen evolution rate of 17.94 mmol m−2 h−1 alongside clean freshwater coproduction while successfully avoiding ionic fouling [94].
Despite their significant engineering potential, the fundamental mechanisms by which hydrogels enhance photocatalytic activity at the molecular scale remain largely underexplored. Key questions regarding the specific hydrogen‐bonding configurations within the gel and their influence on interfacial kinetics are yet to be systematically addressed. Consequently, recent studies have begun to shift their focus toward elucidating these mechanistic contributions from a molecular‐to‐mesoscale perspective.
A representative mechanistic study using a SrTiO3/hydrogel coupled system revealed that an “intermediate” population of weakly hydrogen‐bonded water within the gel plays a pivotal role under photoactivation. In situ Raman spectroscopy revealed that the hydrogen‐bonding networks could be systematically tuned by varying the sample composition. Specifically, adjusting this ratio effectively modulated the proportion of “intermediate water,” with a higher content of these reactive water species directly correlating to enhanced photocatalytic activity (Figure 8e–g). Density functional theory (DFT) calculations indicate that the gel environment can reduce the hydrogen‐bond strength between neighboring water molecules and lower the O─H bond dissociation energy, thereby reducing the activation barrier for interfacial reactions and facilitating coupled proton/electron transfer. Experimentally, in a saturated hydrogel (without photothermal enhancement), the system achieved an H2 evolution rate of 215.62 mmol m−2 h−1 and an O2 evolution rate of 105.95 mmol m−2 h−1 (corresponding to an STH of ≈1.43%); when photothermal coupling was introduced, the STH increased to ≈1.85%, and the hydrogen production rate was ∼3.85 times higher than that measured in bulk water. These combined experimental and theoretical results provide direct support for the concept that tuning water states within a hydrogel can actively modulate interfacial catalytic dynamics—positioning hydrogels as functional microenvironment regulators rather than merely passive water reservoirs [92].
As previously discussed, the physical state of the interfacial water layer plays a decisive role in governing reaction kinetics. Numerous theoretical calculations and experimental investigations focusing on nonimmersed systems have converged on the existence of an optimal window for interfacial water layer thickness. Although specific values vary slightly depending on the intrinsic properties of the catalyst, the prevailing consensus is that photocatalytic activity peaks when the surface water layer is maintained at an ultrathin nanoscale dimension—typically corresponding to approximately 3–4 water molecular layers (Figure 9a,b) [31, 32]. At this length scale, the hydrogen‐bonding network differs distinctively from that of bulk water, creating a favorable reaction environment that ensures adequate substrate supply while simultaneously minimizing the desorption resistance for gaseous products (hydrogen/oxygen) and the energy barriers for proton transport.
FIGURE 9.
(a) Schematic illustration of the evolution from reactive water to nonreactive water under different humidity conditions [Reprinted with permission by Ref. [32]. Copyright 2024, American Chemical Society]. (b) Evolution of the apparent photocatalytic activity of overall photolysis of pure water to H2 and H2O2 as a function of water loading on Pt/TiO2 [Reprinted with permission by Ref. [31]. Copyright 2020, Elsevier]. (c) Attenuated total reflection infrared (ATR−IR) spectra of water adsorbed on silicon oxide at different RHs. (d) Adsorption isotherm of adsorbed water on the silicon oxide surface. (e) Schematic illustrating the structural evolution of water molecules as the adsorbed layer thickness increases with RH [Reprinted with permission by Ref. [65]. Copyright 2005, American Chemical Society]. (f) Scheme illustrating gas‐phase photocatalytic hydrogen production governed by the evolution of chemisorbed water coverage under UV illumination [Reprinted with permission by Ref. [95]. Copyright 2020, Elsevier].
However, in engineering practice, achieving and sustaining such precise “molecular‐level thin layers” via liquid water supply present formidable challenges. Whether employing capillary‐action or hydrogel‐based strategies, the resulting liquid film thickness—constrained by surface tension and fluid rheology—often far exceeds the nanoscale, making it difficult to precisely confine the water layer within the few molecular layers required for optimal activity. Although liquid‐based nonimmersed systems have significantly improved efficiency compared to traditional slurry suspensions, the mass transfer resistance and light scattering losses induced by excessively thick liquid films continue to hinder the full realization of their photocatalytic potential.
To overcome the physical limitations of liquid transport in forming ultrathin films, vapor‐phase supply has emerged as a distinctive nonimmersed paradigm [96, 97]. This strategy delivers water as vapor—often generated in situ via photothermal conversion—which then adsorbs or condenses onto the catalyst to form a precisely controlled, nanoscale “effective water film.” This approach is particularly attractive because it allows for the synergistic utilization of the full solar spectrum: Photocatalysts absorb high‐energy UV–visible photons for carrier excitation, while the abundant low‐energy near‐infrared (NIR) photons are captured by photothermal components to drive steam generation. This “photothermal–photocatalytic” coupling not only provides a continuous supply of high‐purity reactant but can also accelerate surface reaction kinetics through thermal activation effects, representing a critical frontier for boosting STH efficiency [37].
The concept of gas‐phase photocatalytic water splitting is not entirely new. Early exploratory studies had already validated its feasibility on model catalysts such as SrTiO3 [98]. However, constrained by the characterization techniques and mechanistic understanding of the time, this research direction was not systematically expanded. It is the recent explosive development of solar‐driven steam generation (SSG) technology that has truly reinvigorated the field, providing a powerful and efficient means to supply the necessary water vapor [99].
Before delving into specific device architectures, it is imperative to clarify the intrinsic mechanism governing vapor‐fed photocatalysis. Although often termed “water vapor splitting,” the process is not the direct dissociation of free‐space gas molecules but rather a complex reaction occurring at the solid–gas interface [18, 97]. The core sequence is initiated by the adsorption and accumulation of gaseous water molecules onto the photocatalyst surface, governed by a dynamic equilibrium. This equilibrium is profoundly constrained by the catalyst’s intrinsic physicochemical properties, including surface wettability, pore structure, and the polarity of exposed crystal facets, all of which dictate the adsorption energy, surface coverage, and even the hydrogen‐bonding network configuration of the adsorbed water molecules [100].
These material properties, in concert with the ambient relative humidity (RH), determine the microscopic state of the water film on the catalyst surface. For generic photocatalysts devoid of specific hygroscopic modifications, the adsorbed water film manifests as a distinct intermediate phase situated between the gaseous and liquid states (Figure 9c). As the thickness of the adsorbed water layer increases, the interfacial water exhibits a pronounced gradient in phase evolution: Within the first ∼3 molecular layers immediately adjacent to the catalyst interface, water molecules are subject to strong confinement by the interfacial potential field, forming a highly ordered, rigid structure analogous to ice (the “ice‐like water region”) (Figure 9d,e), where the hydrogen‐bonding network is severely restricted. As the coverage extends to ∼4 layers, interfacial constraints weaken and molecular degrees of freedom increase, giving rise to a “transitional region” characterized by a relatively loose hydrogen‐bonding network (Figure 9d,e). Only when the film thickness increases further does the portion distal to the interface gradually exhibit the random hydrogen‐bonding network consistent with bulk water, termed the “liquid water region” (Figure 9d,e). Notably, this evolution of water states driven by varying confinement strength directly mirrors the water state distribution in hydrogel systems, where an analogous transition from bonded water to intermediate water and finally to free water is observed. The proportions and properties of these three regions are not static but remain in a state of dynamic adjustment in response to fluctuations in ambient humidity [65].
Complicating matters further, the introduction of photoexcitation induces drastic perturbations in the physicochemical state of the photocatalyst surface. The generation of photogenerated charge carriers alters the surface electrostatic field distribution and may even induce photoinduced hydrophilicity, thereby significantly modifying the stability of the pre‐existing hydrogen‐bonding network [101, 102]. Concurrently, the photothermal effect accompanying the reaction accelerates the thermal motion of water molecules, disrupting a portion of the rigid hydrogen bonds and shifting the aggregation state of the water layer toward a looser configuration [103]. Ultimately, the entire system achieves a dynamic equilibrium among “adsorption, reaction, and desorption” (Figure 9f) [95]. This highly complex interfacial evolution imposes stringent requirements on the design of photocatalytic water vapor splitting systems: They must ensure adequate substrate adsorption while simultaneously preventing increased mass transfer resistance caused by excessive condensation. Although numerous mysteries regarding the micromechanisms in this domain remain unresolved, recent research has initiated beneficial attempts through interfacial engineering and novel structural designs, achieving significant breakthroughs in performance.
Due to the thermodynamic constraints inherent to semiconductor photocatalysts, their energy utilization efficiency is often limited by spectral mismatch. Most photocatalysts exhibit a response primarily in the ultraviolet or visible regions, remaining largely transparent to the infrared portion of the solar spectrum, which accounts for nearly 50% of solar energy, and is the primary contributor to thermal effects. This mismatch prevents the catalysts themselves from inducing sufficient photothermal action to drive in situ water evaporation. Consequently, to achieve efficient gaseous water supply in non‐immersed systems, researchers have adopted strategies to couple specific photothermal components. By leveraging the high‐efficiency photothermal conversion of auxiliary materials to generate steam, the reaction requirements of the photocatalyst can be sustainably met.
Among various candidates, natural wood has emerged as an ideal substrate for constructing these integrated systems due to its unique biological architecture and cost‐effectiveness. The superiority of wood as a water‐transport and photothermal component stems from its sophisticated evolutionary structure [104, 105, 106]. To enhance light harvesting, the wood is typically subjected to carbonization (carbonized wood, C‐wood). This process preserves the hierarchical porous framework and low thermal conductivity of the original wood while imbuing the material with near‐full‐spectrum absorption. The array of vertical channels acts as a light trap, significantly enhancing broadband absorption through multiple internal reflections while providing a vast surface area for both photocatalyst loading and water evaporation [107]. Utilizing this configuration, Guo et al. [108] reconstructed the traditional “liquid‐rich, solid/gas‐limited” three‐phase system into a highly efficient “quasi–biphase” reaction mode by spin coating photocatalysts onto carbonized wood substrates (Figure 10a). In this architecture, the carbonized wood simultaneously serves as a floatation carrier, a photothermal steam generator, and a catalytic support. By directly exposing the catalyst surface to an expansive gaseous phase, the hydrostatic pressure and desorption resistance for product gases are dramatically reduced. Experimental evidence suggests that this water‐supply mode not only accelerates reaction rates but also significantly bolsters the chemical stability of the catalysts (Figure 10b). For instance, in a vapor‐fed nonimmersed system, CoO catalysts maintained approximately 90% of their hydrogen evolution activity after 5 days of continuous testing, effectively suppressing the photocorrosion issues typically encountered in long‐term reactions. Furthermore, Han et al. [109] delved deeper into the impact of this photothermally induced steam generation mechanism on reaction kinetics (Figure 10c–e). They confirmed that the high‐temperature water vapor generated by the C‐wood substrate not only increases the local reactant concentration but also lowers the surface reaction energy barrier via a thermal activation effect, thereby achieving a hydrogen evolution rate superior to that of traditional liquid‐phase systems. This finding underscores the critical role of “active photothermal evaporation” in overcoming mass transfer limitations within nonimmersed systems.
FIGURE 10.
(a) Schematic of the fabrication process of the wood/photocatalyst structure that generates the water steam and catalyzes its splitting for hydrogen evolution. (b) Rate comparison of H2 evolution in the triphase CoO NPs system and wood/CoO biphase photothermal–photocatalytic system [Reprinted with permission by Ref. [108]. Copyright 2021, Springer Nature]. (c) Schematic photocatalytic splitting of photothermally induced water vapor to evolve hydrogen. (d) H2 production activity of C‐wood/TiO2 under different light sources. (e) H2 evolution rate with AM 1.5G light irradiation for TiO2 powder in water, C‐wood/TiO2 under pure water, C‐wood/TiO2 floating on the water, and iron/TiO2 in vapor (no contact with liquid water) [Reprinted with permission by Ref. [109]. Copyright 2022, Elsevier]. (f) Schematic illustration and microscopic characterization of the photothermal/photocatalytic integrated ZIS–WO/C‐wood system, including scanning electron microscope (SEM), transmission electron microscope (TEM), and high‐resolution transmission electron microscope (HRTEM) images revealing the structural features of WO and ZIS [Reprinted with permission by Ref. [110]. Copyright 2021, Wiley].
To further explore the potential of the C‐wood substrate beyond physical loading, in situ growth strategies have also proven effective for constructing robust interfaces. Han et al. [111] successfully grew graphitic carbon nitride (g‐C3N4) in situ on the surface of carbonized wood via a hydrothermal method. Benefiting from the natural pore structure and abundant oxygen‐containing functional groups on the C‐wood surface, g‐C3N4 was uniformly anchored to the substrate, forming a tight contact interface. This integrated design not only inherits the excellent photothermal evaporation performance of C‐wood but also shortens the diffusion distance of water molecules from transport channels to catalytic sites, significantly enhancing photocatalytic hydrogen production activity. Additionally, to optimize interfacial charge transport, subsequent research introduced nanosecond pulsed fiber laser ablation as a precise surface reconstruction technique. Laser etching induces a “microexplosive” effect that uniformly exposes the hidden internal micropores of the wood, resulting in a superhydrophilic surface that further intensifies capillary‐driven water supply. Beyond structural modification, the high local energy of the laser beam enhances the degree of surface graphitization, creating a highly conductive network. Electrochemical impedance spectroscopy (EIS) confirmed that this optimization significantly lowers the interfacial charge transfer resistance (R ct), thereby synergistically accelerating surface reaction kinetics [112].
This superior conductivity allows carbonized wood to transcend its role as a mere photothermal/water‐supply component, enabling it to function as a sophisticated electron transport mediator (electron bridge) in overall water splitting. In the construction of a Z‐scheme system, the conductive carbon layer of the carbonized wood facilitates an Ohmic contact interface that mediates the recombination of photogenerated electrons from an oxygen‐evolving photocatalyst (e.g., oxygen‐deficient WO3) with the photogenerated holes from a hydrogen‐evolving photocatalyst (e.g., sulfur‐deficient ZnIn2S4) (Figure 10f). This spatial separation preserves charge carriers with stronger redox potentials at their respective active sites. The integrated system demonstrated exceptional performance, yielding photocatalytic H2 and O2 production rates of 169.2 and 82.5 mmol h−1, respectively, with a corresponding STH efficiency of 1.52%. The versatility of this “C‐wood bridge” was further validated across multiple Z‐scheme pairings, including Fe2O3–C3N4 (STH 1.12%) and MoS2–CdS–BiVO4 (STH 0.20%), establishing carbonized wood as a multifunctional platform for engineering complex, high‐efficiency, and stable nonimmersed photocatalytic systems [110].
Beyond directly utilizing the biological scaffolds of natural carbonized wood, designing synthetic matrices with specific chemical compositions and pore morphologies via artificial means represents another critical pathway for constructing efficient photothermal steam systems. This “bottom‐up” synthesis strategy allows researchers to finely tune interfacial connectivity and energy band structures at the molecular scale.
As a potent alternative to natural wood, synthetic porous foams have garnered significant attention due to their highly controllable pore structures. Melamine foam (MF), a commercially available loose and porous material, not only retains its rich three‐dimensional skeleton after carbonization (C‐MF) but also transforms into a highly efficient broadband photothermal absorber. Compared to the highly directional vascular bundles of wood, the disordered yet interconnected pore network of C‐MF ensures capillary water supply while being more conducive to the omnidirectional diffusion of water vapor. Guan et al. [62] leveraged this characteristic to construct a hierarchical coupling system based on Cu0.04In0.25ZnS y @Ru (CIZS@Ru) and C‐MF (Figure 11a). This design effectively achieves the cascaded utilization of the full solar spectrum: The upper CIZS@Ru photocatalyst primarily captures ultraviolet–visible light (UV–vis) to drive electron excitation for hydrogen production, while the penetrating visible and near‐infrared light (vis–NIR) is absorbed by the underlying C‐MF substrate and converted into thermal energy to drive in situ water vapor generation. This spectral splitting and synergistic thermal‐mass management strategy enabled the system to achieve a remarkable hydrogen production rate of 603 mmol h−1 m−2, fully demonstrating the immense potential of artificial foam substrates in constructing efficient gas–solid reaction interfaces (Figure 11b).
FIGURE 11.
(a) Schematic diagram of C‐MF@CIZS@Ru photothermal–photocatalytic coupling solar hydrogen production. (b) Variation of hydrogen production of CIZS@Ru under UV light and full spectrum irradiation in different systems [Reprinted with permission by Ref. [62]. Copyright 2023, American Chemical Society]. (c) Schematic of the fabrication of TiO2 nanoparticles‐decorated Ti3C2T x and GO hybrid foams. (d) Schematic mechanism of light‐trapping structures, photocatalytic degradation, and photothermal conversion for RGO/Ti3C2T x ‐TiO2 foams [Reprinted with permission by Ref. [113]. Copyright 2025, Springer Nature]. (e) Multifunctional MXene membrane interface for integrated seawater desalination and green hydrogen generation. (f) Photothermal and photocatalytic mechanisms of SrTiO3/MX@PVDF membrane: (i) heat and mass transfer through the membrane interface; (ii) photocatalytic vapor splitting at the gas–solid biphasic interface [Reprinted with permission by Ref. [114]. Copyright 2025, Wiley]. (g) Illustration of device construction and working principles of the SVG‐PC sheet [Reprinted with permission by Ref. [17]. Copyright 2023, Springer Nature].
Beyond foam skeletons, constructing microscopically ordered structures via freeze‐drying‐induced self‐assembly (FDISA) strategies is another key approach. Han et al. [113] constructed a semimetallic ternary hybrid foam (rGO/Ti3C2T x ‐TiO2) via an FDISA strategy, providing an excellent paradigm for the microdesign of photothermal materials. Unlike simple physical mixing, this work rationally designed a dual‐purpose in situ redox self‐assembly process: leveraging the strong reducing ability of Ti3C2T x MXene by mixing it with graphene oxide (GO) precursors (Figure 11c). During assembly, the MXene sacrifices part of its structure to reduce insulating GO into highly conductive rGO, while the oxidized MXene is in situ transformed into TiO2 nanoparticles. This synchronous synthesis not only achieves molecular‐level uniform dispersion of components but, more importantly, constructs robust Ti─O─C covalent bonds between rGO and unreacted Ti3C2T x (Figure 11d). These chemical bonds act as critical “electron highways,” allowing hot carriers generated by MXene under irradiation to be efficiently extracted and injected into the graphene‐like quantum states of rGO. This unique electronic structure coupling achieves a synergistic amplification of both photothermal conversion and photocatalytic performance.
In addition to three‐dimensional bulk structures, flexible device designs based on two‐dimensional membrane structures also exhibit unique advantages in interfacial modulation. Li et al. took a distinctive approach, moving away from traditional carbon‐based photothermal materials and opting for glass fiber membranes as substrates. The superior hydrophilicity and porosity of the glass fiber substrate ensure efficient capillary water supply. For the photothermal component, they utilized titanium nitride (TiN) nanoparticles. Unlike the lattice vibration heat generation mechanism of carbon materials, TiN, as a plasmonic material, can trigger strong localized surface plasmon resonance (LSPR) effects across a broad spectrum. This “nonlignin” inorganic membrane construction strategy validates that coupling plasmonic photothermal effects with thermally insulating porous substrates is an equally effective route for constructing efficient vapor‐phase overall water‐splitting interfaces [115].
Furthermore, addressing the potential issues of photocorrosion and mass transfer resistance caused by direct contact between liquid water and photocatalysts, the design of interfacial isolation membranes has become a new research hotspot. Sun et al. [114] proposed a multifunctional interfacial membrane strategy based on MXene/SrTiO3 (Figure 11e). They self‐assembled a hydrophilic SrTiO3/MXene functional layer onto a hydrophobic PVDF substrate membrane via vacuum filtration. In this system, the underlying hydrophobic PVDF membrane acts as a “liquid water barrier,” effectively blocking direct contact between bulk water and the upper photocatalyst; meanwhile, high‐temperature water vapor generated by the photothermal effect at the interface can smoothly pass through the membrane pores to supply the upper photocatalyst (Figure 11f). Simultaneously, the upper hydrophilic photocatalytic layer enhances the affinity for capturing water vapor through capillary condensation. This asymmetric wettability design of “hydrophobic isolation—hydrophilic enrichment” significantly promotes the hydrogen production kinetics of the photocatalyst in a gas‐phase environment.
To solve the mismatch between the rate of photothermal steam generation and the rate of photocatalytic reaction, Pornrungroj et al. [17] turned their attention to the management of macroscopic system flows, designing a hierarchically integrated “photothermal‐photocatalyst sheet” (Figure 11g). In this layered configuration, the bottom layer of hydrophilized carbon paper serves as the photothermal layer, responsible for absorbing penetrating light energy and generating water vapor via capillary action, which then diffuses upward to supply the top‐layer modified photocatalyst for reaction. A notable feature of this design lies in the effective utilization of “excess steam”: Since the rate of steam generation by the photothermal layer far exceeds the consumption rate of photocatalytic water splitting, the research team integrated a condensation collection structure above the reactor. This means that while producing hydrogen, the system can simultaneously condense and recover unreacted water vapor as clean freshwater. This concept of synergistically integrating “photocatalytic hydrogen production” with “solar water purification” not only optimizes the utilization of photothermal energy but also closely aligns with the “water─energy nexus” philosophy, providing a practical engineering paradigm for the development of future multifunctional integrated systems.
Having explored various specific external water supply strategies, we conducted a relative evaluation of the mechanistic advantages and disadvantages across several key engineering and kinetic dimensions for the four major subroutes, namely, capillary fibers, hierarchical foams, polymer hydrogels, and photothermal vapor feeding. This aims to systematically outline the underlying physical logic of these configurations and illustrate their intrinsic trade‐offs.
As shown in Table 2, the macroscopic performance traits of each configuration directly expose the inherent limitations of their underlying mass‐transfer mechanisms. For example, photothermal vapor feeding establishes a pure gas─solid reaction boundary that largely eliminates the adverse effects of macroscopic liquid films, resulting in outstanding gas desorption and thermal management. However, this thermodynamic advantage heavily relies on complex fluidic balances and gastight reaction chambers, significantly compromising system simplicity and scalability potential. In contrast, capillary fiber membranes rely on a fully passive operation mode to deliver unparalleled advantages in scalable deployment. Nevertheless, the relatively thick continuous liquid film remaining on their surfaces introduces substantial mass transfer resistance, which leads to suboptimal gas desorption kinetics. Similarly constrained by pore characteristics, hierarchical foams utilize three‐dimensional macropores to facilitate rapid gas escape, yet their water flux controllability remains poor. This makes them highly susceptible to localized dry‐out or excessive flooding during environmental fluctuations. Finally, polymer hydrogels face severe challenges in macroscopic engineering applications. Actual outdoor wet─dry cycling readily induces irreversible dehydration and structural cracking, while their highly hydrophilic networks exert a severe capillary bubble pinning effect on nascent microbubbles, which drastically hinders gas desorption. Despite these drawbacks, hydrogels provide a customizable chemically confined network at the molecular scale that regulates the reactive water environment, thereby still demonstrating commendable catalytic activity.
TABLE 2.
Systematic engineering evaluation matrix of the four nonimmersed photocatalytic architectures.
| Sub‐route paradigm | Water flux controllability | Thermal management | Stability in media | Scalability | Gas desorption kinetics |
|---|---|---|---|---|---|
| Capillary fiber | Fair | Good | Good | Excellent | Poor |
| Hierarchical foam | Poor | Fair | Good | Good | Good |
| Polymer hydrogel | Fair | Fair | Poor | Fair | Poor |
| Photothermal vapor | Good | Good | Fair | Fair | Excellent |
Note: Ratings (excellent > good > fair > poor) reflect relative mechanistic and engineering trade‐offs. “Excellent” indicates near‐ideal performance or minimal barriers. “Poor” denotes severe intrinsic limitations or high costs.
Regardless of the specific developmental path of these nonimmersed configurations, they all represent technological branches diverging from a single core trajectory. This central premise involves physically decoupling the catalytic reaction interface from the bulk liquid to reconstruct an ideal microscale reaction environment, thereby maximizing photon utilization and eliminating the hydrostatic pressure associated with macroscopic bubble desorption. However, the extreme complexity of multiphysical coupling dictates that resolving the primary bottlenecks of optical shielding and substrate mass transfer during the reconstruction of interfacial water states frequently triggers new systemic contradictions. The overall evaluation clearly indicates that upon comprehensive multidimensional consideration, no single architecture achieves absolute perfection. This demonstrates that within the design of non‐immersed systems, the advantages and disadvantages of different technical pathways are not absolute. Instead, they depend heavily on the limiting factors of reaction kinetics and the engineering robustness of the system in practical applications.
From the perspective of kinetic equilibrium, gas‐phase supply systems face severe challenges regarding the stability of substrate delivery.
Research has indicated that many gas‐phase reactions exhibit exceptionally high initial activity upon illumination, followed by a significant performance decay before reaching a dynamic equilibrium. The fundamental reason lies in the fact that, under dark conditions, the catalyst surface can accumulate a relatively thick layer of adsorbed water sufficient to support rapid initial consumption. Once illumination begins, the intensification of reaction consumption, coupled with the disturbances caused by product desorption, shifts the adsorption equilibrium toward desorption. This often leads the reaction into a “mass‐transfer limited” regime [32, 103]. In practical photothermally coupled systems, the generation of water vapor is highly dependent on ambient temperature and irradiation intensity. Such high sensitivity to external environments makes it difficult for the system to remain stably anchored within the narrow window of “optimal water film thickness.” Conversely, liquid‐phase microsupply systems (such as those driven by hydrogels or capillary channels) may sacrifice some theoretical efficiency due to slightly thicker water layers, yet they exhibit greater environmental robustness, ensuring that the substrate remains saturated at the reaction interface.
Compounding this complexity is the reality that, in practical experimental and operational settings, the boundaries between liquid‐ and gas‐phase supply mechanisms are often indistinct, rendering them difficult to isolate completely. On the one hand, liquid supply systems inevitably generate in situ vapor at the interface driven by the photothermal effect, meaning a portion of the reaction essentially proceeds via “gas‐phase decomposition.” On the other hand, many photothermal substrates (such as carbonized wood or porous membranes) possess strong capillary action; while the photocatalyst is in contact with vapor, it is highly probable that it also contacts liquid water climbing through the substrate’s micropores. This phenomenon of “phase coexistence and interplay” illustrates that liquid‐ and gas‐phase reaction pathways are highly coupled and dynamically interconverted at the actual solid–gas interface.
In summary, nonimmersed systems with external water supply demonstrate reaction activities and research value far exceeding those of traditional systems through the reconstruction of the interfacial microenvironment. While these systems are ingeniously structured and full of opportunities, their complex phase evolution and the requirement for stringent equilibrium control present significant challenges to researchers.
3.2.2. Self‐Supplied Systems
While nonimmersed systems with external feeding demonstrate significant performance gains by optimizing the local reaction interface, a more fundamental, macroscale challenge remains: the global geographical mismatch between solar and water resources (Figure 12a). The reliance on a continuous external water source inevitably implies additional energy and material investment for system operation (e.g., pumping and replenishment). This dependency becomes a critical bottleneck in arid, remote, or water‐scarce regions, where the acquisition of liquid water is itself a high‐cost undertaking. This “solar–water paradox” has motivated researchers to explore truly autonomous systems that can operate without a dedicated liquid water supply.
FIGURE 12.
(a) Superimposed atlas water risk and solar energy potential [Reprinted with permission by Ref. [116]. Copyright 2024, Royal Society of Chemistry]. (b) Gravimetric water uptake of typical moisture sorbents versus the characteristic RH at which half of the total uptake is reached [Reprinted with permission by Ref. [117]. Copyright 2024, Springer Nature]. (c) Schematics of four types of AWH sorbent [Reprinted with permission by Ref. [118]. Copyright 2023, Springer Nature].
To break this deadlock, researchers have begun seeking solutions that can spontaneously resolve the source of substrates without manual intervention. In this context, atmospheric water harvesting (AWH) technology has emerged as a prominent solution, utilizing hygroscopic materials to capture gaseous water molecules directly from the atmosphere [117, 119, 120, 121, 122, 123]. By deeply coupling AWH technology with nonimmersed photocatalysis, the system can transition from “passively waiting for water” to “actively capturing substrates.” This “self‐driven water uptake” nonimmersed photocatalytic system not only achieves a closed loop in terms of energy efficiency but also empowers photocatalytic technology to operate independently across all terrains and climates, thereby becoming a highly forward‐looking research branch in the field.
In the construction of nonimmersed photocatalytic systems, eliminating dependence on liquid water transport pipelines to truly achieve decentralized and globally ubiquitous energy conversion remains a core engineering challenge. In this context, atmospheric water—a pervasive and “universal substrate”—offers a novel solution for building self‐driven reaction systems. It is estimated that the Earth’s atmosphere sustains a constant water vapor reservoir of approximately 1.29 × 1013 tons, a volume more than six times the total discharge of all surface rivers globally, maintaining a dynamic equilibrium with the hydrosphere [124]. More critically, unlike liquid freshwater resources, which are distributed extremely unevenly, atmospheric water is ubiquitous even in arid inland regions or the Sahara Desert. This fundamentally resolves the “solar–water geographical mismatch,” where regions rich in solar energy often suffer from severe water scarcity [116, 125]. Consequently, the development of technologies capable of in situ capture of atmospheric moisture is no longer pursued solely for drinking water production but is increasingly recognized as a vital strategy to provide a continuous, purified reactant source for photocatalysis.
Although traditional AWH technologies, such as fog harvesting and dew condensation, have proven effective under specific conditions, they face stringent environmental constraints: Fog harvesting relies heavily on high‐humidity weather (RH > 90%) and specific topographies, while active condensation technologies, though mature, incur exponentially rising energy costs under low‐humidity conditions (RH < 50%) and are limited by thermodynamic dew‐point constraints [126, 127]. In contrast, sorption‐based atmospheric water harvesting (SAWH) demonstrates unique environmental adaptability utilizing the chemical potential gradient at the surface of hygroscopic materials as the driving force [118, 122].
From a thermodynamic perspective, SAWH is a “capture–concentration–release” cycle driven by chemical potential differences. During the night or periods of low temperature and high humidity, the sorbent is in an unsaturated state where the chemical potential of water molecules on its surface is lower than that of the ambient vapor, driving gaseous water molecules to overcome entropy reduction and spontaneously migrate into the material where they are fixed. Conversely, during the day, the photothermal effect is utilized to raise the material’s temperature, significantly elevating the chemical potential of the adsorbed water above the ambient vapor pressure, thereby driving water molecules to overcome the binding energy and release [117, 127].
It is crucial to note that coupling SAWH with photocatalytic hydrogen production imposes kinetic requirements distinct from those of drinking water production. Traditional water production pursues a “fast‐uptake, fast‐release” strategy to maximize liquid water yield per unit time. However, in a coupled photocatalytic system, water serves not merely as a product but as a critical substrate participating in surface redox reactions. If the desorption rate is too fast, moisture may evaporate instantaneously, or if the supply rate falls below the sum of photocatalytic consumption and evaporative loss, the reaction interface faces the risk of “dry‐out,” interrupting the reaction. Conversely, excessive accumulation of water forming a thick liquid film would reintroduce the mass transfer limitations characteristic of submerged systems. Therefore, photocatalytic coupling requires a more refined “supply–demand matching”: Materials must ensure capture capability at low humidity while enabling controlled release of moisture through pore structure or wettability modulation to maintain long‐term wetting and activity equilibrium at the catalytic interface [123].
To meet these specialized demands, the design of core hygroscopic materials has evolved from early single‐component systems to multifunctional composites. Current mainstream research focuses on several representative material classes: Porous solids, represented by metal‐organic frameworks (MOFs), offer stepwise adsorption isotherms with fast kinetics due to their tunable pore chemistry, making them particularly suitable for environments with large diurnal temperature shifts like deserts [122, 128]. Hygroscopic salts (e.g., LiCl), despite their high water storage density, face risks of liquefaction and leakage and are thus often confined within porous matrices to balance uptake capacity with structural stability [129]. Furthermore, emerging hygroscopic polymer hydrogels synergize hygroscopic salts with thermosensitive polymer networks, not only efficiently “locking” moisture but also reducing desorption energy consumption through phase‐change mechanisms, offering an ideal carrier for constructing “water–energy” coupled interfaces (Figure 12b) [121]. In summary, introducing SAWH technology into nonimmersed photocatalysis essentially constructs a “smart atmospheric water pump” that transcends thermodynamic and kinetic barriers to concentrate dilute gaseous water molecules into highly active reaction substrates, thereby providing a solid material foundation for realizing all‐weather, liquid‐water‐independent “hydrogen from air” production [130].
Among various AWH materials, hygroscopic salts (e.g., LiCl, CaCl2, ZnCl2) have long occupied a central position due to their superior intrinsic hygroscopicity [129, 131]. Leveraging their unique deliquescence properties, these inorganic salts can overcome the physical limitations of pore adsorption, achieving a phase transition from solid salt to saturated solution driven by chemical potential. Consequently, they exhibit water uptake capacities far exceeding those of physical adsorbents (such as silica gel and zeolites) across a broad humidity range, from low to high levels [118]. However, this solid–liquid phase transition presents a significant engineering challenge: The postsorption salt solution exists in a macroscopically flowable state. Without proper confinement, it is prone to leakage and loss, leading to irreversible decay in system performance over repeated cycles, thereby limiting its long‐term stability in nonimmersed devices [131].
To address this issue, researchers have turned to constructing composite systems by coupling hygroscopic salts with porous matrices (Figure 12c). Among these, the “hygroscopic salt–hydrogel composite adsorbent,” formed by integrating hygroscopic salts within a polymeric hydrogel network, is widely recognized as the most promising solution [121, 132]. In this system, the three‐dimensional polymer network acts as a robust “microreservoir”: Although the hygroscopic salt still undergoes microscopic deliquescence and liquefaction, the resulting high‐concentration salt solution is firmly locked within the gel framework via hydrogen bonding and capillary forces, provided the loading amount is appropriately controlled [123]. This effectively realizes a macroscopically “leakage‐free” quasisolid operating mode. Crucially, this coupling is not merely a physical mixture but demonstrates significant synergistic effects: The polymer network provides a physical barrier against leakage and structural support, while its inherent hydrophilic groups contribute additional moisture absorption capacity at high humidity levels; simultaneously, the high concentration of salt ions dispersed within the network significantly lowers the internal water vapor partial pressure of the gel, drastically enhancing capture kinetics under low‐humidity conditions [117]. Driven by this superior water retention capability, processability, and effective confinement of liquid water, the hygroscopic salt–hydrogel composite system has emerged as the carrier of choice for constructing air–water harvesting‐coupled nonimmersed photocatalytic systems [130].
Early AWH‐coupled photocatalytic systems primarily adopted a straightforward “layer‐by‐layer” strategy, utilizing high‐molecular‐weight hydrogels loaded with highly hygroscopic salts (e.g., LiCl, CaCl2) as the bottom “water‐harvesting substrate,” upon which a photocatalyst layer was composited as the “reaction terminal,” thereby constructing a macroscopic coupled device with “bottom water absorption and top hydrogen production.” For instance, a series of works by Huang et al. [133, 134, 135] successfully validated the feasibility of this concept (Figure 13a). They utilized PAAm or PNIPAM hydrogels loaded with CaCl2 and integrated Pt/TiO2 or other catalyst layers onto the gel surface via simple physical adhesion or in situ polymerization. These studies, through laboratory and outdoor prototype testing, robustly demonstrated the engineering viability of utilizing atmospheric water for photocatalytic hydrogen production (Figure 13b,c).
FIGURE 13.
(a) The working principle of the hybrid AWH–PHE device for capturing moisture from the air to generate H2 by solar‐driven water splitting. (b) Comparison of hydrogen evolution performance between interfacial photocatalytic water vapor and volumetric photocatalysis in direct contact with bulk water under one sun illumination [Reprinted with permission by Ref. [133]. Copyright 2024, Elsevier]. (c) Time‐dependent variations of H2 evolution and mass change in a cycle [Reprinted with permission by Ref. [134]. Copyright 2025, Elsevier]. (d) Architecture of the multifunction hygroscopic salt/aerogels/photocatalysts system for AWH at night and water‐energy production in the daytime. (e) Correlation between water collection and H2 production rates of the system. (f) Time‐dependent photocatalytic overall water‐splitting profiles of LiCl/rGO/Nv‐CN/Ov‐BVO system for ten cycles after water sorption under 50% RH [Reprinted with permission by Ref. [136]. Copyright 2025, Springer Nature]. (g) AWH─photothermal water evaporation–photocatalytic hydrogen evolution (AWPH) system. (h) Adsorption and desorption diffusion coefficient of CW550 under different RH at 298 K. (i) Photocatalytic hydrogen evolution performance at different flow rates [Reprinted with permission by Ref. [137]. Copyright 2024, Wiley].
While these configurations effectively demonstrate the feasibility of sustained moisture supply, their kinetic efficiency is fundamentally bounded by mass‐transfer dynamics. AWH inherently relies on the contact area between the material and air. When hydrogels are prepared as dense monolithic blocks, their relatively low surface‐to‐volume ratio and limited internal porosity can moderate rapid air diffusion. Consequently, moisture capture kinetics and overall water uptake are structurally governed, which subsequently influences the continuous reactant supply to the coupled photocatalyst.
To overcome these kinetic bottlenecks, researchers have focused on optimizing the microstructure of the substrate, aiming to construct high‐specific‐surface‐area hydrophobic/hydrophilic channels to enhance mass transfer. Wang et al. [136] proposed an all‐weather multifunctional system utilizing reduced graphene oxide (rGO) aerogel as a scaffold to load LiCl (Figure 13d). The loose, porous hierarchical structure of the rGO aerogel not only significantly increased the contact area with air, improving moisture absorption kinetics, but also provided low‐resistance channels for the rapid evaporation and upward transport of moisture captured at night during the day. Tests showed that photocatalytic activity increased with rising humidity and that performance remained stable over multiple adsorption–desorption cycles (Figure 13e,f).
Similar approaches have been applied to the modification of biomass substrates. Yan et al. [138] developed a nanofibrillated cellulose (NFC) composite bilayer gel (NLS). This design fully leveraged the abundant surface hydroxyl groups and natural hydrophilicity of the NFC network, which not only assisted moisture capture via hydrogen bonding but also effectively dispersed LiCl particles to prevent agglomeration. More importantly, this structure achieved “spectral splitting” of solar energy: The top SrTiO3:Al catalyst utilized ultraviolet/visible light for hydrogen production, while the transmitted infrared light was absorbed by the bottom substrate and converted into thermal energy, driving the continuous supply of moisture to the reaction interface. Furthermore, He et al. [137] took a different approach by utilizing natural wood as a substrate, subjecting it to carbonization after impregnation with ZnCl2 (Figure 13g). In this process, ZnCl2 acted not only as the final hygroscopic component but also as a “pore‐forming agent” during high‐temperature carbonization, opening the occluded pores of the wood. This carbonized wood substrate, retaining its natural vertical channel structure, demonstrated excellent capillary water transport capability and AWH performance.
Although the aforementioned macroscopic coupled systems achieved high water uptake using hygroscopic salts, the hygroscopic components and catalytic active sites remained spatially separated. To achieve ultimate mass transfer efficiency, researchers began exploring the integrated “hygroscopic–catalytic” assembly at the micro–nano scale.
MOFs, with their ultrahigh specific surface area and regular pore channels, are regarded as ideal microscopic hygroscopic carriers. Yan et al. [139] loaded MOF‐801 nanocrystals directly onto the surface of SrTiO3:Al photocatalysts, constructing a microscopic interface with intimate “hygroscopic‐catalytic” contact. The pore structure of the MOF acts like a micro‐pump, feeding captured water molecules directly to adjacent catalytic sites, significantly reducing gas‐phase mass transfer resistance (Figure 14a). However, this configuration places high demands on interfacial bonding quality: Poor adhesion may lead to material separation during the reaction, negating the advantage of direct in situ water supply; additionally, if the MOF coating is too thick or unevenly distributed, it may become a barrier to light absorption or even obscure active sites on the catalyst surface, leading to diminishing returns.
FIGURE 14.
(a) Water adsorption by MOF‐801‐hydrazine from arid air (left) and water supply for SrTiO3:Al photocatalyst toward continuous conversion from atmospheric water to H2 and O2 (right). (b) Time courses of H2 and O2 production over MS‐20 under Xe‐lamp irradiation [Reprinted with permission by Ref. [139]. Copyright 2024, American Chemical Society]. (c) Schematic illustration of amorphous metal oxide‐coated, CoOOH/Rh‐loaded SrTiO3:Al. (d) Durability test of CoOOH/Rh‐loaded SrTiO3:Al coated with TiO x under saturated water vapor feeding balanced with Ar at 24 °C (10 mL min−1, pH2O = 2.9 kPa) [Reprinted with permission by Ref. [140]. Copyright 2022, Springer Nature]. (e) Schematic representation of hygroscopic COFs for photocatalytic H2 evolution from water vapor. (f) Time‐dependent H2 production of solid Py‐HMPA@Pt. (g) The recyclability of photocatalytic H2 production for Py‐HMPA@Pt [Reprinted with permission by Ref. [141]. Copyright 2023, Elsevier]. (h) Schematic illustration of the moisture sorption enabled by TPL. (i) Qualitative comparison of TPL gel with Pt–TiO2/PAAm–CaCl2 (abbreviated as TPC) materials [Reprinted with permission by Ref. [142]. Copyright 2025, Wiley].
To address the issues of light shading and active site obscuration, Suguro et al. [140] proposed a more refined surface engineering strategy. They grew an amorphous TiO x or TaO x layer, only a few nanometers thick, on the surface of SrTiO3:Al particles (Figure 14c). This ultrathin oxide layer, rich in micro–mesopores and hydrophilicity, effectively captured water molecules from the gas phase and maintained a “quasiliquid” microenvironment on the catalyst surface, thereby inhibiting the occurrence of reverse reactions (Figure 14d). Meanwhile, spectral testing confirmed that this nanometer‐thick modification layer had almost no impact on the catalyst’s light absorption, achieving perfect compatibility between hygroscopic function and light absorption function.
Furthermore, covalent organic framework (COF) materials, due to their designable skeletal structures, also demonstrate the potential to realize single‐component multifunctionality. Liu et al. [141] synthesized a series of pyrene‐based one‐dimensional COF materials. By introducing hydrophilic groups into the backbone, the material itself possessed excellent AWH capability, while the pyrene units served as photosensitizing centers to drive hydrogen production (Figure 14e,f). The device demonstrated robust durability, maintaining stable hydrogen production rates over multiple adsorption‐reaction cycles without significant degradation (Figure 14g). This “two‐in‐one” intrinsic design strategy significantly mitigates mass transfer and compatibility issues between heterogeneous interfaces, offering a new avenue for simplifying system design.
In summary, macroscopic systems based on hygroscopic salts excel in capacity and kinetics but struggle to achieve intimate microscopic coupling with photocatalysts, whereas microscopic systems based on MOFs/COFs or thin oxide layers are easy to construct reaction interfaces but are limited by their own water uptake capacities. Therefore, future high‐performance systems urgently require a cross‐scale integration strategy capable of merging the advantages of both. Addressing this challenge, Yang et al. [142] proposed a “multifunction hygroscopic‐photocatalytic hydrogel nanocomposite.” This study creatively introduced photocatalysts with ultrathin amorphous oxide layers grown on their surfaces into a hydrogel network loaded with high‐concentration hygroscopic salts (Figure 14h). In this design, the hygroscopic salt hydrogel serves as a “macroscopic reservoir,” ensuring high moisture capture in low‐humidity environments; meanwhile, the oxide modification layer on the catalyst surface acts as a “microscopic pump station” and “interfacial binder.” Research found that the oxide layer not only significantly improved the interfacial compatibility between inorganic photocatalyst particles and the organic hydrogel network, preventing particle agglomeration, but also further enhanced the hydrophilicity and water retention capability of the local microenvironment. Experimental data indicated that, compared to unmodified catalysts, the composite system introducing the oxide layer achieved further improvements in overall hygroscopic capacity. This hierarchical coupling design effectively balances macroscopic total water uptake with microscopic interfacial reaction rates, providing a highly promising universal paradigm for solving the “supply‐reaction” kinetic mismatch in non‐immersed systems (Figure 14i).
4. Summary and Outlook
This review has systematically traced the evolution of reactor design in photocatalytic water splitting, providing a comprehensive overview of innovative architectures aimed at overcoming the intrinsic limitations of traditional immersed systems. Departing from the conventional focus on catalyst modification, this article highlights the critical, yet often overlooked, role of the “interfacial water layer” as a decisive systemic variable in determining device performance. Centered on this theme, the design philosophy of non‐immersed photocatalytic systems has been elucidated herein, establishing a complete classification framework based on the phase state of water supply—ranging from external supply strategies (liquid microfluidics, photothermal vapor feeding) to self‐supply approaches (AWH coupling). These studies, which achieve synergistic optimization of light, heat, and mass transfer through interfacial reconfiguration, have not only significantly expanded the application boundaries of photocatalytic technology but also offered novel system‐level solutions for realizing efficient, self‐driven solar hydrogen production. However, despite the promising activity and proof‐of‐concept results demonstrated by some prototype devices, research on non‐immersed systems remains in its infancy, and the transition from laboratory exploration to practical engineering application faces numerous deep‐seated scientific and technical challenges.
The mechanisms underlying the overall activity enhancement in nonimmersed systems require a transition from qualitative inference to quantitative resolution. Current explanations for activity improvements are largely attributed to macroscopic factors, such as reduced mass transfer resistance due to water layer thinning or enhanced photothermal effects. However, disentangling the individual impacts of these coupled factors remains a significant challenge, and direct in situ experimental verification remains sparse. Therefore, there is an urgent need to introduce advanced in situ and operando characterization techniques to monitor the dynamic evolution of the interfacial microenvironment during the reaction, enabling a shift from phenomenon description to mechanistic dissection. Furthermore, emerging microscopic theories require in‐depth validation. For instance, while the theory of “interfacial hydrogen‐bond network rigidity” has been preliminarily verified on model catalysts like titanium dioxide, its universal applicability to more compositionally and structurally complex photocatalysts, including metal sulfides, MOFs, and COFs, still demands rigorous experimental validation in the future. A further promising hypothesis suggests that the spontaneous interfacial electric fields inherent to confined water layers may fundamentally steer reaction pathways and accelerate carrier dynamics, mirroring effects observed in microdroplets [34]. Establishing a unified and universal microscopic theoretical framework will not only support the interpretation of experimental phenomena but also feed back into material design, guiding the fundamental enhancement of catalytic efficiency through surface microenvironment modulation.
Constructing efficient nonimmersed systems with scalable production potential remains a critical gap on the path to engineering. Beyond mechanistic understanding, the greatest innovation of nonimmersed systems—particularly those coupled with AWH—lies in their ability to greatly expand the geographical applicability of photocatalytic devices. However, this requires the equipment to be adaptable to real‐world outdoor environments. Most current research is confined to small‐area laboratory tests, lacking the large‐scale outdoor demonstration data available for traditional immersed panel systems. During scale‐up, engineering issues such as the uniformity of multiphysical field distributions (heat and mass), continuous separation and collection of products, and long‐term stability in complex environments will be significantly amplified. Consequently, future research must shift from solely pursuing high activity to balancing “manufacturability” and “system robustness.” Developing low‐cost, easily processable components suitable for industrial production is key to realizing technological implementation.
Through interdisciplinary integration, nonimmersed systems hold promise for broader applications beyond photocatalytic water splitting. The achievements in nonimmersed systems are themselves exemplars of the crossfertilization of multiple disciplines. As global environmental and energy demands evolve, the focus of photocatalysis research is shifting from pure hydrogen production toward more diverse chemical transformations. Among these, photocatalytic carbon dioxide conversion represents a highly promising direction for achieving carbon neutrality. Given that nonimmersed systems inherently possess the advantage of a gas–solid reaction interface, investigating whether gaseous CO2 can be efficiently coupled with trace water films at nonimmersed interfaces to optimize the selectivity and yield of hydrocarbons through triphase interface modulation presents an exciting new frontier. Such organic interdisciplinary integration is expected to spawn entirely new reaction pathways and technological routes, profoundly influencing the future development trends of this field.
With the rapid development of artificial intelligence (AI), utilizing data‐driven methods to accelerate material and system screening has become a critical agenda. The accuracy of AI models relies heavily on high‐quality data input. Despite the vast volume of literature in the photocatalysis field, the extreme lack of standardization in data reporting (e.g., testing conditions, light source parameters, activity calculation methods) constitutes a major barrier. In the future, establishing standardized testing protocols and databases for nonimmersed photocatalysis will be a prerequisite for utilizing AI to discover high‐performance catalyst combinations and optimize reactor structures. Through data‐driven approaches, the rapid identification of optimal solutions within the vast parameter space can be realized, thereby accelerating the research and design process of the entire field.
In summary, the rise of nonimmersed systems has broadened the horizon of photocatalytic water‐splitting research, expanding the dimension of study from single catalyst materials to a holistic consideration of the reaction microenvironment. Although challenges such as unclear mechanisms and scale‐up difficulties persist, their unique advantages in enhancing light utilization, breaking mass transfer limits, and adapting to extreme environments make them a highly promising next‐generation solar energy conversion technology. Ultimately, with deepening mechanistic research, optimized engineering design, and the integration of interdisciplinary technologies, nonimmersed photocatalytic systems are poised to play an irreplaceable role in the grand process of driving the global energy transition and achieving carbon neutrality goals.
Funding
This study was supported by National Natural Science Foundation of China (22102095, 21773153), and National Key Research and Development Program of China (2018YFB1502001).
Conflicts of Interest
The authors declare no conflicts of interest.
Acknowledgments
This work was supported by the Project of National Natural Science Foundation of China (22102095, 21773153) and the National Key Basic Research and Development Program (2018YFB1502001).
Data Availability Statement
Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.
References
- 1. Schleussner C.‐F., Castro P., Lesnikowski A., et al., “Expert Retrospective on a Decade of the Paris Agreement,” Nature Climate Change 15 (2025): 1136–1140. [Google Scholar]
- 2. Lee H., Calvin K., Dasgupta D., et al., IPCC 2023: Climate Change 2023: Synthesis Report, Summary for Policymakers. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change, ed. Lee H., Romero J., and Core Writing Team (Intergovernmental Panel on Climate Change (IPCC), 2023). [Google Scholar]
- 3. Zhu Y., Raimi D., Joiner E., Holmes B., and Prest B. C., Global Energy Outlook 2025: Headwinds and Tailwinds in the Energy Transition (Resources for the Future, 2025). [Google Scholar]
- 4. Boettcher S. W., “Introduction to Green Hydrogen,” Chemical Reviews 124 (2024): 13095–13098. [DOI] [PubMed] [Google Scholar]
- 5. Holmes‐Gentle I., Tembhurne S., Suter C., and Haussener S., “Kilowatt‐Scale Solar Hydrogen Production System Using a Concentrated Integrated Photoelectrochemical Device,” Nature Energy 8 (2023): 586–596. [Google Scholar]
- 6. Kim J. H., Hansora D., Sharma P., Jang J.‐W., and Lee J. S., “Toward Practical Solar Hydrogen Production–an Artificial Photosynthetic Leaf‐to‐Farm Challenge,” Chemical Society Reviews 48 (2019): 1908–1971. [DOI] [PubMed] [Google Scholar]
- 7. Wang Q. and Domen K., “Particulate Photocatalysts for Light‐Driven Water Splitting: Mechanisms, Challenges, and Design Strategies,” Chemical Reviews 120 (2019): 919–985. [DOI] [PubMed] [Google Scholar]
- 8. Lin S., Huang H., Ma T., and Zhang Y., “Photocatalytic Oxygen Evolution From Water Splitting,” Advanced Science 8 (2021): 2002458. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Acar C., Dincer I., and Naterer G. F., “Review of Photocatalytic Water‐Splitting Methods for Sustainable Hydrogen Production,” International Journal of Energy Research 40 (2016): 1449–1473. [Google Scholar]
- 10. Pelicano C. M. and Tong H., “Recent Advances in Cocatalyst Engineering for Solar‐Driven Overall Water Splitting,” Applied Research 3 (2024): e202300080. [Google Scholar]
- 11. Kumar V., Prasad Singh G., Kumar M., et al., “Nanocomposite Marvels: Unveiling Breakthroughs in Photocatalytic Water Splitting for Enhanced Hydrogen Evolution,” ACS Omega 9 (2024): 6147–6164. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Tao X., Zhao Y., Wang S., Li C., and Li R., “Recent Advances and Perspectives for Solar‐Driven Water Splitting Using Particulate Photocatalysts,” Chemical Society Reviews 51 (2022): 3561–3608. [DOI] [PubMed] [Google Scholar]
- 13. Guo L., Chen Y., Su J., Liu M., and Liu Y., “Obstacles of Solar‐Powered Photocatalytic Water Splitting for Hydrogen Production: A Perspective From Energy Flow and Mass Flow,” Energy 172 (2019): 1079–1086. [Google Scholar]
- 14. Abhishek B., Jayarama A., Sanjog S. N., Siddhartha P. D., and Sriganesh S. P., “Challenges in Photocatalytic Hydrogen Evolution: Importance of Photocatalysts and Photocatalytic Reactors,” International Journal of Hydrogen Energy 81 (2024): 1442–1466. [Google Scholar]
- 15. Nishioka S., Osterloh F. E., Wang X., Mallouk T. E., and Maeda K., “Photocatalytic Water Splitting,” Nature Reviews. Methods Primers 3 (2023): 42. [Google Scholar]
- 16. Li R., Luan J., Zhang Y., et al., “A Review of Efficient Photocatalytic Water Splitting for Hydrogen Production,” Renewable & Sustainable Energy Reviews 206 (2024): 114863. [Google Scholar]
- 17. Pornrungroj C., Mohamad Annuar A. B., Wang Q., et al., “Hybrid Photothermal–photocatalyst Sheets for Solar‐Driven Overall Water Splitting Coupled to Water Purification,” Nature Water 1 (2023): 952–960. [Google Scholar]
- 18. Suguro T., Kishimoto F., and Takanabe K., “Photocatalytic Hydrogen Production under Water Vapor Feeding─A Minireview,” Energy & Fuels 36 (2022): 8978–8994. [Google Scholar]
- 19. Fujishima A. and Honda K., “Electrochemical Photolysis of Water at a Semiconductor Electrode,” Nature 238 (1972): 37–38. [DOI] [PubMed] [Google Scholar]
- 20. Qureshi M. and Takanabe K., “Insights on Measuring and Reporting Heterogeneous Photocatalysis: Efficiency Definitions and Setup Examples,” Chemistry of Materials 29 (2017): 158–167. [Google Scholar]
- 21. Takata T., Lin L., Hisatomi T., and Domen K., “Best Practices for Assessing Performance of Photocatalytic Water Splitting Systems,” Advanced Materials 36 (2024): 2406848. [DOI] [PubMed] [Google Scholar]
- 22. Zhao Z., Goncalves R. V., Barman S. K., et al., “Electronic Structure Basis for Enhanced Overall Water Splitting Photocatalysis with Aluminum Doped SrTiO3 in Natural Sunlight,” Energy & Environmental Science 12 (2019): 1385–1395. [Google Scholar]
- 23. Zhou P., Navid I. A., Ma Y., et al., “Solar‐to‐Hydrogen Efficiency of More than 9% in Photocatalytic Water Splitting,” Nature 613 (2023): 66–70. [DOI] [PubMed] [Google Scholar]
- 24. Li J., Sheng B., Chen Y., et al., “Utilizing Full‐Spectrum Sunlight for Ammonia Decomposition to Hydrogen over GaN Nanowires‐Supported Ru Nanoparticles on Silicon,” Nature Communications 15 (2024): 7393. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Hisatomi T. and Domen K., “Reaction Systems for Solar Hydrogen Production via Water Splitting with Particulate Semiconductor Photocatalysts,” Nature Catalysis 2 (2019): 387–399. [Google Scholar]
- 26. Schröder M., Kailasam K., Borgmeyer J., et al., “Hydrogen Evolution Reaction in a Large‐Scale Reactor Using a Carbon Nitride Photocatalyst under Natural Sunlight Irradiation,” Energy Technology 3 (2015): 1014–1017. [Google Scholar]
- 27. Goto Y., Hisatomi T., Wang Q., et al., “A Particulate Photocatalyst Water‐Splitting Panel for Large‐Scale Solar Hydrogen Generation,” Joule 2 (2018): 509–520. [Google Scholar]
- 28. Yamada T., Nishiyama H., Akatsuka H., et al., “Production of Methane by Sunlight‐Driven Photocatalytic Water Splitting and Carbon Dioxide Methanation as a Means of Artificial Photosynthesis,” ACS Engineering Au 3 (2023): 352–363. [Google Scholar]
- 29. Cao F., Wei Q., Liu H., Lu N., Zhao L., and Guo L., “Development of the Direct Solar Photocatalytic Water Splitting System for Hydrogen Production in Northwest China: Design and Evaluation of Photoreactor,” Renewable Energy 121 (2018): 153–163. [Google Scholar]
- 30. Wei Q., Yang Y., Hou J., Liu H., Cao F., and Zhao L., “Direct Solar Photocatalytic Hydrogen Generation with CPC Photoreactors: System Development,” Solar Energy 153 (2017): 215–223. [Google Scholar]
- 31. Yang L., Feng N., Wang Q., Chu Y., Xu J., and Deng F., “Surface Water Loading on Titanium Dioxide Modulates Photocatalytic Water Splitting,” Cell Reports Physical Science 1 (2020): 100013. [Google Scholar]
- 32. Lin Z., Saito H., Sato H., and Sugimoto T., “Positive and Negative Impacts of Interfacial Hydrogen Bonds on Photocatalytic Hydrogen Evolution,” Journal of the American Chemical Society 146 (2024): 22276–22283. [DOI] [PubMed] [Google Scholar]
- 33. Ma X., Shi Y., Liu J., et al., “Hydrogen‐Bond Network Promotes Water Splitting on the TiO2 Surface,” Journal of the American Chemical Society 144 (2022): 13565–13573. [DOI] [PubMed] [Google Scholar]
- 34. Li K., You W., Zhu Y., et al., “Strong Electric Fields on Water Microdroplets Enable Near‐Unity Selectivity in H2O2 Photosynthesis,” Journal of the American Chemical Society 147 (2025): 36136–36145. [DOI] [PubMed] [Google Scholar]
- 35. Yu H., Xuan Y., Zhu Q., and Chang S., “Highly Efficient and Stable Photocatalytic CO2 and H2O Reduction into Methanol at Lower Temperatures through an Elaborate Gas–liquid–solid Interfacial System,” Green Chemistry 25 (2023): 596–605. [Google Scholar]
- 36. Lee W. H., Lee C. W., Cha G. D., et al., “Floatable Photocatalytic Hydrogel Nanocomposites for Large‐Scale Solar Hydrogen Production,” Nature Nanotechnology 18 (2023): 754–762. [DOI] [PubMed] [Google Scholar]
- 37. Su J., Sun N., Zhao Y., et al., “Exploring Novel Strategies and Integrated Systems for Photothermal Catalytic Hydrogen Production,” Chemical Engineering Journal 512 (2025): 162794. [Google Scholar]
- 38. Lei D., Wang L., Lv Y., Luo N., and Wang Z., “A Comprehensive Review of Solar Photocatalysis & Photothermal Catalysis for Hydrogen Production From Biomass: From Material Characteristics to Engineering Application,” Chemistry – A European Journal 30 (2024): e202401486. [DOI] [PubMed] [Google Scholar]
- 39. Nasir A. M., Jaafar J., Aziz F., et al., “A Review on Floating Nanocomposite Photocatalyst: Fabrication and Applications for Wastewater Treatment,” Journal of Water Process Engineering 36 (2020): 101300. [Google Scholar]
- 40. Pinaud B. A., Benck J. D., Seitz L. C., et al., “Technical and Economic Feasibility of Centralized Facilities for Solar Hydrogen Production via Photocatalysis and Photoelectrochemistry,” Energy & Environmental Science 6 (2013): 1983–2002. [Google Scholar]
- 41. Maeda K., “Photocatalytic Water Splitting Using Semiconductor Particles: History and Recent Developments,” Journal of Photochemistry and Photobiology C: Photochemistry Reviews 12 (2011): 237–268. [Google Scholar]
- 42. Chen X., Shen S., Guo L., and Mao S. S., “Semiconductor‐Based Photocatalytic Hydrogen Generation,” Chemical Reviews 110 (2010): 6503–6570. [DOI] [PubMed] [Google Scholar]
- 43. Toroker M. C., Kanan D. K., Alidoust N., Isseroff L. Y., Liao P., and Carter E. A., “First Principles Scheme to Evaluate Band Edge Positions in Potential Transition Metal Oxide Photocatalysts and Photoelectrodes,” Physical Chemistry Chemical Physics 13 (2011): 16644–16654. [DOI] [PubMed] [Google Scholar]
- 44. Wang Y., Silveri F., Bayazit M. K., et al., “Bandgap Engineering of Organic Semiconductors for Highly Efficient Photocatalytic Water Splitting,” Advanced Energy Materials 8 (2018): 1801084. [Google Scholar]
- 45. Maeda K. and Domen K., “New Non‐Oxide Photocatalysts Designed for Overall Water Splitting under Visible Light,” The Journal of Physical Chemistry C 111 (2007): 7851–7861. [Google Scholar]
- 46. Schumacher L. and Marschall R., “Recent Advances in Semiconductor Heterojunctions and Z‐Schemes for Photocatalytic Hydrogen Generation,” Topics in Current Chemistry 380 (2022): 53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47. Xie Z., Gao Q., Shang X., et al., “Mini Review on Electron Mediator in Artificial Photosynthesis: Design, Fabrication, and Perspectives Based on Energy Level Matching,” Green Carbon 2 (2024): 366–382. [Google Scholar]
- 48. Maeda K., “Z‐Scheme Water Splitting Using Two Different Semiconductor Photocatalysts,” ACS Catalysis 3 (2013): 1486–1503. [Google Scholar]
- 49. Hisatomi T., Kubota J., and Domen K., “Recent Advances in Semiconductors for Photocatalytic and Photoelectrochemical Water Splitting,” Chemical Society Reviews 43 (2014): 7520–7535. [DOI] [PubMed] [Google Scholar]
- 50. Kronawitter C. X., Vayssieres L., Shen S., et al., “A Perspective on Solar‐Driven Water Splitting with All‐Oxide Hetero‐Nanostructures,” Energy & Environmental Science 4 (2011): 3889–3899. [Google Scholar]
- 51. Pu Y.‐C., Kibria M., Mi Z., and Zhang J. Z., “Ultrafast Exciton Dynamics in InGaN/GaN and Rh/Cr2O3 Nanoparticle‐Decorated InGaN/GaN Nanowires,” The Journal of Physical Chemistry Letters 6 (2015): 2649–2656. [DOI] [PubMed] [Google Scholar]
- 52. Grigioni I., Stamplecoskie K. G., Jara D. H., et al., “Wavelength‐Dependent Ultrafast Charge Carrier Separation in the WO3/BiVO4 Coupled System,” ACS Energy Letters 2 (2017): 1362–1367. [Google Scholar]
- 53. Corp K. L. and Schlenker C. W., “Ultrafast Spectroscopy Reveals Electron‐Transfer Cascade that Improves Hydrogen Evolution with Carbon Nitride Photocatalysts,” Journal of the American Chemical Society 139 (2017): 7904–7912. [DOI] [PubMed] [Google Scholar]
- 54. Wang X., Xu Q., Li M., et al., “Photocatalytic Overall Water Splitting Promoted by an α‐β Phase Junction on Ga2O3 ,” Angewandte Chemie International Edition 51 (2012): 13089–13092. [DOI] [PubMed] [Google Scholar]
- 55. Ma Y., Pendlebury S. R., Reynal A., Formal F. Le, and Durrant J. R., “Dynamics of Photogenerated Holes in Undoped BiVO4 Photoanodes for Solar Water Oxidation,” Chemical Science 5 (2014): 2964–2973. [Google Scholar]
- 56. Godin R., Wang Y., Zwijnenburg M. A., Tang J., and Durrant J. R., “Time‐Resolved Spectroscopic Investigation of Charge Trapping in Carbon Nitrides Photocatalysts for Hydrogen Generation,” Journal of the American Chemical Society 139 (2017): 5216–5224. [DOI] [PubMed] [Google Scholar]
- 57. Leenheer A. J. and Atwater H. A., “Water‐Splitting Photoelectrolysis Reaction Rate via Microscopic Imaging of Evolved Oxygen Bubbles,” Journal of the Electrochemical Society 157 (2010): B1290. [Google Scholar]
- 58. Sequeira C. A., Santos D. M., Šljukić B., and Amaral L., “Physics of Electrolytic Gas Evolution,” Brazilian Journal of Physics 43 (2013): 199–208. [Google Scholar]
- 59. Hu X., Guo L., and Wang Y., “In Situ Measurement of Local Hydrogen Production Rate by Bubble‐Evolved Recording,” International Journal of Photoenergy 2013 (2013): 568206. [Google Scholar]
- 60. Maksimov A. O., Kaverin A. M., and Baidakov V. G., “Heterogeneous Vapor Bubble Nucleation on a rough Surface,” Langmuir 29 (2013): 3924–3934. [DOI] [PubMed] [Google Scholar]
- 61. Mason J. D., Cone M. T., and Fry E. S., “Ultraviolet (250–550 Nm) Absorption Spectrum of Pure Water,” Applied Optics 55 (2016): 7163–7172. [DOI] [PubMed] [Google Scholar]
- 62. Ding L., Li K., Li J., et al., “Integrated Coupling Utilization of the Solar Full Spectrum for Promoting Water Splitting Activity over a CIZS Semiconductor,” ACS Nano 17 (2023): 11616–11625. [DOI] [PubMed] [Google Scholar]
- 63. Ni G., Zandavi S. H., Javid S. M., Boriskina S. V., Cooper T. A., and Chen G., “A Salt‐Rejecting Floating Solar Still for Low‐Cost Desalination,” Energy & Environmental Science 11 (2018): 1510–1519. [Google Scholar]
- 64. Fu J., Xiao S., Cao J., et al., “Mass Transfer‐Enhanced Photothermal Membranes with Synergistic Light Utilization for High‐Turbidity Wastewater Purification,” Angewandte Chemie International Edition 64 (2025): e202421800. [DOI] [PubMed] [Google Scholar]
- 65. Asay D. B. and Kim S. H., “Evolution of the Adsorbed Water Layer Structure on Silicon Oxide at Room Temperature,” The Journal of Physical Chemistry B 109 (2005): 16760–16763. [DOI] [PubMed] [Google Scholar]
- 66. Wang Y., Hu X., Cao Z., and Guo L., “Investigations on Bubble Growth Mechanism during Photoelectrochemical and Electrochemical Conversions,” Colloids and Surfaces A: Physicochemical and Engineering Aspects 505 (2016): 86–92. [Google Scholar]
- 67. Chen C., Gao Y., Wang F., and Sun C., “A Multiscale Model for Bubble Nucleation Thresholds on Solid Walls in the Presence of Nanometre‐Sized Defects,” Journal of Fluid Mechanics 1015 (2025): A47. [Google Scholar]
- 68. Mondaca‐Medina E., García‐Carrillo R., Lee H., Wang Y., Zhang H., and Ren H., “Nanoelectrochemistry in Electrochemical Phase Transition Reactions,” Chemical Science 14 (2023): 7611–7619. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 69. Fernandez D., Maurer P., Martine M., Coey J., and Möbius M. E., “Bubble Formation at a Gas‐Evolving Microelectrode,” Langmuir 30 (2014): 13065–13074. [DOI] [PubMed] [Google Scholar]
- 70. Zhang Q., Mo D., Moon S., et al., “Bubble Nucleation and Growth on Microstructured Surfaces under Microgravity,” Npj Microgravity 10 (2024): 13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 71. Scardina P. and Edwards M., “Prediction and Measurement of Bubble Formation in Water Treatment,” Journal of Environmental Engineering 127 (2001): 968–973. [Google Scholar]
- 72. Lubetkin S. and Blackwell M., “The Nucleation of Bubbles in Supersaturated Solutions,” Journal of Colloid and Interface Science 126 (1988): 610–615. [Google Scholar]
- 73. Njoka F., Mori S., Ookawara S., and Ahmed M., “Effects of Photo‐Generated Gas Bubbles on the Performance of Tandem Photoelectrochemical Reactors for Hydrogen Production,” International Journal of Hydrogen Energy 44 (2019): 10286–10300. [Google Scholar]
- 74. Liang F., van de Krol R., and Abdi F. F., “Engineering Strategies to Minimize Bubble‐Induced Optical Losses in Photoelectrochemical Water Splitting,” Chemical Engineering Journal 512 (2025): 162513. [Google Scholar]
- 75. Yang X., Chen P., Guo Y., et al., “Emerging Floatable Photothermal Coupled Catalytic System for Solar‐to‐Hydrogen Conversion,” Advanced Functional Materials 35 (2025): 2503851. [Google Scholar]
- 76. Cao F., Liu H., Wei Q., Zhao L., and Guo L., “Experimental Study of Direct Solar Photocatalytic Water Splitting for Hydrogen Production under Natural Circulation Conditions,” International Journal of Hydrogen Energy 43 (2018): 13727–13737. [Google Scholar]
- 77. Nishiyama H., Yamada T., Nakabayashi M., et al., “Photocatalytic Solar Hydrogen Production From Water on a 100‐m2 Scale,” Nature 598 (2021): 304–307. [DOI] [PubMed] [Google Scholar]
- 78. Wang Q., Okunaka S., Tokudome H., et al., “Printable Photocatalyst Sheets Incorporating a Transparent Conductive Mediator for Z‐Scheme Water Splitting,” Joule 2 (2018): 2667–2680. [Google Scholar]
- 79. Zhao Y., Ding C., Zhu J., et al., “A Hydrogen Farm Strategy for Scalable Solar Hydrogen Production with Particulate Photocatalysts,” Angewandte Chemie International Edition 59 (2020): 9653–9658. [DOI] [PubMed] [Google Scholar]
- 80. Patari S. and Mahapatra P. S., “Liquid Wicking in a Paper Strip: An Experimental and Numerical Study,” ACS Omega 5 (2020): 22931–22939. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 81. d. S. Oliveira H., Silva A. C., de Mesquita J. P., et al., “A Novel Floating Photocatalyst Device Based on Cloth Canvas Impregnated with Iron Oxide,” New Journal of Chemistry 37 (2013): 2486–2491. [Google Scholar]
- 82. Gao M., Zhang T., Ng S. W. L., et al., “Polymeric Layered Films for TiO2‐Au/CuS Tandem Photothermal Catalytic H2 Production in Harsh Seawater and Waste Plastic Media,” Advanced Energy Materials 15 (2025): 2404198. [Google Scholar]
- 83. Sun M., Zhou L., Dong T., et al., “Interfacial Design to Enhance Photocatalytic Hydrogen Evolution via Optimizing Energy and Mass Flows,” ACS Applied Materials & Interfaces 13 (2021): 21207–21216. [DOI] [PubMed] [Google Scholar]
- 84. Sun M., Wang Y., Dong T., et al., “Construction of an Interfacial Photocatalytic Mode Based on Carbonized Mushrooms to Enhance Infrared Light‐Assisted Photocatalytic Water Splitting Performance,” Langmuir 38 (2022): 2811–2820. [DOI] [PubMed] [Google Scholar]
- 85. Konch T. J., Dutta T., Buragohain M., and Raidongia K., “Remarkable Rate of Water Evaporation through Naked Veins of Natural Tree Leaves,” ACS Omega 6 (2021): 20379–20387. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 86. Jia C., Li Y., Yang Z., et al., “Rich Mesostructures Derived From Natural Woods for Solar Steam Generation,” Joule 1 (2017): 588–599. [Google Scholar]
- 87. Yuan Y., Zhang Q., Lin S., and Li J., “Water: The Soul of Hydrogels,” Progress in Materials Science 148 (2025): 101378. [Google Scholar]
- 88. Liu Y. and Guo Z., “Small Functional Hydrogels with Big Engineering Applications,” Materials Today Physics 43 (2024): 101397. [Google Scholar]
- 89. Rieth A. J., Hunter K. M., Dincă M., and Paesani F., “Hydrogen Bonding Structure of Confined Water Templated by a Metal‐Organic Framework with Open Metal Sites,” Nature Communications 10 (2019): 4771. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 90. Rockwell P. N., Maneval J. E., Vogel B. M., and Jablonski E. L., “Water Diffusion and Uptake in Injectable ETTMP/PEGDA Hydrogels,” The Journal of Physical Chemistry B 127 (2023): 5055–5061. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 91. Dargaville B. and Hutmacher D., “Water as the Often Neglected Medium at the Interface between Materials and Biology,” Nature Communications 13 (2022): 4222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 92. Li J., Li W., Ding L., Xu R., and Chang K., “Positive Impact of Weak Hydrogen Bonds on O–H Bond Dissociation in Photocatalytic Water Splitting,” ACS Nano 19 (2025): 41319–41329. [DOI] [PubMed] [Google Scholar]
- 93. Lee W. H., Park H., Lee C. W., et al., “Polymeric Stabilization at the Gas–liquid Interface for Durable Solar Hydrogen Production From Plastic Waste,” Nature Nanotechnology 20 (2025): 1–10. [DOI] [PubMed] [Google Scholar]
- 94. Ma N., Dong L., Zhang T., et al., “A Photothermal‐Photocatalytic Layered Aerogel for Harvesting Water and Hydrogen From Seawater,” Journal of Colloid and Interface Science 687 (2025): 573–581. [DOI] [PubMed] [Google Scholar]
- 95. Shearer C. J., Hisatomi T., Domen K., and Metha G. F., “Gas Phase Photocatalytic Water Splitting of Moisture in Ambient Air: Toward Reagent‐Free Hydrogen Production,” Journal of Photochemistry and Photobiology A: Chemistry 401 (2020): 112757. [Google Scholar]
- 96. Guo Y., Tan X., Yu T., and Gong J., “Recent Advances of Photocatalytic Hydrogen Evolution via Water Vapor Splitting,” Advanced Functional Materials 36 (2026): e22276. [Google Scholar]
- 97. Li H., Wahab Ahmed K., Abdelsalam M. A., Fowler M., and Wu X.‐Y., “From Seawater to Hydrogen via Direct Photocatalytic Vapor Splitting: A Review on Device Design and System Integration,” Frontiers in Energy 18 (2024): 291–307. [Google Scholar]
- 98. Domen K., Naito S., Soma M., Onishi T., and Tamaru K., “Photocatalytic Decomposition of Water Vapour on an NiO–SrTiO3 Catalyst,” Journal of the Chemical Society, Chemical Communications 12 (1980): 543–544. [Google Scholar]
- 99. Mao K., Zhang Y., and Tan S. C., “Functionalizing Solar‐Driven Steam Generation towards Water and Energy Sustainability,” Nature Water 3 (2025): 144–156. [Google Scholar]
- 100. Oura K., Katayama M., Zotov A., Lifshits V., and Saranin A., Surface Science: An Introduction (Springer, 2003), 357–387. [Google Scholar]
- 101. Qu M., Huang G., Liu X., et al., “Room Temperature Bilayer Water Structures on a Rutile TiO2 (110) Surface: Hydrophobic or Hydrophilic?,” Chemical Science 13 (2022): 10546–10554. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 102. Wang R., Hashimoto K., Fujishima A., et al., “Light‐Induced Amphiphilic Surfaces,” Nature 388 (1997): 431–432. [Google Scholar]
- 103. Daeneke T., Dahr N., Atkin P., et al., “Surface Water Dependent Properties of Sulfur‐Rich Molybdenum Sulfides: Electrolyteless Gas Phase Water Splitting,” ACS Nano 11 (2017): 6782–6794. [DOI] [PubMed] [Google Scholar]
- 104. Zhu M., Li Y., Chen G., et al., “Tree‐Inspired Design for High‐Efficiency Water Extraction,” Advanced Materials 29 (2017): 1704107. [DOI] [PubMed] [Google Scholar]
- 105. Kuang Y., Chen C., He S., et al., “A High‐Performance Self‐Regenerating Solar Evaporator for Continuous Water Desalination,” Advanced Materials 31 (2019): 1900498. [DOI] [PubMed] [Google Scholar]
- 106. He S., Chen C., Kuang Y., et al., “Nature‐Inspired Salt Resistant Bimodal Porous Solar Evaporator for Efficient and Stable Water Desalination,” Energy & Environmental Science 12 (2019): 1558–1567. [Google Scholar]
- 107. Go M., Hong I., Lee D., et al., “Ultrabroadband Absorptive Refractory Plasmonics for Photocatalytic Hydrogen Evolution Reactions,” NPG Asia Materials 16 (2024): 4. [Google Scholar]
- 108. Guo S., Li X., Li J., and Wei B., “Boosting Photocatalytic Hydrogen Production From Water by Photothermally Induced Biphase Systems,” Nature Communications 12 (2021): 1343. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 109. Han H., Huang K., Yao Y., Li Z., and Meng X., “Enhanced Photocatalytic Splitting of Photothermally Induced Water Vapor to Evolve Hydrogen,” Chemical Engineering Journal 450 (2022): 138419. [Google Scholar]
- 110. Wang Y., Huang W., Guo S., et al., “Sulfur‐Deficient ZnIn2S4/Oxygen‐Deficient WO3 Hybrids with Carbon Layer Bridges as a Novel Photothermal/Photocatalytic Integrated System for Z‐Scheme Overall Water Splitting,” Advanced Energy Materials 11 (2021): 2102452. [Google Scholar]
- 111. Han H. and Meng X., “Hydrothermal Preparation of C3N4 on Carbonized Wood for Photothermal‐Photocatalytic Water Splitting to Efficiently Evolve Hydrogen,” Journal of Colloid and Interface Science 650 (2023): 846–856. [DOI] [PubMed] [Google Scholar]
- 112. Li L., Gong P., Zhang Y., et al., “Boosting Photocatalytic Water Vapor Splitting by Nanosecond Infrared Laser‐Assisted Synthesis of Photothermal Substrate,” ACS Applied Energy Materials 8 (2025): 5190–5198. [Google Scholar]
- 113. Han D.‐D., Wang Q., Chen Z.‐D., et al., “Light‐Propelled Photocatalytic Evaporator for Robotic Solar‐Driven Water Purification,” PhotoniX 6 (2025): 11. [Google Scholar]
- 114. Sun J., Farid M. U., Li X., et al., “MXene Membrane as Multifunctional Interface for Vapor Splitting via Photothermal‐Catalytic Membrane Distillation,” Advanced Materials 38 (2025): e13926. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 115. Li J., Ding L., Su Z., et al., “Non‐Lignin Constructing the Gas–solid Interface for Enhancing the Photothermal Catalytic Water Vapor Splitting,” Advanced Materials 35 (2023): 2305535. [DOI] [PubMed] [Google Scholar]
- 116. Guo J., Butson J. D., Zhang Y., Hu G., Fan X., and Li G. K., “Hydrogen Generation From Atmospheric Water,” Journal of Materials Chemistry A 12 (2024): 12381–12396. [Google Scholar]
- 117. Xu J., Wang P., Bai Z., et al., “Sustainable Moisture Energy,” Nature Reviews Materials 9 (2024): 722–737. [Google Scholar]
- 118. Li R. and Wang P., “Sorbents, Processes and Applications beyond Water Production in Sorption‐Based Atmospheric Water Harvesting,” Nature Water 1 (2023): 573–586. [Google Scholar]
- 119. Zou H., Yang X., Zhu J., et al., “Solar‐Driven Scalable Hygroscopic Gel for Recycling Water From Passive Plant Transpiration and Soil Evaporation,” Nature Water 2 (2024): 663–673. [Google Scholar]
- 120. Xu J., Li T., Chao J., et al., “Efficient Solar‐Driven Water Harvesting From Arid Air with Metal–organic Frameworks Modified by Hygroscopic Salt,” Angewandte Chemie International Edition 59 (2020): 5202–5210. [DOI] [PubMed] [Google Scholar]
- 121. Guo H., Luo Q., Liu D., et al., “Super Moisture‐Sorbent Sponge for Sustainable Atmospheric Water Harvesting and Power Generation,” Advanced Materials 36 (2024): 2414285. [DOI] [PubMed] [Google Scholar]
- 122. Kim H., Yang S., Rao S. R., et al., “Water Harvesting From Air with Metal‐Organic Frameworks Powered by Natural Sunlight,” Science 356 (2017): 430–434. [DOI] [PubMed] [Google Scholar]
- 123. Entezari A., Esan O. C., Yan X., Wang R., and An L., “Sorption‐Based Atmospheric Water Harvesting: Materials, Components, Systems, and Applications,” Advanced Materials 35 (2023): 2210957. [DOI] [PubMed] [Google Scholar]
- 124. Oki T. and Kanae S., “Global Hydrological Cycles and World Water Resources,” Science 313 (2006): 1068–1072. [DOI] [PubMed] [Google Scholar]
- 125. Lord J., Thomas A., Treat N., et al., “Global Potential for Harvesting Drinking Water From Air Using Solar Energy,” Nature 598 (2021): 611–617. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 126. Klemm O., Schemenauer R. S., Lummerich A., et al., “Fog as a Fresh‐Water Resource: Overview and Perspectives,” AMBIO 41 (2012): 221–234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 127. Poredoš P., Shan H., Wang C., Deng F., and Wang R., “Sustainable Water Generation: Grand Challenges in Continuous Atmospheric Water Harvesting,” Energy & Environmental Science 15 (2022): 3223–3235. [Google Scholar]
- 128. Hanikel N., Prévot M. S., Fathieh F., et al., “Rapid Cycling and Exceptional Yield in a Metal‐Organic Framework Water Harvester,” ACS Central Science 5 (2019): 1699–1706. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 129. Wang X., Li X., Liu G., et al., “An Interfacial Solar Heating Assisted Liquid Sorbent Atmospheric Water Generator,” Angewandte Chemie 131 (2019): 12182–12186. [DOI] [PubMed] [Google Scholar]
- 130. Guo J., Zhang Y., Zavabeti A., et al., “Hydrogen Production From the Air,” Nature Communications 13 (2022): 5046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 131. Grekova A., Girnik I., Nikulin V., Tokarev M., Gordeeva L., and Aristov Y. I., “New Composite Sorbents of Water and Methanol “Salt in Anodic Alumina”: Evaluation for Adsorption Heat Transformation,” Energy 106 (2016): 231–239. [Google Scholar]
- 132. Li R., Shi Y., Alsaedi M., Wu M., Shi L., and Wang P., “Hybrid Hydrogel with High Water Vapor Harvesting Capacity for Deployable Solar‐Driven Atmospheric Water Generator,” Environmental Science & Technology 52 (2018): 11367–11377. [DOI] [PubMed] [Google Scholar]
- 133. Huang L., Liu P., Qin C., et al., “Solar‐Driven Hydrogen Production Based on Moisture Adsorption‐Desorption Cycle,” Nano Energy 128 (2024): 109879. [Google Scholar]
- 134. Liu P., Huang L., Qin C., et al., “Scalable Hydrogen Production by Harvesting Moisture From the Air under Natural Sunlight,” Chemical Engineering Journal 510 (2025): 161832. [Google Scholar]
- 135. Qin C., Ren T., Gui C., et al., “Solar‐Driven Hydrogen and Water Co‐Generation Based on Atmospheric Water Harvesting,” Chemical Engineering Journal 504 (2025): 158721. [Google Scholar]
- 136. Wang Y., Zhao P., Zhang Y., et al., “All‐Day Multifunction Hygroscopic Salt/Aerogels/Photocatalysts System for Efficient Freshwater‐Hydrogen‐Oxygen Co‐Production From Air,” Science China Chemistry 68 (2025): 1–11. [Google Scholar]
- 137. He L., Zeng X., Chen H., et al., “A Hybrid Photocatalytic System Splits Atmospheric Water to Produce Hydrogen,” Advanced Functional Materials 34 (2024): 2313058. [Google Scholar]
- 138. Yan X., Tian L., Xue F., et al., “In Situ Hydrogen Production in All‐Level‐Humidity Air: Integrating Atmospheric Water Harvesting with Photocatalysis,” EES Catalysis 3 (2025): 297–304. [Google Scholar]
- 139. Yan X., Tian L., Huang J., et al., “Integrated System for Photocatalytic Overall Water Splitting From Arid Air,” Nano Letters 25 (2024): 71–76. [DOI] [PubMed] [Google Scholar]
- 140. Suguro T., Kishimoto F., Kariya N., et al., “A Hygroscopic Nano‐Membrane Coating Achieves Efficient Vapor‐Fed Photocatalytic Water Splitting,” Nature Communications 13 (2022): 5698. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 141. Liu Y., Han W.‐K., Chi W., et al., “One‐Dimensional Covalent Organic Frameworks with Atmospheric Water Harvesting for Photocatalytic Hydrogen Evolution From Water Vapor,” Applied Catalysis B: Environmental 338 (2023): 123074. [Google Scholar]
- 142. Yang W., Si L., Cheng C., et al., “Synergistic Integration of Atmospheric Water Harvesting and Solar‐Driven Hydrogen Production via Multifunctional Hygroscopic‐Photocatalytic Hydrogel Nanocomposite,” Advanced Functional Materials 36 (2025): e12738. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.