Solar fuel production through concentrating light irradiation

Abstract

The climate crisis necessitates the development of non-fossil energy sources. Harnessing solar energy for fuel production shows promise and offers the potential to utilize existing energy infrastructure. However, solar fuel production is in its early stages of development, constrained by low conversion efficiency and challenges in scaling up production. Concentrated solar energy (CSE) technology has matured alongside the rapid growth of solar thermal power plants. This review provides an overview of current CSE methods and solar fuel production, analyzes their integration compatibility, and delves into the theoretical mechanisms by which CSE impacts solar energy conversion efficiency and product selectivity in the context of photo-electrochemistry, thermochemistry, and photo-thermal co-catalysis for solar fuel production. The review also summarizes approaches to studying the photoelectric and photothermal effects of CSE. Lastly, it explores emerging novel CSE technology methods in the field of solar fuel production.

Headnote

Abstract

The climate crisis necessitates the development of non-fossil energy sources. Harnessing solar energy for fuel production shows promise and offers the potential to utilize existing energy infrastructure. However, solar fuel production is in its early stages of development, constrained by low conversion efficiency and challenges in scaling up production. Concentrated solar energy (CSE) technology has matured alongside the rapid growth of solar thermal power plants. This review provides an overview of current CSE methods and solar fuel production, analyzes their integration compatibility, and delves into the theoretical mechanisms by which CSE impacts solar energy conversion efficiency and product selectivity in the context of photo-electrochemistry, thermochemistry, and photo-thermal co-catalysis for solar fuel production. The review also summarizes approaches to studying the photoelectric and photothermal effects of CSE. Lastly, it explores emerging novel CSE technology methods in the field of solar fuel production.

Keywords: Concentrated solar energy; Solar fuel; Photo-electrochemistry; Thermochemistry; Photo-thermal co-catalysis

1. Introduction

Converting solar energy into usable fuels [1,2], known as solar fuels, holds significant promise in mitigating carbon dioxide emissions resulting from the extensive reliance on fossil fuels. Over the past half-century, numerous techniques and methodologies have been investigated, encompassing areas such as photo-electrochemical processes [3–6], thermochemical fuel synthesis [7–9], photo-thermal co-catalysis [10,11], etc. However, many of these techniques are still being studied in laboratories due to their relatively low efficiency or high costs [12–14]. As concentrated solar-thermal power (CSP) technology advances [15], the matured field of concentrated solar energy (CSE) is increasingly relevant for promoting the use of clean solar fuels. This review concentrates on the application and underlying theories of employing CSE in the prevailing methods of solar fuel production.

The concept of future sustainable energy recycling is illustrated in Fig. 1a. The central challenge lies in converting stable CO2 and H2O into high-energy fuels, such as H2, CO, CH3OH, and CH4, capable of serving as direct fuels or forms of energy storage [16]. Various effective methods exist to realize these transformation processes, all of which demand substantial energy inputs. Embracing renewable energy sources such as solar for these processes is imperative within the future energy system. Solar fuel production typically necessitates high temperatures or substantial electrical currents, both of which inherently require energy concentration. Thus, the utilization of CSE naturally lends itself to the production of solar fuels.

The current mainstream methods of solar concentrating technologies applied in commercial CSP plants are illustrated in Fig. 1b. These methods encompass parabolic trough collector systems, linear Fresnel reflector systems, dish–engine systems, and central receiver systems [17]. The level of concentration can be characterized by the concentration ratio (CR), which is defined as the ratio of the concentrator aperture area (the large mirror area intercepting sunlight) to the receiver aperture area (the small receiver area where the sunlight is redirected) [15]. The CR plays a pivotal role in the application of various CSE technologies, particularly in the process of solar-to-fuel (STF) conversion. A high CR has the potential to significantly elevate the working temperature of the reactor, thereby enhancing the efficiency and dynamics of chemical reactions through the exploitation of thermal effects [18]. However, it is worth noting that a high CR may lead to substantial thermal losses due to thermal convection. Additionally, the elevated temperature could potentially shift the balance towards the reverse reactions [19]. Therefore, careful consideration is essential when selecting the appropriate CSE equipment and determining the optimal CR value for the desired application.

The principal attributes of various CSP plants are enumerated in Table 1. The key performance metric for a CSP plant is the levelized cost of electricity (LCOE). This metric encompasses the overall cost of electrical energy generated by the plant, taking into account both capital expenditures and operational and maintenance costs. According to the data listed in Table 1, apart from the dish-engine, which currently hasn't practical commercial applications, other types of CSE technologies exhibit similar economic characteristics. Therefore, when considering the integration of CSE technology with the STF process, more emphasis should be placed on the technical compatibility of different CSE methods and STF approaches. For instance, as indicated in Table 1, the difference between the annual average efficiency and peak efficiency of parabolic trough collector and dish-engine systems is larger compared to the other two system types. This is because linear Fresnel reflector and central receiver systems possess better irradiance tracking systems for effective focusing, whereas parabolic trough collector and dish-engine systems rely on the inherent characteristics of optical devices for concentration. Insufficient attention to irradiance tracking systems in these systems leads to a significant disparity between their annual average efficiency and peak efficiency. Parabolic trough collector and linear Fresnel reflector are concentrating systems that direct incident solar radiation onto a solar receiver positioned along the focal line using one-axis tracking mirrors. The CR ranges between 30 and 80, with thermal fluid temperatures reaching up to 500 °C. As such, parabolic trough collector and linear Fresnel reflector systems are very well-suited for driving the Rankine cycle to provide dispatchable electricity supply, and they represent the most mature technology in CSP plants. The central receiver system refers to concentrating systems that focus incoming solar radiation onto a solar receiver situated atop a tower, utilizing a large field of two-axis tracking heliostats. Generally, CR falls between 200 and 1000, with thermal fluid temperatures that could reach up to 700 °C. This design of the central receiver system makes it suitable for integration into advanced ther-modynamic cycles or thermochemical processes, which often require high temperatures for efficient operation. Current research on the use of different thermal fluids in central receiver, including saturated or superheated steam, molten salts, and atmospheric or pressurized air, has demonstrated the capability of achieving temperatures within the range of 500- 2000 °C. The dish-engine system consists of smaller two-axis tracking parabolic concentrators that focus incoming solar radiation onto a Stirling engine or Brayton mini-turbine situated at the focal point. CR ranges between 100 and 3000. Due to the high value of CR, it is relatively easy to heat the thermal fluid to temperatures higher than 1500 °C. However, due to the high costs, dish-engine systems are still in the process of prototype development.

To discuss the CSE technology in STF conversion, it is necessary to summarize and classify the prevailing methods for STF conversion. As illustrated in Fig. 2a, the natural photosynthesis process includes a few steps: light absorption, H2O oxidation, CO2 reduction and energy storage [2]. The STF conversion process is essentially simulating natural photosynthesis. The simplest approach is to adopt the photocatalysis scheme shown in Fig. 2b, which involves high-frequency photons absorption by using a wide-bandgap semiconductor, and free electron-hole pair generation that drive oxidation and reduction reactions. The resulting products are typically mixed and require subsequent steps for separation. However, the severe recombination of photogenerated charge carriers often leads to STF efficiency usually below 1%, far from the industrial application requirements (STF efficiency 5%–10%) [20–24]. The combination of photovoltaics and electrocatalysis, represented by PV-EC [25–27] (Fig. 2c) and PEC [28–31] (Fig. 2d) using modern photovoltaic manufacturing processes and III-V semiconductor materials, can achieve efficiencies of 10%–30%. Still, they are constrained by the high manufacturing costs of III-V semiconductor photovoltaic cells, and their hydrogen production cost [13] (10–15 $/kg) is currently unable to compete with the mainstream natural gas reforming scheme for hydrogen production [13] (1–2 $/kg). In order to further efficiently utilize solar energy, photothermal catalysis based on the effective integration of the solar-driven thermal effect can significantly improve STF conversion efficiency [10]. Pure thermochemistry (Fig. 2e) and photothermal co-catalysis (Fig. 2f) are representative, with their typical characteristics being high conversion efficiency. They can be conveniently combined with current CSE technology and the thermal catalysis industry, achieving decarbonization in traditional petrochemical and thermochemical industries. The industrial process of STF conversion primarily involves the oxidation of H2O and the reduction of CO2, just like the natural photosynthesis illustrated in Fig. 2a. This includes processes such as the dissociation of individual molecules of CO2 and H2O, the hydrogenation reduction of CO2, and artificial photosynthesis where both CO2 and H2O participate simultaneously. The common reaction types are listed in Fig. 2g.

In the following context, we will discuss the combinations of different CSE technology and solar fuel production methods.

2. Solar fuel through photo-electrochemistry

2.1. Overview

In terms of the Photo-electrochemical roadmap, the overall water splitting (OWS) is viewed as the most promising avenues for achieving industrial application, three methods are presently being considered: the photocatalytic (PC) [21] routes illustrated in Fig. 3a and b, photoelectrochemical (PEC) [32,33] as depicted in Fig. 3c, and photovoltaic–electrolysis (PV–EC) [34]. Regarding the PV-EC and PEC routes, the primary distinction is whether the photovoltaic (PV) and electrolysis (EC) components are separate or integrated. The industrial application of photo-electrochemical hydrogen production requires achieving a minimum energy conversion efficiency of 10%, transforming solar energy into hydrogen.

Despite the typical solar-to-hydrogen (STH) efficiency ranging between 8% and 14% in the case of PV-EC [24], it remains uncompetitive due to the high manufacturing cost of PV modules. Because the optimal efficiencies are commonly attained through the utilization of costly III-V PV materials, particularly when operating under light concentration. While conventional silicon photovoltaic modules can be manufactured at a low cost and have already been widely adopted, they fail to achieve high efficiency in the production of hydrogen using PV-EC or PEC methods. Hence, employing CSE technology to reduce the consumption of III-V PV materials and electrochemical precious metal catalysts could enhance the competitiveness of photo-electrochemical solar fuels [25,35]. Based on the techno-economic analysis of PEC hydrogen production, increasing the concentration ratio to 20 could potentially result in a 10% cost reduction [36].

The notion of combining PEC with a parabolic trough collector, as shown in Fig. 3d, necessitates a substantial consumption of PV materials structured into a linear form. This presents challenges considering the current manufacture process of the PV industry, and it may also be economically unfeasible. Aside from considerations of manufacture process, the optimal concentration ratio for a III-V PV module typically falls within the range of 100–500 [36,37]. Therefore, parabolic trough collector and linear Fresnel reflector systems might not deliver the most suitable concentration ratio for photo-electrochemical devices, nor provide sufficient localized electric current for the electrochemical process. Concerning the central receiver system, since it requires building a solar tower, if the CR is chosen under 500, then the techno-economy performance is poor, but designing a PV module capable of withstanding such a high energy flux density at 1000 suns (typical value for central receiver system) becomes challenging due to the substantial energy influx into the solar receiver. For the dish concentrator, it is easy to design and build a system with CR between 100 and 1500, so it is very common to choose the dish concentrator to combine with the PEC system. In the case of the PC route, particulate catalysts are dispersed within the reactant solution, allowing for direct integration with the parabolic trough collector and linear Fresnel reflector systems. Taking into account the compatibility between CSE technologies and various photo-electrochemical solar fuel production methods, the recommended combinations are detailed in Table 2. As for the other types of STF processes listed in Table 2, their compatibility will be explained later in the corresponding section.

2.2. Theoretical foundation behind CSE

The photoelectric current and its correlation with irradiation intensity are the most significant physical characteristics in the context of CSE-assisted photoelectrochemical solar fuel production. In the case of a typical PV * EC or PEC system, since most of the PV material used with good performances are III-V semiconductor materials, the photoelectric current is governed by the IV characteristics of the PV material. Numerous studies have been conducted to investigate the response of PV modules to concentrated solar energy. Generally, it has been observed that the photoelectric current at the maximum power point increases linearly with irradiation intensity with sufficient cooling [29].

As depicted in Fig. 4a–c, in a typical PEC device, III-V semiconductor materials serve as light-absorbing agents, while metallic materials function as electrodes. With increasing light intensity, the working photocurrent generally exhibits a linear increase. This is due to the relatively small exponential region of the IV characteristic curve of the photoelectrode commonly employed, which is evident in the IV curves of different-sized Ni foam shown in Fig. 4d. However, for PEC devices utilizing materials such as metal oxides as the integrated lightabsorbing material and electrode, the photoelectrode may operate within the exponential region of the IV curve. Consequently, the impact of light intensity on the photocurrent may be more complex, as illustrated in Fig. 5.

In the case of metal oxide photoelectrodes, as depicted in Fig. 5a, these photoelectrodes serve dual roles as both the photoelectric material and electrocatalyst [38]. Unlike in PV modules where charge carrier separation relies on the P–N junction, in this scenario, charge carrier separation is anticipated to occur at the semiconductor-liquid junction [39]. These factors introduce a greater degree of complexity to the photoelectric current characteristics exhibited by metal oxide photoelectrodes. In the presence of an applied bias voltage, as depicted in Fig. 5a, the photoelectric current varies corresponding to changes in the bias voltage. Typically, the photoelectric current at the maximum power point is considered the characteristic current [38,40], as demonstrated in Fig. 5b. This current also demonstrates a linear increase in tandem with rising irradiation intensity. However, in the absence of an applied bias voltage, the photocurrent remains constant at a fixed irradiation flux and may exhibit a more intricate dependency on irradiation intensity [41], as illustrated in Fig. 5c and d. Under conditions of low irradiation intensity, the photoelectric current may exhibit exponential growth with increasing irradiation flux. This phenomenon arises from the high concentration of surface states on the photoelectrode, which effectively captures photoexcited charge carriers and modifies the photogenerated voltage. Consequently, the photoelectric current experiences exponential augmentation following the Butler–Volmer equation [42] governing electrochemical reactions. Nevertheless, in the presence of high irradiation intensity, the surface state occupation has been saturated by photoexcited charge carriers, more photoexcited charge would induce a shift in band bending at the semiconductor-liquid junction, akin to the behavior observed in the P–N junction of a PV module. Consequently, this alteration leads to a linear increase in the photoelectric current [43,44]. Regarding the particle-mediated PC, complexity arises as there may exist multiple junctions in series, as exemplified in Fig. 5e. The heterojunction, comprising BiVO4 and SrTiO3, exhibits two distinct regimes of photoelectric current enhancement with increasing irradiation intensity, as depicted in Fig. 5f.

Considering all factors, the characteristics of photoelectric current in PEC and PC may exhibit greater complexity than PV + EC. In contrast to the well-established theories for PV + EC and PEC based on III-V semiconductor materials, the development of a comprehensive theory for oxide photoelectrode-based PEC and particle-mediated PC remains in its infancy, potentially presenting a challenge in the utilization of CSE technology.

2.3. Typical devices and techno-economy

Contemporary investigations into the utilization of CSE for photo-electrochemical solar fuel production primarily center around PV-EC and PEC approaches. This preference arises from the fact that the fundamental efficiency of the PC method is insufficient to warrant industrial applications. Two typical devices used in the lab and pilot-scale apparatus are shown in Fig. 6a and b, respectively. In Fig. 6a, a 1.1 m x 1.1 m Fresnel lens was employed to generate concentrated solar irradiation of approximately 16,070 mW/cm² (equivalent to about 160 suns), resulting in a notable STH efficiency of 9.2%. This achievement was attained using pure water and Rh/Cr2O3/ C03O4 photocatalysts loaded onto InGaN/GaN heterostructure nanowires. In this scenario, the concentrated solar energy not only heightened the PV efficiency of InGaN/GaN but also elevated the reactant solution temperature to 70 °C, thereby further enhancing the thermodynamics shift to the forward reaction of the water splitting. The utilization of a concentrating lens in Fig. 6a is frequently employed in the investigation of self-made photo-electrochemical devices. This approach offers notable convenience in terms of concentration ratio adjustments and cost-effectiveness. Additionally, the compatibility of the concentrating lens with existing photo-electrochemical reactors is good, facilitating the integration process. Consequently, all the electrochemical testing methods can be executed with utmost ease and precision. The utmost achieved STH efficiency, obtained through the utilization of PV + EC or PEC for driving the water-splitting reaction, registers at approximately 18%-19% without light concentration [45]. Nonetheless, the application of a straightforward solar energy concentration strategy, as depicted in Fig. 6a, has the potential to elevate the STH efficiency to a remarkable 30% [27].

In addition to the laboratory experiments, the kilowatt-scale solar hydrogen production system shown in Fig. 6b and c utilizing a integrated concentrating solar energy technology and photoelectrochemical (CSE + PEC) device has demonstrated an average STH device-level efficiency of 20.3%, along with a system-level fuel efficiency of 5.5% [28]. A validated theoretical model indicates that enhancements in system-level performance exceeding 16% could be achieved through the optimization of thermally integrated devices, the expansion of PV module area, and the refinement of homogenizer optics. Notably, the observed hydrogen production rate (>2.0 kW) at a solar concentration level of 800 suns underscores the promising potential for the practical implementation of CSE + PEC technology in commercial applications.

The data collected from PEC-driven hydrogen production [46], as presented in Fig. 7, distinctly illustrates that the utilization of CSE has the potential to enhance STH efficiencies and electrochemical current densities. Owing to their capacity to withstand high irradiation flux, III-V semiconductor materials emerge as the most promising candidates for the advancement of CSE + PEC technology. While the CSE + PEC method exhibits a notable level of efficiency, its practical application remains hampered by the considerable expenses associated with manufacturing and the persistent stability challenges stemming from the corrosion susceptibility of III-V semiconductor materials. The documented stability durations, as reported [47], span the range of 1 h–2 months, a duration that falls considerably short of the stringent operational demands placed upon industrial-grade equipment.

To analyze the potential effect of utilizing CSE on the techno-economy of renewable hydrogen production, the levelized cost of hydrogen (LCOH) [13,14] defined in formula (1) is very important and intuitive.

... (1)

where It is the annual investment, Mt is the annual costs for maintenance and operation, Ft is the annual fuel costs, and Ht is the amount of hydrogen production (kg). The last two critical factors in the equation are the project discount rate (r) and the lifetime of the system (n).

There have been a lot of notable studies to assess the techno-economy of solar hydrogen production in comparison with existing technologies of hydrogen production [1214,36,48]. The LCOH of PC, PEC, and PV-EC are listed in Fig. 8 and Table 3. The technical and economic feasibility of photoelectrochemical technology for renewable hydrogen production is evidently the PC route. However, to achieve the target LCOH (1–2 $/kg), the energy conversion efficiency of PC needs to reach 5–10%. Currently, the highest efficiency of PC is only 1–2% [49,50], with most research results showing energy conversion efficiency even below 0.1% [24,51,52]. As a result, while PC methods have significant economic potential, they have not yet been widely accepted as a primary commercial development route. Conventional PEC and PV-EC routes can achieve STH efficiencies of 10–15%. However, they are constrained by the high cost of manufacturing III-V semiconductor materials, resulting in the LCOH (10–15 $/kg) far exceeding the current costs of hydrogen production obtained from fossil fuels (1–2 $/kg). Although the option of producing electricity from PV power plants and then transmitting it to electrochemical hydrogen production facilities via the grid can effectively reduce hydrogen production costs (approximately 5 $/kg) [12,13], it still cannot compete with fossil fuel-based routes. According to the results and analysis in Figs. 7 and 8, the scheme based on CSE combined with PEC can first effectively improve STH energy conversion efficiency. Also, it can significantly reduce the usage of III-V semiconductor materials, demonstrating the potential for large-scale implementation of renewable energy hydrogen production.

In summary, CSE technology indeed fosters the advancement of PEC or PV-EC systems by augmenting overall system efficiency [19,27,29,31,35,53] and mitigating construction expenses [3,4,12–14,36,54]. However, the primary impediments to achieving large-scale photoelectrochemical solar fuel production lie in the inherent challenges related to stability and fabrication costs. Regarding the PC route, while there have been research efforts focused on the utilization of CSE [55–58], the inherent limitation of low efficiency poses a significant hurdle to its viable industrial application.

3. Solar fuel through thermochemistry

3.1. Overview

The utilization of thermochemical processes may be the most straightforward technology within the field of CSE for the production of solar fuels, a concept that has undergone extensive exploration since the 1970s [59,60]. These thermochemical fuel production methods encompass two distinct approaches, as illustrated in Fig. 9a and b, denoting the TwoStep method [61] and the Multistep method [62]. As shown in Fig. 9a, the mechanism of the Two-Step thermochemical dissociation of CO2 and H2O operates akin to a heat engine. Initially, the catalyst is heated to a high temperature, resulting in its reduction and the liberation of O2. Subsequently, the catalyst is cooled to a lower temperature and exposed to CO2 or H2O, facilitating the oxidation of the catalyst and the release of CO or H2. In the case of the Multistep method, it operates under the same fundamental mechanism, but with more intricate procedures, exemplified by the Mn-based reaction steps as depicted in Fig. 9b.

The primary distinction between the Two-Step method and the Multistep method lies in the lower temperature requirement of the latter. Following the thermodynamic analysis conducted by Schreiner [59], it has been deduced that a minimum of three reaction steps is required to drive H2O dissociation within the temperature range of 298–1000 K, as inferred from the necessary entropy change. Numerous Multistep methods have been proposed [7,63], such as the S–I cycle, Fe–Cl cycle, and Cu–Cl cycle. However, a significant challenge persists, as all of these methods yield corrosive intermediates. Despite possessing substantial research experience, the practical industrial application of these methods remains unattained. As for the Two-Step method, the catalyst materials most commonly employed are metal oxides, including CeO2 [64,65], ZnO [66], SnO [67], and Fe3O4 [68]. These materials necessitate operating at high temperatures, typically exceeding 1500 K. Here is the typical reaction process of the Two-step method [69]:

... (2)

where, * denotes the non-stoichiometry – the measure of the redox extent. In principle, the redox cycle can be operated under a temperature-swing mode and/or a pressure-swing mode to control the oxygen exchange capacity of ceria ** = *red **ox, and thereby the fuel yield per cycle.

A central challenge in the domain of solar energy-based thermochemical fuels lies in the conversion of CO2 and H2O. Consequently, research efforts over the past five decades have been primarily concentrated on the efficient dissociation of these stable molecules. Notable research milestones in this area are chronologically cataloged in Fig. 9c [59,66,70–77]. Over the past five decades, the interest in thermochemical cycles for solar fuel production has undergone variations. Initially, the primary emphasis was on efficiently driving water splitting utilizing nuclear energy. Towards the end of the 1990s, this focus transitioned from nuclear to solar heat, and since 2010, there has been a growing emphasis on CO2 splitting as it presents the potential for liquid fuel production. Nevertheless, out of the approximately 3000 possible thermochemical cycles, only a very limited few have been identified as relevant for large-scale production of solar fuels [78]. Current research predominantly centers on the development of new materials aimed at enhancing reaction rates [79–84] and the exploration of pilot-scale demonstrations [69,76,77,85], especially on the Two-Step method.

3.2. Theoretical foundation behind CSE

In the early stages of research concerning thermochemical fuels, foundational principles of thermodynamic analysis and design were established, based on the contributions of Schreiner [59] and Fletcher [60]. Within the framework of reversible thermodynamic cycles, Schreiner [59] determined the essential number of steps necessary for the conversion of H2O into H2, categorizing the predominant methodologies into two distinct categories: Two-Step and Multistep approaches. To optimize the conversion of concentrated solar energy into solar fuels, Fletcher [60] undertook a comprehensive analysis accounting for the energy losses linked to optical deficiencies in the solar energy collector. This analysis led to the derivation of a formula [7] delineating the interplay between ideal efficiency, the maximum attainable operating temperature, and the concentration ratio.

... (3)

where ns, represents the ideal efficiency, I is the intensity of solar radiation, TH is the operating temperature of the receiver, where TL is the temperature of the cold thermal reservoir, C is the concentration ratio of the solar concentrating system, as and * are the effective absorptance for solar radiation and the emittance of the receiver, respectively, and * is the Stefan–Boltzmann constant.

In the context of the Two-Step method, we illustrate a straightforward scheme for the conversion of CO2 and H2O into solar fuels, as depicted in Fig. 10a [86]. Taking the TwoStep method of dissociating H2O as an example, Fig. 10b displays collected data [63] regarding operating temperature and solar-to-hydrogen conversion efficiency, with the solid line representing the relationship defined in formula (3). It is evident that the STF efficiency in thermochemical approaches falls within the range of 10%–80%, a notable improvement over photo-electrochemical methods. Furthermore, this efficiency can be significantly enhanced through flexible adjustments to operating temperature and concentration ratios.

Despite the high thermodynamic efficiency, the production of thermochemical fuels is impeded by the slow reaction rate in the current research, primarily attributed to the sluggish oxygen vacancy formation within the metal oxide in the Two-Step method. The requisite temperatures for thermal reduction (TR) of catalysts and gas-splitting (GS) of CO2 and H2O are depicted in Fig. 10c. Meredig and Wolverton [79] have proposed criteria for evaluating the suitability of materials within the thermochemical fuel production cycle, as illustrated in Fig. 10d. Furthermore, they have compiled data of 105 oxides, revealing that the current oxides do not align with the optimal regimes depicted by the shaded area. Hence, one of the central themes in thermochemical fuel production entails the quest for optimal metal oxide catalysts [80,82,87,88]. The perovskite-type oxides represent the prevailing avenue of investigation, owing to their abundant material diversity and ease of doping, exemplified by compounds such as (La, Sr)(Mn, Al)O3, Ba(Ce, Mn)O3, and (Ca, Ce)(Ti, Mn)O3, as depicted in Fig. 10e [81].

3.3. Pilot-scale demonstration

As depicted in Fig. 10, thermochemical fuel production necessitates operating temperatures ranging from 1500 to 2000 K and a concentration ratio exceeding 500. Based on the characteristics of CSE technology outlined in Table 1, it becomes evident that dish concentrator and central receiver types of CSE devices readily lend themselves to integration with thermochemical reactors, as well as the compatibility assessment between CSE and solar fuel production presented in Table 2.

In the last two decades, demonstrative implementations of CSE for thermochemical fuel production have witnessed substantial advancements, mainly in the field of Two-Step methods. These developments encompass reactor design improvements [18], the increase of concentration ratio in dish concentrator devices [69,89], and successful integration with central receiver technology [77]. Steinfeld and his research group have introduced a practical kilowatt-scale reactor design for integration with dish concentrator technology [17,18,76,85], employing CeO2 as a catalyst, as illustrated in Fig. 11a and b [18]. Experimental results indicate a solar-to-fuel efficiency of 5.25% under a CR of 3000, employing a temperature swing cycle ranging from 1500 °C to 700 °C. The predominant energy loss occurs during the heating phase during the transition from oxidation at Toxidation to reduction at Treductions causing a more than 50% loss of the input thermal energy. This fraction can be significantly mitigated through heat recovery via thermal energy storage. Retrieving only half of this energy could potentially elevate the solar-to-fuel efficiency to values exceeding 20% [18]. This underscores the significance of recuperating sensible heat rejected during temperature-swing cycling, particularly in industrial applications. Furthermore, the successful integration of the reactor with dish concentrator devices has been realized, displaying significant potential for solar fuel production, as depicted in Fig. 11c and d. Techno-economic analyses encompassing the entire process chain, akin to the showcased pathway, have estimated a jet fuel cost within the range of €1.2-2 per liter. These cost assessments are primarily contingent upon the energy efficiencies (with assumed values of ηs = 4.4-11.7%) [69].

The pioneering demonstration of integration between thermochemical fuel production and central receiver system was achieved with the support of the EU Horizon 2020 project SUN-to-LIQUID. Illustrated in Fig. 11e and f [77], a 50-kW reactor, incorporating a reticulated porous structure, was subjected to an average solar flux concentration of 2500 suns. Notably, a solar-to-syngas energy conversion efficiency of 4.1% was attained without the implementation of heat recovery.

In summary, the pilot demonstration of solar energy-driven thermochemical fuel production has showcased its readiness for large-scale application. Looking ahead, the primary challenge lies in devising effective heat recovery systems and thermal insulation techniques to enhance STF efficiency further.

4. Solar fuel through photo-thermal co-catalysis

4.1. Overview

In the realm of solar fuel production, three distinct solar energy response mechanisms have been identified [90]. These mechanisms encompass electron-hole excitation within semiconductors (Fig. 12a), direct conversion of photons into phonons to drive thermal catalysis (Fig. 12b), and the activation of plasmons in metal particles, clusters, and heavily doped semiconductor materials (Fig. 12c). Upon irradiation, the material's response can be analyzed as an interaction involving photons, phonons, and electrons. This interplay induces two distinct pathways for driving solar fuel production, namely the photoelectric and photothermal processes, as depicted in Fig. 12d. It is important to note that both of these pathways are intrinsic to various conversion processes. However, the determination of whether the photothermal or photoelectric effect predominates relies on the unique characteristics of the catalyst and the specific chemical reactions under consideration. For instance, Taking the CO and O2 reaction on the surface of Ru as an example, Ertl et al. [91] have previously demonstrated that irradiation induces the generation of hot electrons and thermal energy, both of which can activate adsorbates but lead to different outcomes. When light irradiates the surface of Ru, the electrons undergo a rapid excitation, resulting in the production of high-energy hot electrons. These hot electrons subsequently interact with the adsorbate on the Ru surface, inducing the production of CO2 within approximately 0.5 ps. Concurrently, the hot electrons interact with the phonons of Ru, leading to a temperature increase and the production of CO at around 4 ps (Fig. 12e). Calculations of transition states reveal that, in order to activate the adsorbed O atom, a greater amount of thermal energy is required compared to the desorption energy of CO. Consequently, in the presence of thermal energy input, CO is preferentially desorbed from the Ru surface rather than undergoing its own oxidation. However, the introduction of hot electrons enables the activation of adsorbed molecules within femtoseconds, resulting in a chemical reaction before adsorbates desorb from the surface (Fig. 12f). Extensive research involving the synergistic integration of the photothermal and photoelectric effects [92–94] has highlighted the utilization of CSE technology as a pivotal approach to enhancing STF efficiency [53,95–97], extending catalyst lifespan [98], and modulating selectivity [91,99], among other factors.

Before delving into the specifics of photo-thermal cocatalysis, it is necessary to categorize different forms of photothermal co-catalytic mechanisms to ensure logical clarity. Although in the actual reaction process, the photoelectric and photothermal responses of catalyst materials will more or less coexist, based on the different reactants and products, these mechanisms can be classified according to reaction types [94], as illustrated in Fig. 13a:

(1) Thermo-assisted photocatalysis: Photocatalysis is typically employed to drive reactions with positive Gibbs free energy changes (*G > 0) because the thermodynamic reaction barriers for such reactions are usually high. Even if a thermodynamic reaction can occur, the reverse reaction tends to dominate due to lower reaction barriers, rendering thermal energy ineffective in driving such reactions. With additional photon energy input, the net Gibbs free energy change can turn negative (*G < 0), making photocatalysis dominant in such reactions. In this case, the thermal energy primarily serves to activate reactants, enhancing the reaction rate, which could be classified as photochemistrydriven photothermal catalysis (P-PTC).

(2) Photo-assisted thermocatalysis: Thermal catalysis is typically used to drive reactions with negative Gibbs free energy changes (*G < 0). Photons play a more significant role in driving chemical reactions through the photothermal effect. Although some photons may directly drive molecular reactions through photocatalysis, the thermodynamically favorable forward reactions usually dominate, making thermal catalysis the primary reaction type. It is important to note that photo-assisted thermal catalysis, as discussed here, differs significantly from pure thermal catalysis described in Section 3. The key distinction lies in the fact that photons, in addition to generating thermal energy through the photothermal effect, can also facilitate the activation of chemical bonds [100,101] and reduce catalyst surface carbon deposition [102], achieving effects that pure thermal catalysis cannot. This case could be classified as thermochemistry-driven photothermal catalysis (T-PTC).

(3) Synergistic photothermocatalysis: The above two classifications differentiate between photocatalysis and thermal catalysis dominating based on reaction types. In certain scenarios, where both photocatalysis and thermal catalysis pathways may coexist and contribute equally, such as plasmon-driven dry reforming of methane [103], making it impossible to distinguish the dominant pathway, these reactions fall under parallel photo-thermal synergistic catalysis. Another example is the artificial photosynthesis reaction of CO2 + H2O [104] shown in Fig. 13b. Generally, the Gibbs free energy change for the water decomposition step is positive (*G > 0), while for CO2 reduction, *G < 0. These reactions can be carried out sequentially by employing photocatalysis and thermal catalysis in tandem. In this case, photocatalysis and thermal catalysis occur in sequence, constituting a series of photo-thermal synergistic catalysis. These cases could be classified as synergetic photochemistry-thermochemistrydriven catalysis (S-PTC).

The energy diagram of the artificial photosynthesis process offers an intuitive means of determining whether a reaction is primarily driven by photocatalysis or thermocatalysis. Given that water splitting is an uphill reaction and CO2 reduction is a downhill reaction, if the reactant includes H2O, the reaction is typically dominated by photocatalysis. Conversely, when there is no H2O in the reactant, the reaction often leans towards being dominated by thermocatalysis. The above statement can be validated by the common STF reaction types and their enthalpy change measurements listed in Table 4.

4.2. Thermo-assisted photocatalysis

Conventional semiconductor photocatalysts predominantly exploit the ultraviolet-visible segment of the solar spectrum, thereby neglecting the considerable infrared wavelength range. Recognizing the potential for performance enhancement, researchers have acknowledged the incorporation of thermal energy as a strategy to augment photocatalytic activity. This can be achieved by utilizing CSE technology to either vaporize the reaction liquid into steam [93,95–97] or convert the heat into electrical energy using thermoelectric materials [53].

Wei et al. [95] introduced a thermo-assisted two-phase photocatalytic system (Fig. 14a). This system utilizes carbonized wood as a thermal-absorbing substrate coated with CoO nanoparticles. The resulting wood/CoO thermo-assisted photocatalytic system employed the carbonized wood substrate to absorb heat, converting liquid water into gaseous water vapor under light conditions. Subsequently, the water vapor goes into contact with the photocatalyst CoO, reducing the barrier for the direct decomposition of hydrogen in water, thus achieving efficient hydrogen production, inducing a 17fold increase compared to the conventional three-phase CoO nanoparticles photocatalytic system (Fig. 14b). Furthermore, by integrating with CSE technology, thermo-assisted photocatalytic could achieve a superlinear increase of hydrogen production rate with the increase in solar intensity, which could be attributed to the enhancement of both the photocatalytic process and the surface temperature of the woodcatalyst interface with increasing light intensity (Fig. 14c).

4.3. Photo-assisted thermocatalysis

The primary condition for utilizing solar energy as a heat source is that the catalyst or the substrate of the supported catalyst exhibits a strong photothermal response, enabling the conversion of a significant portion of photon energy into thermal energy to drive thermochemical reactions. One of the most typical photo-driven thermochemical reaction types is the hydrogenation of CO2, with major products including CO, CH4, and CH3OH, etc. For instance, in Fig. 15a, carbon is used as an absorber of sunlight and heat, with Co particles modified by K ions loaded as the reaction catalyst [100]. Based on the activation energy measurements in Fig. 15b, it is evident that the photo-driven thermochemical reaction exhibits activation energy close to that of conventional thermally driven reactions. This suggests that in this reaction process, the predominant use of photons is for generating thermal energy, and the impact of the photoelectric effect is minimal.

In the same CO2 hydrogenation reaction, the utilization ratio of the photoelectric effect may be increased by employing other types of catalyst materials. For example, in Fig. 15c, CoO and Ir nanoparticles loaded on Al2O3 demonstrate enhanced utilization of the photoelectric effect [102]. Under photon excitation, the nanometal particles generate a strong electromagnetic field on the surface, assisting in the activation of adsorbed molecules. As shown in Fig. 15d, under different conditions, the yield results indicate that sole illumination is insufficient to drive the reaction, suggesting that the electromagnetic field enhancement effect cannot independently drive the chemical reaction. However, under photoassisted heating conditions, the yield exceeds the sum of the yields under both photo-driven and thermally driven conditions, indicating that the electromagnetic field enhancement resulting from light excitation effectively increases the rate of thermochemical reactions.

4.4. Synergetic photothermocatalysis

Thermocatalysis serves as the cornerstone of chemical engineering; however, the significant challenges in terms of high energy consumption and demanding reaction conditions have long hindered the decarbonization efforts within the chemical industry. Given the high selectivity exhibited by photoelectric responses and their potential to reduce the required reaction temperatures in comparison to conventional thermal catalysis, the integration of CSE technology becomes a logical choice for facilitating the retrofitting of traditional thermal catalytic processes. As an illustration, the light-driven dry reforming of methane process depicted in Fig. 16a can manifest two distinct reaction pathways. When subjected to a high light intensity, the reaction not only attains heightened efficiency (Fig. 16b) but also demonstrates enhanced selectivity towards the desired hydrogen product (Fig. 16c). Conversely, when subjected to temperature variations within the range of 650–1000 K, the selectivity exhibits oscillations around a fixed ratio (Fig. 16d) [103]. Except the aforementioned example, it is also possible to employ CSE technology for mitigating the activation energy barrier [105] and mitigating carbon deposition on the catalyst surface during CO2 conversion [98].

Furthermore, the application of synergistic photothermocatalysis can also involve utilizing photochemistry and thermochemistry to drive elementary reactions in different reaction steps. The most typical case is to enhance the pure thermochemical dissociation of CO₂ or H₂O introduced in Section 3. Due to the high temperatures exceeding 1500 °C required for pure thermochemical dissociation of small molecules, significant heat loss poses a challenge to the cooling and protection of reactors and related equipment. By substituting photochemistry for the first step of pure thermochemistry, i.e., the oxygen release process of oxide catalysts, the reaction temperature can be significantly reduced, thereby reducing equipment investment costs. As illustrated in the mechanism diagram in Fig. 17a, during the initial oxygen release process of oxide catalysts stimulated by light, the generated electron-hole pairs act on metal-oxygen chemical bonds, leading to the release of oxygen atoms from the catalyst surface. Subsequently, water molecules are adsorbed, and under the drive of thermal energy, low-temperature dissociation of water molecules is achieved. Experimental results shown in Fig. 17b and e validate the feasibility of the aforementioned approach and strategy.

In summary, the integration of CSE technology and photo-thermal co-catalysis offers novel control and design approaches to enhance the efficiency of solar fuel production. This applies not only to emerging photoelectric catalysis pathways but also to the retrofitting and decarbonization of conventional thermal catalysis processes.

4.5. Typical devices

Considering the various reaction types involved in photothermal catalysis, such as the CO2 hydrogenation reaction, don't require too high temperature. Additionally, since most photo-thermal synergy reactions are gas-phase reactions, the reactors are more compact. Therefore, photo-thermal synergy exhibits good technical compatibility with various CSE technologies, as presented in Table 2. The concept of photo-thermal co-catalysis is still in its early stages of development. Current research primarily focuses on understanding the underlying mechanisms and material selection. Many of the devices employed for photo-thermal co-catalysis are custommade, as illustrated in Fig. 18a [99] and b [106]. Similar to the approach demonstrated in Fig. 6a for photo-electrochemistry, the utilization of Fresnel lenses offers a straightforward and efficient strategy. When considering potential applications of photo-thermal co-catalysis, such as dry reforming of methane, and artificial photosynthesis involving CO2 and H2O, the design of a CSE-driven chemical reactor presents no significant challenge. The primary concern lies in establishing the technical and economic advantages of photo-thermal cocatalysis over traditional photocatalysis and thermocatalysis methods.

5. Methods of studying CSE

The influence of CSE encompasses both photoelectric and photothermal responses. Accurately calculating and measuring these two effects is imperative for the precise prediction and design of CSE-assisted solar fuel production. As depicted in Fig. 12a, b, and c, there are primarily three distinct categories of solar energy irradiation effects on materials: the photoelectric response of semiconductors, the thermal effects by interaction with phonons across various material types, and plasmon excitation within metal particles or clusters. It is noteworthy that plasmon excitation typically entails both photoelectric and photothermal effects. Furthermore, the photoelectric response of semiconductors may transform into a thermal effect through charge carrier recombination.

5.1. Photoelectric response

The measurement of photoelectric response can be achieved through the utilization of standard instruments, including the IV curve measurement of PV materials, photoluminescence spectroscopy, electrochemical impedance spectroscopy, fluorescence spectroscopy, laser-induced current measurement, and others. These methodologies are highly suitable for investigating the photoelectric response of semiconductors and facilitate the acquisition of characteristic responses to irradiation intensity by adjusting the external lighting source. While several methods are available to measure the photoelectric response, a majority of them stem from research in PV and PEC, predominantly offering macroscopic features. Only a select few can provide insights into the in-situ photoelectric response effects on molecular reactions. As an illustrative example of plasmon-induced molecule dissociation, Fig. 19 demonstrates the investigation of the chemical reaction using both theoretical calculations [107] (Fig. 19a and b) and in-situ scanning tunneling microscopy (STM) [108] (Fig. 19c, d and e).

Considering the adsorption of a CO2 molecule onto an Ag20 cluster, the dissociation of CO2 could be induced by a laser with a fixed photon energy of 4.56 eV. Fig. 19a and b depict the maximum charge transfer to CO2 and the rate of CO2 reduction as a function of laser intensity, as determined through time-dependent density functional theory (TDDFT) simulations [107]. Notably, the results illustrate that the charge transfer and reaction rate do not exhibit a linear increase with increasing light intensity. It is noted that the nonlinear dependence can be divided into three distinct regimes. For the charge transfer in the relatively low laser intensity regime (I), a linear correlation is observed; then, a plateau region (II) exists; and finally, a new linear dependence is obtained within the range (III). The same method could also apply to other reactions [109,110]. Another commonly employed approach for analyzing the dissociation of photoexcited molecules is through ab initio nonadiabatic molecular dynamics (NAMD) simulations. For instance, Chu et al. [111] conducted a study on the photo-thermal co-catalytic dissociation of CO2 on TiO2 and proposed a classification scheme encompassing three scenarios for CO2 dissociation.

To directly measure the photoelectric molecule reaction, it is feasible to integrate a STM and a nanoreactor between a metal nanotip and a metal substrate, as depicted in Fig. 19c [108]. Upon laser excitation, the molecules distributed across distinct regions can be dissociated. Furthermore, the timedependent dissociation ratio (Fig. 19d) and rate constant (Fig. 19e) can be precisely determined. This technique provides a strong tool for how variations in light intensity and plasmonic fields influence charge transfer and reaction rates.

In summary, the analysis of photoelectric responses within CSE technology can be carried out through various measurement methods, even at the atomic scale. However, the practical application of this knowledge remains in its nascent stages, primarily confined to III-V semiconductor materials-based photovoltaic and electrochemical (PV + EC or PEC) systems. In the future, the importance lies in efficiently designing CSEassisted solar fuel production based on accurate photoelectric response data.

5.2. Photothermal effect measurement

In addition to photoelectric response, the photothermal effect is a more common phenomenon and accounts for a larger proportion of energy capture. Due to the substantial progress in CSP plant development, the photothermal characteristics of various materials have been extensively examined and analyzed. However, these investigations have primarily centered on a macro scale (>1 cm), which adequately fulfills the heat energy storage or solar thermal power plant requirements. In contrast, for the thermochemical or photothermal catalysis pathway of solar fuel production, where chemical reactions occur at the microscale range of 1 nm to 1 *m, precise control of microscale temperature distribution becomes critically important. This is particularly relevant in the context of CSE-assisted pathways.

Fig. 20a compares the spatial and thermal resolutions of different technologies for microscale temperature measurement [112]. Most of these technologies exhibit a thermal resolution of 0.1 K because they are predominantly based on spectroscopic measurement, thereby constraining the achievable thermal resolution. In the case of Infrared laser interferometric thermometry [113], thermal resolution can reach values as low as 10-4 K. However, a resolution of 0.1 K is typically deemed sufficient for chemical catalysis research. This is primarily because the difference in molecular vibrational energy levels is approximately 1 meV, which corresponds to about 10 K. Consequently, the thermal resolution provided by the current mainstream techniques is notably adequate for such applications.

In comparison to thermal resolution, spatial resolution assumes greater significance, necessitating the implementation of more sophisticated techniques. While conventional thermocouples [112] and infrared images (see Fig. 20b) [105] can solely ascertain temperature within the confines of 0.1 mm, current research has demonstrated that within the range of 5100 nm, temperature differentials can span from 57 to 137 K [114]. Furthermore, various reaction mechanisms may manifest at distinct regimes within this spatial range. Consequently, it becomes evident that the pivotal determinant for quantifying the photothermal effect in solar fuel production lies in achieving a high spatial resolution, from nanometer to atomic scale.

In the pursuit of achieving sub-micrometric resolution, quantum dots (QD) [115,116] and lanthanide-based luminescent nanothermometers [117] emerge as valuable tools, representing techniques founded upon luminescent nanoparticles. In this context, the spatial resolution is inherently constrained by the characteristics of the experimental apparatus, typically manifesting as a microscope, thereby subjecting it to diffraction limitations. Consequently, the exact spatial resolution hinges upon the wavelength of the incident light, resulting in augmented resolution for near-IR probes. Various types of materials and sizes presented in Fig. 21a–c and d-e provide insights into the anticipated average spatial resolution achievable with QD [118] and lanthanide-based nanoparticles [119] falling within the range of 100 nm–1000 nm.

In order to achieve a higher spatial resolution for photothermal effect measurements, the scanning thermal microscope (SThM) (see Fig. 22a) provides a spatial resolution of less than 100 nm [120]. However, because SThM necessitates contact-based measurements (see Fig. 22b), it could impact the incident solar irradiation reaching the samples, potentially affecting the obtained results.

The non-uniform distribution of temperature yields a direct consequence: the excitation of adsorbed molecule vibrations may vary. Therefore, the fundamental approach for assessing photothermal effects lies in monitoring the vibrational dynamics and chemical bond break of adsorbed molecules. In this context, scanning tunneling microscopy (STM), non-contact atomic force microscopy (AFM), and tip-enhanced Raman scattering (TERS) offer spatial resolutions ranging from 5 nm to atomic scales [121,122]. Fig. 22c illustrates the schematic diagram of the scanning Raman picoscopy (SRP) technique. A nanocavity, delineated by the silver tip and substrate, creates a potent and tightly confined plasmonic field. This field is employed to excite and enhance the Raman signals (Fig. 22d) emanating from individual molecules.

In summary, while thermal effect measurements from other research domains [123–125] can offer a thorough analysis and accurate results, it is essential to pay attention to the intricate chemical reaction conditions and building in-situ instrumentation, especially in atomic-scale scenarios, which has become important with the rapid advancements in singleatom catalysts [126–129].

6. Special methods of CSE

In addition to the conventional CSE technology employed in CSP plants, as illustrated in Fig. 1b, numerous distinctive techniques for CSE dedicated to solar fuel generation and chemical reactions have emerged. These methods encompass the utilization of evanescent waves along the external surface of optical fibers (see Fig. 23a and b) [130], the integration of micro condenser lenses at the interface of biphasic emulsion droplets (see Fig. 23c) [131], and the augmentation of a particular optical mode within an optical microcavity (see Fig. 23d) [132]. These approaches exhibit intriguing prospects with respect to the concentration of irradiation energy and afford heightened control over molecular reactions.

Evanescent waves, in the context of optical phenomena involving total internal reflection, refer to the phenomenon wherein the optical wave does not undergo complete reflection back into the first medium at the interface. Instead, it penetrates into the second medium to a depth of approximately one wavelength and travels along the interface for a distance on the order of a wavelength before re-emerging into the first medium, radiating in the direction of the reflected light. This wave, traveling along the surface of the second medium, is commonly referred to as an evanescent wave. As a result, the irradiation intensity within the evanescent wave region significantly exceeds that of the incident light. This phenomenon presents a compelling opportunity for solar energy concentration [133–136]. For instance, in Fig. 23a, a TiO2 photocatalyst is deposited onto the optical fiber, and upon introducing an incident laser into the fiber, it leads to a 32% enhancement in the quantum yield during the degradation of a refractory pollutant (carbamazepine) in water [130].

In Fig. 5a, it is illustrated that a straightforward condenser lens can enhance both the photoelectric reaction rate and efficiency. In the realm of microscale systems, a similar approach has been explored. Zeininger et al. [131] introduced the concept of employing a Janus Emulsion as a solar concentrator within photocatalytic droplet microreactors. This innovation resulted in a substantial enhancement in the performance of a range of homogeneous and heterogeneous photocatalytic reactions, even under diffuse sunlight conditions.

The optical cavity offers flexible control over the internal optical modes. When ensembles of molecules are embedded within appropriate optical cavities, optical/electromagnetic modes strongly couple with delocalized superpositions of molecular transitions, giving rise to hybrid light-matter states (see Fig. 23d) [132]. These polaritons possess the capability to exert an influence on physicochemical properties and a range of thermally or photo-initiated molecular chemical reactions [137].

Due to the fundamental characteristics of polaritons, we can categorize them as a unique CSE technology that exclusively enhances specific optical modes of irradiation or, more simply, specific wavelengths of irradiation. The resulting effects are particularly intriguing, especially when considering the presence of molecules within the cavity; adjusting the optical modes can significantly impact the reaction rate [137,138]. When two-dimensional or three-dimensional photocatalysts, such as C3N4 or TiO2 particles, are introduced into the cavity, the dynamics of charge carriers become influenced. By selecting the appropriate optical mode that collaborates with the electron-hole pairs, the hybrid lightmatter states can significantly extend the lifetimes of hot electrons or holes, consequently elevating the quantum yield efficiency [139].

7. Conclusions and outlook

In Table 5, we have compiled and summarized exemplary studies of three types of reactions: photo-electrochemistry, thermochemistry, and photo-thermal co-catalysis. This includes the catalyst materials used, CSE devices, STF efficiency, and other relevant information. Based on the data presented, it can be observed that the efficiency of photoelectrochemistry is currently quite dispersed. The best efficiency for PC remains at the level of 1–2%. However, efficiencies for PEC and PV-EC can reach 15–30%. Despite this, the high cost of III-V semiconductor materials used in these processes makes them less competitive compared to hydrogen production schemes based on fossil fuels. Nevertheless, when combined with CSE technology, the amount of III-V semiconductor materials can be significantly reduced, leading to cost reduction.

As for pure thermochemistry, despite its high theoretical efficiency, current experimental setups have not considered the heat losses in the recovery and recycling process. As a result, the highest experimental efficiency is currently 5.25%. Despite this, commercialization is progressing rapidly, with commercial demonstrations combining dish concentrator systems and central receiver systems. The key to the next stage of development lies in improving system processes and recovering lost heat, aiming to further improve energy efficiency and reduce costs.

For photo-thermal co-catalysis, if primarily based on thermochemistry or does not include the H2O decomposition step, the efficiency is high, reaching 55%. With the appropriate combination of CSE technology, this may be the first practical STF technology. However, if the reaction involves H2O decomposition or requires photochemistry as a key step, the efficiency is similar to PC, ranging only between 1 and 2%. Therefore, enhancing the energy conversion efficiency of photochemical steps is a key issue when combining CSE technology with photo-thermal co-catalysis. This requires a more detailed investigation into the relationships between factors such as photoelectric response, light intensity, materials, and surface active sites.

In summary, this paper offers a comprehensive review of the prevailing techniques associated with CSE technology within the realm of solar fuel production. This thorough review encompasses CSE-assisted photoelectrochemical, thermochemical, and photo-thermal co-catalytic pathways, as well as the methods for studying the CSE effect and specialized approaches to harnessing CSE. CSE technology not only serves as a technical approach but also signifies an excitation method for investigating reaction systems in response to varying light intensity. This technology offers the dual advantage of achieving high-density energy concentration and enhancing energy quality parameters, including photoelectric response (voltage) and photothermal effect (temperature), with the flexibility to tailor these effects to specific materials. Furthermore, the distinction between photoelectric and photothermal responses governs the reaction pathways and reaction conditions. The utilization of concentrated light also exhibits substantial potential in micro-device functionality and the regulation of chemical reactions within specialized environments. Simultaneously, for a comprehensive understanding of the solar energy concentration responses at the microscopic level, such as photoelectric and thermal effects at atomic resolution, the development of corresponding in-situ response characterization methodologies becomes imperative.

In the context of future prospects, the following recommendations and areas for further investigation can be outlined:

1) Mechanism insights of photoelectrochemical pathway: A more in-depth exploration of the mechanisms is warranted. This entails extensive data validation of optimal light intensities across different materials and systems, accompanied by the development of characterization methods for surface states. Furthermore, a more comprehensive analysis of the impact of light intensity on particle photocatalysis, including distinctions between semiconductorliquid junctions and solid heterojunctions, is essential.

2) It is imperative to acknowledge that the advantageous heat exchange between electrocatalysts and light-absorbing components within PEC devices has been insufficiently explored thus far. This aspect stands as a pivotal advantage of the close integration between light-absorbing and catalytic components within a PEC device. The potential impact of such beneficial heat exchange becomes particularly pronounced under light concentration conditions, promising exciting avenues for further exploration and optimization in the field. By delving into the heat exchange mechanisms, we can unlock enhanced efficiencies and pave the way for the next generation of advanced PEC technologies.

3) Thermochemical solar fuel: Investigate a broader range of low-temperature, non-corrosive cycling processes to alleviate engineering implementation challenges. Additionally, explore a wider range of oxide materials, particularly delving into the exploration of the abundant material repository represented by ABO3 compounds.

4) Precise measurement of photothermal effects and analysis of excited states in hot electrons: This segment involves thorough investigations encompassing performance assessments under varying light intensities and frequencies, in-situ temperature measurements, and atomic-scale optoelectronic property testing.

5) Special methods of CSE: The design and application of microcavities for photocatalysis, the development of microscale devices for solar fuel production, and the elucidation of the precise mechanisms governing the coupling between electromagnetic phonons and molecular reactions.

CRediT authorship contribution statement

Yiwei Fu: Writing – review & editing, Writing – original draft, Investigation, Conceptualization. Yi Wang: Investigation, Data curation. Jie Huang: Visualization. Kejian Lu: Validation. Maochang Liu: Supervision, Conceptualization, Writing – review & editing, Funding acquisition.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgment

We gratefully acknowledge the financial support by the National Key Research and Development Program of China (2022YFB3803600), the National Natural Science Foundation of China (No. 52276212), the Natural Science Foundation of Jiangsu Province (No. BK20231211), the Suzhou Science and Technology Program (SYG202101), the Key Research and Development Program in Shaanxi Province of China (No. 2023YBGY-300), and the China Fundamental Research Funds for the Central Universities.

Sidebar

Received 16 September 2023; revised 27 December 2023; accepted 1 January 2024

Available online 3 January 2024

Footnote

* Corresponding author. International Research Center for Renewable Energy, State Key Laboratory of Multiphase Flow, Xi'an Jiaotong University, Xi'an, Shaanxi, 710049, China.

E-mail address: [email protected] (M. Liu).

1 These authors contributed equally

References

References

[1] G.A. Ozin, Adv. Mater. 27 (2015) 1957–1963.

[2] J. Lv, J. Xie, A.G.A. Mohamed, X. Zhang, Y. Feng, L. Jiao, E. Zhou, D. Yuan, Y. Wang, Nat. Rev. Chem. 7 (2023) 91–105.

[3] Y. Zhao, C. Ding, J. Zhu, W. Qin, X. Tao, F. Fan, R. Li, C. Li, Angew. Chem. Int. Ed. 59 (2020) 9653–9658.

[4] H. Nishiyama, T. Yamada, M. Nakabayashi, Y. Maehara, M. Yamaguchi, Y. Kuromiya, Y. Nagatsuma, H. Tokudome, S. Akiyama, T. Watanabe, R. Narushima, S. Okunaka, N. Shibata, T. Takata, T. Hisatomi, K. Domen, Nature 598 (2021) 304–307.

[5] J. Li, J. Ren, S. Li, G. Li, M.M.J. Li, R. Li, Y.S. Kang, X. Zou, Y. Luo, B. Liu, Y. Zhao, Green Energy Environ. 9 (2024) 859–876.

[6] K. Wu, M. Cao, Q. Zeng, X. Li, Green Energy Environ. 8 (2023) 383405.

[7] T. Kodama, N. Gokon, Chem. Rev. 107 (2007) 4048–4077.

[8] A. Steinfeld, Sol. Energy 78 (2005) 603–615.

[9] C. Yuan, X. Liu, X. Wang, C. Song, H. Zheng, C. Tian, K. Gao, N. Sun, Z. Jiang, Y. Xuan, Y. Ding, Green Energy Environ. 9 (2024) 1290–1305.

[10] S. Fang, Y.H. Hu, Chem. Soc. Rev. 51 (2022) 3609–3647.

[11] C. Lv, X. Bai, S. Ning, C. Song, Q. Guan, B. Liu, Y. Li, J. Ye, ACS Nano 17 (2023) 1725–1738.

[12] M.R. Shaner, H.A. Atwater, N.S. Lewis, E.W. McFarland, Energy Environ. Sci. 9 (2016) 2354–2371.

[13] R.J. Detz, J.N.H. Reek, B.C.C. van der Zwaan, Energy Environ. Sci. 11 (2018) 1653–1669.

[14] Z. Li, S. Fang, H. Sun, R.J. Chung, X. Fang, J.H. He, Adv. Energy Mater. 13 (2023) 2203019.

[15] L.A. Weinstein, J. Loomis, B. Bhatia, D.M. Bierman, E.N. Wang, G. Chen, Chem. Rev. 115 (2015) 12797–12838.

[16] J.A. Herron, J. Kim, A.A. Upadhye, G.W. Huber, C.T. Maravelias, Energy Environ. Sci. 8 (2015) 126–157.

[17] M. Romero, A. Steinfeld, Energy Environ. Sci. 5 (2012) 9234–9245.

[18] D. Marxer, P. Furler, M. Takacs, A. Steinfeld, Energy Environ. Sci. 10 (2017) 1142–1149.

[19] P. Zhou, I.A. Navid, Y. Ma, Y. Xiao, P. Wang, Z. Ye, B. Zhou, K. Sun, Z. Mi, Nature 613 (2023) 66–70.

[20] X. Chen, S. Shen, L. Guo, S.S. Mao, Chem. Rev. 110 (2010) 65036570.

[21] S. Nishioka, F.E. Osterloh, X. Wang, T.E. Mallouk, K. Maeda, Nat. Rev. Methods Primers 3 (2023) 42.

[22] D. Jing, H. Liu, X. Zhang, L. Zhao, L. Guo, Energy Convers. Manag. 50 (2009) 2919–2926.

[23] K. Villa, X. Domenech, S. Malato, M.I. Maldonado, J. Peral, Int. J. Hydrogen Energy 38 (2013) 12718–12724.

[24] J.H. Kim, D. Hansora, P. Sharma, J.W. Jang, J.S. Lee, Chem. Soc. Rev. 48 (2019) 1908–1971.

[25] S.A. Bonke, M. Wiechen, D.R. Macfarlane, L. Spiccia, Energy Environ. Sci. 8 (2015) 2791–2796.

[26] K. Fujii, S. Nakamura, K. Watanabe, MRS Online Proc. Libr. 1491 (2012) 52–57.

[27] J. Jia, L.C. Seitz, J.D. Benck, Y. Huo, Y. Chen, J.W.D. Ng, T. Bilir, J.S. Harris, T.F. Jaramillo, Nat. Commun. 7 (2016) 13237.

[28] I. Holmes-Gentle, S. Tembhurne, C. Suter, S. Haussener, Nat. Energy 8 (2023) 586–596.

[29] M.A. Khan, I. Al-Shankiti, A. Ziani, N. Wehbe, H. Idriss, Angew. Chem. Int. Ed. 59 (2020) 14802–14808.

[30] A. Ziani, I. Al-Shankiti, M.A. Khan, H. Idriss, Energy Fuels 34 (2020) 13179–13185.

[31] E. Boutin, M. Patel, E. Kecsenovity, S. Suter, C. Janaky, S. Haussener, Adv. Energy Mater. 12 (2022) 2200585.

[32] S.C. Warren, K. Voïtchovsky, H. Dotan, C.M. Leroy, M. Cornuz, F. Stellacci, C. Hebert, Nat. Mater. 12 (2013) 842–849.

[33] T.W. Kim, K.-S. Choi, Science 343 (2014) 990–994.

[34] J. Gao, F. Sahli, C. Liu, D. Ren, X. Guo, J. Werner, Q. Jeangros, S.M. Zakeeruddin, C. Ballif, M. Gr€atzel, J. Luo, Joule 3 (2019) 29302941.

[35] Q. Wang, C. Pornrungroj, S. Linley, E. Reisner, Nat. Energy 7 (2022) 13–24.

[36] B.D. James, G.N. Baum, J. Perez, K.N. Baum, DOE report (2009).

[37] M.A. Green, K. Emery, Y. Hishikawa, W. Warta, Prog. Photovoltaics Res. Appl. 24 (2016) 3–11.

[38] M.V.N.S. Gupta, H. Baig, K.S. Reddy, T.K. Mallick, B. Pesala, A.A. Tahir, ACS Appl. Energy Mater. 3 (2020) 9002–9009.

[39] Y. Yang, S. Niu, D. Han, T. Liu, G. Wang, Y. Li, Adv. Energy Mater. 7 (2017) 1700555.

[40] G. Segev, H. Dotan, K.D. Malviya, A. Kay, M.T. Mayer, M. Gr€atzel, A. Rothschild, Adv. Energy Mater. 6 (2016) 1500817.

[41] A.B. Bocarsly, J.M. Bolts, P.G. Cummins, M.S. Wrighton, Appl. Phys. Lett. 31 (1977) 568–570.

[42] A.J. Bard, L.R. Faulkner, H.S. White, Electrochemical Methods: Fundamentals and Applications, John Wiley & Sons, New York, 2001.

[43] K. Uosaki, H. Kita, J. Electrochem. Soc. 128 (1981) 2153.

[44] M.A. Butler, J. Appl. Phys. 48 (2008) 1914–1920.

[45] W.-H. Cheng, M.H. Richter, M.M. May, J. Ohlmann, D. Lackner, F. Dimroth, T. Hannappel, H.A. Atwater, H. Lewerenz, ACS Energy Lett. 3 (2018) 1795–1800.

[46] S. Tembhurne, F. Nandjou, S. Haussener, Nat. Energy 4 (2019) 399407.

[47] J. Tournet, Y. Lee, S.K. Karuturi, H.H. Tan, C. Jagadish, ACS Energy Lett. 5 (2020) 611–622.

[48] B.A. Pinaud, J.D. Benck, L.C. Seitz, A.J. Forman, Z. Chen, T.G. Deutsch, B.D. James, K.N. Baum, G.N. Baum, S. Ardo, H. Wang, E. Millere, T.F. Jaramillo, Energy Environ. Sci. 6 (2013) 1983–2002.

[49] Q. Wang, T. Hisatomi, Q. Jia, H. Tokudome, M. Zhong, C. Wang, Z. Pan, T. Takata, M. Nakabayashi, N. Shibata, Y. Li, L.D. Sharp, A. Kudo, T. Yamada, K. Domen, Nat. Mater. 15 (2016) 611–615.

[50] J. Liu, Y. Liu, N. Liu, Y. Han, X. Zhang, H. Huang, Y. Lifshitz, S.T. Lee, J. Zhong, Z. Kang, Science 347 (2015) 970–974.

[51] J.W. Ager, M.R. Shaner, K.A. Walczak, I.D. Sharp, S. Ardo, Energy Environ. Sci. 8 (2015) 2811–2824.

[52] L. Liao, Q. Zhang, Z. Su, Z. Zhao, Y. Wang, Y. Li, X. Lu, D. Wei, G. Feng, Q. Yu, X. Cai, J. Zhao, Z. Ren, H. Fang, F.R. Hernandez, S. Baldelli, J. Bao, Nat. Nanotechnol. 9 (2014) 69–73.

[53] C. Pornrungroj, V. Andrei, E. Reisner, J. Am. Chem. Soc. 145 (2023) 13709–13714.

[54] T. Hisatomi, K. Domen, Nat. Catal. 2 (2019) 387–399.

[55] S.M. Rodrıguez, J.B. Galvez, M.M. Rubio, P.F. Iba~nez, D.A. Padilla, M.C. Pereira, J.F. Mendes, J.C. de Oliveira, Sol. Energy 77 (2004) 513524.

[56] Q. Wei, Y. Yang, J. Hou, H. Liu, F. Cao, L. Zhao, Sol. Energy 153 (2017) 215–223.

[57] R. Ma, J. Sun, D.H. Li, J.J. Wei, J. Catal. 392 (2020) 165–174.

[58] A. Vilanova, P. Dias, J. Azevedo, M. Wullenkord, C. Spenke, T. Lopes, A. Mendes, J. Power Sources 454 (2020) 227890.

[59] B.M. Abraham, F. Schreiner, Ind. Eng. Chem. Fundam. 13 (1974) 305310.

[60] E.A. Fletcher, R.L. Moen, Science 197 (1977) 1050–1056.

[61] C.N.R. Rao, S. Dey, Proc. Natl. Acad. Sci. U.S.A. 114 (2017) 1338513393.

[62] B. Xu, Y. Bhawe, M.E. Davis, Proc. Natl. Acad. Sci. U.S.A. 109 (2012) 9260–9264.

[63] O. Oruc, I. Dincer, Fuel 286 (2021) 119325.

[64] S. Abanades, G. Flamant, Sol. Energy 80 (2006) 1611–1623.

[65] J.R. Scheffe, A. Steinfeld, Energy Fuels 26 (2012) 1928–1936.

[66] A. Steinfeld, Int. J. Hydrogen Energy 27 (2002) 611–619.

[67] S. Abanades, P. Charvin, F. Lemont, G. Flamant, Int. J. Hydrogen Energy 33 (2008) 6021–6030.

[68] R.B. Diver, J.E. Miller, M.D. Allendorf, N.P. Siegel, R.E. Hogan, J. Sol. Energy Eng. 130 (2008) 041001.

[69] R. Sch€appi, D. Rutz, F. D€ahler, A. Muroyama, P. Haueter, J. Lilliestam, A. Patt, P. Furler, A. Steinfeld, Nature 601 (2022) 63–68.

[70] D. O'keefe, C. Allen, G. Besenbruch, L. Brown, J. Norman, R. Sharp, K. McCorkle, Int. J. Hydrogen Energy 7 (1982) 381–392.

[71] H. Kaneko, T. Miura, H. Ishihara, S. Taku, T. Yokoyama, H. Nakajima, Y. Tamaura, Energy 32 (2007) 656–663.

[72] A.L. Duigou, J.-M. Borgard, B. Larousse, D. Doizi, R. Allen, B.C. Ewan, G.H. Priestman, R. Elder, R. Devonshire, V. Ramos, G. Cerri, C. Salvini, A. Giovannelli, G. de Maria, C. Corgnale, S. Brutti, M. Roeb, A. Noglik, P.M. Rietbrock, S. Mohr, L. de Oliveira, N. Monnerie, M. Schmitz, C. Sattler, A.O. Martinez, D. de L. Manzano, J.C. Rojas, S. Dechelotte, O. Baudouin, Int. J. Hydrogen Energy 32 (2007) 1516–1529.

[73] M. Roeb, J.P. S€ack, P. Rietbrock, C. Prahl, H. Schreiber, M. Neises, L. de Oliveira, D. Graf, M. Ebert, W. Reinalter, M. Meyer-Gru¨nefeldt, C. Sattler, A. Lopez, A. Vidal, A. Elsberg, P. Stobbe, D. Jones, A. Steele, S. Lorentzou, C. Pagkoura, A. Zygogianni, C. Agrafiotis, A.G. Konstandopoulos, Sol. Energy 85 (2011) 634–644.

[74] T. Kodama, N. Gokon, H.S. Cho, S. Bellan, K. Matsubara, K. Inoue, AIP Conf. Proc. 2033 (2018) 130009.

[75] M. Orfila, M. Linares, R. Molina, J.A. Botas, R. Sanz, J. Marug an, Int. J. Hydrogen Energy 41 (2016) 19329–19338.

[76] W.C. Chueh, C. Falter, M. Abbott, D. Scipio, P. Furler, S.M. Haile, A. Steinfeld, Science 330 (2010) 1797–1801.

[77] S. Zoller, E. Koepf, D. Nizamian, M. Stephan, A. Patane, P. Haueter, M. Romero, J. Gonzalez-Aguilar, D. Lieftink, E. de Wit, S. Brendelberger, A. Sizmann, A. Steinfeld, Joule 6 (2022) 1606–1616.

[78] C. Sattler, J. Sol. Energy Eng. 141 (2019) 020304.

[79] B. Meredig, C. Wolverton, Phys. Rev. B 80 (2009) 245119.

[80] J.M. Naik, C. Ritter, B. Bulfin, A. Steinfeld, R. Erni, G.R. Patzke, Adv. Energy Mater. 11 (2021) 2003532.

[81] R.B. Wexler, G.S. Gautam, E.B. Stechel, E.A. Carter, J. Am. Chem. Soc. 143 (2021) 13212–13227.

[82] R.B. Wexler, E.B. Stechel, E.A. Carter, in: N.D. Sankir, M. Sankir (Eds.), Solar Fuels, Scrivener Publishing LLC, 2023, pp. 1–63.

[83] J.R. Scheffe, A. Steinfeld, Mater. Today 17 (2014) 341–348.

[84] C. Muhich, A. Steinfeld, J. Mater. Chem. A 5 (2017) 15578–15590.

[85] M. Tou, R. Michalsky, A. Steinfeld, Joule 1 (2017) 146–154.

[86] Y. Hao, A. Steinfeld, Sci. Bull. 62 (2017) 1099–1101.

[87] J.E. Miller, M.D. Allendorf, R.B. Diver, L.R. Evans, N.P. Siegel, J.N. Stuecker, J. Mater. Sci. 43 (2008) 4714–4728.

[88] R.J. Carrillo, J.R. Scheffe, Sol. Energy 156 (2017) 3–20.

[89] R.F. Service, Science 326 (2009) 1472–1475.

[90] Z.J. Wang, H. Song, H. Liu, J. Ye, Angew. Chem. Int. Ed. 59 (2020) 8016–8035.

[91] M. Bonn, S. Funk, C. Hess, D.N. Denzler, C. Stampfl, M. Scheffler, M. Wolf, G. Ertl, Science 285 (1999) 1042–1045.

[92] S.W.L. Ng, M. Gao, W. Lu, M. Hong, G.W. Ho, Adv. Funct. Mater. 31 (2021) 2104750.

[93] H. Liu, H.-G. Ye, M. Gao, Q. Li, Z. Liu, A.-Q. Xie, L. Zhu, G.W. Ho, S. Chen, Adv. Sci. 8 (2021) 2101232.

[94] M. Gao, T. Zhang, G.W. Ho, Nano Res. 15 (2022) 9985–10005.

[95] W.H. Lee, C.W. Lee, G.D. Cha, B.-H. Lee, J.H. Jeong, H. Park, J. Heo, M.S. Bootharaju, S.-H. Sunwoo, J.H. Kim, K.H. Ahn, D.-H. Kim, T. Hyeon, Nat. Nanotechnol. 18 (2023) 754–762.

[96] S. Guo, X. Li, J. Li, B. Wei, Nat. Commun. 12 (2021) 1343.

[97] Y. Wang, W. Huang, S. Guo, X. Xin, Y. Zhang, P. Guo, S. Tang, X. Li, Adv. Energy Mater. 11 (2021) 2102452.

[98] Q. Hu, Y. Li, J. Wu, Y. Hu, H. Cao, Y. Yang, Adv. Energy Mater. 13 (2023) 2300071.

[99] Z. Zhang, Z. Gao, H. Liu, S. Abanades, H. Lu, ACS Appl. Energy Mater. 2 (2019) 8376–8380.

[100] H. Wang, S. Fu, B. Shang, S. Jeon, Y. Zhong, N.J. Harmon, C. Choi, E.A. Stach, H. Wang, Angew. Chem. Int. Ed. 62 (2023) e202305251.

[101] Y. Tang, Z. Yang, C. Guo, H. Han, Y. Jiang, Z. Wang, J. Liu, L. Wu, F. Wang, J. Mater. Chem. A 10 (2022) 12157–12167.

[102] Y. Tang, T. Zhao, H. Han, Z. Yang, J. Liu, X. Wen, F. Wang, Adv. Sci. 10 (2023) 2300122.

[103] L. Zhou, J.M.P. Martirez, J. Finzel, C. Zhang, D.F. Swearer, S. Tian, H. Robatjazi, M. Lou, L. Dong, L. Henderson, P. Christopher, E.A. Carter, P. Nordlander, N.J. Halas, Nat. Energy 5 (2020) 61–70.

[104] H. Lin, S. Luo, H. Zhang, J. Ye, Joule 6 (2022) 294–314.

[105] L. Xu, Y. Ren, Y. Fu, M. Liu, F. Zhu, M. Cheng, J. Zhou, W. Chen, K. Wang, N. Wang, N. Li, Chem. Eng. J. 468 (2023) 143831.

[106] S. Bai, W. Jing, G. He, C. Liao, F. Wang, Y. Liu, L. Guo, ACS Nano 17 (2023) 10976–10986.

[107] Y. Zhang, L. Yan, M. Guan, D. Chen, Z. Xu, H. Guo, S. Hu, S. Zhang, X. Liu, Z. Guo, S. Li, S. Meng, Adv. Sci. 9 (2022) 2102978.

[108] E. Kazuma, J. Jung, H. Ueba, M. Trenary, Y. Kim, Science 360 (2018) 521–526.

[109] Y. Zhang, W. Meng, D. Chen, L. Zhang, S. Li, S. Meng, Nano Res. 15 (2022) 3894–3900.

[110] L. Yan, F. Wang, S. Meng, ACS Nano 10 (2016) 5452–5458.

[111] W. Chu, Q. Zheng, O.V. Prezhdo, J. Zhao, J. Am. Chem. Soc. 142 (2020) 3214–3221.

[112] M. Quintanilla, L.M. Liz-Marzan, Nano Today 19 (2018) 126–145.

[113] V.M. Donnelly, J.A. Mccaulley, J. Vac. Sci. Technol., A 8 (1990) 84–92.

[114] C. Mao, H. Li, H. Gu, J. Wang, Y. Zou, G. Qi, J. Xu, F. Deng, W. Shen, J. Li, S. Liu, J. Zhao, L. Zhang, Chem 5 (2019) 2702–2717.

[115] F. Haupt, A. Imamoglu, M. Kroner, Phys. Rev. Appl. 2 (2014) 024001.

[116] F. Seilmeier, M. Hauck, E. Schubert, G.J. Schinner, S.E. Beavan, A. H€ogele, Phys. Rev. Appl. 2 (2014) 024002.

[117] K. Binnemans, Chem. Rev. 109 (2009) 4283–4374.

[118] W.R. Algar, K. Susumu, J.B. Delehanty, I.L. Medintz, Anal. Chem. 83 (2011) 8826–8837.

[119] M. Runowski, P. Wozny, N. Stopikowska, I.R. Martín, V. Lavín, S. Lis, ACS Appl. Mater. Interfaces 12 (2020) 43933–43941.

[120] K. Kim, J. Chung, G. Hwang, O. Kwon, J.S. Lee, ACS Nano 5 (2011) 8700–8709.

[121] Y. Zhang, B. Yang, A. Ghafoor, Y. Zhang, Y.-F. Zhang, R.-P. Wang, J.-L. Yang, Y. Luo, Z.-C. Dong, J.G. Hou, Natl. Sci. Rev. 6 (2019) 1169–1175.

[122] J. Xu, X. Zhu, S. Tan, Y. Zhang, B. Li, Y. Tian, H. Shan, X. Cui, A. Zhao, Z. Dong, J. Yang, Y. Luo, B. Wang, J.G. Hou, Science 371 (2021) 818–822.

[123] D. Zhang, C. Li, C. Zhang, M.N. Slipchenko, G. Eakins, J.-X.Cheng, Sci. Adv. 2 (2016) e1600521.

[124] S. Adhikari, P. Spaeth, A. Kar, M.D. Baaske, S. Khatua, M. Orrit, ACS Nano 14 (2020) 16414–16445.

[125] J. Zhao, A. Matlock, H. Zhu, Z. Song, J. Zhu, B. Wang, F. Chen, Y. Zhan, Z. Chen, Y. Xu, X. Lin, L. Tian, J.-X. Cheng, Nat. Commun. 13 (2022) 7767.

[126] B.-H. Lee, S. Park, M. Kim, A.K. Sinha, S.C. Lee, E. Jung, W.J. Chang, K.-S. Lee, J.H. Kim, S.-P. Cho, H. Kim, K.T. Nam, T. Hyeon, Nat. Mater. 18 (2019) 620–626.

[127] C. Gao, J. Low, R. Long, T. Kong, J. Zhu, Y. Xiong, Chem. Rev. 120 (2020) 12175–12216.

[128] P. Zhou, H. Chen, Y. Chao, Q. Zhang, W. Zhang, F. Lv, L. Gu, Q. Zhao, N. Wang, J. Wang, S. Guo, Nat. Commun. 12 (2021) 4412.

[129] Z.H. Xue, D. Luan, H. Zhang, X.W. (David) Lou, Joule 6 (2022) 92133.

[130] Y. Song, L. Ling, P. Westerhoff, C. Shang, Nat. Commun. 12 (2021) 4101.

[131] P. Simon Marques, B.D. Frank, A. Savateev, L. Zeininger, Adv. Opt. Mater. 9 (2021) 2101139.

[132] J.B. Perez-Sanchez, A. Koner, N.P. Stern, J. Yuen-Zhou, Proc. Natl. Acad. Sci. U.S.A. 120 (2023) e2219223120.

[133] L.W. Miller, M.I. Tejedor-Tejedor, M.A. Anderson, Environ. Sci. Technol. 33 (1999) 2070–2075.

[134] J.J. Madhukeswara, V.V. Petrunin, Chem. Phys. Lett. 445 (2007) 309314.

[135] L. Ling, H. Tugaoen, J. Brame, S. Sinha, C. Li, J. Schoepf, K. Hristovski, J.-H. Kim, C. Shang, P. Westerhoff, Environ. Sci. Technol. 51 (2017) 13319–13326.

[136] Y. Huang, A. Xiao, G. Hou, H. Li, T. Guo, B.-Q. Guan, J. Mater. Chem. A 6 (2018) 20513–20522.

[137] J. Yuen-Zhou, W. Xiong, T. Shegai, J. Chem. Phys. 156 (2022) 030401.

[138] B. Xiang, R.F. Ribeiro, Y. Li, A.D. Dunkelberger, B.B. Simpkins, J. Yuen-Zhou, W. Xiong, Sci. Adv. 5 (2019) eaax5196.

[139] Q. Ou, Y. Shao, Z. Shuai, J. Am. Chem. Soc. 143 (2021) 17786–17792.

[140] Z. Zeng, B. Luo, D. Jing, L. Guo, Sol. Energy 230 (2021) 538–548.

[141] S. Tang, X. Xing, W. Yu, J. Sun, Y. Xuan, L. Wang, Y. Xu, H. Hong, H. Jin, iScience 23 (2020) 101012.

[142] G. Peharz, F. Dimroth, U. Wittstadt, Int. J. Hydrogen Energy 32 (2007) 3248–3252.

[143] F. Sastre, C. Versluis, N. Meulendijks, J. Rodríguez-Fernandez, J. Sweelssen, K. Elen, M.K. van Bael, T.D. Hartog, M.A. Verheijen, P. Buskens, ACS Omega 4 (2019) 7369–7377.

[144] S. Docao, A.R. Koirala, M.G. Kim, I.C. Hwang, M.K. Song, K.B. Yoon, Energy Environ. Sci. 10 (2017) 628–640.

[145] X. Liu, H. Shi, X. Meng, C. Sun, K. Zhang, L. Gao, Y. Ma, Z. Mu, Y. Ling, B. Cheng, Y. Li, Y. Xuan, Y. Ding, Sol. RRL 5 (2021) 2100185.

[146] Y. Liu, F. Wang, Z. Jiao, S. Bai, H. Qiu, L. Guo, Electrochem. Energy Rev. 5 (2022) 5.

[147] R. Han, M.A. Melo Jr., Z. Zhao, Z. Wu, F.E. Osterloh, J. Phys. Chem. C 124 (2020) 9724–9733.