ABSTRACT
Indoor volatile organic compounds (VOCs) remain a major health concern, yet conventional photocatalysts show limited effectiveness under low‐intensity indoor lighting. Here, a spectrally coordinated plasmonic–photothermal–photocatalytic architecture integrating Cu(II)‐grafted TiO2 (CuT), yolk–shell Au@SiO2 (AuSi), and reduced graphene oxide sheets (GS) is presented for efficient mixed‐VOC mineralization under white light‐emitting diode (LED) illumination. Rather than acting as a simple multicomponent composite, this architecture couples partially overlapping but preferential spectral contributions across the LED spectrum: CuT drives visible‐light photocatalytic oxidation, AuSi contributes localized surface‐plasmon‐resonance‐induced electric‐field enhancement near CuT, and GS converts longer‐wavelength photons into localized heat to accelerate interfacial oxidation kinetics. Compared with CuT and binary controls, the composite enhances reactive oxygen species generation, CO2 formation, and humidity tolerance. CuT/AuSi‐GS enables near‐stoichiometric CO2 formation during CH3CHO/C7H8 mixed‐VOC mineralization at 10 ppmv within 1 h and maintains over 80% CO2 yield throughout 168 h of continuous‐flow operation at 2 ppmv. Chamber‐scale validation in a 64‐L reactor indicates the feasibility of VOC mineralization under a dedicated white LED source. Operando infrared spectroscopy traces stepwise oxidation of oxygenated and aromatic VOCs toward CO2, while wavelength‐dependent activity, spectral‐overlap analysis, thermal profiling, complementary spectroscopy, and simulations support coordinated energy utilization and deep mineralization under indoor white LED illumination.
Keywords: indoor light‐emitting diode, localized surface plasmon resonance, photothermal conversion, plasmonic–photothermal photocatalyst, volatile organic compounds
A plasmonic–photothermal–photocatalytic hybrid comprising Cu(II)‐grafted TiO2, yolk–shell Au@SiO2, and reduced graphene oxide sheets (CuT/AuSi‐GS) harvests full‐spectrum white light‐emitting diode (LED) light via complementary photocatalytic oxidation, localized surface‐plasmon‐resonance‐induced electric‐field enhancement, and photothermal acceleration. The tri‐functional architecture drives mineralization of aliphatic and aromatic volatile organic compounds under white LED illumination, highlighting its promise as an LED‐driven photocatalyst platform for indoor air purification.

1. Introduction
Volatile organic compounds (VOCs) are among the most pervasive indoor air pollutants and tend to accumulate indoors, posing substantial health and environmental risks [1, 2]. VOCs, including acetaldehyde (CH3CHO) and toluene (C7H8), are particularly concerning due to their toxicity and persistence, necessitating the development of effective and sustainable purification technologies [3]. Photocatalysis, an attractive route due to its low energy consumption and environmental compatibility, has been extensively explored for VOC abatement under solar or UV illumination [4]. However, conventional photocatalysts such as TiO2 exhibit limited activity under low‐intensity visible indoor lighting (e.g., LED illumination) because of their wide bandgap, inefficient utilization of the visible spectrum, and their inability to convert weak indoor photon flux into coupled electronic and thermal driving forces for deep VOC mineralization [5]. Consequently, achieving efficient indoor VOC purification under realistic lighting conditions remains a critical challenge [6]. This challenge becomes more stringent for mixed‐VOC streams, where readily oxidized oxygenated compounds and more recalcitrant aromatic molecules must be mineralized simultaneously rather than merely removed from the gas phase.
Numerous strategies have been employed to overcome this limitation, including doping, heterojunction formation, and defect engineering, which aim to extend light absorption and promote reactive oxygen species (ROS) generation [7, 8, 9]. However, these approaches often remain insufficient under long‐wavelength or low‐photon‐density illumination typical of indoor LEDs. While photon upconversion techniques, such as triplet–triplet annihilation, offer a conceptual solution, their poor operational stability and weak ROS generation hinder practical use [10]. More importantly, many visible‐light photocatalytic systems are evaluated using single VOCs under laboratory light sources, leaving unresolved whether the same materials can sustain CO2‐selective oxidation of chemically distinct VOC mixtures under low‐power white LED operation.
Recently, the integration of plasmonic materials with semiconductors has emerged as a promising solution to achieve extended light harvesting under visible‐light illumination [11, 12]. In particular, gold nanoparticles (Au NPs), due to their localized surface plasmon resonance (LSPR) properties, significantly enhance photocatalytic activity by generating strong electric fields within the 500–600 nm range [13, 14]. To maximize their stability and LSPR effects, Au NPs are often encapsulated within core‐shell or yolk–shell nanostructures [15]. These architectures facilitate enhanced light interactions by promoting multiple internal light scatterings and efficient energy transfer to adjacent semiconductor catalysts [16]. In particular, the yolk–shell structure, where functional Au NPs are freely suspended within a hollow shell, offers key structural and catalytic advantages over conventional bulk or core‐shell architectures [17]. The movable core ensures maximal surface exposure and preserves catalytic functionality, while the shell prevents aggregation, leakage, and deactivation [18].
Concurrently, carbon‐based materials such as graphene and black carbon have gained attention for their broadband absorption and high photothermal conversion efficiencies in the near‐infrared (NIR) region (>650 nm), enabling significant temperature elevation and efficient adsorption of VOC molecules [19, 20, 21, 22]. Such plasmonic‐photothermal hybrid structures can thus potentially overcome the limitations of conventional photocatalysts under white LED illumination by effectively utilizing the longer‐wavelength visible‐to‐NIR region (≥500 nm) [23, 24]. However, simply combining plasmonic and photothermal components does not guarantee improved mineralization, because plasmonic field enhancement, photothermal heating, and semiconductor photoactivation must be spatially and spectrally coordinated to avoid nonselective heating, charge recombination, or incomplete oxidation. Although multi‐component photocatalysts integrating photocatalytic, plasmonic, and photothermal units have been widely explored, most reported systems emphasize broad light absorption, charge‐transfer promotion, or photothermal heating as additive or cooperative effects [25, 26, 27]. In contrast, efficient indoor VOC mineralization under weak white LED illumination requires a more deliberate spectral energy‐management strategy, in which each optical function is preferentially matched with the emission profile of the practical light source and coupled to the VOC oxidation interface. While both LSPR‐induced electric‐field enhancement and NIR‐responsive photothermal effects have individually indicated potential for extending photocatalytic activity into the visible and NIR regions, their integrated application for efficient photocatalysis under practical white LED conditions remains largely unexplored. In particular, a spectrally resolved materials architecture that assigns distinct optical and catalytic functions to photocatalytic, plasmonic, and photothermal components, while validating mixed‐VOC mineralization and reaction pathways under white LED irradiation, has not been systematically established [28].
Here, we present a spectrally coordinated hybrid photocatalytic architecture that strategically integrates Cu(II)‐grafted TiO2 (CuT), yolk–shell structured Au@SiO2 (AuSi), and reduced graphene oxide sheets (GS) within a single composite architecture (CuT/AuSi‐GS). This composite is specifically designed to maximize the utilization of low‐intensity indoor LED illumination through a trifunctional photocatalytic, plasmonic, and photothermal energy‐management scheme (Figure 1a). The conceptual spectral allocation shown in Figure 1b was guided by the manufacturer‐provided white LED emission profile and by the component‐specific optical, photocatalytic, electromagnetic‐field, and photothermal responses. Specifically, the assignments were based on the optical response and wavelength‐dependent photocatalytic CO2 formation of CuT, the LSPR extinction and FEM‐simulated electromagnetic‐field response of AuSi, and the broadband optical absorption and wavelength‐dependent photothermal temperature response of GS. CuT primarily contributes to visible‐light photocatalytic oxidation, AuSi provides an LSPR‐induced electric‐field enhancement and light scattering near the CuT‐containing matrix, and GS converts longer‐wavelength photons into localized heat to accelerate interfacial reaction kinetics. The comparative activity summarized in Figure 1c highlights the advantage of integrating these components within a single ternary architecture rather than simply relying on pristine CuT or binary combinations. To elucidate the synergistic plasmonic‐photothermal effect within the CuT/AuSi‐GS hybrid, wavelength‐dependent photocatalytic experiments, spectral‐overlap analysis, finite element method (FEM) simulations using COMSOL Multiphysics, time‐resolved photoluminescence (PL) spectroscopy, operando Fourier transform infrared (FT‐IR) spectroscopy, and electron spin resonance (ESR) spectroscopy are combined. Together, these complementary analyses support a mechanistic framework linking preferential spectral contributions, carrier dynamics, ROS generation, and photothermal acceleration to enhanced VOC oxidation under white LED illumination. Operando FT‐IR analysis is used as a central mechanistic tool to track the oxidation of acetaldehyde and toluene through oxygenated intermediates toward CO2, thereby linking the spectral energy‐management design to deep mineralization. Under white LED irradiation, CuT/AuSi‐GS achieved near‐stoichiometric CO2 formation during CH3CHO/C7H8 mixed‐VOC mineralization at a total concentration of 10 ppmv within 1 h, maintained CO2 yields above 80% throughout 168 h of continuous‐flow operation at a total concentration of 2 ppmv, and was further evaluated in a 64‐L chamber. These performance outcomes, together with the observed humidity tolerance, support the potential of this composite for sustainable and efficient indoor air purification.
FIGURE 1.

Concept and functional design of the CuT/AuSi‐GS ternary hybrid for indoor LED‐driven VOC mineralization. (a) Schematic illustration of the plasmonic–photothermal–photocatalytic hybrid designed to harvest complementary spectral regions of indoor white LED light for VOC mineralization through coupled visible‐light photocatalytic oxidation, LSPR‐induced electric‐field enhancement, and light‐to‐heat conversion. (b) Conceptual spectral allocation of the preferential CuT‐driven photocatalytic oxidation, AuSi‐mediated LSPR electric‐field enhancement, and GS‐assisted photothermal acceleration across the indoor LED spectrum, based on the manufacturer‐provided white LED emission and component‐specific optical, photocatalytic, electromagnetic‐field, and photothermal responses. (c) Schematic comparison of indoor LED‐driven VOC mineralization performance for CuT, CuT/GS, and CuT/AuSi‐GS, highlighting the synergistic advantage of the ternary hybrid architecture.
2. Results and Discussion
2.1. Structural and Chemical Characterization of CuT‐Based Composites
High‐resolution transmission electron microscopy (HR‐TEM) images of the hierarchical architecture of the synthesized CuT/AuSi‐GS composite are shown in Figure 2a–c and Figure S1a. Energy‐dispersive x‐ray spectroscopy (EDS) mapping corroborates the encapsulation of Au NPs within a SiO2 shell and their subsequent envelopment by GS layers and CuT NPs (Figure 2d–j). The carbon elemental map in Figure 2g indicates that GS was first adsorbed onto the AuSi surface and CuT was then distributed on the surface of the AuSi‐GS composite (Figure 2i,j). This sequential assembly is critical because it places the AuSi plasmonic domains and GS photothermal domains in close proximity to the CuT photocatalytic matrix, enabling optical energy harvested at different wavelengths to converge at the VOC oxidation interface. For pristine CuT (Figure S1b), Cu2+ ions (2–4 nm) were grafted onto the TiO2 surface, which comprised anatase and rutile phases. The isolated AuSi exhibited a yolk–shell structure (∼600 nm) enclosing Au NPs (∼15 nm) within a SiO2 shell of ∼40 nm thickness (Figure S1c). The HR‐TEM images revealed that, in the CuT/AuSi‐GS composite, CuT and GS enveloped AuSi. Inductively coupled plasma optical emission spectroscopy (ICP‐OES) analysis indicated that the Au content in the AuSi component was approximately 35 wt% (Table S1), which is consistent with the nominal Au loading reported during AuSi synthesis [29] and the yolk–shell structure containing numerous embedded Au nanoparticles observed by HR‐TEM. X‐ray powder diffraction (XRD) analysis (Figure 2k) indicated that CuT/AuSi‐GS retains the characteristic peaks associated with the anatase (25.4° from JCPDS #21‐1272) and rutile (27.6° from JCPDS #21‐1276) phases of CuT [30], along with strong peaks corresponding to metallic Au (39.5° from JCPDS #04‐0784) from AuSi [31]. No additional crystalline phases of copper or GS were detected in these XRD patterns.
FIGURE 2.

Structural and compositional characterization of the CuT/AuSi‐GS hybrid. (a–c) HR‐TEM images and (d–j) elemental EDS mappings of CuT/AuSi‐GS. (k) XRD patterns and (l,m) XPS spectra corresponding to Cu 2p and Au 4f, respectively, for the CuT‐based composites.
To further clarify the chemical states of Cu and Au, x‐ray photoelectron spectroscopy (XPS) was performed. The Cu 2p XPS spectra exhibited clear peaks at approximately 932.1 and 952.1 eV, together with the expected Cu(II) spectral characteristics, supporting the stable oxidation state of Cu(II) on the TiO2 surface throughout the synthesis process (Figure 2l) [32]. Furthermore, the Au 4f XPS spectra are consistent with metallic Au, with a slight positive shift from 83.6 eV for pristine AuSi to 84.0 eV for CuT/AuSi‐GS (Figure 2m). This marginal shift is more likely attributable to alterations in the local dielectric screening and photoelectron attenuation within the heterogeneous matrix following encapsulation by the outer CuT/GS layers, rather than direct inter‐component electronic transfer across the insulating SiO2 shell [33]. This interpretation supports the preservation of zero‐valent Au NPs while avoiding overinterpretation of long‐range charge transfer through the SiO2 shell. Further details on the other XPS elements are provided in Figures S2–S4.
The Brunauer–Emmett–Teller (BET) analysis indicated a significantly enhanced specific surface area for the CuT/AuSi‐GS composite (121.7 m2 g−1) compared to pristine CuT alone (60.9 m2 g−1), as summarized in Figure S5 and Table S2. This increase in surface area is attributed primarily to the incorporation of GS and the hollow‐structured AuSi NPs, suggesting enhanced potential for VOC adsorption and catalytic surface reactions. Collectively, the HR‐TEM, EDS, XRD, XPS, and BET results confirm a well‐defined CuT/AuSi‐GS structure in which CuT and GS encapsulate yolk–shell AuSi, thereby increasing accessible surface area and enabling plasmonic‐photothermal coupling. This structural validation provides the basis for evaluating whether the designed architecture, rather than simple physical mixing, can translate spectrally coordinated light harvesting into enhanced photocatalytic VOC mineralization under indoor LED illumination.
2.2. Wavelength‐Dependent Photothermal and Photocatalytic Behavior of CuT‐Based Composites
The wavelength‐ and temperature‐controlled photocatalytic tests were conducted using customized optical filters and an external thermoelectric module, as schematically illustrated in Figure 3a. These experiments were designed to evaluate spectrally coordinated energy utilization in the CuT/AuSi‐GS hybrid, rather than nonspecific light absorption or simple heating. To rationalize these spectrally filtered experiments, the optical absorption properties of the CuT‐based composites were examined by UV–vis diffuse reflectance spectroscopy (DRS). UV–vis DRS analysis revealed that CuT features an extended absorption edge around 500 nm, attributed to direct interfacial charge transfer (IFCT) from the valence band of TiO2 to Cu2+ ions, as well as a pronounced shoulder in the 700–800 nm region, corresponding to d–d transitions of surface‐grafted Cu2+ (Figure 3b and Figure S6) [34, 35]. These assignments are further corroborated by XPS, which confirms the presence of Cu2+ states (Figure 2l), in agreement with previous reports on Cu‐modified TiO2 catalysts [36, 37]. The incorporation of AuSi and GS significantly broadens the optical response of the CuT matrix, enabling efficient light absorption across the entire visible‐to‐near‐NIR spectrum. Specifically, AuSi containing ∼15 nm Au NPs introduces a distinct LSPR band centered at ∼530 nm, while GS provides broadband absorption well into the NIR (>650 nm), thereby supporting plasmonic and photothermal functions to complementary spectral regions.
FIGURE 3.

Photothermal and wavelength‐dependent photocatalytic activity of CuT‐based composites for CH3CHO removal under xenon lamp irradiation (λ > 400 nm). (a) Schematic illustration of wavelength‐ and temperature‐controlled photocatalytic tests using customized optical filters and an external thermoelectric module. (b) UV–vis DRS spectra of CuT‐based composites. (c) CH3CHO removal after 1 h and concurrent CO2 yield after 3 h over CuT/AuSi‐GS under photocatalysis (P) and thermocatalysis (T), with temperature controlled by an external thermoelectric module. (d) Pseudo‐first‐order rate constants (k) for CH3CHO degradation and (e) CO2 formation rate (r) under P, T, and combined photo‐thermocatalysis (P+T) conditions. (f) CH3CHO removal and CO2 yield under 2‐sun irradiation and/or external heating at 120°C. (g) CH3CHO removal, CO2 yield, and surface temperature of the CuT‐based composites under 1‐sun irradiation. (h) Wavelength‐dependent CO2 yield and surface temperature during photocatalytic CH3CHO degradation under spectrally filtered 1‐sun light (400–700 nm). Triangles indicate photothermal surface‐temperature variations. i) Comparison of CO2 yield over CuT/Si and CuT/AuSi under full‐spectrum (λ > 400 nm) and externally heated short‐wavelength (400 < λ < 450 nm) 1‐sun irradiation.
To quantitatively assess the light‐to‐heat conversion efficiency of the composite, we systematically measured surface temperatures of CuT and CuT/AuSi‐GS under varying irradiation intensities (λ > 400 nm, 60–200 mW cm−2) using a xenon lamp. Pristine CuT showed only a gradual temperature increase, reaching around 50°C at 200 mW cm−2, whereas the CuT/AuSi‐GS composite achieved a surface temperature of 120°C under identical conditions (Figure S7). This pronounced photothermal effect is attributed to the integrated broadband absorption and efficient photothermal conversion, with GS providing the major broadband photothermal contribution and AuSi contributing plasmonic absorption in the visible region.
The influence of photothermal properties on catalytic performance was systematically investigated by monitoring the degradation of CH3CHO under three distinct conditions: (i) photocatalysis (light only, P), (ii) thermocatalysis (external heating only, T), and (iii) combined photo‐thermocatalysis (P+T) (Figure 3a). The P+T condition was applied only to CuT to evaluate the synergistic effect of simultaneous light and external heat inputs relative to the self‐heated CuT/AuSi‐GS hybrid. The thermocatalytic experiments were carefully designed using an external thermoelectric module to achieve temperatures equivalent to those induced by light irradiation in CuT/AuSi‐GS samples (60°C, 80°C, 100°C, and 120°C at 60, 100, 155, and 200 mW cm−2, respectively). In both CuT and CuT/AuSi‐GS catalysts, photocatalytic (P) conditions consistently resulted in greater CH3CHO removal and CO2 formation than purely thermal (T) activation, indicating that visible‐light‐induced photocatalysis represents a major reaction pathway, with a substantial contribution from the 400–500 nm IFCT photoactivation region of CuT (Figure 3c and Figure S8). Notably, CuT/AuSi‐GS under P conditions outperformed CuT under P+T, indicating that the ternary hybrid does not merely reproduce the effect of external heating, but creates a coupled photonic and thermal environment that improves mineralization efficiency.
Further investigation of photocatalytic performance across varying light intensities revealed that pristine CuT exhibited minimal photothermal synergy, even when external heating was applied up to 120°C, as reflected by only marginal increases in both CH3CHO removal (k = 0.07 min−1) and CO2 formation (r = 2.4 ppm min−1) (Figure 3d,e). In contrast, CuT/AuSi‐GS displayed marked enhancements in photocatalytic activity with increasing light intensity, without external heating, achieving surface temperatures up to 120°C under 2‐sun irradiation (200 mW cm−2). The corresponding CH3CHO removal rate constants (k = 0.10 min−1) and CO2 formation rate (r = 3.1 ppm min−1) were significantly higher, resulting in elevated CO2 concentrations (∼350 ppm) compared to those observed with pristine CuT (∼250 ppm) (Figure 3f), indicating that the enhanced activity of CuT/AuSi‐GS cannot be ascribed solely to photothermal effects. The simultaneous increase in CO2 formation is particularly important because VOC disappearance can arise from adsorption or partial oxidation, whereas CO2 evolution reflects deeper oxidative conversion. Full experimental details and supporting data are provided in Figures S9–S11.
To clarify the mechanisms underlying this enhancement, binary composites (CuT/AuSi, CuT/GS) were compared to the ternary CuT/AuSi‐GS under 1‐sun xenon illumination (100 mW cm−2). Although all catalysts showed similar CH3CHO removal rates (∼90–95%), their CO2 yields differed: CuT (43%), CuT/GS (48%), CuT/AuSi (57%), and CuT/AuSi‐GS (68%) (Figure 3g and Figure S12). Notably, despite CuT/GS attaining the highest surface temperature (∼87°C), its CO2 formation remained lower than that of CuT/AuSi (∼78°C), further highlighting that photothermal effects alone are insufficient to explain the observed catalytic enhancements. Instead, the higher CO2 yields in AuSi‐containing composites indicate an AuSi‐associated contribution beyond simple macroscopic heating, consistent with LSPR‐induced electric‐field and light‐scattering effects around the CuT‐containing matrix. These findings were consistent under higher light intensities (2‐sun), where CuT/AuSi achieved higher CO2 yields (68.8%) at substantially lower surface temperatures (∼90°C) than CuT/GS (59.8% at 140°C), indicating more efficient utilization of incident light (Figures S13–S15 and Table S3). Thus, the superior performance of CuT/AuSi‐GS arises from the integration of plasmonic and photothermal functions, rather than from a simple increase in catalyst temperature.
To examine wavelength‐dependent contributions, spectrally filtered photocatalytic experiments were performed using long‐pass and short‐pass optical filters (Figure 3h and Figures S16 and S17). Since these irradiation conditions progressively extend the spectral window rather than isolate individual wavelength bands, the resulting activity and temperature trends are interpreted as cumulative wavelength‐dependent responses and not as independent band‐specific contributions. While all composites exhibited efficient light‐to‐heat conversion, with GS‐containing samples showing slightly superior efficiency, significant differences were noted in photocatalytic performance. Specifically, CO2 yields from CuT/AuSi and CuT/GS only modestly exceeded those of CuT under <500 nm illumination. By contrast, extending the irradiation spectrum through the 500–600 nm region was accompanied by a marked increase in CO2 formation in the AuSi‐containing composites. Although this cumulative spectral trend does not isolate an independent band‐specific contribution, it is consistent with an AuSi‐associated spectral contribution beyond macroscopic photothermal heating alone.
To assess whether the enhanced CO2 formation in AuSi‐containing catalysts can be explained solely by macroscopic photothermal heating, we further compared CuT/Si and CuT/AuSi under two scenarios: (1) full‐spectrum visible light (>400 nm), and (2) 400–450 nm irradiation with external heating to replicate the temperature observed under full‐spectrum conditions (Figure 3i and Figure S18). CuT/Si, which incorporates a hollow SiO2 shell without Au NPs, exhibited negligible heat generation and did not produce a comparable enhancement, indicating that the SiO2 shell itself is not responsible for the improved activity. More importantly, for CuT/AuSi, the CO2 yield decreased from 56% under full‐spectrum visible‐light irradiation to 38% under 400–450 nm irradiation with external heating to 78°C, despite reaching a comparable reaction temperature. Since the reaction temperature was matched in these two conditions, this decrease highlights the importance of the incident spectral window, rather than macroscopic heating alone, in promoting CO2 formation over CuT/AuSi. The result suggests that the visible‐light region overlapping with the AuSi LSPR response contributes to mineralization under full‐spectrum irradiation. Accordingly, AuSi is discussed as providing LSPR‐induced electric‐field and light‐scattering effects around the CuT‐containing matrix. These effects operate together with CuT‐driven photocatalytic oxidation and thermal activation.
2.3. Plasmonic and Photothermal Contributions to Charge Dynamics and ROS Generation
To elucidate the impact of LSPR‐induced electric‐field enhancement on photocatalytic charge dynamics, FEM simulations were performed using COMSOL Multiphysics with simplified AuSi and CuT/AuSi model structures (Figure S19) [38]. To focus on the plasmonic electromagnetic response of the AuSi‐containing structure while minimizing interference from GS‐driven photothermal absorption, the simulations were performed without GS, allowing the LSPR‐induced electromagnetic response of AuSi to be evaluated. Electromagnetic field distributions were modeled across the 350–800 nm wavelength range for both AuSi and CuT/AuSi. The isolated AuSi exhibited a moderate normalized electric‐field response (∼3 V m−1) up to 500 nm (Figure 4a), which increased to 6.004 V m−1 at 580 nm, corresponding to the LSPR excitation of the embedded Au NPs (Figure 4b). By contrast, the CuT/AuSi composite exhibited a pronounced enhancement in the localized electric field, reaching 10.659 V m−1 at 640 nm (Figure 4c). Notably, substantial field localization was also observed under short‐wavelength illumination within the 400–500 nm range, corresponding to the IFCT‐associated photoactivation region of CuT. This intensified electromagnetic response is attributed to the synergistic coupling between the LSPR of AuSi and the interfacial light scattering/absorption properties of the CuT matrix, resulting in an enhanced local electromagnetic‐field response that can support carrier‐dynamics modulation and photocatalytic activity [39, 40].
FIGURE 4.

Plasmonic and photothermal contributions to charge dynamics and ROS generation in CuT‐based composites. (a) Simulated normalized electric‐field response over a wavelength range of 340–800 nm and the corresponding electric‐field distributions of (b) AuSi at 580 nm and (c) CuT/AuSi at 640 nm calculated using COMSOL Multiphysics. (d) TCSPC plots of CuT‐based composites (λ ex = 375 nm and λ em = 420 nm). Fluorescence spectra of (e) 7‐hydroxycoumarin (7‐hCOU) and (f) 2‐hydroxyterephthalic acid (2‐hTA) as probes of •OH generation. (g) UV–vis absorption spectra of NBT solution as a probe of •O2 − generation. (h) ESR spectra after subtraction of the corresponding control signals. Experimental details: For (e–g), [COU]0 and [TA]0 = 10 mM, [NBT]0 = 20 µM, pH = 7.0, and [catalyst]0 = 1.5 g L−1. For (h), [DMPO]0 and [TEMP]0 = 10 mM, and the irradiation time was 15 min under 1‐sun xenon irradiation (λ > 400 nm).
As the initial simulations were conducted under non‐heated conditions to evaluate the plasmonic electromagnetic response, an additional thermo‐optic sensitivity analysis was further carried out to examine whether the LSPR‐induced electric‐field enhancement is preserved under catalyst temperatures relevant to photothermal operation under 1‐sun irradiation (Text S10 and Figure S20). Using literature‐based temperature‐dependent optical parameters and thermo‐optic corrections [41, 42, 43, 44], the recalculated normalized electric‐field response showed only a limited decrease under heating, from 6.004 to 5.716 V m−1 for AuSi and from 10.659 to 9.823 V m−1 for CuT/AuSi, corresponding to losses of 4.8% and 7.8%, respectively. These results indicate that the LSPR‐induced electric‐field response is largely retained even under photothermal conditions induced by the hybrid system, and that the moderate thermo‐optic attenuation is unlikely to eliminate the AuSi‐mediated field contribution under operating conditions. This interpretation is also consistent with the photocatalytic CH3CHO removal results in Figure 3, where the AuSi‐containing composites outperform thermally matched controls, indicating that the enhancement cannot be explained solely by macroscopic photothermal heating. Thus, LSPR‐induced electric‐field enhancement and photothermal heating can operate concurrently in the CuT/AuSi‐GS architecture, with AuSi modulating the local electromagnetic‐field environment and carrier dynamics, while GS primarily accelerates interfacial reaction kinetics.
To further assess the data‐guided spectral allocation in Figure 1b, the component‐specific optical and functional responses were analyzed together with the white LED emission profile (Figure S21). Similar light‐source/absorber spectral‐overlap approaches have been used to rationalize wavelength‐dependent photocatalytic performance under defined irradiation conditions [45, 46]. The LED‐weighted optical‐overlap analysis showed partially overlapping spectral distributions, while the wavelength‐dependent functional responses indicated a substantial short‐visible contribution to CuT photocatalysis, the most pronounced AuSi electric‐field response in the mid‐visible region, and an increasing GS photothermal response toward longer wavelengths. Collectively, these complementary analyses support the preferential and spectrally coordinated component functions illustrated in Figure 1b rather than wavelength‐exclusive operation.
To experimentally connect these simulated electric‐field responses with charge‐carrier behavior, temperature‐dependent PL spectroscopy and time‐correlated single‐photon counting (TCSPC) analysis were performed (Figure 4d and Figure S22). PL intensity generally serves as an indicator of electron‐hole recombination, with lower intensity being associated with suppressed radiative recombination [47, 48]. Under both room temperature and heating conditions, GS‐containing composites exhibited lower PL intensities than pristine CuT, indicating suppressed radiative recombination and/or emission quenching [49, 50]. However, TCSPC analysis revealed that CuT/AuSi (5.12 ns) exhibited the longest average charge carrier lifetime, approximately 3‐fold longer than that of CuT (1.75 ns) or CuT/GS (1.65 ns) (τ avg values and fitting parameters provided in Figure S23 and Table S4). The shortened lifetime in CuT/GS, despite its low PL intensity, is attributed not only to rapid interfacial electron extraction but also to the intrinsic PL quenching properties of graphene [51]. In contrast, the extended carrier lifetime in CuT/AuSi is consistent with the AuSi‐associated plasmonic field environment contributing to prolonged photoexcited‐state dynamics, thereby supporting more efficient IFCT‐mediated carrier utilization [52, 53]. Given that CuT inherently enables visible‐light‐induced IFCT due to Cu(II) surface grafting [30, 54], the PL and TCSPC analyses show that AuSi and GS regulate charge dynamics through distinct but complementary pathways. AuSi is associated with changes in carrier lifetime and carrier utilization, consistent with LSPR‐induced field effects, whereas GS contributes to carrier‐dynamics modulation, emission quenching, and photothermal reaction acceleration. These complementary effects collectively support the superior photocatalytic efficiency of CuT/AuSi‐GS.
To evaluate ROS generation under visible light (λ > 400 nm), we performed aqueous‐phase radical detection tests under conditions where thermal energy contributions were minimized (Figure S24). The formation of hydroxyl radicals (•OH) was quantified using coumarin (COU) and terephthalic acid (TA) as probes, which generate the fluorescent products 7‐hydroxycoumarin (7‐hCOU) at 450 nm and 2‐hydroxyterephthalic acid (2‐hTA) at 425 nm, respectively (Figure 4e,f). CuT/AuSi‐GS yielded markedly higher fluorescent signals than CuT, indicating enhanced •OH generation under visible‐light irradiation, consistent with more efficient utilization of photoinduced oxidative species. Superoxide radical (•O2 –) formation was monitored by nitro blue tetrazolium (NBT) assay, with CuT/AuSi‐GS showing a more pronounced absorbance decrease at 260 nm, consistent with more efficient formation of superoxide‐related oxidative species (Figure 4g). Complementary ESR experiments using 5,5‐dimethyl‐1‐pyrroline‐N‐oxide (DMPO) and 4‐oxo‐2,2,6,6‐tetramethyl‐1‐piperidinyloxy (TEMP) provided additional evidence consistent with enhanced formation of •OH, •O2 –, and singlet oxygen (1O2) in CuT/AuSi‐GS relative to CuT (Figure 4h and Figure S25), whereas both systems showed no detectable hydrogen peroxide (H2O2) by N,N‐diethyl‐1,4‐phenylenediamine sulfate (DPD) colorimetry (Figure S26). This absence suggests that ROS production occurs predominantly through single‐electron transfer pathways (e.g., O2 → •O2 –), rather than via two‐electron peroxide pathways [55]. Overall, the carrier‐dynamics and ROS analyses support the mechanistic basis for describing CuT/AuSi‐GS as a coordinated energy‐management photocatalyst, where AuSi‐mediated LSPR field effects and GS‐assisted photothermal acceleration operate together with CuT‐driven IFCT‐mediated oxidation to promote deep VOC mineralization rather than simple pollutant disappearance.
2.4. Mixed‐VOC Mineralization and Operando FT‐IR Insights
To evaluate the practical applicability of the CuT/AuSi‐GS hybrid under chemically more realistic conditions, we investigated the simultaneous photocatalytic mineralization of a mixed‐VOC stream composed of CH3CHO and C7H8 at a 1:1 ratio under white LED irradiation (100 mW cm−2) using both batch and flow‐type reactor systems (Figure 5a) [56, 57]. Mixed‐VOC mineralization was selected as a more stringent performance criterion than single‐VOC removal because indoor air typically contains chemically diverse pollutants, and efficient oxidation of a small oxygenated VOC does not necessarily indicate effective mineralization of a more persistent aromatic VOC. Although the higher‐concentration batch tests were primarily used to establish clear mechanistic and kinetic differences among the catalysts, additional experiments were conducted at substantially lower VOC levels to better reflect indoor‐air application scenarios, where VOCs are typically present as low‐concentration mixtures rather than single high‐concentration components [58, 59].
FIGURE 5.

Mixed‐VOC mineralization and operando FT‐IR analysis of surface intermediates over CuT/AuSi‐GS under white LED irradiation (100 mW cm−2). (a) Schematic illustration of the batch and flow‐type reactors used for photocatalytic mixed‐VOC removal. (b) Time‐dependent profiles of CH3CHO, C7H8, and CO2 during the photocatalytic oxidation of mixed‐VOC (CH3CHO + C7H8, 1:1 ratio) at a total concentration of 10 ppmv. (c) Long‐term continuous removal of mixed‐VOC (CH3CHO + C7H8, total 2 ppmv, 1:1 ratio, 300 mL min−1) at RH 50%. (d–i) Operando FT‐IR spectra for (d–f) CH3CHO and (g–i) C7H8, indicating (d,g) dark adsorption/partial oxidation, (e,h) light‐driven mineralization, and (f,i) the corresponding enlarged spectral regions. In the enlarged spectra (f,i), gray labels denote adsorbate‐related bands assigned to acetaldehyde/crotonaldehyde or toluene, whereas black labels indicate more oxidized intermediate species.
At a total VOC concentration of 30 ppmv (Figure S27a), both pollutants were continuously consumed after light irradiation, accompanied by substantial CO2 formation approaching the stoichiometric value, indicating efficient mineralization within 3 h. When the total concentration was lowered to 10 ppmv (Figure 5b), ca. 1 ppmv of CH3CHO was adsorbed under dark conditions, whereas C7H8 showed negligible dark adsorption. Upon white LED irradiation, the hybrid exhibited more rapid degradation kinetics and achieved near‐stoichiometric CO2 formation within 1 h, indicating that the ternary architecture remains effective even at lower VOC levels that are more relevant to indoor environments. Importantly, whereas many previously reported photocatalytic systems exhibit relatively high activity toward small oxygenated VOCs such as acetaldehyde but more limited performance or faster deactivation toward aromatic VOCs such as toluene, the present hybrid catalyst showed high mineralization performance for both aliphatic and aromatic VOCs under the same photocatalytic conditions [60, 61]. To further determine whether this enhancement originates from the integrated hybrid architecture rather than from the simple physical mixing of the three functional components, a physical mixture of CuT, AuSi, and GS with the same nominal component ratio as CuT/AuSi‐GS (90:8:2) was additionally evaluated under identical mixed‐VOC batch conditions (total VOC concentration = 10 ppmv; Figure S27b). Although the physical mixture eventually achieved near‐complete gas‐phase removal of both VOCs, CO2 formation remained at ca. 70% of the stoichiometric yield, indicating incomplete mineralization. This result indicates that simple physical mixing can promote VOC disappearance, but is insufficient for efficient mineralization toward CO2, supporting the importance of the integrated CuT/AuSi‐GS architecture for deep VOC mineralization. To further evaluate applicability to chemically diverse gas‐phase pollutants, an additional ternary gas‐phase pollutant test was performed using CH3CHO, C7H8, and methanethiol (CH3SH), each at 5 ppmv, under white LED irradiation (Figure S27c,d). CH3SH was selected as a representative odor‐relevant volatile organic sulfur compound (VOSC) [62]. CuT/AuSi‐GS achieved complete gas‐phase removal of all three pollutants within 2 h of light irradiation, while CO2 formation approached the carbon‐stoichiometric level calculated from the total carbon content of the ternary mixture. The slower and more synchronized removal of CH3CHO and CH3SH suggests substrate‐dependent kinetics, possibly involving competitive adsorption/oxidation of oxygenated and sulfur‐containing pollutants on CuT‐centered oxidation sites. These results indicate that CuT/AuSi‐GS is not limited to single‐component model reactions, but can simultaneously process chemically diverse mixed gas‐phase pollutants under indoor white LED light. The high CO2 yield further indicates that the hybrid promotes mineralization rather than simply enhancing gas‐phase pollutant disappearance.
To assess long‐term performance under continuous operation, a flow‐type mixed‐VOC test was further conducted using CH3CHO and C7H8 at a total concentration of 2 ppmv (1:1 ratio) and a flow rate of 300 mL min−1 under RH 50% (Figure 5c). Under these low‐concentration flow conditions, the CuT/AuSi‐GS catalyst sustained continuous VOC removal without abrupt performance decay for at least 100 h, supporting the robustness of the hybrid architecture beyond batch‐scale experiments. From 100 to 168 h, the CH3CHO removal efficiency remained nearly 100%, while C7H8 removal was maintained at approximately 80%, with the total CO2 yield remaining in the range of ∼80%–90%. This result is particularly meaningful because it bridges the gap between conventional xenon lamp‐based laboratory tests and practical indoor purification scenarios, in which pollutants are continuously introduced at low concentrations. Together, the batch and flow data indicated that the coupled plasmonic–photothermal–photocatalytic effects remain operative across a broad concentration range and under both batch and continuous‐flow operation. The sustained mineralization efficiency is a notable advantage of the CuT/AuSi‐GS hybrid because particulate TiO2‐based photocatalysts often undergo surface passivation by partially oxidized aromatic intermediates during VOC oxidation [63]. In contrast, the present hybrid maintains high CO2 yield during prolonged mixed‐VOC operation, suggesting continuous oxidative turnover of strongly adsorbed aromatic intermediates rather than their progressive accumulation on the catalyst surface.
Post‐reaction characterization of the used CuT/AuSi‐GS catalyst further supported its structural robustness after prolonged mixed‐VOC operation. As shown in Figure S28, XPS analysis revealed no major changes in the principal chemical states of Au, Cu, Ti, and Si, whereas the C 1s and O 1s spectra showed that the peaks assigned to C═O, HO─C═O, and C─O/Si─OH became substantially weaker or nearly disappeared on the used catalyst surface compared with the fresh catalyst. This result indicates a decrease in weakly bound oxygenated O/OH‐containing surface species after the long‐term stability test, without causing noticeable performance loss [64, 65]. Additional evidence is provided by the TGA/DSC analysis in Figure S29. In the low‐temperature region below ∼350°C, the used catalyst exhibited a smaller weight loss than the fresh catalyst in the TGA profile (Figure S29a), which is consistent with the XPS results. More importantly, below 140°C, corresponding to the maximum catalyst surface temperature observed even for CuT/GS under the 2‐sun photothermal condition, the weight loss of both fresh and used catalysts remained below ∼3%, suggesting that thermally induced mass loss is unlikely to significantly influence the measured photocatalytic CO2 yield [66, 67]. A similar trend was also observed in the DSC profiles (Figure S29b). These results support that the measured CO2 originates predominantly from VOC oxidation rather than from thermally induced degradation of the catalyst.
To gain mechanistic insight into the surface reaction pathways, operando FT‐IR measurements were performed for CH3CHO and C7H8 under dark (ca. 30 min) and white LED‐illuminated conditions (ca. 180 min). Operando FT‐IR was used to track the time‐dependent evolution of adsorbed species and partially oxidized intermediates, providing mechanistic information that cannot be obtained from outlet VOC concentration alone. For CH3CHO (Figure 5d and Figure S30a–c), the dark spectra revealed adsorption of CH3CHO on the catalyst surface together with the formation of crotonaldehyde‐like species via aldol condensation (1683 and 1653 cm−1) [68, 69] and partially oxidized intermediates, including acetate/formate‐related carbonyl and carboxylate groups in the 1635–1435 cm−1 region [70, 71], indicating that adsorption and initial surface activation proceed even in the absence of light. Upon white LED irradiation (Figure 5e,f), the aldehydic bands progressively diminished, while more oxidized carbonyl‐ and carboxylate‐type species evolved together with increasing CO2 signal at 2350 cm−1 [72], consistent with stepwise oxidation of CH3CHO toward CO2 on the CuT/AuSi‐GS surface. Comparable operando trends were also observed for CuT, CuT/AuSi, and CuT/GS, as shown in Figure S30d–f. Notably, although the depletion of acetaldehyde‐derived bands was broadly comparable among the CuT‐based catalysts, the decrease in surface hydroxyl bands was less pronounced over CuT/AuSi‐GS than over CuT, CuT/AuSi, and CuT/GS, while oxidized intermediate bands still evolved clearly and CO2 formation was most pronounced on the ternary hybrid surface. This suggests that CuT/AuSi‐GS promotes more efficient oxidative turnover from adsorbed CH3CHO to oxygenated intermediates and final CO2 while retaining a larger fraction of surface hydroxyl species (3600–3300 cm−1). Such behavior is consistent with the enhanced ROS generation and charge‐carrier utilization observed in Figure 4 and may also contribute to the robust photocatalytic performance of the hybrid under humid conditions. These observations indicate that the overall CH3CHO mineralization pathway is preserved across the CuT‐based catalysts, whereas the CuT/AuSi‐GS hybrid drives the sequence more efficiently toward deeper oxidation and higher CO2 yield.
A similar but more complex trend was observed for C7H8, reflecting the more demanding oxidation pathway of the aromatic substrate. Under dark conditions (Figure 5g and Figure S31a–c), C7H8 bands (1500 and 1380 cm−1) were observed together with partially oxidized oxygenated surface intermediates such as benzaldehyde‐like species (1650 and 1636 cm−1), indicating that adsorption and initial activation of the aromatic substrate already proceed before illumination [73, 74]. After light irradiation (Figure 5h), these bands evolved into more oxidized species, including benzoic acid/benzoate‐like intermediates (1560, 1520, and 1452 cm−1) [75, 76] and ring‐opened carboxylate compounds (1300 and 1220 cm−1) [56], together with a concomitant increase in CO2 (2350 cm−1). The enlarged spectra further support the temporal evolution of these oxygenated intermediates during light‐driven oxidation (Figure 5i). A similar qualitative tendency was observed for CuT, CuT/AuSi, and CuT/GS during C7H8 oxidation (Figure S31d–f). However, noticeable differences were observed in the evolution of surface hydroxyl bands (3600–3300 cm−1), aromatic toluene‐derived bands (3200–2700 cm−1), and oxygenated intermediates within the 1700–1200 cm−1 region, indicating catalyst‐dependent turnover kinetics of surface aromatic species. In particular, more pronounced depletion of surface OH species was observed for CuT and CuT/AuSi, whereas the GS‐containing catalysts showed less extensive OH depletion under illumination, suggesting more efficient regeneration or retention of surface hydroxyl species during C7H8 oxidation. This behavior is consistent with the enhanced •OH generation and accelerated surface reaction kinetics observed for the hybrid system in Figure 4. Although the decrease in toluene‐related bands was most evident for CuT/AuSi, this trend did not directly correlate with the extent of final mineralization, indicating that the simple disappearance of aromatic adsorbate bands cannot by itself represent complete oxidative conversion. Rather, the catalyst‐dependent differences are more appropriately reflected in the subsequent progression of oxidation intermediates: both AuSi‐ and GS‐containing catalysts promoted the conversion of benzaldehyde‐like species toward benzoate‐type intermediates more effectively than pristine CuT, whereas the CuT/AuSi‐GS hybrid most efficiently advanced the sequence from benzoate‐type species to ring‐opened carboxylates and final CO2. These observations indicate that the overall aromatic oxidation pathway is preserved across the CuT‐based catalysts, while the ternary CuT/AuSi‐GS architecture most effectively drives the reaction toward deeper oxidation and final mineralization.
Operando FT‐IR analysis further links the photophysical energy‐management functions of CuT/AuSi‐GS to the surface oxidation chemistry of VOCs. Based on the structural and spectroscopic analyses, the Cu(II)‐grafted TiO2 surface is considered the primary photocatalytic oxidation interface, where surface Cu(II) species enable visible‐light‐induced IFCT and adjacent TiO2 oxygen/hydroxyl species participate in VOC adsorption, initial activation, and oxidative turnover. The retention of surface hydroxyl features together with the evolution of carbonyl, carboxylate, and ring‐opened intermediates toward CO2 supports the involvement of surface oxygen/hydroxyl species in the oxidation sequence. In this architecture, AuSi is not assigned as a direct Au‐Cu catalytic interface because the Au NPs are encapsulated within the SiO2 shell; rather, AuSi provides LSPR‐induced electric‐field and light‐scattering effects around the CuT‐containing matrix. Meanwhile, GS‐containing contact regions can assist VOC adsorption, localized heat generation, and interfacial reaction acceleration. Thus, the catalytic architecture is best described as a CuT‐centered oxidation interface coupled with adjacent AuSi and GS domains that modulate the optical and thermal reaction environment. Further site‐selective poisoning, isotope‐labeling, or operando surface‐sensitive spectroscopic studies could provide more definitive turnover‐site assignments.
The mixed‐VOC experiments and operando FT‐IR analysis collectively indicate that CuT/AuSi‐GS enables efficient and sustained mineralization of both aliphatic and aromatic VOCs under white LED illumination. More importantly, the catalyst remains active under mixed‐pollutant and continuous‐flow conditions, while the spectroscopic evidence is consistent with the progressive oxidation of surface‐bound oxygenated intermediates toward CO2 rather than their persistent accumulation. These findings provide mechanistic and application‐level support for the practical relevance of the ternary hybrid architecture for indoor air purification.
2.5. Indoor LED Application and Practical Validation
The conceptual basis for the practical performance of CuT/AuSi‐GS is summarized in Figure 6a. Under broad‐spectrum light irradiation, CuT functions as the primary photocatalytic oxidant component, AuSi enhances visible‐light utilization through LSPR‐induced electric‐field enhancement, and GS converts the longer‐wavelength portion of the spectrum into localized heat. This spectrally coordinated functional integration enables the CuT/AuSi‐GS hybrid to utilize broad‐spectrum light more effectively than the individual or binary components, thereby sustaining both efficient charge utilization and accelerated surface reaction kinetics under low‐intensity light conditions.
FIGURE 6.

Indoor LED application and practical validation of CuT/AuSi‐GS. (a) Schematic illustration of spectrally coordinated utilization of visible‐to‐NIR light by CuT/AuSi‐GS, highlighting the complementary roles of CuT, AuSi, and GS under xenon lamp, halogen lamp, and white LED irradiation. (b) Surface temperatures of CuT and CuT/AuSi‐GS under xenon lamp, white LED, and halogen lamp irradiation. (c,d) Pseudo‐first‐order rate constants (k) for photocatalytic degradation and concurrent CO2 formation rates of (c) CH3CHO and (d) C7H8 over CuT/AuSi‐GS under different light sources. (e) Photocatalytic CH3CHO mineralization under different relative humidity conditions and comparison with commercial photocatalysts (P25 and WO3) under white LED irradiation. (f) Photocatalytic chamber equipped with a CuT/AuSi‐GS‐coated acrylic plate, together with the corresponding thermal image showing photothermal heating under white LED irradiation. Time‐dependent profiles of VOC removal and CO2 formation over CuT/AuSi‐GS‐coated acrylic plates under white LED irradiation for (g) CH3CHO and (h) C7H8. Experimental details: White LED and halogen lamp irradiations were conducted at 25 mW cm−2, whereas xenon lamp (λ > 400 nm) irradiation in (b–d) was applied at 100 mW cm−2. Initial VOC concentrations were (c,e) CH3CHO 200 ppmv, (d) C7H8 55 ppmv, and (g,h) 10 ppmv for each VOC.
To further evaluate the applicability of CuT/AuSi‐GS under lower‐intensity white LED and halogen illumination, its photothermal and photocatalytic performance under white LED and halogen irradiation (25 mW cm−2) was benchmarked against a simulated solar reference (xenon lamp, 100 mW cm−2). The spectral profiles of the light sources are shown in Figure S32, whereas the corresponding temperature profiles are summarized in Figure 6b. The hybrid consistently exhibited substantially higher surface temperatures than pristine CuT under all light sources, consistent with a substantial photothermal contribution from AuSi and GS. Notably, even under white LED illumination at 25 mW cm−2 (Figure S33), CuT/AuSi‐GS reached approximately 75°C, whereas CuT remained near 35°C. Under halogen irradiation, which contains a larger long‐wavelength fraction, the hybrid attained an even higher surface temperature (Figure S34).
The photocatalytic degradation performance of representative aliphatic (CH3CHO) and aromatic (C7H8) VOCs was then evaluated under the same light sources (Figure 6c,d and Figures S35 and S36). Under white LED illumination, CuT/AuSi‐GS achieved high removal rate constants for CH3CHO (k = 0.037 min−1) and C7H8 (k = 0.036 min−1), with corresponding CO2 formation rates (r = 2.2 and 1.8 ppm min−1, respectively). These results closely matched performance under xenon lamp irradiation, with minor deviations attributed to differences in the spectral overlap with the LSPR of AuSi and the photothermal absorption of GS, which respond to visible and NIR wavelengths, respectively. Conversely, halogen lamp illumination resulted in a lower CH3CHO rate constant (k = 0.02 min−1) and reduced CO2 formation (r = 1.6 ppm min−1), primarily due to the limited visible light emission in the 400–500 nm region, which contributes substantially to CuT photoactivation. A similar trend was observed for C7H8 degradation, further supporting visible‐light‐driven photocatalysis as a major contributor to the overall CuT/AuSi‐GS response. These results indicate that photocatalytic performance does not scale simply with temperature and that high temperature alone is insufficient to maximize mineralization. Instead, efficient white LED photocatalysis requires an appropriate balance between visible‐light‐driven excitation of CuT, plasmonic enhancement by AuSi, and photothermal assistance from GS. Accordingly, the lower activity under halogen irradiation is attributed to its weaker overlap with the 400–500 nm CuT photoactivation region, despite its stronger photothermal contribution.
Stability and durability assessments showed that CuT/AuSi‐GS maintained high catalytic activity over repeated cycles at low, environmentally relevant VOC concentrations (10 ppmv CH3CHO and C7H8), which are substantially lower than those used in the mechanistic tests and closer to application‐relevant low‐concentration conditions [77, 78]. Under white LED illumination, the composite achieved nearly complete CH3CHO mineralization (>98% CO2 yield) across 10 consecutive cycles (Figure S37a) and similarly robust performance for C7H8 over 5 cycles (Figure S37b). In contrast, pristine CuT showed rapid deactivation, emphasizing the superior stability and reusability of the hybrid system (Figure S37c). These results indicate that the ternary hybrid architecture not only improves activity but also markedly enhances long‐term operational stability under low‐intensity white LED conditions. The enhanced reusability may be associated with the structural and functional stability of the integrated CuT/AuSi‐GS architecture, where Cu‐grafted TiO2 provides a robust photocatalytic framework, the SiO2 shell stabilizes the embedded Au NPs and helps preserve the AuSi LSPR response during light irradiation, and GS serves as a stable photothermal component. The retained CO2 formation also suggests that VOC‐derived intermediates are effectively oxidized rather than strongly accumulated on the catalyst surface, thereby mitigating rapid deactivation during repeated operation.
The robustness of the hybrid was assessed under varying relative humidity and against benchmark commercial photocatalysts (Figure 6e and Figures S38 and S39). Across RH 25%–80%, CuT/AuSi‐GS maintained consistently higher CO2 yields than commercial P25 and WO3 NPs, indicating that the ternary hybrid remains effective under humidity levels relevant to indoor environments. In particular, the sustained mineralization performance at elevated humidity indicates that the present architecture is less susceptible to severe deactivation by competitive adsorption of water [56], while still benefiting from the combined photocatalytic and photothermal functions of the hybrid. This comparison further highlights the performance advantage of CuT/AuSi‐GS over representative commercial photocatalysts under white LED illumination.
To evaluate chamber‐scale performance, an acrylic plate coated with CuT/AuSi‐GS was installed in a 64‐L indoor photocatalytic chamber equipped with a dedicated white LED source (25 mW cm−2) (Figure 6f). The irradiance at the catalyst surface was adjusted by varying the LED‐to‐catalyst distance and therefore represents a catalyst‐directed illumination condition rather than ordinary ambient room lighting. The coated plate contained 0.88 g of dry catalyst, corresponding to a catalyst loading of ca. 75.1 g m−2 over the coated acrylic area, and was tested under controlled ambient conditions (23°C ± 3°C and 50% ± 10% RH), following standardized protocols for gaseous pollutant removal (SPS‐KACA002‐132, Republic of Korea) [79]. Owing to the photothermal effect of the hybrid, the coated plate reached approximately 84.6°C under the dedicated LED irradiation condition. The reported 84.6°C corresponds to the local catalyst‐coated surface temperature, while the chamber air temperature and relative humidity were separately monitored during operation (Figure S40a,b). Control experiments using an uncoated acrylic plate exhibited negligible VOC removal, indicating the catalytic origin of the observed activity (Figure 6g,h). Under chamber conditions, CuT/AuSi‐GS achieved complete mineralization of CH3CHO, as evidenced by a 100% CO2 yield within 90 min of white LED illumination (Figure 6g). For C7H8, nearly complete gas‐phase removal was observed within 150 min; however, the CO2 concentration reached approximately 65 ppmv at 180 min, slightly below the theoretical maximum of 70 ppmv expected for full mineralization. This small carbon deficit may be associated with closed‐chamber effects, such as adsorption or condensation of unreacted/intermediate species on internal chamber surfaces, retention of carbonaceous intermediates on the catalyst surface, or incomplete conversion during chamber‐scale operation [80, 81]. Although the present chamber experiment cannot quantitatively distinguish these contributions, the higher CO2 yields obtained in the lab‐scale Pyrex reactor under more controlled conditions support the intrinsic capability of CuT/AuSi‐GS for deep C7H8 mineralization (Figure 5b,c and Figure S37b). The CuT/AuSi‐GS coating retained 100% of its initial mass after chamber‐scale photocatalytic operation, and no visible peeling, cracking, or catalyst detachment was observed (Figure S40c,d), supporting the mechanical stability of the immobilized catalyst layer under the chamber‐test conditions. To quantify the electrical energy requirement of the chamber configuration, the energy input was estimated using the manufacturer‐rated electrical power consumption of the dedicated white LED source (13 W electrical input; 100 W‐equivalent). The resulting electrical energy inputs were 19.5 Wh for CH3CHO and 39.0 Wh for C7H8, corresponding to CO2‐based mineralization energy yields of 2.68 and 4.36 µmol CO2 Wh−1, respectively (Table S5). These values provide a quantitative energy benchmark for the present chamber configuration based on the LED electrical input. The elevated local catalyst‐surface temperature further indicates that heat dissipation and thermal management should be considered in future device‐scale implementation. Collectively, these results support the feasibility of translating CuT/AuSi‐GS from powder‐based batch tests to immobilized plate configurations that are more compatible with practical indoor air purification devices.
Overall, the white LED validation results indicate that CuT/AuSi‐GS is not merely an efficient model photocatalyst under xenon irradiation, but also a robust hybrid system that performs effectively under dedicated white LED irradiation at 25 mW cm−2. By integrating visible‐light photocatalysis, AuSi‐mediated LSPR electric‐field/light‐scattering effects, and photothermal conversion in a single architecture, the catalyst achieves sustained VOC mineralization across different lighting environments, humidity levels, and reactor formats. The chamber experiment specifically evaluates larger‐area operation using an immobilized catalyst coating, whereas the separate 168 h continuous‐flow experiment assesses sustained mineralization under continuous pollutant delivery. Although the present 64‐L chamber configuration was not operated under continuous airflow, these complementary experiments separately address immobilized chamber‐scale operation and long‐term flow performance, while airflow rate, residence time, and thermal management remain important parameters for future device‐scale optimization. To further clarify the novelty of this strategy relative to previously reported multi‐component photocatalytic systems, we compared the present CuT/AuSi‐GS system with representative visible‐light, plasmonic/LSPR‐driven, photothermal/thermally assisted, and hybrid photocatalysts for VOC removal (Table S6). This comparison highlights that most previous systems were evaluated using single VOCs, high‐intensity laboratory light sources, limited reactor formats, or VOC removal without comprehensive CO2‐based mineralization analysis. In contrast, CuT/AuSi‐GS achieves mixed‐VOC mineralization under white LED irradiation, sustained continuous‐flow operation, practical humidity tolerance, and chamber‐scale validation. Notably, few previous studies have demonstrated mixed‐VOC mineralization under white LED irradiation together with long‐term continuous‐flow operation and chamber‐scale validation, highlighting the originality of the CuT/AuSi‐GS platform. Therefore, CuT/AuSi‐GS should be understood as a model implementation of a spectrally coordinated functional‐integration strategy rather than as a claim that this exact material formulation is universally applicable. The principle may be extendable to other systems when analogous photocatalytic oxidation, LSPR electric‐field/light‐scattering, and photothermal components are appropriately matched with the incident light source and target reaction environment. Taken together, these results support CuT/AuSi‐GS as an LED‐driven energy‐management photocatalyst that retains its spectrally coordinated function across powder, flow, and immobilized chamber configurations.
3. Conclusion
This study presents a rationally designed hybrid photocatalyst platform for efficient and durable VOC mineralization under white LED illumination. By integrating photocatalytic CuT, plasmonic AuSi, and photothermal GS within a single composite, a spectrally coordinated architecture is achieved, allowing visible and near‐infrared photons to be converted into complementary photocatalytic, plasmonic, and photothermal driving forces. This spectrally coordinated light utilization enables synergistic enhancement of charge‐carrier dynamics, ROS generation, and surface reaction kinetics. As a result, the CuT/AuSi‐GS hybrid achieves near‐stoichiometric CO2 formation during mixed‐VOC mineralization at a total concentration of 10 ppmv within 1 h under white LED irradiation, sustains >80% CO2 yield for 168 h under continuous‐flow mixed‐VOC operation at a total concentration of 2 ppmv, and maintains effective mineralization of both CH3CHO and C7H8 in a 64‐L chamber. These results highlight the ability of the hybrid to process both oxygenated and aromatic VOCs under LED‐driven conditions. Mechanistic investigations further support differentiated but cooperative roles of CuT‐driven photocatalytic oxidation, AuSi‐mediated LSPR field effects, and GS‐assisted photothermal acceleration. FEM simulations using COMSOL Multiphysics indicate local electromagnetic‐field enhancement around AuSi, while TCSPC and PL analyses are consistent with altered photoexcited‐state dynamics in AuSi‐containing composites. In parallel, GS‐assisted photothermal conversion at longer wavelengths accelerates interfacial reaction kinetics. Operando FT‐IR measurements trace the progressive transformation of both aliphatic and aromatic VOCs through oxygenated surface intermediates toward CO2, linking the hybrid energy‐management design to deep mineralization rather than simple VOC disappearance. Together, these results support the integration of photocatalytic oxidation, plasmonic enhancement, and photothermal conversion within a single hybrid architecture as a promising design strategy for indoor air purification under dedicated white LED operation.
Author Contributions
M.L. conceived the study, designed the experiments, and drafted the manuscript. G.Y.K. performed the catalytic experiments and analyzed the resulting data. G.S. contributed to the experimental set‐up, conducted additional experiments, and assisted with interpretation of results. H.C. performed the catalytic experimental analysis. J.L. and J.W. conducted experimental characterization studies. W.K. helped with the experimental set‐up. C.L. and K.C. discussed the results and commented on the manuscript. J.K. and H.K. supervised the whole project, revised and reviewed the manuscript. All authors discussed the results and approved the final version of the manuscript.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Supporting File: adma74944‐sup‐0001‐SuppMat.docx.
Acknowledgements
This work was supported by the National Research Foundation of Korea (NRF) grants funded by the Korean government (MSIT) (RS‐2025‐00514164) and by the Ministry of Education (RS‐2025‐25436909). This work was also supported by the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korean government (MOTIE) (RS‐2024‐00421291, Clean Hydrogen and Ammonia Innovation Research Center). Additional support was provided by the Yonsei Fellowship (funded by Lee Youn Jae).
Contributor Information
Jae‐Hyuk Kim, Email: jaehyuk.kim@pusan.ac.kr.
Hyoung‐il Kim, Email: hi.kim@yonsei.ac.kr.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Fortenberry C., Walker M., Dang A., et al., “Analysis of Indoor Particles and Gases and Their Evolution With Natural Ventilation,” Indoor Air 29 (2019): 761–779, 10.1111/ina.12584. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Nazaroff W. W. and Goldstein A. H., “Indoor Chemistry: Research Opportunities and Challenges,” Indoor Air 25 (2015): 357–361, 10.1111/ina.12219. [DOI] [PubMed] [Google Scholar]
- 3. Chin J. Y., Godwin C., Parker E., et al., “Levels and Sources of Volatile Organic Compounds in Homes of Children With Asthma,” Indoor Air 24 (2014): 403. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Tsang C. H. A., Li K., Zeng Y., et al., “Titanium Oxide Based Photocatalytic Materials Development and Their Role of in the Air Pollutants Degradation: Overview and Forecast,” Environment International 125 (2019): 200–228, 10.1016/j.envint.2019.01.015. [DOI] [PubMed] [Google Scholar]
- 5. Weon S., He F., and Choi W., “Status and Challenges in Photocatalytic Nanotechnology for Cleaning Air Polluted With Volatile Organic Compounds: Visible Light Utilization and Catalyst Deactivation,” Environmental Science: Nano 6 (2019): 3185. [Google Scholar]
- 6. Lee M., Yim H., Kim B., et al., “Harnessing Waste Heat From Indoor Lamps for Sustainable Thermocatalytic Mineralization of Acetaldehyde Using Platinized TiO2 ,” Chemosphere 308 (2022): 136350, 10.1016/j.chemosphere.2022.136350. [DOI] [PubMed] [Google Scholar]
- 7. Gao X., Liu X., Zhu Z., et al., “Enhanced Visible Light Photocatalytic Performance of CdS Sensitized TiO2 Nanorod Arrays Decorated With Au Nanoparticles as Electron Sinks,” Scientific Reports 7 (2017): 973, 10.1038/s41598-017-01124-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Jing L., Zhou W., Tian G., and Fu H., “Surface Tuning for Oxide‐Based Nanomaterials as Efficient Photocatalysts,” Chemical Society Reviews 42 (2013): 9509, 10.1039/c3cs60176e. [DOI] [PubMed] [Google Scholar]
- 9. Chen M., Wang H., Chen X., et al., “High‐Performance of Cu‐TiO2 for Photocatalytic Oxidation of Formaldehyde Under Visible Light and the Mechanism Study,” Chemical Engineering Journal 390 (2020): 124481, 10.1016/j.cej.2020.124481. [DOI] [Google Scholar]
- 10. Kim H.‐I., Weon S., Kang H., et al., “Plasmon‐Enhanced Sub‐Bandgap Photocatalysis via Triplet–Triplet Annihilation Upconversion for Volatile Organic Compound Degradation,” Environmental Science & Technology 50 (2016): 11184–11192, 10.1021/acs.est.6b02729. [DOI] [PubMed] [Google Scholar]
- 11. Sheldon M. T., van de Groep J., Brown A. M., Polman A., and Atwater H. A., “Plasmoelectric Potentials in Metal Nanostructures,” Science 346 (2014): 828–831, 10.1126/science.1258405. [DOI] [PubMed] [Google Scholar]
- 12. Lv S., Du Y., Wu F., Cai Y., and Zhou T., “Review on LSPR Assisted Photocatalysis: Effects of Physical Fields and Opportunities in Multifield Decoupling,” Nanoscale Advances 4 (2022): 2608–2631, 10.1039/D2NA00140C. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Amendola V., Pilot R., Frasconi M., Maragò O. M., and Iatì M. A., “Surface Plasmon Resonance in Gold Nanoparticles: A Review,” Journal of Physics: Condensed Matter 29 (2017): 203002, 10.1088/1361-648X/aa60f3. [DOI] [PubMed] [Google Scholar]
- 14. Ghosh S. K. and Pal T., “Interparticle Coupling Effect on the Surface Plasmon Resonance of Gold Nanoparticles: From Theory to Applications,” Chemical Reviews 107 (2007): 4797–4862, 10.1021/cr0680282. [DOI] [PubMed] [Google Scholar]
- 15. Li C. and Jin Y., “Shell‐Isolated Plasmonic Nanostructures for Biosensing, Catalysis, and Advanced Nanoelectronics,” Advanced Functional Materials 31 (2021): 2008031, 10.1002/adfm.202008031. [DOI] [Google Scholar]
- 16. Li A., Zhu W., Li C., Wang T., and Gong J., “Rational Design of Yolk–Shell Nanostructures for Photocatalysis,” Chemical Society Reviews 48 (2019): 1874–1907, 10.1039/C8CS00711J. [DOI] [PubMed] [Google Scholar]
- 17. Xiong S., Tang R., Gong D., et al., “Yolk‐Shell Catalyst: From Past to Future,” Applied Materials Today 21 (2020): 100798, 10.1016/j.apmt.2020.100798. [DOI] [Google Scholar]
- 18. Purbia R. and Paria S., “Yolk/Shell Nanoparticles: Classifications, Synthesis, Properties, and Applications,” Nanoscale 7 (2015): 19789–19873, 10.1039/C5NR04729C. [DOI] [PubMed] [Google Scholar]
- 19. Lim S. T., Kim J. H., Lee C. Y., et al., “Mesoporous Graphene Adsorbents for the Removal of Toluene and Xylene at Various Concentrations and Its Reusability,” Scientific Reports 9 (2019): 10922, 10.1038/s41598-019-47100-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Yu L., Wang L., Xu W., et al., “Adsorption of VOCs on Reduced Graphene Oxide,” Journal of Environmental Sciences 67 (2018): 171–178, 10.1016/j.jes.2017.08.022. [DOI] [PubMed] [Google Scholar]
- 21. Lin K.‐T., Lin H., Yang T., and Jia B., “Structured Graphene Metamaterial Selective Absorbers for High Efficiency and Omnidirectional Solar Thermal Energy Conversion,” Nature Communications 11 (2020): 1389, 10.1038/s41467-020-15116-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Yu X., Cheng H., Zhang M., Zhao Y., Qu L., and Shi G., “Graphene‐Based Smart Materials,” Nature Reviews Materials 2 (2017): 17046, 10.1038/natrevmats.2017.46. [DOI] [Google Scholar]
- 23. Wei H., Loeb S. K., Halas N. J., and Kim J.‐H., “Plasmon‐Enabled Degradation of Organic Micropollutants in Water by Visible‐Light Illumination of Janus Gold Nanorods,” Proceedings of the National Academy of Sciences 117 (2020): 15473–15481, 10.1073/pnas.2003362117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Li A., Zhang Q., Zhao S., et al., “A Dual Plasmonic Core—Shell Pt/[TiN@TiO2] Catalyst for Enhanced Photothermal Synergistic Catalytic Activity of VOCs Abatement,” Nano Research 15 (2022): 7071–7080, 10.1007/s12274-022-4380-9. [DOI] [Google Scholar]
- 25. Liu D. and Xue C., “Plasmonic Coupling Architectures for Enhanced Photocatalysis,” Advanced Materials 33 (2021): 2005738, 10.1002/adma.202005738. [DOI] [PubMed] [Google Scholar]
- 26. Ramos‐Fernandez E. V., Rendon‐Patiño A., Mateo D., et al., “Photothermal Catalysts, Light and Heat Management: from Materials Design to Performance Evaluation,” Advanced Energy Materials 15 (2025): 2405272, 10.1002/aenm.202405272. [DOI] [Google Scholar]
- 27. Li S., Yang J., Ruan X., Cui X., and Ravi S. K., “Plasmonic Nanomaterials in Photothermal Catalysis and Artificial Photosynthesis: Hot Electron Dynamics, Design Challenges, and Future Prospects,” Advanced Functional Materials 36 (2026): 2503186, 10.1002/adfm.202503186. [DOI] [Google Scholar]
- 28. Zhang X., Li X., Zhang D., et al., “Product Selectivity in Plasmonic Photocatalysis for Carbon Dioxide Hydrogenation,” Nature Communications 8 (2017): 14542, 10.1038/ncomms14542. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Kim G. Y., Lee D., Choe H.‐S., et al., “Yolk–Shell‐Type Gold Nanosphere‐Encapsulated Mesoporous Silica for Catalytic Oxidation of Organic Pollutants in the Presence of Persulfate,” Environmental Science: Nano 9 (2022): 2510. [Google Scholar]
- 30. Liu M., Qiu X., Hashimoto K., and Miyauchi M., “Cu(II) Nanocluster‐Grafted, Nb‐Doped TiO2 as an Efficient Visible‐Light‐Sensitive Photocatalyst Based on Energy‐Level Matching Between Surface and Bulk States,” Journal of Materials Chemistry A 2 (2014): 13571–13579, 10.1039/C4TA02211D. [DOI] [Google Scholar]
- 31. Kim G. Y., Park J.‐M., Choe H.‐S., et al., “Yolk‐Shell Nanoparticle‐Loaded Multifunctional Membrane With Photothermal and Advanced Oxidation Properties,” Chemical Engineering Journal 511 (2025): 162091, 10.1016/j.cej.2025.162091. [DOI] [Google Scholar]
- 32. Irie H., Kamiya K., Shibanuma T., et al., “Visible Light‐Sensitive Cu(II)‐Grafted TiO2 Photocatalysts: Activities and X‐Ray Absorption Fine Structure Analyses,” Journal of Physical Chemistry C 113 (2009): 10761–10766, 10.1021/jp903063z. [DOI] [Google Scholar]
- 33. Li Y., Jin C., Yuan G., et al., “Implantation of Fe3O4 Nanoparticles in Shells of Au@M‐SiO2 Yolk@Shell Nanocatalysts With Both Improved Recyclability and Catalytic Activity,” Langmuir 33 (2017): 7486. [DOI] [PubMed] [Google Scholar]
- 34. Dias Filho N. L., “Adsorption of Cu(II) and Co(II) Complexes on a Silica Gel Surface Chemically Modified with 2‐mercaptoimidazole,” Mikrochimica Acta 130 (1999): 233–240, 10.1007/BF01242911. [DOI] [Google Scholar]
- 35. Polliotto V., Livraghi S., Krukowska A., et al., “Copper‐Modified TiO2 and ZrTiO4: Cu Oxidation State Evolution during Photocatalytic Hydrogen Production,” ACS Applied Materials & Interfaces 10 (2018): 27745–27756, 10.1021/acsami.8b05528. [DOI] [PubMed] [Google Scholar]
- 36. Irie H., Miura S., Kamiya K., and Hashimoto K., “Efficient Visible Light‐sensitive Photocatalysts: Grafting Cu(II) Ions onto TiO2 and WO3 Photocatalysts,” Chemical Physics Letters 457 (2008): 202–205, 10.1016/j.cplett.2008.04.006. [DOI] [Google Scholar]
- 37. Miyauchi M., Irie H., Liu M., et al., “Visible‐Light‐Sensitive Photocatalysts: Nanocluster‐Grafted Titanium Dioxide for Indoor Environmental Remediation,” Journal of Physical Chemistry Letters 7 (2016): 75–84, 10.1021/acs.jpclett.5b02041. [DOI] [PubMed] [Google Scholar]
- 38. Chen J.‐J., Wu J. C. S., Wu P. C., and Tsai D. P., “Improved Photocatalytic Activity of Shell‐Isolated Plasmonic Photocatalyst Au@SiO2/TiO2 by Promoted LSPR,” The Journal of Physical Chemistry C 116 (2012): 26535–26542, 10.1021/jp309901y. [DOI] [Google Scholar]
- 39. Kumar M. K., Krishnamoorthy S., Tan L. K., Chiam S. Y., Tripathy S., and Gao H., “Field Effects in Plasmonic Photocatalyst by Precise SiO2 Thickness Control Using Atomic Layer Deposition,” ACS Catalysis 1 (2011): 300–308, 10.1021/cs100117v. [DOI] [Google Scholar]
- 40. Torimoto T., Horibe H., Kameyama T., et al., “Plasmon‐Enhanced Photocatalytic Activity of Cadmium Sulfide Nanoparticle Immobilized on Silica‐Coated Gold Particles,” The Journal of Physical Chemistry Letters 2 (2011): 2057–2062, 10.1021/jz2009049. [DOI] [Google Scholar]
- 41. Shen P.‐T., Sivan Y., Lin C.‐W., Liu H.‐L., Chang C.‐W., and Chu S.‐W., “Temperature‐ and Roughness‐Dependent Permittivity of Annealed/Unannealed Gold Films,” Optics Express 24 (2016): 19254, 10.1364/OE.24.019254. [DOI] [PubMed] [Google Scholar]
- 42. Matsuoka J., Kitamura N., Fujinaga S., Kitaoka T., and Yamashita H., “Temperature Dependence of Refractive Index of SiO2 Glass,” Journal of Non‐Crystalline Solids 135 (1991): 86–89, 10.1016/0022-3093(91)90447-E. [DOI] [Google Scholar]
- 43. Gülşen G. and Inci M. N., “Thermal Optical Properties of TiO2 Films,” Optical Materials 18 (2002): 373. [Google Scholar]
- 44. Ito T., Yamaguchi H., Masumi T., and Adachi S., “Optical Properties of CuO Studied by Spectroscopic Ellipsometry,” Journal of the Physical Society of Japan 67 (1998): 3304–3309, 10.1143/JPSJ.67.3304. [DOI] [Google Scholar]
- 45. Daskalova D., Aguila Flores G., Plachetka U., et al., “Combined Structural and Plasmonic Enhancement of Nanometer‐Thin Film Photocatalysis for Solar‐Driven Wastewater Treatment,” ACS Applied Nano Materials 6 (2023): 15204–15212, 10.1021/acsanm.3c02867. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Rauch M., Schmidt S., Arends I. W. C. E., et al., “Photobiocatalytic Alcohol Oxidation Using LED Light Sources,” Green Chemistry 19 (2017): 376. [Google Scholar]
- 47. Lu T., Ma Z., Du C., et al., “Temperature‐Dependent Photoluminescence in Light‐Emitting Diodes,” Scientific Reports 4 (2014): 6131, 10.1038/srep06131. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Yuan Q., Liang B., Zhou C., et al., “Interplay Effect of Temperature and Excitation Intensity on the Photoluminescence Characteristics of InGaAs/GaAs Surface Quantum Dots,” Nanoscale Research Letters 13 (2018): 387, 10.1186/s11671-018-2792-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49. Iwase A., Ng Y. H., Ishiguro Y., Kudo A., and Amal R., “Reduced Graphene Oxide as a Solid‐State Electron Mediator in Z‐Scheme Photocatalytic Water Splitting under Visible Light,” Journal of the American Chemical Society 133 (2011): 11054–11057, 10.1021/ja203296z. [DOI] [PubMed] [Google Scholar]
- 50. Yu H., Tian J., Chen F., et al., “Synergistic Effect of Dual Electron‐Cocatalysts for Enhanced Photocatalytic Activity: RGO as Electron‐Transfer Mediator and Fe(III) as Oxygen‐Reduction Active Site,” Scientific Reports 5 (2015): 13083. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51. Rahman M. Z., Maity P., Mohammed O. F., and Gascon J., “Insight into the Role of Reduced Graphene Oxide in Enhancing Photocatalytic Hydrogen Evolution in Disordered Carbon Nitride,” Physical Chemistry Chemical Physics 24 (2022): 11213–11221, 10.1039/D2CP00200K. [DOI] [PubMed] [Google Scholar]
- 52. Cushing S. K., Li J., Meng F., et al., “Photocatalytic Activity Enhanced by Plasmonic Resonant Energy Transfer from Metal to Semiconductor,” Journal of the American Chemical Society 134 (2012): 15033–15041, 10.1021/ja305603t. [DOI] [PubMed] [Google Scholar]
- 53. Liu S., Wu Z., Zhu Z., et al., “Quantifying the Distinct Role of Plasmon Enhancement Mechanisms in Prototypical Antenna‐Reactor Photocatalysts,” Nature Communications 16 (2025): 2245, 10.1038/s41467-025-57569-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54. Zhang H., Guo L.‐H., Wang D., Zhao L., and Wan B., “Light‐Induced Efficient Molecular Oxygen Activation on a Cu(II)‐Grafted TiO2/Graphene Photocatalyst for Phenol Degradation,” ACS Applied Materials & Interfaces 7 (2015): 1816–1823, 10.1021/am507483q. [DOI] [PubMed] [Google Scholar]
- 55. Nosaka Y., Takahashi S., Sakamoto H., and Nosaka A. Y., “Reaction Mechanism of Cu(II)‐Grafted Visible‐Light Responsive TiO2 and WO3 Photocatalysts Studied by Means of ESR Spectroscopy and Chemiluminescence Photometry,” Journal of Physical Chemistry C 115 (2011): 21283–21290, 10.1021/jp2070634. [DOI] [Google Scholar]
- 56. Shin G., Shit S. C., Koo M. S., et al., “Plasmonic Enhancement Enables Deactivation Resilient TiO2 for Sustainable VOCs Remediation Under Practical Conditions,” Advanced Functional Materials 36 (2026): 26193, 10.1002/adfm.202526193. [DOI] [Google Scholar]
- 57. Park C., Shin G., Chung M. W., et al., “Time‐Resolved Spectroscopic Investigation for the Practical Application of a Photocatalytic Air Purifier,” Journal of Hazardous Materials 472 (2024): 134382, 10.1016/j.jhazmat.2024.134382. [DOI] [PubMed] [Google Scholar]
- 58. Wang W., Lin F., An T., et al., “Photocatalytic Mineralization of Indoor VOC Mixtures Over Unique Ternary TiO2/C/MnO2 With High Adsorption Selectivity,” Chemical Engineering Journal 425 (2021): 131678, 10.1016/j.cej.2021.131678. [DOI] [Google Scholar]
- 59. He F., Muliane U., Weon S., and Choi W., “Substrate‐Specific Mineralization and Deactivation Behaviors of TiO2 as an Air‐Cleaning Photocatalyst,” Applied Catalysis B: Environmental 275 (2020): 119145, 10.1016/j.apcatb.2020.119145. [DOI] [Google Scholar]
- 60. Bianchi C. L., Gatto S., Pirola C., et al., “Photocatalytic Degradation of Acetone, Acetaldehyde and Toluene in Gas‐phase: Comparison Between Nano and Micro‐sized TiO2 ,” Applied Catalysis B: Environmental 146 (2014): 123–130, 10.1016/j.apcatb.2013.02.047. [DOI] [Google Scholar]
- 61. Xu Z., Ren Y., Deng X., et al., “Recent Developments on Gas‐Phase Volatile Organic Compounds Abatement Based on Photocatalysis,” Advanced Energy and Sustainability Research 3 (2022): 2200105, 10.1002/aesr.202200105. [DOI] [Google Scholar]
- 62. Bentley R. and Chasteen T. G., “Environmental VOSCs‐Formation and Degradation of Dimethyl Sulfide, Methanethiol and Related Materials,” Chemosphere 55 (2004): 291–317, 10.1016/j.chemosphere.2003.12.017. [DOI] [PubMed] [Google Scholar]
- 63. Weon S. and Choi W., “TiO2 Nanotubes With Open Channels as Deactivation‐Resistant Photocatalyst for the Degradation of Volatile Organic Compounds,” Environmental Science & Technology 50 (2016): 2556–2563, 10.1021/acs.est.5b05418. [DOI] [PubMed] [Google Scholar]
- 64. Abbas N., Hussain M., Russo N., and Saracco G., “Studies on the Activity and Deactivation of Novel Optimized TiO2 Nanoparticles for the Abatement of VOCs,” Chemical Engineering Journal 175 (2011): 330–340, 10.1016/j.cej.2011.09.115. [DOI] [Google Scholar]
- 65. Ardizzone S., Bianchi C. L., Cappelletti G., Naldoni A., and Pirola C., “Photocatalytic Degradation of Toluene in the Gas Phase: Relationship Between Surface Species and Catalyst Features,” Environmental Science & Technology 42 (2008): 6671–6676, 10.1021/es8009327. [DOI] [PubMed] [Google Scholar]
- 66. Shahin R., Yadav R. K., Verma R. K., Mishra S., Singh C., and Baeg J.‐O., “Solar‐Driven Photocatalytic System for CO2 Fixation and Conversion of Dopamine into Indole Derivative,” Photochemistry and Photobiology 102 (2026): 79–92, 10.1111/php.70007. [DOI] [PubMed] [Google Scholar]
- 67. Ribbens S., Beyers E., Schellens K., et al., “Systematic Evaluation of Thermal and Mechanical Stability of Different Commercial and Synthetic Photocatalysts in Relation to Their Photocatalytic Activity,” Microporous and Mesoporous Materials 156 (2012): 62–72, 10.1016/j.micromeso.2012.01.036. [DOI] [Google Scholar]
- 68. Singh M., Zhou N., Paul D. K., and Klabunde K. J., “IR Spectral Evidence of Aldol Condensation: Acetaldehyde Adsorption Over TiO2 Surface,” Journal of Catalysis 260 (2008): 371–379, 10.1016/j.jcat.2008.07.020. [DOI] [Google Scholar]
- 69. Rekoske J. E. and Barteau M. A., “Competition Between Acetaldehyde and Crotonaldehyde During Adsorption and Reaction on Anatase and Rutile Titanium Dioxide,” Langmuir 15 (1999): 2061–2070, 10.1021/la9805140. [DOI] [Google Scholar]
- 70. Liao L.‐F., Lien C.‐F., and Lin J.‐L., “FTIR Study of Adsorption and Photoreactions of Acetic Acid on TiO2 ,” Physical Chemistry Chemical Physics 3 (2001): 3831–3837, 10.1039/b103419g. [DOI] [Google Scholar]
- 71. Chen T., Wu G., Feng Z., et al., “In Situ FT‐IR Study of Photocatalytic Decomposition of Formic Acid to Hydrogen on Pt/TiO2 Catalyst,” Chinese Journal of Catalysis 29 (2008): 105. [Google Scholar]
- 72. Iskra A., Gentleman A. S., Kartouzian A., Kent M. J., Sharp A. P., and Mackenzie S. R., “Infrared Spectroscopy of Gas‐Phase M+(CO2)n (M = Co, Rh, Ir) Ion‐ Molecule Complexes,” Journal of Physical Chemistry A 121 (2017): 133–140, 10.1021/acs.jpca.6b10902. [DOI] [PubMed] [Google Scholar]
- 73. Wang H., Zhao Q., Li D., et al., “Boosting Photothermocatalytic Oxidation of Toluene Over Pt/N‐TiO2: The Gear Effect of Light and Heat,” Environmental Science & Technology 58 (2024): 7662–7671, 10.1021/acs.est.3c10459. [DOI] [PubMed] [Google Scholar]
- 74. Liang Y., Zhang Y., Fábri C., et al., “High‐resolution FTIR Spectroscopy of Benzaldehyde in the Far‐Infrared Region: Probing the Rotational Barrier,” Physical Chemistry Chemical Physics 23 (2021): 8549–8556, 10.1039/D1CP00188D. [DOI] [PubMed] [Google Scholar]
- 75. Guo M., Kong X., Li C., and Yang Q., “Hydrogenation of Benzoic Acid Derivatives Over Pt/TiO2 Under Mild Conditions,” Communications Chemistry 4 (2021): 54, 10.1038/s42004-021-00489-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 76. Yu J., Wang X., Chen L., et al., “Enhanced Adsorption and Visible‐Light Photocatalytic Degradation of Toluene by CQDs/UiO‐66 MOG With Hierarchical Pores,” Chemical Engineering Journal 435 (2022): 135033, 10.1016/j.cej.2022.135033. [DOI] [Google Scholar]
- 77. Bai L., Dai H., Wang J., and Li G., “Source Apportionment and Health Risk Assessment of Indoor Volatile Organic Compounds,” Indoor and Built Environment 31 (2022): 1564–1576, 10.1177/1420326X211065043. [DOI] [Google Scholar]
- 78. González‐Martín J., Kraakman N. J. R., Pérez C., Lebrero R., and Muñoz R., “A State–of–the‐Art Review on Indoor Air Pollution and Strategies for Indoor Air Pollution Control,” Chemosphere 262 (2021): 128376, 10.1016/j.chemosphere.2020.128376. [DOI] [PubMed] [Google Scholar]
- 79. Kwon H. J., Yang D. S., Koo M. S., et al., “Long‐Lifetime Water‐Washable Ceramic Catalyst Filter for Air Purification,” Nature Communications 14 (2023): 520, 10.1038/s41467-023-36050-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 80. Zhang X., Cappa C. D., Jathar S. H., et al., “Influence of Vapor Wall Loss in Laboratory Chambers on Yields of Secondary Organic Aerosol,” Proceedings of the National Academy of Sciences 111 (2014): 5802–5807, 10.1073/pnas.1404727111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 81. Jørgensen R. B., Bjørseth O., and Malvik B., “Chamber Testing of Adsorption of Volatile Organic Compounds (VOCs) on Material Surfaces,” Indoor Air 9 (1999): 2–9, 10.1111/j.1600-0668.1999.t01-3-00002.x. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting File: adma74944‐sup‐0001‐SuppMat.docx.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
