Skip to main content
NCBI home page
Chemical Science logoLink to Chemical Science
. 2026 Sep 30. Online ahead of print. doi: 10.1039/d6sc06755g

Epitaxial Ga/ZSM-5@S-1 core–shell zeolites boost CO2-coupled propane aromatization

Luyuan Yang a,†, Yitao Yang a,†, Meizan Jing a, Kun Yang a, Guilin Liu a, Yumeng Fo a, Xiangyang Ji a, YuHui Xia a, Konstantin Yuryevich Koltunov b, Zhen Zhao a, Jian Liu a,✉, Weiyu Song a,c,d,✉
PMCID: PMC13643530  PMID: 42846272

Abstract

Mitigating climate change necessitates efficient carbon capture and utilization (CCU) strategies. However, current CO2 conversion technologies are often constrained by their reliance on molecular hydrogen (H2), which is energy-intensive and economically burdensome to produce. Coupling CO2 conversion with alkane aromatization offers a promising solution by utilizing in situ hydrogen species from alkanes. However, the efficiency of such systems is often compromised by the random distribution of active sites in conventional catalysts, which leads to competitive adsorption and inefficient H species utilization, ultimately restricting CO2 conversion. Here, this study reports a Ga/ZSM-5@S-1 catalyst synthesized via epitaxial growth, achieving a high CO2 conversion of ∼75% alongside ∼73% BTX selectivity in propane (C3H8) aromatization. XPS sputtering, in situ X-ray absorption near-edge structure (XANES), wide-angle X-ray scattering (WAXS), and in situ spectroscopic characterization reveal two distinct (Ga2O2)2+ species in Ga/ZSM-5@S-1: one at the BAS and another at surface hydroxyl sites on the S-1 shell, spatially separated from the BAS. The latter species provides additional adsorption sites for CO2 during the reaction. Density functional theory (DFT), ab initio molecular dynamics (AIMD) simulations, and isotope-labeled kinetics demonstrate that hydrogen species (H*) preferentially accumulate on (Ga2O2)2+ sites, which correlates with enhanced CO2 conversion. Furthermore, GC-MS, in situ MS, and FTIR reveal a novel pathway in which CO2 serves as an oxygen source, inserting into hydrocarbon chains to form oxygenated intermediates that boost aromatic production. This study establishes a new paradigm for the rational design of bifunctional catalysts, enabling highly efficient, value-added conversion of CO2 in the presence of alkanes.

Epitaxial Ga/ZSM-5@S-1 core–shell zeolites spatially separate active sites, enhancing CO2 conversion and BTX formation in CO2-coupled propane aromatization.graphic file with name d6sc06755g-ga.webp

Introduction

Mitigating climate change and achieving carbon-neutrality goals require the development of efficient strategies for carbon capture and utilization (CCU).1 Within the diverse range of carbon dioxide (CO2) utilization technologies, the catalytic conversion of CO2 into high-value chemicals represents a pivotal frontier.2,3 However, many established CO2 conversion routes, such as reverse water–gas shift (RWGS),4,5 methanation,6,7 and methanol synthesis,8–10 typically rely on externally supplied H2 and are constrained by thermodynamic equilibria. In this context, coupling CO2 with light alkanes (e.g., propane, C3H8) offers an appealing alternative, wherein C3H8 provides in situ hydrogen and CO2 functions as a mild hydrogen acceptor to suppress coke formation while promoting the formation of high-value aromatics.1

Despite the conceptual soundness of this strategy, its practical implementation remains severely hindered by sluggish CO2 activation, with CO2 conversion typically stagnating below 40% in most reported CO2-coupled alkane conversion systems.11–13 This persistent bottleneck stems primarily from competitive adsorption between reactants14,15 and from inefficient utilization of in situ generated hydrogen species.16 A straightforward way to address these limitations is to increase the number of accessible CO2 activation sites while maintaining alkane aromatization. In typical Ga/ZSM-5 catalysts, CO2 adsorbs weakly and competes with hydrocarbons for surface sites, particularly under high hydrocarbon partial pressures.15 The resulting poor CO2 coverage creates a kinetic mismatch: hydrogen species generated during alkane dehydrogenation cannot be efficiently coupled to CO2. Instead, a large fraction of this hydrogen recombines into H2 and exits the reactor or is diverted to side reactions.17 Improving CO2 conversion in CO2-coupled alkane aromatization therefore requires introducing additional CO2 activation sites positioned to intercept hydrogen species.

Herein, this study reports the core–shell-like Ga/ZSM-5@S-1 catalyst. After epitaxial growth of an S-1 shell over ZSM-5, Ga species were incorporated by impregnation and subsequently dispersed and stabilized by H2 treatment.4,18,19 XPS depth profiling, in situ X-ray absorption near-edge structure (XANES), and wide-angle X-ray scattering (WAXS) confirmed the formation of the core–shell structure with dispersed (Ga2O2)2+ species. Catalytic tests show that Ga/ZSM-5@S-1 achieves ∼75% CO2 conversion with ∼73% BTX selectivity. To clarify the reaction pathway, we conducted isotope-tracing and in situ FTIR measurements. These results indicate that CO2 is converted at Ga–OH sites and that the hydrogen on Ga–O originates from C3H8. Breakthrough and CO2 temperature-programmed surface reaction (CO2-TPSR) measurements further showed that the S-1 shell bearing uniformly distributed (Ga2O2)2+ species preserves sufficient CO2 uptake sites, alleviating competitive adsorption between C3H8 and CO2 at (Ga2O2)2+-BAS. In the core–shell catalyst, H2 generated in the inner layer during C3H8 aromatization diffuses outward through the S-1 shell region enriched in (Ga2O2)2+ species, increasing the likelihood of reaction with CO2. Ab initio molecular dynamics (AIMD), density functional theory (DFT), and isotope-labeled kinetic analyses suggest that in situ hydrogen species (H*) preferentially accumulate on (Ga2O2)2+ clusters, facilitating CO2 conversion. The energy barrier for CO2-assisted H* consumption is substantially low. In situ MS and FTIR spectroscopy revealed that CO2 acts not only as a hydrogen acceptor but also as a carbon source: CO2 and derived CO are inserted into hydrocarbon chains to form oxygenated intermediates, which subsequently enter the aromatic cycle and directly promote BTX formation. Collectively, these findings demonstrate that the additional (Ga2O2)2+-S-1 sites introduced by the epitaxial growth strategy enhance CO2 utilization in coupled alkane conversion, offering a general design principle for bifunctional catalysts with improved CO2 conversion and high-value product selectivity.

Results

Evidence for the successful construction of nanoscale spatially engineered Ga/ZSM-5@S-1

This study employs an epitaxial growth strategy to encapsulate ZSM-5 within silicalite-1 (S-1). TEM images, WAXS (Fig. 1a–c), Raman spectra, and N2 adsorption–desorption measurements proved that ZSM-5@S-1 maintained the MFI framework structure. The Raman spectrum (Fig. S1b) exhibited a characteristic MFI framework peak at 387 cm−1, which remained virtually unchanged after coating.20,21 N2 adsorption–desorption measurements (Fig. S1c, d and Table S1) further indicated that the epitaxial S-1 did not introduce additional mesoporosity. Energy-dispersive X-ray (EDX) point analysis confirms the elemental distribution, showing that the ZSM-5@S-1 exterior consists exclusively of Si and O, whereas Si, O, and Al are detected in the interior (Fig. 1b).

Fig. 1. TEM images of (a) ZSM-5 and (b) ZSM-5@S-1; (c) WAXS patterns of ZSM-5, Ga/ZSM-5, and Ga/ZSM-5@S-1; (d) in situ normalized XANES spectrum of Ga-MFI at the Ga K-edge; elemental maps of Ga/ZSM-5@S-1 (e) before and (f) after H2 treatment, acquired by EDX; (g) in situ H2 FTIR of Ga/ZSM-5@S-1 (600 °C), p = 1 atm, pH2 : pN2 = 1 : 1, WHSV = 24.11 h−1; (h) elemental depth profiles of Ga/ZSM-5@S-1, acquired by XPS-sputtering spectra; (i) structure of Ga/ZSM-5@S-1 (schematic); (j) Ga 2p3/2 XPS spectra of Ga/ZSM-5 and Ga/ZSM-5@S-1; (k) normalized Ga K-edge XANES spectra collected at ambient temperature of Ga foil, Ga2O3, Ga/ZSM-5, Ga/ZSM-5@S-1 (without H2) and Ga/ZSM-5@S-1 samples; (l) the k3 weighted Ga K-edge EXAFS spectra and corresponding FT k3-weighted EXAFS spectra of samples, experiment (dots) and best-fit (solid line).

Fig. 1

Open in a new tab

To achieve uniform dispersion of Ga species and stabilize the active-site structure, this study exploited the inherent mobility of Ga. Specifically, under high-temperature H2, surface Ga2O3 can be partially reduced to more mobile Ga-suboxides, which promotes Ga dispersion.19,22 To probe the specific evolution, we performed TEM, EDX analysis, in situ FTIR, and XANES on Ga/ZSM-5@S-1 before and after H2 pretreatment. TEM and EDX analyses (Fig. 1e and f) revealed a redistribution of Ga species within the catalyst. This phenomenon is mechanistically supported by in situ XANES (Fig. 1d), which identified the formation of low-valent Ga species at 600 °C.19 In the impregnation-prepared Ga/ZSM-5@S-1, Ga species are initially anchored to different hydroxyl groups, including Si–OH, Al–OH, and Si–O(H)–Al (Fig. S2b). After H2 treatment, in situ H2-FTIR and –OH FTIR spectra (Fig. 1g and S2) revealed the appearance of a Si–OH peak and a corresponding negative feature for Si–OH⋯Ga, indicating the liberation of silanol groups due to Ga migration. Concurrently, the intensity of the Brønsted acid site (BAS) peak at 3610 cm−1 (ref. 23–25) decreased significantly (Fig. 1g). This supports the hypothesis that Ga species, initially anchored on silanol groups, are activated by H2 and subsequently migrate to the BAS. As illustrated in XPS sputtering depth profiling (Fig. 1h and S4), the near-surface region showed negligible Al but relatively high Ga content. With increasing sputtering time from 0 to 200 s, the Al content increased from ∼0 to ∼1.5 atom%, while the Ga concentration decreased from ∼1.2 to ∼0.8 atom%. Consequently, the Ga species in Ga/ZSM-5@S-1 were distributed across two specific sites: the hydroxyl groups on the epitaxial S-1 shell and the BAS in the core. The local coordination and chemical state of Ga species were elucidated through combined XAS and XPS analyses. Ga 2p3/2 XPS indicated that the as-prepared catalyst contained Ga3+ species (Fig. 1j). The Ga K-edge XANES spectra of H2-treated Ga/ZSM-5@S-1 closely resembled Ga2O3, suggesting no significant change in average Ga valence (Fig. 1k).12 Fitting of the R-space gave a first-shell Ga–O coordination number of ∼2.2 and a second-shell Ga–O–Ga coordination number of ∼0.7 (Fig. 1l and Table S2). Based on the average coordination numbers derived from EXAFS, we infer that the Ga species in Ga/ZSM-5@S-1 are predominantly present as (Ga2O2)2+.

Catalytic performance

The Ga/ZSM-5@S-1 catalyst achieved a CO2 conversion of 74.5% and BTX selectivity of 72.9%, significantly outperforming currently reported propane-coupled CO2 aromatization (PCA) catalysts (Fig. 2a and d). Upon reaching a steady state at 2 h, the CO2 conversion substantially surpassed the thermodynamic equilibrium conversion (34%) of the reverse water–gas shift (RWGS) reaction at the same temperature, with CO as the predominant product (Fig. 2a). In contrast, the Ga/ZSM-5 catalyst reached a maximum CO2 conversion of only 50%, followed by a rapid decline, and a lower BTX selectivity of 55% (Fig. 2b). Furthermore, the epitaxially grown structure suppressed carbon deposition. This is supported by the reduction in UV-vis absorption bands above 640 nm, which are assigned to carbon species accumulated outside the channels and regarded as precursors to carbon deposition (Fig. 2e).26–28 Consequently, the carbon deposition rate on Ga/ZSM-5@S-1 (2.71 mg gcat−1 h−1) was lower than that on Ga/ZSM-5 (4.03 mg gcat−1 h−1).

Fig. 2. (a) Product distribution (bars), C3H8 and CO2 conversion (black and green points, respectively) for C3H8-coupled CO2 aromatization reaction over (a) Ga/ZSM-5@S-1 and (b) Ga/ZSM-5; pC3H8 : pCO2 : pN2 = 1 : 1 : 1, WHSV = 2.62 h−1; (c) schematic of the shell structure for coke suppression; (d) comparison of Ga/ZSM-5@S-1 with the established catalysts at their maximum reported catalytic performance; (e) UV-vis spectra of coke species over used catalysts.

Fig. 2

Open in a new tab

The role of epitaxial growth structures in high CO2 conversion

To uncover the mechanistic origin of the high CO2 conversion in Ga/ZSM-5@S-1, this study first sought to distinguish the respective activation sites for CO2 and C3H8. First, the activation sites were distinguished using in situ FTIR spectroscopy of C3H8 and CO2 after H/D exchange. As shown in Fig. 3a and S8, H/D exchange led to the formation of new –OD species.29,30 After D2 treatment, the broad band at 2670 cm−1, assigned to Ga–OD species,23,29,30 decreased significantly with increasing temperature during CO2 introduction (Fig. 3b), accompanied by the appearance of DCOO− species.31 This behavior indicated that CO2 was activated at Ga–O sites of (Ga2O2)2+ through consumption of D* species. In contrast, after introducing C3H8, the intensity of the Ga–OD peak remained nearly unchanged, while the bands at 3610 cm−1 (Si–OH–Al)32,33 and 3665 cm−1 (Ga–OH)30 both increased markedly (Fig. 3d). Since C3H8 is the only hydrogen source in this system, this result identified (Ga2O2)2+-BAS as the activation site for C3H8. The simultaneous decrease in the –CH3 (1471 cm−1) and –CH2– (1375 cm−1), assigned to alkane species,34,35 further confirmed the consumption of C3H8. Building on this foundation, Na+ exchange was employed to decrease the BAS content in Ga/ZSM-5@S-1,36 and the resulting changes in catalytic performance were examined (Fig. 3c). The experimental results validated the rationality of the above distinction of active sites. After the PCA reaction reached a steady state, the C3H8 conversion over Ga/Na1-ZSM-5@S-1 (a single cycle of Na+ exchange) decreased by ∼20%, and the BTX selectivity decreased by ∼23%, while the CO2 conversion remained nearly unchanged. In contrast, Ga/Na2-ZSM-5@S-1 (two cycles of Na+ exchange) exhibited almost negligible conversion of both C3H8 and CO2. These results indicated that a decrease in BAS abundance severely suppresses C3H8 conversion and BTX selectivity, further confirming that (Ga2O2)2+-BAS was responsible for C3H8 activation.37,38 In comparison, CO2 conversion was governed by the availability of H species in the system. When C3H8 conversion on Ga/Na2-ZSM-5@S-1 became extremely low, the amount of H2 generated was insufficient to sustain CO2 conversion, leading to a decline.

Fig. 3. (a) H/D exchange in situ FTIR spectra of Ga/ZSM-5@S-1, P = 1 atm, T = 30–600 °C, 57 °C between samples, pD2 : pN2 = 1 : 29, total flow rate = 30 ml min−1; in situ FTIR spectra of (b) CO2 and (d) C3H8 introduced into Ga/ZSM-5@S-1 after H/D exchange, P = 1 atm, pCO2/(pC3H8) : pN2 = 1 : 29, total flow rate: 30 mL min−1, T = 30–600 °C, 57 °C between samples; (c) comparison of the catalytic performance of the Ga/ZSM-5@S-1 catalyst prepared by Na exchange on ZSM-5 with different numbers of exchange cycles.

Fig. 3

Open in a new tab

The above experiments suggest that CO2 conversion is dictated by the concentration of hydrogen species in the system, which in turn governs the overall synergy of the coupled reactions.39 Increasing the H2 partial pressure led to a concomitant increase in both CO2 conversion and CO production rates (Fig. S9a and b), indicating that H species consumption via CO2 reduction was a primary reaction pathway.40,41 Conversely, the BTX production rate decreased (Fig. S9c), suggesting that excessive H2 concentrations may exacerbate undesirable side reactions, such as hydrogenolysis and cracking.42 Consequently, achieving this delicate H balance relies strictly on the efficient transfer of H2 between its source and sink. By varying the catalyst integration methods, we observed that catalysts Ⓐ and Ⓑ, which possessed larger active site spacings, exhibited lower CO2 conversion and BTX selectivity (Fig. 4a and Table S4). This suggested that excessive distance restricts H2 diffusion from Al–O–Si to (Ga2O2)2+ sites, thereby hindering its synergistic conversion with CO2. While physical mixing (catalyst Ⓒ) partially mitigated these diffusion limitations, its performance remained inferior to the catalyst Ⓓ (Ga/ZSM-5@S-1). Although physical mixing brought Ga/S-1 and Ga/ZSM-5 into proximity (Fig. S9d and e), it did not enforce a direct reaction sequence. In this random configuration, a fraction of the CO2 inevitably bypasses the Ga/S-1 domain and directly accesses the Ga/ZSM-5 sites, thereby allowing competitive adsorption to persist. This behavior can be understood from the intrinsic adsorption properties of the two reactants. Owing to its larger kinetic diameter and higher polarizability, C3H8 generally exhibits a much stronger thermodynamic propensity for adsorption.14,15 On the conventional Ga/ZSM-5 catalyst, breakthrough profiles revealed this unfavorable competition: C3H8 acted as the dominant adsorbate, aggressively displacing pre-adsorbed CO2 (Fig. 4b). This displacement gave rise to a pronounced “roll-off” feature in the CO2 signal,15,43 indicating that CO2 molecules were rapidly evicted from the active sites before undergoing effective activation. By contrast, introducing the Ga/S-1 outer region suppressed this unfavorable competitive adsorption. The Ga/S-1 layer created a tailored microenvironment that strengthened the interaction with CO2 (Fig. S7a and b), thereby significantly improving the resistance of adsorbed CO2 to displacement by C3H8. This enhanced affinity was evidenced by a broader CO2 uptake front, with the half-peak width prior to reaching adsorption equilibrium increasing from 75 s to 104 s (Fig. 4b and c). Consistently, the CO2 breakthrough time was delayed from 287 s to 300 s. Together, these results demonstrated that the Ga/S-1 region provided additional adsorption sites with preferential affinity for CO2.44,45 As a result, this architectural advantage maintained a higher local CO2 surface coverage under hydrocarbon-rich conditions, ensuring a sufficient reactant supply for subsequent CO2 conversion.

Fig. 4. (a) Tandem catalysts in different modes: Ⓐ (dual-bed): upper bed: Ga/S-1, lower bed: Ga/ZSM; Ⓑ: (dual-bed): upper bed: Ga/ZSM, lower bed: Ga/S-1; Ⓒ: (powder-mix): Ga/S-1 and Ga/ZSM-5 physically mixed; Ⓓ: Ga/ZSM-5@S-1 catalyst; breakthrough curve of (b) Ga/ZSM-5 and (c) Ga/ZSM-5@S-1 on a 1 : 1 C3H8/CO2 mixture; (d) CO production rates and CO2 conversion rate on Ga/S-1 and Ga/ZSM-5; (e) BTX production rates and C3H8 conversion rate on Ga/S-1 and Ga/ZSM-5. C3H8 : CO2 : N2 = 1 : 1, total flow rate = 5.23 mL min−1, T = 550 °C; (f) AIMD simulation; (g) reaction energy diagrams of the H* coupling path and CO2-assisted H* extraction path on the Ga/ZSM-5@S-1 catalyst.

Fig. 4

Open in a new tab

Beyond mitigating competitive adsorption, the ordered arrangement of active sites also promoted more efficient H2 utilization by establishing a directed tandem pathway. The random spatial distribution led to a significant portion of the in situ generated H2 being lost to parallel pathways, such as hydrocracking, rather than being efficiently intercepted by CO2 at the intended Ga/S-1 site. By contrast, the ordered tandem structure of the catalyst Ⓓ physically shielded the internal Ga/ZSM-5 from direct CO2 exposure. This structural confinement dictated that the abundant H2 species generated during aromatization must permeate through the outer Ga/S-1 shell, ensuring their efficient consumption by CO2 before escaping the catalyst. As shown in Fig. 4d, Ga/S-1 preferentially activates CO2, yielding a higher proportion of CO. Conversely, Ga/ZSM-5 is more active for C3H8 conversion, resulting in higher BTX selectivity (Fig. 4e). Ultimately, the additional (Ga2O2)2+ active sites anchored on S-1 expand the potential for CO2 conversion.

In addition to H2 diffusion and dissociation, the active hydrogen species (H*) supplied for CO2 conversion on (Ga2O2)2+ may also stem from the surface diffusion of –OH species within the system.

At the reaction temperature of 550 °C, AIMD simulations captured the dynamic migration of H* from the BAS to Ga–O sites to form stable Ga–OH species (Fig. 4f and S10b). DFT calculations further supported these findings, yielding a modest activation barrier of 1.06 eV for the migration step (Fig. 4g). DFT calculations (Fig. 4g) revealed that the CO2-assisted H* extraction pathway proceeded with significantly lower reaction barriers (C Created by potrace 1.16, written by Peter Selinger 2001-2019 O bond cleavage: Ea = 0.80 eV; H* removal: Ea = 1.32 eV). This pathway involved initial CO2 dissociation to CO* and O*, followed by O* capturing H* to form –OH.46 Subsequently, a low-energy dehydration step removed the –OH group, thereby regenerating the active site.

Another clue to explain the high CO2 conversion is its direct participation in the aromatization cycle. Kinetic experiments revealed that increasing the partial pressure of CO2 markedly accelerated the formation rates of both CO and BTX (Fig. 5a and b). This suggested that, in addition to driving the reverse water–gas shift (RWGS) reaction, CO2 actively participated in the aromatization cycle, likely by generating oxygenated intermediates. Based on experimental observations, we hypothesized that, within the Ga/ZSM-5@S-1 catalyst, hydrogen species migrate outward and are consumed in the Ga/S-1 layer, whereas CO2 is activated upon passing through this region, and the resulting reactive oxygen species subsequently diffuse inward. Because a hydrocarbon pool composed of bulky aromatization intermediates was formed at the BAS of the inner Ga/ZSM-5 domain, these inwardly diffusing oxygen-containing species can enter the hydrocarbon pool and participate in the aromatization cycle. To substantiate this spatially coupled insertion mechanism, we utilized GC-MS, in situ MS, and FTIR to track the evolution of oxygen-containing intermediates. As shown in Fig. 5c and g, the fragment peak intensities for 2-cyclopenten-1-one, 2,2,5,5-tetramethyl-3-cyclopenten-1-one, and 4,5-dihydro-2(1H)-pentalenone increased concurrently with rising CO2 partial pressure. Since CO2 is the sole oxygen source in the system, this observation indicated that it generated cyclic oxygenated species via a complex pathway involving insertion into hydrocarbon chains, followed by dehydration and condensation steps.3 In addition, semi-quantitative GC-MS analysis (Fig. S10) showed that, under reaction conditions, the concentrations of several detectable oxygen-containing intermediates were higher on Ga/ZSM-5@S-1 than on Ga/ZSM-5, supporting the proposed spatially coupled insertion mechanism.

Fig. 5. (a) CO production rate and (b) BTX production rate of Ga/ZSM-5@S-1 with the change in partial pressure of CO2, 0.05 g catalyst, T = 550 °C, P = 1 atm, pC3H8 : pCO2 = 10 : x, (x = 5, 10, 15, 20, 25), N2 as an equilibrium gas, total flow rate: 40 mL min−1; (c) some intermediate species content with the change in partial pressure of CO2; (d) in situ MS of Ga/ZSM-5@S-1 at 550 °C, p = 1 atm, pC3H8 : pCO2 : pN2 = 1 : 1 : 10, total flow rate: 30 mL min−1; (e) in situ FTIR of Ga/ZSM-5@S-1 at varying temperature (100–550 °C), 50 °C between samples, p = 1 atm, C3H8 : CO2 : N2 = 1 : 1 : 10, total flow rate: 30 mL min−1; (f) schematic diagram of CO2-promoted aromatic hydrocarbon recycling; (g) GC-MS results for the retained species during the coupling reaction at different temperatures, 0.05 g catalyst, T = 550 °C, P = 1 atm, pC3H8 : pCO2 = 1 : 1 : 1, total flow rate: 15 mL min−1.

Fig. 5

Open in a new tab

In situ MS, 13C isotope tracing experiments, and GC-MS measurements provided further insight into how CO2 promoted the aromatic cycle. The abundance of specific species was observed to rise sequentially with reaction temperature (Fig. 5d and g): the initial detection of pentenoic acid (141 °C) suggested that CO2 first inserts into long-chain hydrocarbons. Subsequently, the appearance of cyclopentanone species (2-cyclopenten-1-one, methylcyclopentanone, and 2,2,5,5-tetramethyl-3-cyclopenten-1-one: about 165 °C, 4,5-dihydro-2(1H)-pentalenone: 368 °C) confirmed that these oxygenated hydrocarbons underwent ring closure via condensation reactions, ultimately leading to the formation of aromatic products (benzene: 388 °C, toluene: 408 °C). By comparing MS spectra from reactions of C3H8 with 13CO2 and 12CO2,47,48 this study determined the 13C distribution in representative products (Fig. S11); the presence of 13C in benzene and toluene provided evidence for the incorporation of carbon from CO2. This transformation pathway was corroborated by in situ FTIR analysis (Fig. 5e). With increasing temperature, the peaks at 1745 cm−1 and 1689 cm−1 intensified, corresponding to lactone and unsaturated ketone species,47,49–51 respectively. The peak at 1654 cm−1 was assigned to C Created by potrace 1.16, written by Peter Selinger 2001-2019 C stretching vibrations, confirming the presence of olefinic products.47 Furthermore, the signal at 1604 cm−1 indicated the formation of aromatic carbenium ions and the cyclic coke precursor,52,53 while bands at 1485 cm−1 and 1386 cm−1 were attributed to carboxylate (COO−) species, providing direct evidence of CO2 insertion.34,51,54 A simplified schematic of CO2 incorporation into the aromatic cycle via oxygenated intermediates is shown in Fig. 5f.

Conclusions

In summary, a spatially engineered Ga/ZSM-5@S-1 catalyst was synthesized via epitaxial growth, followed by impregnation and H2 treatment. XPS sputtering, in situ XANES, and WAXS confirmed the successful construction of a catalyst with a Ga/S-1 exterior and a Ga/ZSM-5 interior. In C3H8-coupled CO2 aromatization, the Ga/ZSM-5@S-1 catalyst exhibited outstanding performance, delivering a CO2 conversion of ∼75% together with a BTX selectivity of ∼73%, while also showing remarkable resistance to coke formation, with a coking rate of only 2.71 mg gcat−1 h−1.

The additional (Ga2O2)2+ active sites anchored on S-1 are key to achieving high CO2 conversion. In situ FTIR and isotope-labeled kinetic experiments revealed a synergistic mechanism in which the outer Ga/S-1 region promoted CO2 conversion, while the inner Ga/ZSM-5 region drove C3H8 aromatization. This structure introduced additional CO2 adsorption sites in the outer Ga/S-1 region. As evidenced by adsorption breakthrough and TPSR results, Ga/ZSM-5@S-1 alleviated competitive adsorption between CO2 and C3H8. More importantly, it established an ordered reaction pathway that enabled the efficient utilization of in situ generated hydrogen species. Both DFT calculations and AIMD simulations consistently indicate that H* preferentially accumulates at (Ga2O2)2+ sites, which exhibit a lower energy barrier for CO2-assisted H* consumption, thereby promoting CO2 conversion and suppressing hydrogenation side reactions. Beyond hydrogen species consumption, GC-MS, in situ MS, 13C isotope tracing, and in situ FTIR further demonstrated that CO2 played an additional role: it inserted into aliphatic hydrocarbons to form oxygenated intermediates that participated in the aromatic cycle, thereby promoting BTX formation.

Overall, this study demonstrates efficient CO2 conversion and high BTX selectivity by introducing (Ga2O2)2+/S-1 active sites via epitaxial growth, offering insights for the design of bifunctional catalysts to enhance CO2 utilization.

Author contributions

L. Y. and Y. Y. share first authorship. L. Y. and W. S. conceived the idea. L. Y. carried out catalyst synthesis and performance evaluations. Y. Y. performed the computational studies and advanced spectroscopic analyses. K. Y., G. L., Y. F., and M. J. aided in experimental data collection. X. J., Y. X., K. Y. and Z. Z. provided intellectual insights and data interpretation. J. L., K. Y. K. and W. S. supervised the project and secured funding. L. Y. and Y. Y. drafted the manuscript, which was extensively reviewed and finalized by J. L. and W. S. All authors have approved the final version.

Conflicts of interest

The authors declare no competing interests.

Supplementary Material

SC-OLF-D6SC06755G-s001

Acknowledgments

We gratefully acknowledge the financial support from the Natural Science Foundation of Xinjiang Uygur Autonomous Region (2026D01E68), National Natural Science Foundation of China (22668063), Xinjiang Uygur Autonomous Region Key S&T Special Project (2025A01008-2), Frontier Interdisciplinary Exploration Research Program of China University of Petroleum, Beijing (2462025XKQY006), Research Foundation of China University of Petroleum-Beijing at Karamay (XQZX20260013).

Data availability

All data generated or analysed during this study are included in this article and its supplementary information (SI) files. Supplementary information: detailed catalyst preparation, characterisation, catalytic-performance evaluation and computational methods, together with experimental and computational data, including in situ spectroscopic, isotope-labelling, kinetic, DFT and AIMD results, as well as Fig. S1–S17 and Tables S1–S6. See DOI: https://doi.org/10.1039/d6sc06755g.

References

  1. Wang D. Xie Z. Porosoff M. D. Chen J. G. Recent advances in carbon dioxide hydrogenation to produce olefins and aromatics. Chem. 2021;7:2277–2311. [Google Scholar]
  2. Biswas A. N. Xie Z. Chen J. G. Can CO2-assisted alkane dehydrogenation lead to negative CO2 emissions? Joule. 2022;6:269–273. doi: 10.1016/j.joule.2021.12.008. [DOI] [Google Scholar]
  3. Xie J. Olsbye U. The Oxygenate-Mediated Conversion of COx to Hydrocarbons-On the Role of Zeolites in Tandem Catalysis. Chem. Rev. 2023;123:11775–11816. doi: 10.1021/acs.chemrev.3c00058. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Wang J. Li R. Zhang G. Dong C. Fan Y. Yang S. Chen M. Guo X. Mu R. Ning Y. Li M. Fu Q. Bao X. Confinement-Induced Indium Oxide Nanolayers Formed on Oxide Support for Enhanced CO(2) Hydrogenation Reaction. J. Am. Chem. Soc. 2024;146:5523–5531. doi: 10.1021/jacs.3c13355. [DOI] [PubMed] [Google Scholar]
  5. Zhang B. Huang S. Li Y. Shen J. Tian X. Ding M. Highly dispersed Cu on hollow spherical CeO2: an efficient and stable catalyst for the RWGS reaction. Appl. Catal., B. 2025;366:125003. doi: 10.1016/j.apcatb.2024.125003. [DOI] [Google Scholar]
  6. Li Y. Yuan J. Ji Y. Yan S. Liu J. Liu B. Liu X. Reversal of Selectivity to CO in CO2 Hydrogenation by Forming an InCo Alloy and H-ZSM-5 Zeolite-Driven Suppression of Cobalt Oxidation Deactivation in CO2 Methanation Reaction. ACS Catal. 2025:19823–19839. doi: 10.1021/acscatal.5c06917. [DOI] [Google Scholar]
  7. West T. Gold grafting for CO2 methanation. Nat. Synth. 2023;2:707. doi: 10.1038/s44160-023-00380-w. [DOI] [Google Scholar]
  8. Xu J. Wang K. Zhao M. Zhang R. Cui W. Liu L. Wang X. Wang J. Song S. Zhang H. Boosting CO2 Hydrogenation by Synergistic Incorporation of Pure Silica Silicalite-1 Zeolite and CeO2 into Cu Catalysts. Angew. Chem., Int. Ed. 2025;64:e202423438. doi: 10.1002/anie.202423438. [DOI] [PubMed] [Google Scholar]
  9. Han X. Li M. Hao Z. Chen J. Tang S. Zhang Z. Xiao T. Jia W. Chen M. Gao P. Bai Y. Wang S. Xi S. Lim K. H. Wang Y. Kawi S. Ma X. Unraveling the Role of Atomically Dispersed Ga Species in the Selective Hydrogenation of CO2 to Methanol over Cu/ZrO2. ACS Catal. 2025;15:15141–15152. doi: 10.1021/acscatal.5c02841. [DOI] [Google Scholar]
  10. Chang H. Gao F. Ma S. Zhu Y. Liu Z. Liu J. He H. Zhang K. Liu Y. Cao Y. Unlocking Methanol Synthesis from CO2 and H2 on ZnO/ZrO2 Catalysts: Surface Hydroxyl-Mediated Activation. ACS Catal. 2025:6005–6017. doi: 10.1021/acscatal.5c01585. [DOI] [Google Scholar]
  11. Li C. Fang X. Li B. Yan S. Chen Z. Yang L. Hao S. Liu H. Liu J. Zhu W. Efficient carbon integration of CO2 in propane aromatization over acidic zeolites. Chin. J. Catal. 2025;72:314–322. doi: 10.1016/S1872-2067(25)64680-8. [DOI] [Google Scholar]
  12. Feng H. Hu Z.-P. Chen E. Wu Y. Li M. Han J. Wei Y. Liu Z. Dynamic Interconversion of Ga(III) and GaHx in Ga-MFI Zeolites for Efficient CO2-Cyclohexane Coupling Conversion to Aromatics. J. Am. Chem. Soc. 2025;147:32915–32923. doi: 10.1021/jacs.5c09737. [DOI] [PubMed] [Google Scholar]
  13. Fan G. Liu R. Zhao Y. Chen E. Yan S. Han J. Liu J. Yu Z. Liu Z. CO2-enhanced alkane aromatization over Cu-ZSM-5 zeolite: insights into active sites and catalytic performance. J. Energy Chem. 2025;106:600–607. doi: 10.1016/j.jechem.2025.02.048. [DOI] [Google Scholar]
  14. Zhai P. Xie Z. Huang E. Aireddy D. R. Yu H. Cullen D. A. Liu P. Chen J. G. Ding K. CO2-mediated oxidative dehydrogenation of propane enabled by Pt-based bimetallic catalysts. Chem. 2023;9:3268–3285. [Google Scholar]
  15. Sun Y.-C. He Z.-H. Wang S.-W. Zhu Z.-J. Wang K. Wang W. Liu J. Wang H. Yang Y. Liu Z.-T. Boosting CO2 oxidative propane dehydrogenation via separate activation CO2 and propane by in-framework and extra-framework Co species on ZSM-5 zeolite. Appl. Catal., B. 2025;373:125324. doi: 10.1016/j.apcatb.2025.125324. [DOI] [Google Scholar]
  16. Zhang X. Shen Q. Li X. Wen L. Sun J. Liu J. Liao C. Liu S. Jiang G. Liu R. Reactive Hydrogen Species Behaviors on Pd/TiN: in Situ SERS Guided Regulation for Chemoselective Hydrogenation. Angew. Chem., Int. Ed. 2025:e202503279. doi: 10.1002/anie.202503279. [DOI] [PubMed] [Google Scholar]
  17. Chen X. Ren D. Yan H. Li Z. Zhang X. Luan B. Wang Y. Shi H. Jiang H. Hu Y. Hu X. Zhang A. Dong M. Zhou X. Feng X. Liu Y. Yang C. CO2-Assisted in situ hydrogen extraction for highly selective aromatization of n-Hexane over Zn modified HZSM-5 catalyst. Fuel. 2023;349:128712. doi: 10.1016/j.fuel.2023.128712. [DOI] [Google Scholar]
  18. Fang R. Xu L. Cui Y. Wang Q. Qin Y. Xu Y. Zhang Q. Wen Y. Zhao J. Feng F. Sun T. Xu X. Zhu Y. Lu C. Zhang Q. Lin L. Li X. Liquid Coordination Environment-Induced Liquid-Like Metal Behavior: Mobile Single-Atom Copper Catalytic Centers. J. Am. Chem. Soc. 2026;148:11215–11227. doi: 10.1021/jacs.5c22819. [DOI] [PubMed] [Google Scholar]
  19. Nowak I. Effect of H2-O2 pre-treatments on the state of gallium in Ga/H-ZSM-5 propane aromatisation catalysts. Appl. Catal., A. 2003;251:107–120. doi: 10.1016/S0926-860X(03)00299-0. [DOI] [Google Scholar]
  20. Yan Y. Zhang Z. Bak S.-M. Yao S. Hu X. Shadike Z. Do-Thanh C.-L. Zhang F. Chen H. Lyu X. Chen K. Zhu Y. Lu X. Ouyang P. Fu J. Dai S. Confinement of Ultrasmall Cobalt Oxide Clusters within Silicalite-1 Crystals for Efficient Conversion of Fructose into Methyl Lactate. ACS Catal. 2018;9:1923–1930. doi: 10.1021/acscatal.8b03230. [DOI] [Google Scholar]
  21. Lv X. Yang M. Song S. Xia M. Li J. Wei Y. Xu C. Song W. Liu J. Boosting Propane Dehydrogenation by the Regioselective Distribution of Subnanometric CoO Clusters in MFI Zeolite Nanosheets. ACS Appl. Mater. Interfaces. 2023;15:14250–14260. doi: 10.1021/acsami.2c21076. [DOI] [PubMed] [Google Scholar]
  22. Bhan A. Nicholas Delgass W. Propane Aromatization over HZSM-5 and Ga/HZSM-5 Catalysts. Catal. Rev. 2008;50:19–151. doi: 10.1080/01614940701804745. [DOI] [Google Scholar]
  23. Yuan Y. Brady C. Annamalai L. Lobo R. F. Xu B. Ga speciation in Ga/H-ZSM-5 by in situ transmission FTIR spectroscopy. J. Catal. 2021;393:60–69. doi: 10.1016/j.jcat.2020.11.004. [DOI] [Google Scholar]
  24. Yuan Y. Brady C. Lobo R. F. Xu B. Understanding the Correlation between Ga Speciation and Propane Dehydrogenation Activity on Ga/H-ZSM-5 Catalysts. ACS Catal. 2021;11:10647–10659. doi: 10.1021/acscatal.1c01497. [DOI] [Google Scholar]
  25. Rodrigues V. d. O. Faro Júnior A. C. On catalyst activation and reaction mechanisms in propane aromatization on Ga/HZSM5 catalysts. Appl. Catal., A. 2012;435–436:68–77. doi: 10.1016/j.apcata.2012.05.036. [DOI] [Google Scholar]
  26. Ma P. Zhou H. Li Y. Wang M. Nastase S. A. F. Zhu M. Cui J. Cavallo L. Cheng K. Dutta Chowdhury A. Selectivity descriptors of the catalytic n-hexane cracking process over 10-membered ring zeolites. Chem. Sci. 2024;15:11937–11945. doi: 10.1039/D4SC00603H. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Nordvang E. C. Borodina E. Ruiz-Martínez J. Fehrmann R. Weckhuysen B. M. Effects of Coke Deposits on the Catalytic Performance of Large Zeolite H-ZSM-5 Crystals during Alcohol-to-Hydrocarbon Reactions as Investigated by a Combination of Optical Spectroscopy and Microscopy. Chem. – Eur. J. 2015;21:17324–17335. doi: 10.1002/chem.201503136. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Palumbo L. Bonino F. Beato P. Bjørgen M. Zecchina A. Bordiga S. Conversion of Methanol to Hydrocarbons: Spectroscopic Characterization of Carbonaceous Species Formed over H-ZSM-5. J. Phys. Chem. C. 2008;112:9710–9716. doi: 10.1021/jp800762v. [DOI] [Google Scholar]
  29. Yuan Y. Lobo R. F. Xu B. Ga2O22+ Stabilized by Paired Framework Al Atoms in MFI: A Highly Reactive Site in Nonoxidative Propane Dehydrogenation. ACS Catal. 2022;12:1775–1783. doi: 10.1021/acscatal.1c05724. [DOI] [Google Scholar]
  30. Yang C. Ma S. Liu Y. Wang L. Yuan D. Shao W.-P. Zhang L. Yang F. Lin T. Ding H. He H. Liu Z.-P. Cao Y. Zhu Y. Bao X. Homolytic H2 dissociation for enhanced hydrogenation catalysis on oxides. Nat. Commun. 2024;15:540. doi: 10.1038/s41467-024-44711-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Bai B. Ye Y. Jiao F. Xiao J. Pan Y. Cai Z. Chen M. Pan X. Bao X. Surface Structure Dependent Activation of Hydrogen over Metal Oxides during Syngas Conversion. J. Am. Chem. Soc. 2024;146:34909–34915. doi: 10.1021/jacs.4c14395. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Osuga R. Yokoi T. Kondo J. N. IR observation of activated ether species on acidic OH groups on H-ZSM-5 zeolites. Mol. Catal. 2019;477:110535. [Google Scholar]
  33. Yu Q. Sun H. Sun H. Li L. Zhu X. Ren S. Guo Q. Shen B. Highly mesoporous IM-5 zeolite prepared by alkaline treatment and its catalytic cracking performance. Microporous Mesoporous Mater. 2019;273:297–306. doi: 10.1016/j.micromeso.2018.08.016. [DOI] [Google Scholar]
  34. Song S. Chen X. Fo Y. Yang M. Su H. Yang K. Ji X. Lv X. Li Z. Wei Y. Huang G. Xu C. Liu J. Song W. Lewis acid–base modulation for efficient CO2 oxidative propane dehydrogenation: a case study of a La-modified binuclear Feoxo site. Chem Catal. 2023;3:100663. [Google Scholar]
  35. Kolyagin Y. G. Ordomsky V. V. Khimyak Y. Z. Rebrov A. I. Fajula F. Ivanova I. I. Initial stages of propane activation over Zn/MFI catalyst studied by in situ NMR and IR spectroscopic techniques. J. Catal. 2006;238:122–133. doi: 10.1016/j.jcat.2005.11.037. [DOI] [Google Scholar]
  36. Zhang A. Tao K. Wang X. Liu J. Dai Y. Dai B. Kong L. Zhou S. Pu X. Liu J. Li J. Li J. Evolution of active Fe species in ZSM-5 induced by charge-compensated Na ions during catalyst synthesis for cellulose pyrolysis. Chem. Eng. J. 2025;525:170776. doi: 10.1016/j.cej.2025.170776. [DOI] [Google Scholar]
  37. Iglesia E. Baumgartner J. E. Hydrogen transfer and activation of propane and methane on ZSM5-based catalysts. Catal. Lett. 1993;21:55–70. doi: 10.1007/BF00767371. [DOI] [Google Scholar]
  38. Biscardi J. A. Iglesia E. Structure and function of metal cations in light alkane reactions catalyzed by modified H-ZSM5. Catal. Today. 1996;31:207–231. doi: 10.1016/S0920-5861(96)00028-4. [DOI] [Google Scholar]
  39. Wu X. Wang C. Zhao S. Wang Y. Zhang T. Yao J. Gao W. Zhang B. Arakawa T. He Y. Chen F. Tan M. Yang G. Tsubaki N. Dual-engine-driven realizing high-yield synthesis of Para-Xylene directly from CO2-containing syngas. Nat. Commun. 2024;15:8064. doi: 10.1038/s41467-024-52482-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Gomez E. Nie X. Lee J. H. Xie Z. Chen J. G. Tandem Reactions of CO2 Reduction and Ethane Aromatization. J. Am. Chem. Soc. 2019;141:17771–17782. doi: 10.1021/jacs.9b08538. [DOI] [PubMed] [Google Scholar]
  41. Tu C. Fan H. Wang D. Rui N. Du Y. Senanayake S. D. Xie Z. Nie X. Chen J. G. CO2-assisted ethane aromatization over zinc and phosphorous modified ZSM-5 catalysts. Appl. Catal., B. 2022;304:120956. doi: 10.1016/j.apcatb.2021.120956. [DOI] [Google Scholar]
  42. Aguayo A. T. Castaño P. Mier D. Gayubo A. G. Olazar M. Bilbao J. Effect of Cofeeding Butane with Methanol on the Deactivation by Coke of a HZSM-5 Zeolite Catalyst. Ind. Eng. Chem. Res. 2011;50:9980–9988. doi: 10.1021/ie200946n. [DOI] [Google Scholar]
  43. Yu X. Li B. Zhang B. Wang J. Shi X. Li L. Pore channel regulation via organic amine-modified 13X zeolite for CO2 capture from C2H2. J. Environ. Chem. Eng. 2025;13:119627. doi: 10.1016/j.jece.2025.119627. [DOI] [Google Scholar]
  44. Wang C.-T. Li W.-C. Wang M. Zhao G.-H. Liu R.-S. Hao G.-P. Lu A.-H. Wood-structured carbon with reassembled pores showing high propylene adsorption rate for efficient separation of propylene/propane. Sep. Purif. Technol. 2025;354:128649. doi: 10.1016/j.seppur.2024.128649. [DOI] [Google Scholar]
  45. Cui Y. Su W. Xing Y. Hao L. Sun Y. Cai Y. Experimental and simulation evaluation of CO2/CO separation under different component ratios in blast furnace gas on zeolites. Chem. Eng. J. 2023;472:144579. doi: 10.1016/j.cej.2023.144579. [DOI] [Google Scholar]
  46. Yang L. Yang Y. Yang M. Fo Y. Ji X. Yang K. Liu G. Xia Y. Zhao Z. Liu J. Song W. Carbon Dioxide as a Hydrogen-Transport Modulator-Disrupting Surface Hydrogen Accumulation during Propane Aromatization. ACS Catal. 2025;15:16771–16781. doi: 10.1021/acscatal.5c03541. [DOI] [Google Scholar]
  47. Yang K. Li J. Wei C. Zhao Z. Liu Z. Coupling Conversion of CO2 and n-Butane Over Modified ZSM-5: Incorporation of the Carbon from CO2 into Hydrocarbon Products. ACS Catal. 2023;13:10405–10417. doi: 10.1021/acscatal.3c02158. [DOI] [Google Scholar]
  48. Wei C. Yu Q. Li J. Liu Z. Coupling Conversion of n-Hexane and CO over an HZSM-5 Zeolite: Tuning the H/C Balance and Achieving High Aromatic Selectivity. ACS Catal. 2020;10:4171–4180. doi: 10.1021/acscatal.9b05619. [DOI] [Google Scholar]
  49. Lievens C. Mourant D. He M. Gunawan R. Li X. Li C.-Z. An FT-IR spectroscopic study of carbonyl functionalities in bio-oils. Fuel. 2011;90:3417–3423. doi: 10.1016/j.fuel.2011.06.001. [DOI] [Google Scholar]
  50. Luzgin M. Propane carbonylation on sulfated zirconia catalyst as studied by 13C MAS NMR and FTIR spectroscopy. J. Catal. 2004;223:290–295. doi: 10.1016/j.jcat.2004.01.034. [DOI] [Google Scholar]
  51. Blasco T. Boronat M. Concepción P. Corma A. Law D. Vidal-Moya J. A. Carbonylation of Methanol on Metal–Acid Zeolites: Evidence for a Mechanism Involving a Multisite Active Center. Angew. Chem., Int. Ed. 2007;46:3938–3941. doi: 10.1002/anie.200700029. [DOI] [PubMed] [Google Scholar]
  52. Zhu D. Wang Z. Meng F. Zhao B. Kanitkar S. Tang Y. Catalytic Conversion of Chloromethane to Olefins and Aromatics Over Zeolite Catalysts. Catal. Lett. 2020;151:1038–1048. doi: 10.1007/s10562-020-03364-z. [DOI] [Google Scholar]
  53. Zhang J. Zhou A. Gawande K. Li G. Shang S. Dai C. Fan W. Han Y. Song C. Ren L. Zhang A. Guo X. b-Axis-Oriented ZSM-5 Nanosheets for Efficient Alkylation of Benzene with Methanol: Synergy of Acid Sites and Diffusion. ACS Catal. 2023;13:3794–3805. doi: 10.1021/acscatal.2c06384. [DOI] [Google Scholar]
  54. Zhang Z. Tian H. Sun J. Meira D. M. Zhang M. Ding X. Ji D. Qiu C. Lu Z. Sun L. Zhang Y. Tu W. Zhou Y. Yang X. Howe J. Wang L. Tong S.-Y. Zou Z. Light-enabled coupling of tandem ethane dehydrogenation and CO2 hydrogenation. Chem Catal. 2023;3:100644. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

SC-OLF-D6SC06755G-s001
SC-OLF-D6SC06755G-s001.pdf (2.9MB, pdf)

Data Availability Statement

All data generated or analysed during this study are included in this article and its supplementary information (SI) files. Supplementary information: detailed catalyst preparation, characterisation, catalytic-performance evaluation and computational methods, together with experimental and computational data, including in situ spectroscopic, isotope-labelling, kinetic, DFT and AIMD results, as well as Fig. S1–S17 and Tables S1–S6. See DOI: https://doi.org/10.1039/d6sc06755g.

Articles from Chemical Science are provided here courtesy of Royal Society of Chemistry

RESOURCES