# Fileset

[ChemSusChem - 2025 - Jin - Superior Performance and Catalytic Mechanism of an Icosahedral Quasicrystal Al‐Cu‐Fe in CO2 (2).pdf](https://mdr.nims.go.jp/filesets/4071ab78-d75b-4489-97ee-cde706c99249/download)

## Creator

Huixin Jin, [Wenyang Zhang](https://orcid.org/0000-0001-7957-0708), Haruka Yoshikawa, Asuka Ishikawa, Farid Labib, Boya Zhang, Jintao Zhou, Rongkai Kang, Jingyu Qin, Jianxin Zhang, [Ya Xu](https://orcid.org/0000-0001-9067-5244), [Ryuji Tamura](https://orcid.org/0000-0001-8589-4311)

## Rights

[Creative Commons BY-NC-ND Attribution-NonCommercial-NoDerivs 4.0 International](https://creativecommons.org/licenses/by-nc-nd/4.0/)

## Other metadata

[Superior Performance and Catalytic Mechanism of an Icosahedral Quasicrystal Al‐Cu‐Fe in CO                    <sub>2</sub>                    Reduction to CO](https://mdr.nims.go.jp/datasets/9ebc5a50-eda0-4e7d-aec5-cb0c8faf73b2)

## Fulltext

Superior Performance and Catalytic Mechanism of an Icosahedral Quasicrystal Al‐Cu‐Fe in CO2 Reduction to COwww.chemsuschem.orgSuperior Performance and Catalytic Mechanism of anIcosahedral Quasicrystal Al-Cu-Fe in CO2 Reduction to COHuixin Jin, Wenyang Zhang,* Haruka Yoshikawa, Asuka Ishikawa, Farid Labib, Boya Zhang,Jintao Zhou, Rongkai Kang, Jingyu Qin, Jianxin Zhang, Ya Xu,* and Ryuji Tamura*Dissimilar to long-range periodic crystals, quasicrystals featurelong-range aperiodic order and noncrystallographic rotationalsymmetry. These structural peculiarities and complexities endowit with unique properties and untold research values, togetherwith significant challenges in decoding the structure–propertyrelationship. Herein, the application of icosahedral quasicrystal(iQC) Al63Cu25Fe12 as a high-performance catalyst is investigatedfor reducing CO2 to CO, known as the reverse water–gas shiftreaction, which is a key reaction for producing useful chemicals.Compared with the samples with similar compositions but differ-ent structures, it shows superior CO2 conversion rate and COselectivity. Combined with density functional theory calculations,the origin of the high catalytic activity of iQC Al63Cu25Fe12 hasbeen deciphered. It is found that for the approximant crystalof the quasicrystal, the reaction tends to occur near the icosahe-dral cluster. Compared to other control groups, it exhibits muchlower reaction barriers during CO2 reduction to CO. This demon-strates that the high catalytic performance of iQC Al63Cu25Fe12stems from its internally rich icosahedral cluster content. Theresearch exemplifies the potential of quasicrystals in the fieldof catalysis and sheds light on the structure–property relationshipof complex structures.1. IntroductionQuasicrystals, characterized by their unique aperiodic order,exhibit a fascinating structural complexity that sets them apartfrom conventional crystalline materials. Unlike traditional crystals,which possess a repeating unit cell and exhibit long-range peri-odicity, quasicrystals showcase a well-defined, yet nonrepetitivearrangement of atoms, together with unusual rotational symme-tries, such as fivefold or eightfold symmetry, which are not foundin classical crystallography. The distinctive structural characteristicsof quasicrystals provides unique properties and lucrative opportu-nities that are unobtainable elsewhere, leading to the develop-ment of multifarious applications, such as superconductivity,[1–3]magnetism,[4–7] hydrogen storage,[8] and corrosion and oxidationresistance.[9,10] However, past study on quasicrystals is ratherinadequate. As for the application in catalysis, researchers havemostly focused on using quasicrystals as a substrate, with thecatalytic centers being other materials attached to the surface,and the reactions involved are also quite limited. For example,Tsai and his coworkers[11–14] did a series of studies on the applica-tion of quasicrystals in catalysis and found that leachates obtainedafter etching the Al-Cu-Fe quasicrystal with an alkaline solutionshowed high catalytic activity and the excellent thermal stabilityin steam reforming of methanol. Research on using quasicrystalsthemselves as catalysts in various catalytic reactions is still farfrom sufficient.The missing research pertains to the structural peculiarities andcomplexities of quasicrystals, especially the non-3D periodicity, hasalways been the formidable challenge that dissuades researchersfrom further investigation. To date, the most extensive and in-depth research aimed at uncovering structure–property relation-ships is limited to the low-index planes of simple structures.[15,16]The study on intricate systems, such as complex intermetallic com-pounds, whose unit cells are quite large, is sorely missing.[17]Quasicrystals, a special type of complex intermetallic compoundfeaturing infinitely large unit cells, face the insurmountable chal-lenge of deciphering their structure with theoretical approachesH. Jin, J. Zhou, R. TamuraDepartment of Materials Science and TechnologyTokyo University of ScienceTokyo 125-8585, JapanE-mail: tamura@rs.tus.ac.jpW. ZhangKagami Memorial Research Institute for Materials Science and TechnologyWaseda University2-8-26 Nishiwaseda, Shinjuku, Tokyo 169-0051, JapanE-mail: w.iac23290@kurenai.waseda.jpH. Yoshikawa, A. Ishikawa, F. LabibResearch Institute of Science and TechnologyTokyo University of ScienceTokyo 125-8585, JapanB. Zhang, R. Kang, J. Qin, J. ZhangSchool of Materials Science and EngineeringShandong UniversityNo. 17923, Jingshi Road, Jinan 250061, P. R. ChinaY. XuResearch Center for Energy and Environmental MaterialsNational Institute for Materials Science3-13 Sakura, Tsukuba, Ibaraki 305-0003, JapanE-mail: XU.Ya@nims.go.jpSupporting information for this article is available on the WWW under https://doi.org/10.1002/cssc.202501424© 2025 The Author(s). ChemSusChem published by Wiley-VCH GmbH. This isan open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in anymedium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.ChemSusChem 2025, 18, e202501424 (1 of 11) © 2025 The Author(s). ChemSusChem published by Wiley-VCH GmbHChemSusChemResearch Articledoi.org/10.1002/cssc.202501424http://www.chemsuschem.orghttps://orcid.org/0000-0001-7957-0708https://orcid.org/0000-0001-7957-0708https://orcid.org/0000-0001-9067-5244https://orcid.org/0000-0001-9067-5244https://orcid.org/0000-0001-8589-4311https://orcid.org/0000-0001-8589-4311mailto:tamura@rs.tus.ac.jpmailto:w.iac23290@kurenai.waseda.jpmailto:XU.Ya@nims.go.jphttp://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://doi.org/10.1002/cssc.202501424http://crossmark.crossref.org/dialog/?doi=10.1002%2Fcssc.202501424&domain=pdf&date_stamp=2025-10-28and study the structure–property relationships and underlyingmechanisms. Therefore, past research on the properties of quasi-crystals has mostly focused on comparative analysis of experimen-tal data, with little emphasis on theoretical analysis.Here, we report the icosahedral quasicrystal (iQC) Al63Cu25Fe12as a high-performance thermal catalyst for the reverse water–gasshift (RWGS) reaction. The RWGS reaction can effectively use CO2,a major contributor to the greenhouse effect and climate change,as a raw material to produce CO, which can then be convertedinto various value-added chemicals through subsequent Fischer–Tropsch synthesis.[18–22] In recent years, Cu- and Fe-containingcatalysts have been widely investigated[20,23–30] as promising alter-natives to noble metals (e.g., Pt) for the RWGS reaction due to theirfavorable activity-to-cost ratio. The iQC Al63Cu25Fe12 in this workshows a CO2 conversion rate of 43% at 500 °C and a weight hourlyspace velocity (WHSV) of 36 000mL g�1 h�1, much higher thanthose of other samples with similar compositions but differentstructures. The experimental characterizations demonstrate thatthe states and contents of different elements on the surfaces ofiQC Al63Cu25Fe12 and its control groups are generally similar, sug-gesting that the quasicrystal structure itself played a major role inthe catalysis. Combining with density functional theory (DFT)calculations,[17,31–33] lower reaction barriers of iQC Al63Cu25Fe12toward RWGS are uncovered. The reactions aremore likely to occurnear the icosahedral clusters abundant within the iQC Al63Cu25Fe12,thereby demonstrating that the quasicrystal structure itself con-tributes to the high catalytic activity of the reactions.2. Results2.1. Synthesis and CharacterizationsThe alloy of iQC Al63Cu25Fe12 was prepared by mixing pure metalsof Al, Cu, and Fe and melting it in an electric arc furnace under anargon atmosphere. After heat treatments, the pure phase ingot ofiQC Al63Cu25Fe12 was obtained. Under high-purity argon protec-tion, the ingot was milled for different durations using a planetaryball mill (Experimental Section). In addition to the iQCAl63Cu25Fe12 sample, two ternary complex intermetallic com-pounds (β-Al55Cu25Fe20 and ω-Al70Cu20Fe10) and two binary inter-metallic compounds (Al2Fe and θ-Al2Cu) were selected as controlgroups. Compared with the quasicrystal, these control groupsfeature similar contents of active elements (Cu and Fe) and totallydifferent crystal structures (Table 1, Supporting Information), pro-viding the necessary basis to uncover the mechanism by whichcrystal structure influences the RWGS catalytic reaction.X-ray diffraction (XRD) patterns were measured for iQCAl63Cu25Fe12, β-Al55Cu25Fe20, ω-Al70Cu20Fe10, Al2Fe, and θ-Al2Cubefore and after ball-milling to show the crystalline phases(Figure 1a–e). The XRD patterns of iQC Al63Cu25Fe12, β-Al55Cu25Fe20,ω-Al70Cu20Fe10, Al2Fe, and θ-Al2Cu before ball-milling are wellindexed to quasicrystal, cubic intermetallic (Pm-3m), tetragonalintermetallic (P4/mnc), triclinic intermetallic (P-1), and tetragonalintermetallic (I4/mcm). The standard powder diffraction file cardsare marked in the XRD patterns. The XRD patterns of all samplesbefore ball milling exhibit sharp peaks with no obvious impuritypeaks observed, indicating that these samples are single-phasewith good crystallinity. It can be clearly observed that the peaksof all samples after ball-milling are broadened compared to thosebefore ball-milling. It is the significant reduction in particle andgrain sizes caused by the ball milling processes that results inthe increase in the full width at half maximum of the diffractionpeaks. In addition, an impurity phase different from the matrixcan be observed on the XRD of θ-Al2Cu after ball milling, corre-sponding to the cubic Al4Cu9 intermetallic phase (P43m). This indi-cates that the binary intermetallic compound has lower wearresistance and stability compared to ternary complex intermetalliccompounds and quasicrystals, leading to phase separation duringthe ball-milling process.Taking iQC Al63Cu25Fe12 as an example, the scanning electronmicroscopy (SEM) images and particle size distribution histo-grams more intuitively reflect the changes in particle size withmilling time (Figure 1, Supporting Information). As the ball-millingtime increases, the particle size gradually decreases from tens ofmicrons to submicron levels. However, when the ball milling timereaches 8 h, extending the milling time can hardly reduce the par-ticle size any further, which is also evident from the graph show-ing particle size changes with ball-milling period (Figure 2,Supporting Information). For the other control groups, the sameball-milling process of 8 h was applied, and the resulting particlesizes are similar to that of iQC Al63Cu25Fe12, all at the submicronlevel (Figure 3, Supporting Information). The nitrogen adsorp-tion–desorption isotherm curves of iQC Al63Cu25Fe12 after differ-ent time of ball-milling processes and the correspondingcalculated specific surface areas in Figure 4, SupportingInformation, indicate that with the increase in ball-milling time,the specific surface areas of iQC Al63Cu25Fe12 also increases.However, after 8 h of ball milling, since further milling doesnot significantly reduce the particle size, the change in specificsurface area becomes less noticeable. This trend can be observedmore clearly in Figure 5, Supporting Information. Figure 6,Supporting Information, shows the nitrogen adsorption–desorption isotherm curves of β-Al55Cu25Fe20, ω-Al70Cu20Fe10,Al2Fe, and θ-Al2Cu before ball-milling for 8 h. After 8 h of ball-milling, iQC Al63Cu25Fe12, β-Al55Cu25Fe20, ω-Al70Cu20Fe10, Al2Fe, andθ-Al2Cu exhibit similar specific surface areas (Figure 7, SupportingInformation), and the catalyst performance variation induced bydifferences in specific surface area is accordingly excluded.The morphology and crystal structures of all the samples werefurther characterized by high resolution transmission electronmicroscopy (HRTEM). Figure 1f–i presents the TEM results ofiQC Al63Cu25Fe12 before ball milling, where f–h show the selectedarea electron diffraction (SAED) patterns and correspondingHRTEM images at specific locations, respectively. The incidentaxes correspond to the quasicrystal’s twofold, threefold, andfivefold rotational symmetry. These results further confirmthe quasicrystal structure of the iQC Al63Cu25Fe12. Figure 1ishows a locally magnified HRTEM image along the fivefoldsymmetry axis, revealing a richness of the icosahedral atomicstructures within the quasicrystal. Figure 2 shows the HRTEMimages, SAED patterns, high-angle annular dark field (HAADF)ChemSusChem 2025, 18, e202501424 (2 of 11) © 2025 The Author(s). ChemSusChem published by Wiley-VCH GmbHChemSusChemResearch Articledoi.org/10.1002/cssc.202501424 1864564x, 2025, 24, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/cssc.202501424 by National Institute For, Wiley Online Library on [15/12/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://doi.org/10.1002/cssc.202501424images, and elemental energy disperse spectroscopy (EDS) map-pings of iQC Al63Cu25Fe12, β-Al55Cu25Fe20, ω-Al70Cu20Fe10, Al2Fe,and θ-Al2Cu after 8 h ball-milling. The consistency betweenthe SAED patterns of iQC Al63Cu25Fe12 after ball-milling and thediffraction pattern observed along the twofold symmetry axisof the quasicrystal before ball-milling verify that such post-processing procedure will not disrupt the quasicrystal structure(Figure 2 and 8, Supporting Information). Such structure consis-tency can also be found in the SAED patterns of iQC Al63Cu25Fe12along the threefold symmetry axis and that before ball milling(Figure 8, Supporting Information). For the powders ofβ-Al55Cu25Fe20, ω-Al70Cu20Fe10, Al2Fe, and θ-Al2Cu after ball mill-ing, their SAED patterns were obtained along [111], [7-9-1],[3-4-1], and [313] axis, all corresponding to the crystal phasesFigure 1. Phase structures of iQC Al63Cu25Fe12 and its control groups (β-Al55Cu25Fe20, ω-Al70Cu20Fe10, Al2Fe, and θ-Al2Cu) before and after ball-milling.a–e) Atomic arrangements and XRD patterns of Al63Cu25Fe12 and its control groups. The icosahedral clusters in (a) are used as a simplified representationof the quasicrystal structure. f–h) The SAED patterns and the HRTEM images of the corresponding areas of iQC Al63Cu25Fe12 before ball-milling, with inci-dence axes along the twofold, threefold, and fivefold rotational symmetry. i) The locally enlarged HRTEM image of (h). Note that the inset in the top-rightcorner of (i) is a model of icosahedral clusters viewed from the fivefold axis. Based on this model, the icosahedral clusters in the image are marked withyellow lines.ChemSusChem 2025, 18, e202501424 (3 of 11) © 2025 The Author(s). ChemSusChem published by Wiley-VCH GmbHChemSusChemResearch Articledoi.org/10.1002/cssc.202501424 1864564x, 2025, 24, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/cssc.202501424 by National Institute For, Wiley Online Library on [15/12/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://doi.org/10.1002/cssc.202501424obtained from XRD, further confirming the crystal structuresof these materials. The HAADF and EDS mappings of iQCAl63Cu25Fe12, β-Al55Cu25Fe20, ω-Al70Cu20Fe10, Al2Fe, and θ-Al2Cushow uniform distributions of the constituent elements of thecorresponding material without phase separation. The quantita-tive data of EDS in Table 2, Supporting Information, show the rel-ative abundance of elements in all samples, which is generallyconsistent with the elemental composition of the correspondingmaterials.X-ray photoelectron spectroscopy (XPS) spectra of iQCAl63Cu25Fe12, β-Al55Cu25Fe20, ω-Al70Cu20Fe10, Al2Fe, and θ-Al2Cuwere collected to reveal the element composition and furtherstudy the chemical state by conducting deconvolution (Figure9, Supporting Information). The Al 2p spectrum of iQCAl63Cu25Fe12 exhibits three peaks at 77.5, 74.6, and 72.2 eV, cor-responding to hydroxide, Al2O3, and metal Al.[34] In Cu 2p spec-trum shows only one peak at 932.4 eV, which is attributed tometal Cu. In Fe 2p spectrum, there are two main peaks locatedat 709.8 and 706.2 eV, which can be attributed to Fe2O3 and metalFe.[35] Before and during the RWGS catalytic reaction, the catalystswere continuously protected by hydrogen, making the surfaceoxidation of the active elements Cu and Fe in the alloy powdernegligible. However, during the XPS tests, the catalysts wereexposed to air and inevitable oxidation of surface elementsoccurred, so the appearance of oxide peaks was quite reason-able.[36] The oxidation of the inactive (in terms of the RWGS reac-tion) element Al is inevitable when exposed to air.[34] Thechemical states of the corresponding elements (particularly theactive elements Cu and Fe) in the other control groups inFigure 9b–e, Supporting Information are essentially the sameas those in iQC Al63Cu25Fe12, which excludes the effects of thisfactor on catalytic performance.2.2. Catalytic Performance in the RWGS ReactionThe thermal catalytic performance of iQC Al63Cu25Fe12 and itscontrol groups in RWGS reaction were evaluated at various tem-peratures. The activities in RWGS reaction are shown in Figure 3a.iQC Al63Cu25Fe12 demonstrates the highest activity at allFigure 2. Microstructure and composition distributions of iQC Al63Cu25Fe12 and its control groups after ball-milling. a,b) TEM image of iQC Al63Cu25Fe12after 8 h ball-milling and the SAED pattern observed along the twofold symmetry axis. c) HAADF images and corresponding EDS elemental mappingsof iQC Al63Cu25Fe12 after ball-milling. d,e) TEM image of β-Al55Cu25Fe20 after ball-milling and the SAED pattern observed along the [111] zone axis.f ) HAADF images and EDS mappings of β-Al55Cu25Fe20 after ball-milling. g–o) TEM images, SAED patterns, HAADF images, and EDS mappings of othercontrol groups (ω-Al70Cu20Fe10, Al2Fe, and θ-Al2Cu) after ball-milling.ChemSusChem 2025, 18, e202501424 (4 of 11) © 2025 The Author(s). ChemSusChem published by Wiley-VCH GmbHChemSusChemResearch Articledoi.org/10.1002/cssc.202501424 1864564x, 2025, 24, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/cssc.202501424 by National Institute For, Wiley Online Library on [15/12/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://doi.org/10.1002/cssc.202501424temperatures, and at a reaction temperature of 500 °C, the CO2conversion of it reaches 43%. The differences in activity betweeniQC Al63Cu25Fe12 and the other four samples are quite significant.Although the activities of the control groups showed someincrease with rising temperatures, they remained at much lowerlevels. The CO2 conversions of β-Al55Cu25Fe20, ω-Al70Cu20Fe10,Al2Fe, and θ-Al2Cu at 500 °C are 17%, 22%, 23%, and 6%, respec-tively. In addition, the CO selectivity of iQC Al63Cu25Fe12 and itscontrol groups were nearly 100% at all temperatures, with almostno CH4 produced as a byproduct (Figure 3b). To further investi-gate the origin of the superior activity of iQC Al63Cu25Fe12 in com-parison to other control groups with similar compositions butdifferent structures, the apparent activation energy Ea of eachcatalyst was analyzed from a kinetic perspective (Figure 3c).The Ea of iQC Al63Cu25Fe12 is 42.51 kJ mol�1, much lower thanβ-Al55Cu25Fe20 (50.22 kJ mol�1), ω-Al70Cu20Fe10 (45.50 kJ mol�1),Al2Fe (68.70 kJ mol�1), and θ-Al2Cu (74.79 kJ mol�1), indicatingthe much faster kinetics of iQC Al63Cu25Fe12 catalyst in RWGSreaction.To analyze the stability of catalysts with different structures,samples that were cooled to 100 °C after completing the firstround of performance tests were subjected to a second roundof RWGS performance tests. The CO2 conversion and CO selectiv-ity, as well as the corresponding rate of changes of iQCAl63Cu25Fe12 and its control groups at 450 and 500 °C are shownin Figure 3d. It can be observed that, compared to the β and ωphases, the CO2 conversion rate of the iQC Al63Cu25Fe12 maintainsat a relatively stable level at all temperatures, while the other twophases exhibit significant fluctuations. The activity differencesbetween the quasicrystal and the β/ω phases are quite pro-nounced, with all catalysts demonstrating good selectivity forCO. These results indicate that the quasicrystal catalyst possessessuperior stability compared to the β and ω phases. In addition, thestability of the quasicrystal was tested through five consecutivecycles and a long-term reaction at 500 °C, and the results indi-cated that the quasicrystal exhibited good stability (Figure 10,Supporting Information). To facilitate the comparison with litera-ture data[37] on other catalysts used for the RWGS reaction, weemployed the same feed ratios, space velocities, and other reac-tion conditions (Experimental Section) for the RWGS catalytic per-formance tests of both the iQC Al63Cu25Fe12 and ω-Al70Cu20Fe10phases (Figure 3e). It can be observed that as the temperaturerises, the difference in CO production rates between iQCAl63Cu25Fe12 and its control groups becomes increasingly moreobvious. At 400 °C, the reaction rate of iQC Al63Cu25Fe12 reaches96.8 mmol gcat�1 h�1, ≈4.7 times that of ω-Al70Cu20Fe10(20.6 mmol gcat�1 h�1). We also made a comparison betweenthe CO production rate and selectivity of iQC Al63Cu25Fe12 andvarious other reported non-noble or even noble metal catalysts(Figure 3f and Table 3, Supporting Information), where iQCAl63Cu25Fe12 shows a certain superiority in both. Additionally,by varying the partial pressures of CO2 and H2, the reaction orderswith respect to CO2 and H2 were determined to be 0.39 and 0.74for iQC Al63Cu25Fe12, and 0.50 and 0.49 for ω-Al70Cu20Fe10, respec-tively (Figure 11, Supporting Information). This indicates that foriQC Al63Cu25Fe12, the activation or participation step of hydrogenFigure 3. Evaluation of the catalytic performance of iQC Al63Cu25Fe12 and its control groups in the RWGS reaction. a,b) CO2 conversion and CO selectivityas a function of reaction temperature for different catalysts. c) Apparent activation energy values of iQC Al63Cu25Fe12, β-Al55Cu25Fe20, ω-Al70Cu20Fe10, Al2Fe,and θ-Al2Cu catalysts. d) The performance of different catalysts obtained after completing the first round of RWGS testing, followed by cooling and con-ducting the second round of testing. The numbers above the bars represent the change in CO2 conversion in the second round compared to the firstround. e) The CO production rate and selectivity of iQC Al63Cu25Fe12 and its control group obtained under the same feed ratio, space velocity, and otherRWGS reaction conditions as used in Ref. [37]. f ) Comparison of iQC Al63Cu25Fe12 with previously reported non-noble catalysts and represented noble metalcatalysts (numbers correspond to the line numbers in Table 3, Supporting Information).ChemSusChem 2025, 18, e202501424 (5 of 11) © 2025 The Author(s). ChemSusChem published by Wiley-VCH GmbHChemSusChemResearch Articledoi.org/10.1002/cssc.202501424 1864564x, 2025, 24, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/cssc.202501424 by National Institute For, Wiley Online Library on [15/12/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://doi.org/10.1002/cssc.202501424is likely one of the rate-determining steps of the reaction.For ω-Al70Cu20Fe10, the reaction rate is jointly controlled bythe coadsorption of CO2 and H2 or their subsequent surfacereaction step.We also studied how the catalytic performance of iQCAl63Cu25Fe12 changed with the ball-milling time and particle sizes(Figure 12, Supporting Information). Due to the ordered atomicoccupancy and lack of translational symmetry in quasicrystals,Figure 4. The adsorption characteristics of different adsorbates and intermediates related to the RWGS process at various adsorption sites of AC of the iQC-AlCuFe, β, and ω phases. a–c) The atomic structures and nonequivalent adsorption sites of the close-packed planes of the (a) AC, (b) β, and (c) ω phases. Notedthat although the numbers n and n’ represent nonequivalent sites, after structural optimization, the final positions of the different adsorbates are identical.Therefore, only one site is counted in the analysis. The dashed circle highlights the icosahedral cluster within the AC structure. d–f ) The adsorption energies ofCO2, H, CO, O, and OH at various adsorption sites of the (d) AC, (e) β phase, and (f ) ω phase. N/A represents that the adsorbate does not adsorb at this site.ChemSusChem 2025, 18, e202501424 (6 of 11) © 2025 The Author(s). ChemSusChem published by Wiley-VCH GmbHChemSusChemResearch Articledoi.org/10.1002/cssc.202501424 1864564x, 2025, 24, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/cssc.202501424 by National Institute For, Wiley Online Library on [15/12/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://doi.org/10.1002/cssc.202501424there are no well-defined slip systems, and dislocation motionrequires the reconstruction of chemical bonds. When externalstress is applied, the absence of effective plastic deformationmechanisms in these intermetallic compounds prevents stressrelaxation. The inability to release stress through dislocationmotion results in significantly higher brittleness compared topure metals. As the ball-milling time increases from 0 to 10 h,the CO2 conversion rises at different reaction temperatures (from2% to 44% at 500 °C), with an increasing trend gradually slowingdown, for the reduction of catalyst particle sizes with the increas-ing ball-milling time mitigates. This trend can be observed moreclearly in Figure 12c, Supporting Information. Regardless of theball-milling time, the CO selectivity of the iQC Al63Cu25Fe12 cata-lyst consistently remained around 100%.2.3. Reaction Mechanism of Different CatalystsThe above characterizations indicate the similar specific surfaceareas, chemical states, and contents of surface-active elements ofthese catalysts despite their different phase structures, so the dif-ference in their phase structures is assumed to be the primaryfactor contributing to the significant differences in RWGS catalyticperformances, for phase structures could exert major influenceon the energy barriers during the conversion of reactants to prod-ucts. To verify this hypothesis and further explore the intrinsicmechanisms by which different structures influence catalystactivity, we studied the potential mechanisms of the RWGS pro-cess on the surfaces of three catalysts: iQC, β, and ω, using first-principles calculations[31,32,38] based on DFT. Since quasicrystaldoes not have periodicity in three dimensions, we chose the sim-plest periodic approximant crystal (AC) of the iQC-AlCuFe, the 1/1cubic approximant, as the model for calculations.[39] This modelstructure is the ideal structure obtained from the shear process ofthe iQC using the Katz, Gratias and Elser method.[39,40] It containsvarious cluster compositions, including the icosahedral clusters,and has been confirmed to be very similar in geometric and phys-ical properties to iQC-AlCuFe.[41]For the surfaces of the 1/1 AC, β, and ω phases, we selectedthe close-packed planes of each structure for analysis and calcu-lated the adsorption energies of different adsorbates and inter-mediates related to the RWGS process at various adsorption sites,as shown in Figure 4a–c. The adsorption states of these adsor-bates at each site after optimization are shown in Figure 13–27,Supporting Information. The adsorption energy of H at the moststable site of 1/1 AC is �0.55 eV, which is lower than that of βphase and higher than that of ω phase; the adsorption energyof CO2 at the most stable site of 1/1 AC is �0.70 eV, which isFigure 5. CO2-TPD and H2-TPD results of QC-AlCuFe, β, and ω phases, as well as the corresponding adsorption capacities, and the energy of the most sta-ble adsorption sites of CO2 and H on different structural surfaces. a–c) CO2-TPD of QC-AlCuFe, β, and ω phases, as well as the adsorption energy of themost stable sites of CO2 on surfaces of different phases. d–f ) H2-TPD of QC-AlCuFe, β, and ω phases, as well as the adsorption energy of the most stablesites of H on surfaces of different phases.ChemSusChem 2025, 18, e202501424 (7 of 11) © 2025 The Author(s). ChemSusChem published by Wiley-VCH GmbHChemSusChemResearch Articledoi.org/10.1002/cssc.202501424 1864564x, 2025, 24, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/cssc.202501424 by National Institute For, Wiley Online Library on [15/12/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://doi.org/10.1002/cssc.202501424higher than that of β and lower than that of ω. In other words,compared to the β and ω catalysts featuring similar componentsbut different structures, the quasicrystal exhibits moderateadsorption capabilities for each reactant, consistent with theSabatier principle.[42] The CO2 and H2 temperature-programeddesorption (TPD) results of the iQC-AlCuFe, β, and ω phases alsoexperimentally support this point (Figure 5). The adsorptioncapacity of iQC-AlCuFe for CO2 and H2 lies between that of βand ω phases, consistent with the DFT calculation results. TheCO adsorption capacity is also evaluated by CO-TPD tests(Figure 28, Supporting Information). The results reveals thatthe CO adsorption capacity of both the ω phase and iQC-AlCuFe is weaker than that of the β phase. Additionally, the com-parison of the adsorption energy values at different sites on the 1/1 AC reveals that the stable adsorption sites for the reactants Hand CO2 are both located near the icosahedral clusters, indicatingthat the RWGS process is more likely to occur in proximity tothese clusters.Numerous studies[20] have shown that catalysts containingelement Fe and Cu typically proceed via two pathways duringthe RWGS reaction: one is the direct dissociation (redox) mecha-nism (CO2*þ H*! CO*þO*þ H*, CO*þO*þ H*! CO*þOH*),and the other is the carboxylate mechanism (CO2*þ H*! COOH*,COOH*! CO*þOH*). Thus, the energy barriers for both pathwayson different catalyst surfaces were calculated, and the structuresand energies of the various adsorption states and transition statesare illustrated in Figure 6. It can be deduced that on the 1/1 AC andω phase surfaces, the RWGS reaction prefers the direct dissociationmechanism, while on the β phase surface, it is more inclined tofollow the carboxylate mechanism. By comparison, the energy bar-rier for the direct dissociation mechanism on the 1/1 AC surface issignificantly lower than the barriers of β and ω phases for both ofthe two pathways. The barrier for the direct dissociation of CO2 intoCO and O on 1/1 AC surface is reduced to 0.35 eV, which facilitatesthe conversion from CO2 to CO and is consistent with the experi-mentally measured significant increase in the CO formation rate.Therefore, on the 1/1 AC surface, the possibility of the RWGS reac-tions occurring near the icosahedral clusters is greater than atother locations, and the abundance of such structures (seeFigure 1i) in iQC leads to lower energy barriers, which explainits superior CO formation rates and CO2 conversion efficiency com-pared to the similarly composed β and ω phases. To verify whetherthe RWGS reaction on the quasicrystal surface tends to follow adirect dissociation mechanism, we conducted in situ time-resolveddiffuse reflectance infrared fourier transform spectroscopy(DRIFTS) tests (Figure 7). It was observed that upon the introduc-tion of CO2, a significant amount of gaseous CO was formed, asindicated by two symmetrical peaks at ≈2140 cm�1 in the spec-trum.[37] After switching to H2, the band intensities for carboxylate(COOH, 1,666 cm�1) and bidentate carbonate species (bi-CO32�,1545–1307 cm�1)[37] greatly increased. In contrast, the intensityof the gaseous CO band did not increase (as seen in the blue spec-tra after 1 and 5min) but gradually decreased with the continuedflow of H2. These results suggest that during the RWGS reaction onthe quasicrystal surface, CO2 tends to dissociate directly into COand O, which is consistent with the theoretical calculationsFigure 6. Reaction path and energy barriers for the RWGS reaction on dif-ferent catalyst surfaces. a–c) Energy profiles for the direct dissociation andcarboxylate mechanism on the (a) (001) AC, (b) (110) β, and (c) (110) ω sur-faces. The values in parentheses indicate the energy barrier of the reaction.The red and black lines represent the direct dissociation pathway and thecarboxylate pathway, respectively.Figure 7. In situ time-resolved DRIFTS obtained over QC-AlCuFe when thefeed gas was switched between CO2 and H2.ChemSusChem 2025, 18, e202501424 (8 of 11) © 2025 The Author(s). ChemSusChem published by Wiley-VCH GmbHChemSusChemResearch Articledoi.org/10.1002/cssc.202501424 1864564x, 2025, 24, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/cssc.202501424 by National Institute For, Wiley Online Library on [15/12/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://doi.org/10.1002/cssc.202501424(Figure 6). Furthermore, the amount of linearly adsorbed CO(COlinear) was relatively weak, in agreement with the CO-TPDresults (Figure 28, Supporting Information).In this study, an iQC Al63Cu25Fe12 is found to be a non-noblehigh-performance catalyst for the thermal catalytic RWGS reac-tion. By comparing the samples with similar compositions but dif-ferent structures, the influence of catalyst surface area andelemental chemical states on catalytic performance are ruledout, highlighting the crucial role of the quasicrystal structurein this catalysis. Compared to these control groups, as well asother reported non-noble or even noble metal catalysts, the qua-sicrystal exhibits a higher CO production rate and selectivity. DFTcalculations reveal lower reaction barriers of AC of iQC-AlCuFeduring CO2 reduction to CO compared with other samples.Paths with lower energy barriers are all near the icosahedral clus-ters, indicating that the high activity of iQC-AlCuFe for RWGSreaction is closely associated with the abundant icosahedral clus-ters in the quasicrystal. The discovery of structure–performancerelationship of iQC also provides a reference for the intensivestudy of other complex structures.3. Experimental SectionPreparation of iQC Al63Cu25Fe12 Ingot and Its Control GroupsThe Al-Cu-Fe alloys were prepared by melting a mixture of pure ele-ments with purities of 99.9 wt% Al, 99.9 wt% Cu, and 99.9 wt% Fe inan electric arc furnace under an argon atmosphere. The phase struc-tures and compositions of the alloys are shown in Table 1, SupportingInformation. The cast ingots were sealed in quartz tubes filled withargon and annealed at 800, 700, 800, 1050, and 500 °C for 10, 24, 24,72, and 72 h, respectively. They were then quenched in liquid nitro-gen to obtain the pure phases of the iQC Al63Cu25Fe12 and its controlgroups (β-Al55Cu25Fe20, ω-Al70Cu20Fe10, Al2Fe, and θ-Al2Cu).Preparation of iQC Al63Cu25Fe12 Powder and Its ControlGroupsThe heat-treated iQC Al63Cu25Fe12 ingot and milling pot were placedin a glove box. Before crushing the ingot in a mortar, the containerwas vacuumed and filled with high-purity argon at least three times.The crushed 2 g ingot was placed in the milling jar along with 1mLethanol. Under the protection of high-purity argon, the ingot wasmilled for different durations (2, 4, 6, 8, and 10 h) using a planetaryball-milling machine (Pulverisette 6, Fritsch) to obtain alloy powdersof varying sizes. During the milling process, the rotation speed wasset to 400 r.p.m., and after every 10min of milling, the rotation waspaused for 2 min to release heat. After milling, the pot was opened inthe glove box under high-purity argon protection, and the sampleswere collected for subsequent characterization and performancetesting. The same method was used to mill the control groups for8 h to collect alloy powders.Materials CharacterizationThe phase structures of these materials before and after ball millingwere characterized by XRD on a Rigaku SmartLab diffractometerequipped with a Cu Kα radiation (λ= 0.15406 nm). The particle sizesand surface morphologies of catalysts were observed with a ZEISSGemini 560 SEM. Surface areas of alloy powders were measured witha Micromeritics ASAP 2020 instrument. These samples were firstdegassed at 150 °C for 3 h under vacuum. By using the N2 adsorp-tion–desorption isotherm at 77 K, surface areas were calculated bythe Brunauer–Emmett–Teller method. The phase identification,microstructure observation, and composition analysis of these alloypowders were carried out using HAADF scanning transmission elec-tron microscopy (JEOL JEM-2100F, with an acceleration voltage of200 kV) and TEM-EDS (JEOL JED-2300, operating at 200 kV). XPS spec-tra were conducted on a JEOL JPS-9100.Temperature-Programed ExperimentThe CO2, CO, or H2 adsorption properties of the catalyst were evalu-ated by TPD (CO2-TPD, CO-TPD, or H2-TPD) on an AutoChem 2920(Micromeritics Instrument) equipped with a thermal conductivitydetector. The catalyst (100mg) was pretreated at 430 °C for 1 h undera 10% H2/Ar atmosphere, followed by purging with Ar for 30 min at430 °C to remove the adsorbed H species. After cooling to 50 °C,high-purity CO2, CO, or 10% H2/Ar was preadsorbed for 30 min.Subsequently, the insert gas (30mLmin�1, He for CO2-TPD andCO-TPD, and Ar for H2-TPD) was purged until the baseline stabilized.The TPD profile was then recorded as the temperature increased from50 to 700 °C at a rate of 10 °C min�1.Catalytic Performance EvaluationThe catalytic performance of RWGS reaction was tested in afixed-bed reactor (quartz; inner diameter, 8 mm). 100 mg catalystwas pretreated with H2 (30 mL min�1) at 703 K for 1h before thecatalytic test. The composition of the reaction gas is 40vol%H2/10vol% CO2/50vol% N2, and the total gas flow rate was setat 60 mL min�1. The RWGS reaction test was conducted at the tem-perature range of 250–500 °C (measured at every 50 °C interval,with each temperature held for 30 min). The composition of theoutlet product was studied by two on-line gas chromatographsequipped with thermal conductivity detectors (GL Science,GC323). H2, N2, CH4, and CO were separated using a zeolite-packedcolumn (Molecular Sieve 13X), while H2O, CO2, CH3OH, and otherbyproducts were separated using a column packed with a porouspolymer adsorbent (Porapak Q). The pipelines connecting the reac-tor and the gas chromatograph were heated to >393 K during thetests to make sure that all the products existed in gas form. Theproduct mixture went through a cooling device to get rid of theliquid products, and the total flow rate of the remaining gaseousproducts was recorded using a flow meter (Mesa Laboratories,Defender530). The flow rate of each product was determined byanalyzing the results of gas chromatographic and total flow ratemeasurements.After completing the first round of performance testing, the sampleswere allowed to cool to 100 °C before conducting a second round ofperformance testing on the iQC Al63Cu25Fe12 and its control groupsto evaluate the stability of the quasicrystal and other catalysts.Additionally, to facilitate comparison with literature data[14] on othercatalysts used in the RWGS reaction, the same feed ratios (72 vol% H2/24 vol% CO2/4 vol% N2), space velocity (total gas flow rate was 100mL min�1, WHSV= 60 000 mL g�1 h�1), atmospheric pressure andother reaction conditions were used for performance testing ofthe iQC Al63Cu25Fe12 and its control groups. The CO formation ratesof the iQC Al63Cu25Fe12 are measured and compared with somereported catalysts, as detailed in Table 3, Supporting Informationand Figure 3f.ChemSusChem 2025, 18, e202501424 (9 of 11) © 2025 The Author(s). ChemSusChem published by Wiley-VCH GmbHChemSusChemResearch Articledoi.org/10.1002/cssc.202501424 1864564x, 2025, 24, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/cssc.202501424 by National Institute For, Wiley Online Library on [15/12/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://doi.org/10.1002/cssc.202501424The CO2 conversion (CCO2), CO selectivity (SCO), and CO productionrates (RCO) were calculated as followsCCO2¼ FCO2, in � FCO2, outFCO2, in� 100% (1)SCO ¼ FCO;outFCO2, in � FCO2, out� 100% (2)RCO ¼ FCO;outω� 100% (3)FX is the flow rate of gas X (mmol h�1), and ω is the mass of thecatalyst.The activation energy for the RWGS reaction was calculated based onperformance data obtained during temperature-programed reac-tions in the range of 250–400 °C, with a CO2 conversion of <15%(except iQC at 400 °C). The activation energy was estimated usingthe Arrhenius equation.[12]Parameter Setting of DFT CalculationsAll structural relaxation and energetic calculations were conducted viathe DFT code—Vienna Ab Initio Simulation Package (VASP)[31] withperiodic conditions and the plane-wave basis sets.[32] The interactionsbetween electron ions were addressed using the method of projector-augmented wave theory.[43] Generalized gradient approximationmethod was used to analyze exchange-correlation energy.[44,45] Thecut-off energy was set to 500 eV in all calculations. K-point samplinggrids determined byMonkhorst–Packmethod[46] were set to 3� 3� 3,11� 11� 11, and 7� 7� 3 for bulk iQC Al63Cu25Fe12, β-Al55Cu25Fe20,and ω-Al70Cu20Fe10, respectively. Broyden Fletcher Goldfarb Shannonalgorithm was utilized to relax the models and optimize the structures.For iQC Al63Cu25Fe12, β-Al55Cu25Fe20, and ω-Al70Cu20Fe10 slabs, thek-points samplings were 2� 2� 1, 3� 2� 1, and 2� 3� 1, respec-tively. 15 Å was used as the vacuum layer between periodicallyrepeated slabs to avoid interactions among slabs. The energychanges in the structural optimization process and the maximumstress converge to 1.0� 10�5 eV atom�1 and 0.01 eV Å�1, respec-tively. The climbing image nudged elastic band method isemployed to obtain the transition state and energy barrier fordifferent pathways. Three images between initial and final config-urations are inserted and relaxed until the maximum residual forcefalls below 0.03 eV Å�1.The adsorption energies Eads of different adsorbates and intermedi-ates were calculated as followsEads ¼ Eadsorbate=slab � Eslab � Eadsorbate (4)where Eadsorbate/slab is the total energy of the slab with the adsorbedspecies, Eslab is the total energy of the bare slab, and Eadsorbate is thetotal energy of the adsorbate. According to this definition, a smallervalue of Eads corresponds to a stronger adsorption of a given specieson the slab.In situ DRIFTS ExperimentPrior to spectrum acquisition, the catalyst was reduced in a flow of10% H2/He (30 mLmin�1) at 703 K for 1 h. Subsequently, the catalystwas thoroughly purged with He for 2 h at the same temperature toremove any residual adsorbed H2. Following this, the temperaturewas raised to 500 °C under a He flow. Upon reaching the target tem-perature, the catalyst was exposed to a flow of 10% CO2/He(30 mLmin�1) for 30min, and then switched to a flow of 10%H2/He (30 mLmin�1).AcknowledgementsThis research was supported by the Japan Society for thePromotion of Science (JSPS Kakenhi grant no. 22F22348, JSPSKakenhi grant no. 24F24036) and JST, CREST grant no.JPMJCR22O3, Japan.Conflict of InterestThe authors declare no conflict of interest.Data Availability StatementThe data that support the findings of this study are available fromthe corresponding author upon reasonable request.Keywords: catalyst · CO2 conversion · density functional theory ·quasicrystal · reverse water–gas shift reaction[1] Y. Tokumoto, K. Hamano, S. Nakagawa, Y. Kamimura, S. Suzuki, R. Tamura,K. Edagawa, Nat. Commun. 2024, 15, 1529.[2] K. Kamiya, T. Takeuchi, N. Kabeya, N. Wada, T. Ishimasa, A. Ochiai,K. Deguchi, K. Imura, N. Sato, Nat. Commun. 2018, 9, 154.[3] R. N. Araújo, E. C. Andrade, Phys. Rev. B 2019, 100, 014510.[4] R. Takeuchi, F. Labib, T. Tsugawa, Y. Akai, A. Ishikawa, S. Suzuki, T. Fujii,R. Tamura, Phys. Rev. Lett. 2023, 130, 176701.[5] R. Tamura, A. Ishikawa, S. Suzuki, T. Kotajima, Y. Tanaka, T. Seki, N. Shibata,T. Yamada, T. Fujii, C.-W. Wang, J. Am. Chem. Soc. 2021, 143, 19938.[6] A. I. Goldman, T. Kong, A. Kreyssig, A. Jesche, M. Ramazanoglu,K. W. Dennis, S. L. Bud’ko, P. C. Canfield, Nat. Mater. 2013, 12, 714.[7] R. Tamura, T. Abe, S. Yoshida, Y. Shimozaki, S. Suzuki, A. Ishikawa, F. Labib,M. Avdeev, K. Kinjo, K. Nawa, Nat. Phys. 2025, 1.[8] S. K. Verma, A. Bhatnagar, M. A. Shaz, T. P. Yadav, Int. J. Hydrog. Energy2023, 48, 9762.[9] S. Ryabtsev, V. Polonskyi, O. Sukhova, Sci. Mater. 2020, 56, 263.[10] C. Zhou, F. Cai, H. Xu, S. Gong, Mater. Sci. Eng. A 2004, 386, 362.[11] A. Tsai, M. Yoshimura, Appl. Catal. A-Gen. 2001, 214, 237.[12] S. Kameoka, T. Tanabe, A. P. Tsai, Catal. Today 2004, 93, 23.[13] T. Tanabe, S. Kameoka, A. P. Tsai, Catal. Today 2006, 111, 153.[14] T. Tanabe, S. Kameoka, A. P. Tsai, Appl. Catal. A-Gen. 2010, 384, 241.[15] J.r. Hafner, M. Krajci, Acc. Chem. Res 2014, 47, 3378.[16] A. Jain, Y. Shin, K. A. Persson, Nat. Rev. Mater. 2016, 1.[17] H. X. Jin, J. X. Zhang, P. Li, Y. J. Zhang, W. Y. Zhang, J. Y. Qin, L. H. Wang,H. B. Long, W. Li, R. W. Shao, E. Ma, Z. Zhang, X. D. Han, Nat. Commun.2022, 13.[18] Z. Xiao, C. Zhang, J. Gu, E. Yuan, G. Li, J.-J. Zou, D. Wang, Chem. Eng. J.2025, 507, 160529.[19] C. Zhang, Z. Xiao, X. Guo, X. Tan, J. Gu, J. Li, G. Li, J.-J. Zou, D. Wang, Chem.Eng. J. 2025, 165414.[20] E. Pahija, C. Panaritis, S. Gusarov, J. Shadbahr, F. Bensebaa, G. Patience,D. C. Boffito, ACS Catal 2022, 12, 6887.[21] Z. Xiao, H. Zhang, X. Tan, F. Ye, Y. Zhang, J. Gu, J. Li, K. Sun, S. Zhang,J. J. Zou, Adv. Energy Mater. 2025, 15, 2500988.[22] Z. Xiao, L. Zhang, X. Tan, K. Sun, J. Li, L. Pan, J. J. Zou, G. Li, D. Wang, Adv.Funct. Mater. 2025, 2500339.[23] H. Wang, M. S. Bootharaju, J. H. Kim, Y. Wang, K. Wang, M. Zhao, R. Zhang,J. Xu, T. Hyeon, X. Wang, J. Am. Chem. Soc. 2023, 145, 2264.[24] K. de Kock, S. Raseale, W. Marquart, T. Verfaille, M. Claeys, N. Fischer, ACSCatal. 2025, 15, 5835.ChemSusChem 2025, 18, e202501424 (10 of 11) © 2025 The Author(s). ChemSusChem published by Wiley-VCH GmbHChemSusChemResearch Articledoi.org/10.1002/cssc.202501424 1864564x, 2025, 24, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/cssc.202501424 by National Institute For, Wiley Online Library on [15/12/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://doi.org/10.1002/cssc.202501424[25] A. I. Rabee, H. Abed, T. H. Vuong, S. Bartling, L. Kraußer, H. Atia,N. Rockstroh, E. V. Kondratenko, A. Brückner, J. Rabeah, ACS Catal.2024, 14, 10913.[26] B. Lu, Z. Han, X. Zhi, L. Zhang, Chem. Eng. J. 2024, 500, 156844.[27] A. M. Bahmanpour, F. Héroguel, M. Kılıç, C. J. Baranowski, P. Schouwink,U. Röthlisberger, J. S. Luterbacher, O. Kröcher, Appl. Catal. B Environ. 2020,266, 118669.[28] A. M. Bahmanpour, F. Héroguel, M. Kılıç, C. J. Baranowski, L. Artiglia,U. Röthlisberger, J. S. Luterbacher, O. Kröcher, ACS Catal. 2019, 9, 6243.[29] M. Zhu, P. Tian, R. Kurtz, T. Lunkenbein, J. Xu, R. Schlögl, I. E. Wachs,Y. F. Han, Angew. Chem. 2019, 131, 9181.[30] G. Kim, S. H. Ryu, H. Jeong, Y. Choi, S. Lee, J. H. Choi, H. Lee, Angew. Chem.2023, 135, e202306017.[31] G. Kresse, D. Joubert, Phys. Rev. B 1999, 59, 1758.[32] G. Kresse, J. Furthmüller, Phys. Rev. B 1996, 54, 11169.[33] H. Jin, J. Zhang, Y. Zhang, W. Zhang, S. Ma, S. Mao, Y. Du, Z. Wang, J. Qin,Q. Wang, Mater. Charact. 2022, 183, 111609.[34] P. Barua, V. Srinivas, S. Dhabal, T. Ghosh, J. Mater. Res. 2002, 17, 1892.[35] M. C. Biesinger, B. P. Payne, A. P. Grosvenor, L. W. M. Lau, A. R. Gerson,R. S. Smart, Appl. Surf. Sci. 2011, 257, 2717.[36] C. Hansen, W. Zhou, E. Brack, Y. Wang, C. Wang, J. Paterson, J. Southouse,C. Copéret, J. Am. Chem. Soc. 2024.[37] H. Kang, L. Zhu, S. Li, S. Yu, Y. Niu, B. Zhang, W. Chu, X. Liu, S. Perathoner,G. Centi, Nat. Catal. 2023, 6, 1062.[38] H. Jin, J. Zhang, W. Zhang, Y. Zhang, S. Mao, Y. Du, S. Ma, J. Qin, Q. Wang,Intermetallics 2023, 152, 107768.[39] F. Puyraimond, M. Quiquandon, D. Gratias, M. Tillard, C. Belin, A. Quivy,Y. Calvayrac, Found. Crystallogr. 2002, 58, 391.[40] H. Yamada, T. Takeuchi, U. Mizutani, N. Tanaka, Mater. Res. Soc. Sympos.Proc. 1999, 553, 117.[41] A. Quivy, M. Quiquandon, Y. Calvayrac, F. Faudot, D. Gratias, C. Berger,R. Brand, V. Simonet, F. Hippert, J. Phys. Condens. Matter. 1996, 8, 4223.[42] P. Sabatier, La catalyse en chimie organique, University of MichiganLibrary 1913.[43] P. E. Blöchl, Phys. Rev. B 1994, 50, 17953.[44] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77, 3865.[45] N. Troullier, J. L. Martins, Phys. Rev. B 1991, 43, 1993.[46] D. J. Chadi, Phys. Rev. B 1977, 16, 1746.Manuscript received: July 5, 2025Revised manuscript received: September 26, 2025Version of record online: October 29, 2025ChemSusChem 2025, 18, e202501424 (11 of 11) © 2025 The Author(s). ChemSusChem published by Wiley-VCH GmbHChemSusChemResearch Articledoi.org/10.1002/cssc.202501424 1864564x, 2025, 24, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/cssc.202501424 by National Institute For, Wiley Online Library on [15/12/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://doi.org/10.1002/cssc.202501424 Superior Performance and Catalytic Mechanism of an Icosahedral Quasicrystal Al-Cu-Fe in CO2 Reduction to CO 1. Introduction 2. Results 2.1. Synthesis and Characterizations 2.2. Catalytic Performance in the RWGS Reaction 2.3. Reaction Mechanism of Different Catalysts 3. Experimental Section Outline placeholder Preparation of iQC Al63Cu25Fe12 Ingot and Its Control Groups Preparation of iQC Al63Cu25Fe12 Powder and Its Control Groups Materials Characterization Temperature-Programed Experiment Catalytic Performance Evaluation In&thinsp;situ DRIFTS Experiment