Engineering ultrafine Cu nanoparticles supported on zeolites via solvent-free inter-zeolite transformation for bioethanol dehydrogenation

Abstract

Although Cu nanoparticles (Cu-NPs) supported on zeolites have been widely used in sustainable catalysis, they still suffer from accelerated deactivation due to metal sintering. To overcome this problem, inter-zeolite transformation (IZT) under solvent-free conditions offers advantages in the redispersion of Cu confined inside zeolite structures, ultimately resulting in enhanced catalytic activity. Herein, we report the benefits ofsolvent-free IZT, which not only transforms the original zeolite framework (FAU) into other zeolites (BEA) but also simultaneously redisperses the aggregated Cu-NPs on FAU surfaces into highly dispersed Cu-NPs on the transformed BEA (BEA-IZT) structure. The PXRD patterns illustrate that FAU has been completely transformed into BEA. The Cu clusters are redispersed on BEA-IZT with a size of ∼2.43 nm, eventually facilitating the formation of uniform metallic Cu, as confirmed by time-resolved X-ray absorption near edge spectroscopy (TR-XANES). Interestingly, the highly dispersed Cu-NPs deposited on the transformed BEA zeolite promote superior catalytic dehydrogenation of renewable feedstock, bioethanol, to acetaldehyde, achieving ethanol conversion, acetaldehyde selectivity, and yield of approximately 90, 70, and 60%, respectively. This first example opens up the perspective of material design through the solvent-free IZT process for redispersing sintered metal particles to produce a highly reactive catalyst for sustainable ethanol dehydrogenation.

---

1 Introduction

Nanoparticles (NPs) have been widely applied in diverse functions, such as energy conversion and storage, chemical manufacturing, biological applications, and environmental technologies. In addition, it is well known that NPs are among the most important catalysts in several catalytic processes as their unique sizes, shapes, and compositions provide distinctive quantum properties, resulting in different catalytic activities compared to those of bulk materials.NPs derived from earth-abundant and inexpensive metals have garnered substantial interest because of their potential as an alternative to expensive and noble metal catalysts, which have been typically used in many conventional chemical processes, such as hydroisomerization, hydrocracking, and fluid catalytic cracking (FCC).^1–5 In particular, copper nanoparticles (Cu-NPs) are considered highly promising because of their high natural abundance, low cost, and superior chemical catalytic properties.^6,7 Moreover, Cu is one of the promising metals with a wide range of oxidation states, such as Cu⁰, Cu^I, Cu^II, and Cu^III, which can be easily prepared by traditional methods, including impregnation, atomic layer deposition (ALD), and ion exchange.^1,6,7 Therefore, Cu-based catalysts are frequently employed in various sustainable catalytic reactions such as ethanol dehydrogenation,^7,8 water–gas shift,⁹ CO₂ hydrogenation,^10,11 and hydrogenolysis of esters and alcohols.¹² In recent years, acetaldehyde production from bioethanol, a renewable feedstock, has received heightened attention due to its being an environmentally friendly and economical alternative to conventional chemical processes. In addition, acetaldehyde has been increasingly used for the production of many chemicals, such as acetic acid, ethyl acetate, n-butanol, pyridine, and acetic anhydride.¹³ To achieve a high yield of acetaldehyde, the key to acetaldehyde formation depends on the electronic properties of Cu-NPs.¹⁴ Interestingly, the electronic properties of Cu-NPs can be controlled by adjusting their particle size. For example, a decrease in the size of metal particles changes their physical, chemical, and mechanical properties, as well as the preference for being low-coordinated atoms, resulting in changes in bond breaking and forming in catalysis.^14–19

However, the design of Cu-NP-based catalysts, which are highly active, selective, stable, and robust, is still challenging. To improve the stability of Cu-NPs, several strategies for designing Cu-NPs have been developed. Among them, the most crucial one is to immobilize metals onto solid surfaces to enhance the stability and distribution of metal NPs.^6,20 In this regard, several support materials can be used to immobilize Cu-NPs, for example, iron oxides,^6,21,22 silica,^23–26 carbon-based materials,²⁷ and zeolites.^28–32 It is well known that zeolite-supported Cu-NPs are widely used in heterogeneous catalysis due to their bifunctional characteristics. Typically, incorporating Cu-NPs on zeolite surfaces using the impregnation method has been widely used to prepare Cu-NPs supported on zeolite materials,^33–37 which can produce a wide range of Cu-NP particle sizes ranging from ∼1 to 20 nm.^33,36,37 Unfortunately, loading a high content of Cu-NPs typically causes Cu-NP aggregation, resulting in zeolite pore blockage, thus consequently decreasing the surface area of zeolite support, eventually lowering the catalytic activity.³⁸ Moreover, the loaded Cu-NPs are commonly located at the external surface of zeolite particles, causing further processes through migration and Ostwald ripening, resulting in fast catalyst deactivation and loss of catalytic activity.^39,40 To circumvent these limitations, special attention has been given to the preparation method to improve catalyst lifetime by preventing rapid catalyst deactivation due to metal sintering.

Over the past decade, several preparation methods have been applied to enhance metal-sintering resistance, such as metal encapsulation, steam-assisted, ligand-protected, and organo-ligand stabilization approaches to promote the high dispersion of metals on a zeolite surface.^30,41–43 However, these approaches often suffer from complicated preparation procedures or require special preparation of metal complexes. The previous work reported that small (∼1 nm) metal clusters were formed after sequential treatment in O₂ and H₂.⁴¹ However, the synthesis method requires forming a water-soluble metal–sulfur complex via a mercaptosilane ligand and post-sequential treatment of O₂ and H₂ to achieve a small metal cluster. Moreover, achieving the ultrasmall or sub-nanometer size of metal-incorporated zeolites typically needs an additional process to prevent metal aggregation during the hydrothermal crystallization process.^44–46 It often requires the pre-preparation of the metal complex before using it in the hydrothermal crystallization of zeolite. Thus, these additional preparation steps would make metal-supported zeolite synthesis more complicated and require using a high volume of chemicals, eventually generating substantial waste.^45,46

To overcome these limitations, the inter-zeolite transformation (IZT) process, one of the most promising methods applied for the conversion of parent zeolites to other daughter frameworks, with a higher framework density compared to the original zeolite,^47–50 would be an interesting approach to simultaneously encapsulate metal nanoparticles within the framework. For example, platinum (Pt), ruthenium (Ru), and rhodium (Rh) nanoclusters encapsulated in beta (BEA) and faujasite (FAU) zeolites with cluster sizes of less than 2 nm could be used as the starting parent zeolites for the IZT process into the MFI zeolite framework. The IZT process successfully preserves the encapsulated metal (Pt, Ru, and Rh) nanoclusters within the MFI zeolite framework. It should be mentioned that the nanocluster size of the encapsulated metal was not significantly different from the parent zeolite, as illustrated in the available literature.⁵¹ In addition, the IZT approach can be applied to many metal clusters, for example, nickel (Ni),^52,53 cobalt (Co),^52,54 zinc (Zn),^52,54 tin (Sn),⁵⁴ iron (Fe), and silver (Ag) supported on zeolite structures.⁵⁵

Although the above-mentioned IZT in the presence of metals has been applied to synthesize metal-supported zeolites, there has been no report so far that illustrates the redispersion of aggregated metal clusters supported on a transformed zeolite via the solvent-free IZT process. To illustrate the first example of the redispersion of highly dispersed Cu nanoparticles supported on zeolite surfaces, the solvent-free IZT approach is introduced using deactivated/sintered or large metal particles supported on zeolites as starting materials, which can be further converted to target zeolites containing highly dispersed metals. Both freshly prepared metal-support zeolites and post-catalytic reaction catalysts can be used as the starting zeolite for IZT, which may show a new perspective for upgrading the deactivated catalyst to a new high-performance catalyst. In addition, the IZT process allows the confinement of metals in the secondary pore of recrystallized zeolites. This concept illustrates the rebirth of the deactivated metal-supported zeolites to produce a highly dispersed metal catalyst, which is beneficial over the conventional in situ encapsulation process.

In this contribution, the redispersion of Cu-NPs via solvent-free inter-zeolite transformation (IZT) of Cu-NPs-based FAU into Cu-NPs-based BEA zeolite is demonstrated. Interestingly, not only the conversion of the original zeolite framework to the target zeolite structure but also the redispersion of Cu-NPs on a solid supporter was significantly observed, eventually preventing metal sintering and controlling the high portion of Cu metallic active species under environmentally friendly conditions. Moreover, insights into theIZT mechanism of Cu-FAU into Cu-BEA were addressed to elucidate the transformation behaviors. Subsequently, the real-time measurement of the change in the oxidation state of Cu-NPs was also investigated under hydrogen pretreatment at 450 °C to explore the advantages and robustness of catalysts prepared by the solvent-free IZT process compared to the one prepared by the simple impregnation method. These characters also benefit the catalytic performance in bioethanol dehydrogenation to produce acetaldehyde selectively. This work opens up new perspectives on catalyst redispersion by transforming deactivated/sintered catalysts or large metal particles on a zeolite with low catalytic activity into a new type of sustainable catalyst containing highly dispersed metal species as greatly efficient catalysts for bioethanol upgrading applications. Moreover, this work establishes a new concept that shows the opportunity for scaling up the synthesis of a new type of sustainable catalyst using a simple, uncomplicated, and environmentally friendly process.

2 Results and discussion

2.1 Structural and physicochemical properties of the as-prepared zeolites and Cu-NP supported zeolites

The preparation procedure of the Cu-NPs supported on BEA zeolite using a solvent-free IZT process (Cu-BEA-IZT) is illustrated in Fig. 1A. Briefly, Cu was first impregnated on commercial FAU zeolite (FAU-Com) by the incipient wetness impregnation (IWI) method (Cu-FAU-IWI). Subsequently, the as-prepared Cu-FAU-IWI was mixed with tetraethyl ammonium hydroxide (TEAOH) and sodium hydroxide (NaOH), followed by hydrothermal recrystallization at 140 °C for 3 days. During the recrystallization process, the parent FAU zeolite started to dissolve into an amorphous phase to provide building block units, which are essential for the formation of a daughter zeolite (BEA) structure.⁵⁵

Fig. 1 (A) Schematic illustration of solvent-free inter-zeolite transformation (IZT) of Cu-FAU-IWI into Cu-BEA-IZT, (B) PXRD patterns of the as-prepared samples, (C) FESEM, and (D) global particle size distribution (PSD) of the as-prepared samples including: (a) Cu-FAU-IWI, (b) Cu-BEA-IZT, and (c) Cu-BEA-IWI.

For the first step of the synthesis of Cu-BEA-IZT (step (i)), the parent Cu-FAU-IWI zeolite was prepared using the IWI method to load Cu nanoparticles on a commercial acidic FAU (FAU-Com). The preparation process was followed by the dropwise addition of Cu solution onto FAU-Com zeolite, combined with a drying process at 100 °C overnight. Subsequently, the obtained product of Cu-FAU-IWI was calcined and then applied as the parent zeolite for the IZT process. Indeed, the Cu clusters would be deposited on the external surface of the parent FAU-Com zeolite, and their size distribution was non-homogeneous. Subsequently, the parent Cu-FAU-IWI zeolite was mixed with NaOH and TEAOH solutions (step (ii)). The mixing process was performed for 5 minutes to obtain the mixed gel, which was then transferred to a Teflon-lined hydrothermal autoclave for hydrothermal recrystallization at 140 °C for 3 days. In this step, the Cu-FAU-IWI crystal structure was dissolved, and amorphous precursors were formed under a high alkalinity environment; therefore, the remaining FAU crystal structure and the formation of an amorphous phase could be obtained together with Cu detachment from the zeolite surface. Consequently, in step (iii), Cu decorated on BEA (Cu-BEA-IZT) was initially formed, and its crystalline structure was continuously developed when prolonging the crystallization period. In this step, the Cu-NPs were migrated and encapsulated in the secondary porous structures of BEA crystals. However, the dissolution of the Cu-FAU-IWI crystalline structure and the formation of the amorphous precursor were competitive. Hence, a mixture of the amorphous phase and the BEA framework was obtained while a small portion of the starting FAU zeolite remained. Finally (step (iv)), the BEA zeolite (Cu-BEA-IZT) was completely formed by utilizing all amorphous precursors derived from Cu-FAU-IWI zeolite. In this step, the Cu-NPs were hypothesized to be encapsulated in the secondary porous structure of BEA zeolite. To confirm this hypothesis, a reference sample, Cu impregnated on the BEA obtained by the IZT approach (BEA-IZT) prepared by the conventional impregnation method (Cu-BEA-IWI), was made.

To confirm the crystalline structures of related materials, Fig. 1B shows the powder X-ray diffraction (PXRD) patterns of the as-synthesized samples before and after the IZT process. Obviously, the PXRD pattern of the starting FAU or commercial FAU (FAU-Com) showed characteristic diffraction peaks of the FAU topology at 2θ of 6.36°, 10.35°, 12.10°, 15.93°, 19.01°, 20.72°, 23.20°, 24.03°, 26.23°, 27.50°, 30.14°, 31.26°, 31.93°, and 34.65°, corresponding to plane indices (hkl) of (111), (220), (311), (331), (333), (440), (620), (533), (551), (642), (733), (660), (555), and (664), respectively (#PDF 01-073-2313).⁵⁶ Importantly, after the IZT process, the characteristic diffraction peaks at 2θ of 7.79° and 22.44° were observed, corresponding to plane indices (hkl) of (101) and (302), which reveal the presence of the BEA framework (#PDF 01-074-8795).⁵⁶ In addition, the characteristic diffraction peaks of the FAU framework are absent after the IZT process. These observations confirm that the parent FAU is completely transformed into the BEA framework. In fact, the FAU starts to dissolve, and the BEA can be formed during hydrothermal recrystallization. Typically, the FAU, which is categorized as one of the low-density zeolite frameworks, can be transformed into various kinds of zeolites, for example, beta (BEA), chabazite (CHA), mordenite (MOR), and ferrierite (FER).⁵⁷ Particularly, the presence of a high alkalinity environment promotes the dissolution of parent FAU zeolite into the amorphous phase. However, during the decomposition of the parent FAU zeolite, the formation of the daughter BEA zeolite takes place. Basically, the BEA zeolite can be formed from the building units of 12-, 6-, and 4-membered rings (MRs) from the parent FAU zeolite. However, the BEA framework generally consists of 12-, 6-, 5-, and 4-MRs building units. Instead of sharing 12-, 6-, and 4-MRs building units from the parent FAU framework, the rearrangement of the 5-MRs building unit has proceeded during the dissolution of the FAU framework into amorphous precursors to eventually form the BEA framework.^47,56,58 In addition, using specific structure directing agents (SDAs) can selectively control the formation of the BEA framework by providing the locally ordered aluminosilicate species that readily build up the BEA zeolite framework during the dissolution of FAU zeolite. Moreover, the formation of the BEA zeolite framework via the solvent-free IZT process proceeded through the amorphous fragment and particle assembly and/or attachment. Hence, it could be categorized as a non-classical crystallization process.⁴⁷ Straightforwardly, Cu-FAU-IWI can be further used as the starting source for the IZT process, eventually producing the Cu-BEA-IZT sample. Additionally, the PXRD pattern of Cu-FAU-IWI showed the sharp characteristic peaks of the FAU framework, which are similar to those of the FAU-Com sample with a relative crystallinity of 91% and 100% for Cu-FAU-IWI and FAU-Com, respectively. After that, Cu-FAU-IWI is successfully used as the starting source for the IZT process into BEA zeolite (Cu-BEA-IZT), implying that Cu-FAU-IWI can be used as the starting source containing both metal nanoclusters and the parent zeolite precursor for the solvent-free IZT process.

The different morphological structures of the as-prepared samples between the parent and daughter zeolite are clearly observed. It should be mentioned that FAU-Com shows a large particle size with a global particle size distribution (PSD) of 0.68 ± 0.20 μm (Fig. S1). After the IZT process to eventually produce the BEA zeolite (BEA-IZT), the nanosized structure of BEA zeolite was formed with a particle size distribution (PSD) of 88.57 ± 24.70 nm (Fig. S2). These observations clearly reveal that both the crystalline zeolite structure and the morphological structure can be changed after the IZT process. Interestingly, when combined with Cu nanoclusters, Cu-FAU-IWI shows a similar trend to bare FAU-Com.A large global particle size with a PSD of 0.64 ± 0.19 μm was obtained for Cu-FAU-IWI (Fig. 1C and D of a). Subsequently, the smaller particle size of Cu-BEA-IZT with a PSD of 90.30 ± 19.67 nm was obtained after the IZT process (Fig. 1C and D of b). In addition, the reference sample, Cu-BEA-IWI, prepared by conventional Cu impregnation onto BEA (Cu-BEA-IWI), shows similar zeolite particle sizes compared to Cu-BEA-IZT with a PSD of 93.37 ± 24.52 nm (Fig. 1C and D of c). These behaviors clearly confirm that the IZT approach changes the zeolite crystalline and morphological structure, while the Cu incorporation did not significantly affect the global particle size of zeolites compared to the one without metal clusters. However, the relative crystallinity of Cu-BEA-IZT was lower than that of metal-free BEA-IZT, with a relative crystallinity of 74% and 100% for Cu-BEA-IZT and bare BEA-IZT, respectively. This relates to the fact that the incorporation and insertion of Cu clusters into the secondary porous structure hinder the growth of the lattice plane index of (101), which appears at a 2θ of 7.79° (Fig. 1B and S3).

To further illustrate the metal nanocluster size distribution, Fig. 2 illustrates high-resolution transmission electron microscopy (HRTEM) and high-angle annular dark field scanning transmission electron microscopy (HAADF-STEM) images corresponding to the particle size of Cu-NPs deposited on zeolite surfaces. The HRTEM and HAADF-STEM images of Cu-FAU-IWI showed non-homogeneous dispersion of Cu-NPs supported on zeolite surfaces (Fig. 2A and B of a). The HRTEM images illustrate a lattice fringe of 1.28 nm for Cu-FAU-IWI corresponding to the FAU lattice plane index of (111) (Fig. 2A of a). In addition, as can be seen, obviously, the Cu-NPs were predominantly located at the outermost surface of FAU, eventually forming clusters.^59,60 Then, a broad PSD of Cu-NPs with 7.45 ± 3.20 nm was obtained (Fig. 2C of a). Moreover, EDS mapping confirms that Cu species are located on the outermost surfaces of the FAU zeolite and formed as clusters, as illustrated in Fig. S4A.

Fig. 2 (A) HRTEM images, (B) HAADF-STEM images, (C) PSD of the as-prepared samples, and (D) schematic illustration of Cu-NP supported on zeolites, including: (a) Cu-FAU-IWI, (b) Cu-BEA-IZT, and (c) Cu-BEA-IWI samples.

As aforementioned, after the IZT process, the redispersion of Cu-NPs could be addressed as shown in HRTEM and HAADF-STEM images (Fig. 2A and B of b). As stated above, a lattice fringe of 1.07 nm was obtained corresponding to the BEA lattice plane indices of (101) (Fig. 2A of b).^61–63 The highly dispersed Cu-NPs were observed over the BEA zeolite with a PSD of 2.43 ± 0.58 nm for Cu-BEA-IZT (Fig. 2A–C of b). Note that EDS mapping also confirms the absence of Cu clusters with a high dispersion of Cu species (Fig. S4B). These observations indicate that the IZT process can effectively redisperse Cu-NPs in the simultaneous transformation of one zeolite framework to another. Previously, the IZT process of a 2D zeolite into a 3D zeolite proved that metals (Au and Pt) could be confined and encapsulated inside transformed MWW supercage pores. The particle number of Pt of approximately 10–12 atoms and a cluster size of approximately 0.2–0.7 nm was obtained and confined in the cage of MWW-22 zeolite, while Pt impregnated in MWW-22 zeolite exhibits a large particle size of up to 30–50 nm.⁶⁴ In addition, Au has been encapsulated in MWW-22 zeolite pores with an Au size ranging from sub-nanoclusters of 0.5 to 1 nm and a nanoparticle size of 1 to 2 nm.⁶⁵ However, metal encapsulation was successfully obtained by adding the metal precursor during the transformation process. This did not demonstrate the redispersion of metal NP clusters from large metal particles deposited on the external surface of the parent zeolite into highly dispersed metals on the daughter zeolite surface. In this work, for the first time we report Cu-NP redispersion and encapsulation in the secondary porous structure of a daughter zeolite obtained via the IZT process.

In strong contrast to this, Cu-BEA-IWI shows non-uniformity and poor dispersion of Cu-NPs supported on BEA surfaces. The Cu-NPs are primarily located at the external BEA surface, which is similar to what has been observed for the Cu-FAU-IWI catalyst (Fig. 2A and B of c). The PSD of Cu-NPs is in the range of 7.48 ± 4.39 nm (Fig. 2C of c).⁵⁷ EDS mapping also illustrates that the Cu clusters are critically observed, confirming the aggregation of large Cu nanoparticles (Fig. S4C). To summarize, the schematic illustration in Fig. 2D indicates the location and distribution of Cu-NPs of Cu-FAU-IWI, Cu-BEA-IZT, and Cu-BEA-IWI catalysts. The Cu-NPs are mainly located on the outermost surfaces of the zeolite; this phenomenon is observed in the cases of Cu-FAU-IWI and Cu-BEA-IWI prepared by the direct impregnation process. In contrast, highly dispersed Cu-NPs were observed and were mostly located inside the secondary porous structure of BEA when the IZT process was applied.

To further visualize the lattice fringe of Cu-NPs, the Fast Fourier Transform (FFT) and Inverse Fast Fourier Transform (iFFT) of HRTEM images were applied. It should be noted that the HRTEM image of Cu-FAU-IWI (Fig. S5A) was used for FFT and iFFT investigation in selected areas of (I) (Fig. S5B), (II) (Fig. S5C), and (III) (Fig. S5D). The lattice fringes of the Cu-NPs in Cu-FAU-IWI were observed at 0.20 (Fig. S5B), 0.24 (Fig. S5C), and 0.20 nm (Fig. S5D) corresponding to Cu (111), (002), and (111) plane indices, respectively.⁶⁶ In addition, the HRTEM image of Cu-BEA-IZT (Fig. S6A) was used to investigate the lattice plane indices of Cu-NPs of the selected areas of (I) (Fig. S6B), (II) (Fig. S6C), and (III) (Fig. S6D). A Cu-NP lattice plane index on Cu-BEA-IZT of 0.21 was obtained in all selected areas of (I), (II), and (III), corresponding to the (111) plane index of Cu. Moreover, the Cu-NP lattice plane indices on the HRTEM image of Cu-BEA-IWI (Fig. S7A) showed a similar spacing of approximately 0.21 nm for area (I) (Fig. S7B), 0.21 nm for area (II) (Fig. S7C), and 0.21 nm for area (III) (Fig. S7D) corresponding to the (111) plane index.^66–68 It could be seen that the Cu lattice plane indices of Cu for Cu-BEA-IZT predominantly corresponded to the (111) plane index, while Cu-FAU-IWI showed (111) and (002) plane indices.

The physicochemical properties of the as-synthesized samples were also verified, as shown in Table 1 and S1. The starting FAU-Com with a Si/Al ratio of 15.8 was applied. After the IZT process, a slightly decreased Si/Al ratio of 13.4 was obtained in the transformed BEA-IZT samples (Table S1). Similarly, Cu-BEA-IZT and Cu-BEA-IWI samples showed a similar Si/Al trend to BEA-IZT with Si/Al ratios of 13.4 and 13.7, respectively, while the Si/Al ratio of Cu-FAU-IWI was 15.2 (Table 1). However, the Si/M ratio, where M is the summation of Al and Cu, of these three samples was not significantly different from the Si/M ratios of 9.6, 9.5, and 9.1 for Cu-FAU-IWI, Cu-BEA-IZT, and Cu-BEA-IWI, respectively.

Table 1 The physicochemical properties of the as-prepared catalysts

Sample	Si/Al^a	Si/M^b	Cu^c (wt%)	S BET (m^2 g^−1)	S micro (m^2 g^−1)	S ext (m^2 g^−1)	V total (cm^3 g^−1)	V micro (cm^3 g^−1)	V ext (cm^3 g^−1)	S ext/SBET^j	Average pore diameter^k (nm)
a The average Si/Al was measured from WDXRF. b The average Si/M (Si/(Al + Cu)) ratio was calculated from Si, Al, and Cu content measured from WDXRF. c Cu content was measured from WDXRF. d Specific BET surface area. e micropore surface area. f External surface area. g Total pore volume. h Micropore volume. i External volume. j External surface area fraction. k Average pore diameter was obtained from the BJH plot.
Cu-FAU-IWI	15.2	9.6	4.5	717	624	93	0.414	0.254	0.160	0.130	3.6
Cu-BEA-IZT	13.4	9.5	3.8	443	319	124	0.467	0.126	0.341	0.280	5.8
Cu-BEA-IWI	13.7	9.1	4.0	559	447	112	0.439	0.179	0.260	0.200	5.4

---

As expected, the N₂ adsorption/desorption isotherm of type I combined with the H4 hysteresis loop was obtained over the bare FAU-Com zeolite corresponding to a microporous structure with narrow-slit pore characteristics (Fig. S8) with a specific surface area (S_BET) and a total pore volume (V_total) of 773 m² g⁻¹ and 0.519 cm³ g⁻¹, respectively (Table S1).^69–71 After the IZT process, combined type I and IV isotherms with an H1 hysteresis loop were obtained over bare BEA-IZT corresponding to the combination of micro- and meso-porous characteristics with a well-defined narrow distribution of cylindrical pore channels with an S_BET of 680 m² g⁻¹ (Fig. S8 and Table S1).^69–71 Although the S_BET of bare BEA-IZT decreased with respect to the bare FAU-Com zeolite, an improvement of the external surface area (S_ext) and external pore volume (V_ext) from 106 to 153 m² g⁻¹ and 0.191 to 0.360 cm³ g⁻¹, respectively, was clearly seen, indicating enhanced mesoporosity and external surface area after the IZT process. The enhancement of the mesopore and external surface area fraction of BEA-IZT was observed at 0.225, while the mesopore and external surface area fraction of FAU-Com was 0.137. In addition, it could be seen that an increase in the average BJH pore diameter from 3.0 to 8.3 nm was obtained over FAU-Com and BEA-IZT samples, respectively (Fig. S9 and Table S1).

In the case of Cu-NP incorporated samples prepared by the IWI method, a N₂ adsorption/desorption isotherm of type I with an H4 hysteresis loop was obtained over Cu-FAU-IWI, which is similar to that of the FAU-Com sample, indicating the presence of micropores with narrow-slit pore characters (Fig. S8). In contrast, Cu-BEA-IZT and Cu-BEA-IWI samples exhibited combined type I and IV N₂ adsorption/desorption isotherms together with an H2 hysteresis loop, indicating disordered pore structures due to pore blocking when loading Cu species into microporous structures.^69–71 It should be mentioned that the pore diameter remarkably reduces from 8.3 nm (Fig. S9 and Table S1) for BEA-IZT to 5.8 and 5.4 nm (Fig. S10 and Table 1) for Cu-BEA-IZT and Cu-BEA-IWI, respectively. Likewise, an S_BET and V_total of 717 m² g⁻¹ and 0.414 cm³ g⁻¹, as well as 559 m² g⁻¹ and 0.439 cm³ g⁻¹, were obtained over Cu-FAU-IWI and Cu-BEA-IWI, respectively. Therefore, the significant reduction of S_BET and V_total of Cu-FAU-IWI and Cu-BEA-IWI, prepared by the IWI technique, was affected by the pore blocking of Cu-NPs, mostly located at the outermost surfaces or pore mouths of zeolites. In the case of Cu-BEA-IZT, an S_BET and V_total of 443 m² g⁻¹ and 0.467 cm³ g⁻¹, respectively, were obtained. Interestingly, the external surface area fraction (S_ext/S_BET) of Cu-BEA-IZT was significantly higher than that of Cu-FAU-IWI and Cu-BEA-IWI, with values of 0.130, 0.280, and 0.200 for Cu-FAU-IWI, Cu-BEA-IZT, and Cu-BEA-IWI, respectively. It again confirms that the IZT process can be applied to convert bulk zeolite to hierarchical zeolite structures.

2.2 Insights into the acidic properties of the as-prepared Cu-NP supported zeolites

The acidic properties of the prepared samples were examined to explore the related properties of catalysts that could affect the catalytic performance of the bioethanol dehydrogenation reaction. The NH₃-TPD profiles and pyridine-adsorbed FTIR spectra of the non-reduced sample are shown in Fig. 3. The deconvoluted NH₃-TPD profiles showed that the freshly prepared sample contained three types of acid sites, including weak, medium, and strong acid sites (Fig. 3A), corresponding to the desorption peaks around 200–250, 350–400, and 550–600 °C, respectively. In addition, the acidic properties are shown in Table 2. The Cu-FAU-IWI sample showed a moderate total acid density of 0.943 mmol g⁻¹. However, the Cu-BEA-IZT sample showed the lowest total acid density of 0.755 mmol g⁻¹. In strong contrast to this, the Cu-BEA-IWI sample showed the highest total acid density of 1.357 mmol g⁻¹. Moreover, the Lewis acid (L) to Brønsted acid (B) ratio (L/B) was investigated (Fig. 3B). The peak at wavenumbers of approximately 1450, 1490, and 1540 cm⁻¹ was assigned to Lewis, combined Lewis/Brønsted acid, and Brønsted acid sites, respectively. The sharp response peak corresponding to the Lewis acid site was obtained over Cu-FAU-IWI, Cu-BEA-IZT, and Cu-BEA-IWI samples. However, the small response peak corresponding to the Brønsted acid site was observed for the Cu-BEA-IWI sample due to the very low amount of Brønsted acid sites. L/B ratios of 6.14, 26.64, and 5.76 were obtained over Cu-FAU-IWI, Cu-BEA-IZT, and Cu-BEA-IWI samples. It would be indicated that the Lewis acid site plays a key role in the Cu-BEA-IZT sample, which is necessary for acetaldehyde production from ethanol.

Fig. 3 (A) NH₃-TPD profiles and (B) pyridine adsorbed FTIR spectra of the as-prepared catalysts including Cu-FAU-IWI, Cu-BEA-IZT, and Cu-BEA-IWI.

Table 2 The acidic properties of the as-prepared catalysts

Sample	Acid density^a (mmol g^−1)				L/B^b
	Weak	Medium	Strong	Total
a Acid density was measured using the NH3-TPD technique. b Lewis-to-Brønsted ratio was measured using the pyridine adsorption FTIR technique.
Cu-FAU-IWI	0.375	0.331	0.237	0.943	6.14
Cu-BEA-IZT	0.324	0.319	0.112	0.755	26.64
Cu-BEA-IWI	0.187	0.744	0.462	1.357	5.76

---

2.3 Inter-zeolite transformation behavior of Cu-FAU-IWI into Cu-BEA-IZT

According to the IZT process mechanism, the IZT behavior of Cu-FAU-IWI into Cu-BEA-IZT as a function of synthesis periods was verified by analysis of the collected samples at different synthesis times (Fig. 4). The PXRD patterns illustrate that the FAU zeolite framework starts to dissolve at the early stage of 2 and 4 h of the IZT process (Fig. 4A). This is evident from the decreased relative crystallinity of the FAU characteristic peaks from 100% to 55.1% and 54.1% for 0, 2, and 4 h, respectively (Fig. S10). In addition, a broad hump of the PXRD pattern in the 2θ range from 17.50° to 32.50° was observed, confirming the formation of the amorphous phase (Fig. 4A). Subsequently, after prolonging the crystallization time to 6 h, the intense PXRD also confirms the formation of the BEA zeolite framework with the characteristic diffraction peak at 2θ of 7.79° and 22.44°. However, the presence of a characteristic PXRD pattern of FAU zeolite was still obtained, especially at a 2θ of 6.36°, confirming that the FAU zeolite was not completely dissolved. The relative crystallinities of 36.0% and 64.0% were obtained for FAU and BEA zeolite frameworks, respectively (Fig. S11). Thereafter, upon prolonging the IZT process to 72 h, a sharp and intense PXRD pattern corresponding to the BEA zeolite was observed, while the PXRD pattern of the FAU framework completely disappeared. The relative crystallinities of 66.1%, 87.5%, 89.9%, 72.3%, and 74.1% were obtained for BEA zeolite within synthesis times of 8, 12, 24, 48, and 72 h, respectively. It should be noted that during the IZT process, the characteristic PXRD pattern corresponding to Cu was not detected. This may relate to a low loading of Cu species in the as-synthesized samples (Table S2).

Fig. 4 (A) PXRD patterns and (B) FESEM images of samples obtained from the IZT process of Cu-FAU-IWI as a function of recrystallization time: 2, 4, 6, 8, 12, 24, 48, and 72 h.

To visualize the morphological development during the IZT process, the morphological structure change was observed during the IZT process (Fig. 4B). The Cu-FAU-IWI sample shows a large crystal structure with a PSD of 0.64 ± 0.19 μm (Fig. 1C, D of a, and 4B). Subsequently, the morphological structure of Cu-FAU-IWI was destroyed and dissolved into solution in the simultaneous formation of an amorphous phase in an alkali environment. The morphological structure of the sample taken at 2 h illustrates the collapsed structure (Fig. 4B; 2 h). However, in this stage, the BEA framework was not developed, as indicated by the above-mentioned PXRD results (Fig. 4A). Then, the FAU structure after 4 h was dissolved continuously, and a collapsed structure with a broad PSD of 275.17 ± 182.27 nm was obtained (Fig. 4B; 4 h and S12; 4 h). This stage indicates the partial dissolution of the FAU framework into the amorphous phase and rearrangement into building blocks of the BEA zeolite framework. It could be confirmed that the peak characteristic of the parent FAU framework (600–400 cm⁻¹) from ATR-FTIR spectra disappeared during the IZT process at 2 and 4 h (Fig. S13).⁷²

Thereafter, the BEA framework was suddenly formed at 6 h of IZT corresponding to the PXRD pattern. The morphological structure was completely changed from a collapsed structure into nano-sized crystals with a PSD of 95.88 ± 22.47 nm (Fig. S12; 6 h). In this case, the FAU structure remained with a very small amount, as confirmed by the above-mentioned PXRD pattern and relative crystallinity (Fig. 4A and S11). Moreover, prolonging the IZT period would provide the rearrangement of the BEA zeolite framework. The PXRD pattern confirms the complete formation of BEA (Fig. 4A). The PSD of the obtained BEA was obtained in the range of 82.59 ± 20.05 (Fig. S12; 8 h), 103.90 ± 24.56 nm (Fig. S12; 12 h), 88.34 ± 21.08 (Fig. S12; 24 h), 76.69 ± 17.24 (Fig. S12; 48 h), and 90.30 ± 19.67 nm (Fig. S12; 72 h) upon prolonging recrystallization times of 8, 12, 24, 48, and 72 h, respectively.

From all the above-mentioned results, the schematic illustration of the IZT process of Cu-FAU-IWI into Cu-BEA-IZT was proposed, as shown in Scheme S1. For the mechanistic point of view, there are three main steps involved in the transformation processes: (i) the parent FAU framework was dissolved in a high alkalinity environment at the early stage of the IZT process and Cu particle detachment (at 2–4 h synthesis time), (ii) the dissolution of the parent FAU framework was continued with the formation of the daughter BEA framework accompanied by Cu confinement in the secondary pore structure, and (iii) the formation, development, and rearrangement of the daughter BEA framework with the Cu confinement process. The overall IZT process is illustrated in Scheme S1. At first, the FAU structure would be partially dissolved into an amorphous phase with the simultaneous formation of secondary building units. It could be confirmed that the characteristic structural double ring (D6R) occurring around 560 cm⁻¹, representing the FAU framework, disappeared at an early stage of the IZT process (Fig. S13). Moreover, the remaining ATR-FTIR peak characteristic around 455 cm⁻¹ corresponds to structure-insensitive tetrahedral bending, indicating the formation of amorphous and secondary building units.⁷² In addition, it could be noted that the Cu-NPs were detached and lodged in the solid amorphous phase, as can be seen from the increment of Cu content from 4.3 to 5.7% at recrystallization times of 0 and 4 h, respectively (Table S2). The HAADF-STEM image showed that at 4 h of the IZT process, Cu was present in a solid amorphous phase as well as in a liquid fraction (Fig. S14). However, Cu particles had a chance to migrate and aggregate with each other, as can be seen in Fig. S14B, where a large Cu particle size was observed. Fortunately, overall EDS mapping exhibited a uniform distribution of Cu over the amorphous solid. In this stage, the interaction between the solid amorphous phase and the Cu particle could physically interact with each other. After 8 h of the IZT process, the BEA zeolite was formed (Fig. 4A), and the Cu particles were observed (Fig. S14C and D). In the liquid phase, a huge and dense particle was observed, indicating the suspension of a remaining amorphous phase in deionized water during the preparation. After that, the size of the Cu particle was reduced because the Cu remained in a high alkali environment during exposure to high temperature, resulting in dissolution of the aggregated Cu particle. Moreover, Cu was confined in the secondary pore of BEA zeolite during the formation of BEA zeolite, as can be seen from Fig. S14D. The Cu particles are confined in the BEA zeolite particle, while some Cu particles are located at the outermost surface of the zeolite. Fortunately, the formation of BEA was still ongoing, so some Cu particles at the external surface and surrounding Cu particles can migrate inside into the secondary pore of the zeolite. After finishing the IZT process at 72 h, the HAADF-STEM images showed a very small Cu particle size of around 2.43 ± 0.58 nm (Fig. 2). Therefore, this phenomenon indicates that Cu first physically interacted with the FAU zeolite through the impregnation process. During the IZT process, the FAU dissolved into an amorphous form, and Cu was detached from the surface of the FAU zeolite and surrounded by the matrix. After that, during the formation of the BEA zeolite, the Cu was confined inside the secondary pore of the BEA zeolite. In this stage, the interaction between Cu and BEA zeolite through the IZT process would be stronger than that obtained through the impregnation process.

The Cu amount was increased from approximately 4.3% to 5.8% and 5.7% at the early stage of the IZT process of 2 and 4 h, respectively, indicating that Cu was in the solid amorphous phase. After that, Cu was partially confined in the secondary pore of the BEA structure during the formation of the BEA framework. Secondly, the dissolution of the FAU zeolite structure continued to dissolve. However, the number of building block units was enough to further recrystallize into the BEA zeolite framework using a specific SDA. Finally, the rearrangement and development of the BEA zeolite framework was pronounced. However, in this stage, the development of Cu confinement in the secondary pore of BEA zeolite continued to proceed. The final Cu content of 4.0% was obtained after 72 h of the IZT process, which was similar to the starting Cu content (4.3%) before the IZT process (0 h) (Table S2).

2.4 Insight into the oxidation state and reducibility of the as-prepared Cu-NP supported zeolites

To further investigate the structures of Cu-NPs, X-ray absorption spectroscopy (XAS) was performed over different prepared samples. The absorption edge and white line characteristics of Cu K-edge X-ray absorption near edge structure (XANES) spectra of all fresh samples were in between the characteristics of CuO and Cu₂O (Fig. S15). However, this distinctive feature is most similar to that of CuO. In addition, linear combination fitting (LCF) reveals that the fresh Cu-BEA-IZT sample contains 77.7% CuO and 22.3% Cu₂O , respectively (Fig. S15C). Moreover, fresh Cu-FAU-IWI (Cu-FAU-IWI-fresh) and fresh Cu-BEA-IWI (Cu-BEA-IWI-fresh) (samples prepared by the IWI method contain 90.2% and 91.9% CuO with 9.8% and 8.1% Cu₂O, respectively). H₂ temperature program reduction (H₂-TPR) was performed to monitor the reduction behavior of Cu-NPs. The H₂-TPR profiles of the as-prepared samples are illustrated in Fig. S16. The Cu-FAU-IWI-fresh sample shows two characteristic peaks regarding reduction temperature zones of Cu-NPs at approximately 300 and 450 °C, which are assigned to the reduction of CuO to Cu₂O species and the reduction of Cu₂O to metallic Cu species, respectively. However, after the IZT process, the reduction temperature of Cu-NPs on the Cu-BEA-IZT-fresh sample is higher than that of the Cu-FAU-IWI-fresh sample. A broad reduction peak illustrating the transformation of CuO to Cu₂O species is obtained, ranging approximately from 300 to 600 °C, while the transformation of Cu₂O to metallic Cu is obtained in the temperature range of 500 to 650 °C. This can be ascribed to the fact that improved interaction between small Cu-NPs and BEA surfaces with high dispersion is achieved after the IZT process.⁷³ In strong contrast to this, the Cu-BEA-IWI-fresh sample shows the broad single-reduction peak of CuO to metallic Cu species, appearing in the range from 200 to 400 °C. These observations relate to the formation of large Cu-NPs with poor dispersion, attributed to the weak interaction between Cu species and BEA surfaces.

Straightforwardly, the change in the oxidation state of Cu-NPs supported on zeolite after H₂ reduction was determined. The reduction process was performed under 10% H₂ at 450 °C for 30 min, using three samples Cu-FAU-IWI-fresh, Cu-BEA-IZT-fresh, and Cu-BEA-IWI-fresh, and after hydrogen pretreatment the samples were labelled as reduced Cu-FAU-IWI (Cu-FAU-IWI-red), reduced Cu-BEA-IZT (Cu-BEA-IZT-red), and reduced Cu-BEA-IWI (Cu-BEA-IWI-red), respectively. The Cu-NP species were discussed after reduction at 450 °C for 30 min under 10% H₂ (Cu-FAU-IWI-red, Cu-BEA-IZT-red, and Cu-BEA-IWI-red). The wide scan X-ray photoelectron spectroscopy (XPS) spectra and the XPS spectra of Cu 2p_3/2 are illustrated in Fig. S17 and 5A, respectively. However, the XPS spectra of Cu 2p_3/2 could be used to confirm the presence of CuO and the combination of Cu₂O with metallic Cu (Cu₂O + Cu) species. The binding energy over Cu-FAU-IWI-red was obtained to be 933.5 and 933.2, corresponding to CuO and (Cu₂O + Cu), respectively. However, after the IZT process, the binding energy shifted to 933.1 and 932.3. Typically, the binding energy of Cu 2p_3/2 below 933 eV is assigned to Cu₂O and metallic Cu species, and the one above 933 eV is assigned to the oxidation states of Cu^II.⁷⁴ From these observations, the binding energy of Cu-BEA-IZT-red shifts to a lower binding energy compared to Cu-FAU-IWI-red, suggesting that less positive charge is present in Cu-BEA-IZT-red.⁷⁵ As expected, a binding energy shift is also observed in the case of Cu-BEA-IWI-red. Binding energies of 933.9 and 932.7 eV were obtained, corresponding to CuO and (Cu₂O + Cu) species, respectively, corresponding to the more positive charge present in Cu-BEA-IWI-red compared to Cu-BEA-IZT-red.

Fig. 5 (A) XPS spectra of Cu 2p_3/2, (B) XPS spectra of Cu LMM Auger, (C) wide scan XAS spectra of the Cu K-edge, (D) narrow scans of the XAS spectra of the Cu K-edge, and (E) phase composition of Cu-FAU-IWI-red, Cu-BEA-IZT-red, and Cu-BEA-IWI-red samples. Note that the spectra of XPS and XAS were recorded after the samples were reduced at 450 °C for 30 min with 10% H₂ gas. The oxidation phase composition was calculated by linear combination fitting of XAS spectra.

To further identify Cu₂O and metallic Cu species of the samples after reduction at 450 °C for 30 min, Cu LMM Auger must be taken into account for proper discrimination of Cu-NP species. As illustrated in Fig. 5B, the Cu Auger LMM XPS signals show kinetic energies of 918.2 and 913.7 eV for Cu-FAU-IWI-red, assigned to metallic Cu and Cu₂O, respectively. In addition, the XPS spectra of Cu Auger LMM of Cu-BEA-IWI-red illustrate a kinetic energy shift to 918.6 and 914.0 eV, assigned to the characteristics of metallic Cu and Cu₂O, respectively. It should be noted that Cu species obtained by the IWI method (Cu-FAU-IWI-red and Cu-BEA-IWI-red samples) showed similar kinetic energies of approximately 918 and 914 eV for metallic Cu and Cu^I, respectively.^30,75–77 Interestingly, after the IZT process, the kinetic energy of Cu-BEA-IZT-red shifts to 921.3 and 919.3 eV, corresponding to the characteristics of metallic Cu and Cu₂O, respectively. This relates to the fact that the shift of kinetic energy to higher energy indicates a high fraction of metallic Cu contained in the Cu-BEA-IZT-red sample.

To further verify the Cu-NP species of reduced samples, the Cu K-edge X-ray absorption near edge structure (XANES) measurement and linear combination fitting (LCF) of XAS spectra were performed. The measured XANES spectra of reduced samples, Cu-FAU-IWI-red, Cu-BEA-IZT-red, and Cu-BEA-IWI-red, were compared with the reference XANES spectra of metallic Cu, Cu₂O, and CuO (Fig. 5C). The adsorption edge and white line characteristics of metallic Cu and Cu₂O are observed, indicating that the reduced samples contained mainly metallic Cu and Cu₂O species after H₂ pretreatment at 450 °C for 30 min. To gain insights into the oxidation state composition of reduced samples, the linear combination fitting (LCF) of XAS spectra was performed (Fig. 5D). Cu-FAU-IWI-red exhibits approximately 70.4% metallic Cu and 29.6% Cu₂O. Cu-BEA-IWI-red also shows approximately 73.0% metallic Cu and 27.0% Cu₂O. Remarkably, the Cu-BEA-IZT-red catalyst contains the metallic Cu and Cu₂O oxidation states in proportions of approximately 80.2% and 19.8%, respectively. In addition, considering the oxidation phase composition, the Cu₂O to metallic Cu (Cu₂O/Cu) ratio was calculated. Cu₂O/Cu ratios of 0.25, 0.42, and 0.37 were obtained for Cu-BEA-IZT-red, Cu-FAU-IWI-red, and Cu-BEA-IWI-red, respectively. Obviously, the Cu₂O/Cu ratio of Cu-FAU-IWI-red decreased from 0.42 to 0.25 after the IZT process. This indicates that the Cu-NPS of the Cu-BEA-IZT-red sample could be reduced to the highest content of metallic Cu via the H₂ reduction process at 450 °C for 30 min compared to other samples.

To investigate the change of Cu species during the H₂ pretreatment step, the time-resolved Cu K-edge X-ray absorption near edge structure (TR-XANES) was further examined under in situ 10% H₂/N₂ pretreatment at 450 °C for 30 min at a heating rate of 10 °C from 40 to 450 °C (Fig. 5). From this observation, two steps of the pretreatment process are included: (i) during heating under a N₂ atmosphere and (ii) during 10% H₂/N₂ pretreatment at 450 °C for 30 min. For the first step (stage (i)), during heating to 450 °C under the N₂ atmosphere, the self-reduction/autoreduction of Cu-NPs during elevated temperature occured.^78,79 The self-reduction could generally proceed via oxygen-bridged [Cu–O–Cu]²⁺ by hydration and/or O₂ release from decomposition of [Cu–O–Cu]²⁺.^78–80

As illustrated in Fig. 6, the peak intensity of the white line gradually decreases during heating under the N₂ atmosphere to 450 °C (Fig. 6A), as observed with the Cu-FAU-IWI-red sample, and the phase composition of CuO sharply decreased from 95.2 to 36.0% during heating to 300 °C (Fig. 6B). After heating to 450 °C under the N₂ atmosphere, the Cu-NP species comprised 75.3% Cu₂O and 24.7% metallic Cu. In the case of the Cu-BEA-IZT-red sample, self-reduction of Cu-NPs proceeded with a gradual decrease in the white line intensity (Fig. 6C), resulting in 28.5% CuO, 71.5% Cu₂O, and no detectable metallic Cu (Fig. 6D). It is evident that the self-reduction of the Cu-BEA-IZT-red sample required more effort than the Cu-FAU-IWI-red sample during heating to 450 °C under the N₂ atmosphere, which might be due to the strong interaction between the Cu-NPs and BEA zeolite support. It is important to note that the Cu-NP-derived IWI method provides a weak interaction between the Cu-NPs and zeolite support and is more prone to the self-reduction process of Cu-NPs. This behavior was strongly confirmed by the Cu-BEA-IWI-red sample, where the gradual decrease in white line intensity was obtained during heating to 450 °C under the N₂ atmosphere (Fig. 6E). Cu-NP species comprising 61.8% Cu₂O and 38.2% metallic Cu were obtained at 450 °C under the N₂ atmosphere (Fig. 6F). The observation of high metallic Cu (38.2%) indicates the self-reduction behavior of the Cu-NPs derived from the IWI preparation method.

Fig. 6 (A, C, and E) TR-XANES spectra and (B, D, and F) phase composition obtained from linear combination fitting (LCF) of (A and B) Cu-FAU-IWI-red, (C and D) Cu-BEA-IZT-red, and (E and F) Cu-BEA-IWI-red catalysts.

In the second step (stage (ii)), in situ H₂ pretreatment using 10% H₂/N₂ at 450 °C for 30 min was applied. Obviously, the characteristic spectra of the Cu-FAU-IWI-red sample during the in situ H₂ reduction process at 450 °C differed from those pretreated under N₂ (Fig. 6A). Cu-FAU-IWI-red showed that the Cu₂O content linearly decreased from 75.3% to 31.2% obtained with 0 to 12 min of the in situ H₂ reduction process, respectively (Fig. 6B). In other words, the metallic Cu content from 24.7% to 68.8% was obtained with 0 and 12 min of the reduction process, respectively. After that, the content of Cu₂O and metallic Cu of Cu-FAU-IWI-red remained constant until 30 min of the reduction process, with an absence of CuO species. In contrast, Cu-BEA-IZT-red demonstrated the highest content of metallic Cu during H₂ reduction within 12 min. The white line characteristic spectra during H₂ pretreatment were different from the white line characteristic spectra during heating under the N₂ atmosphere with a large decrease in white line intensity (Fig. 6C). Cu-NP species comprising 22.4% Cu₂O and 77.6% metallic Cu were obtained after 12 min of the in situ H₂ reduction process, respectively (Fig. 6D). After that, the Cu₂O and metallic Cu contents remained constant at approximately 20.0% and 80.0% until 30 min of the in situ H₂ reduction process.

In the case of Cu-BEA-IWI-red during the in situ H₂ reduction process, the white line characteristic spectra were predominantly attributed to the presence of both Cu₂O and metallic Cu (Fig. 6E). However, the linear decrease of Cu₂O was observed while the linear increase of metallic Cu was found. The Cu₂O species content decreased from 61.8% to 27.0%, while metallic Cu content increased from 38.2% to 73.0%, which was obtained with 0 and 30 min of the in situ H₂ reduction process (Fig. 6F).

From the results of TR-XAS observation, it could be seen that the freshly prepared samples contained CuO as the major composition. However, the weak interaction between support and Cu-NPs obtained by the IWI preparation method leads to easy occurrence of the auto-reduction of Cu-NPs species of Cu-FAU-IWI-red and Cu-BEA-IWI-red samples by presenting metallic Cu species during heating to 450 °C under an N₂ atmosphere.^78–80 In the case of Cu-BEA-IZT-red, it occurs only in CuO and Cu₂O phase composition during heating to 450 °C under an N₂ atmosphere, which might possibly be due to the strong interaction between Cu and the support, and also, the smaller Cu-NP size restricts the ability to be reduced by hydrogen. However, after the H₂ reduction process at 450 °C for 30 min was performed, metallic Cu species were obtained with Cu-BEA-IZT-red. The highest phase composition of metallic Cu was obtained with the Cu-BEA-IZT-red sample. These observations confirm that the IZT process illustrates the advantage of high metal dispersion and also provides a high amount of uniformly metallic Cu species after H₂ reduction compared to the other samples.

2.5 Catalytic dehydrogenation of bioethanol

To illustrate the advantages of the as-prepared catalysts obtained via anin situ solvent-free IZT process, the catalytic reaction of bioethanol dehydrogenation to acetaldehyde was chosen as the model reaction to evaluate the catalytic performance of the Cu-NP supported on zeolites. Prior to the catalytic performance testing, the catalyst was reduced at 450 °C for 30 min under a 10% H₂ atmosphere. The catalytic activity of bioethanol conversion to acetaldehyde via dehydrogenation over different catalysts is illustrated in Fig. 7. At first, the catalytic activity of the Cu-FAU-IWI-red catalyst was studied as the benchmarking experiment. It was found that a complete ethanol conversion of 100.0% was observed. However, the ethanol conversion slightly decays as a function of time-on-stream (TOS) (Fig. 7A). Upon prolonging the reaction time to 15 h, an ethanol conversion of 89.4% was obtained. Obviously, a major product selectivity of 64.2% of ethylene was obtained with a small portion of 18.0% acetaldehyde, 2.4% butylene, and 12.6% diethyl ether. It should be noted that ethylene selectivity was approximately higher than 80.0% at an early stage of the reaction period (1–2 h); subsequently, a decrease in ethylene selectivity was observed with the pronounced production of diethyl ether was observed after 5 h of TOS.

Fig. 7 Catalytic dehydrogenation of bioethanol to acetaldehyde over (A) Cu-FAU-IWI-red, (B) Cu-BEA-IZT-red, and (C) Cu-BEA-IWI-red catalysts, and (D) acetaldehyde yield over different catalysts.

In the case of the Cu-BEA-IZT-red catalyst, at the initial reaction time (at 1 h), a remarkable product selectivity of 76.1% acetaldehyde, 10.3% ethylene, 3.0% diethyl ether, 1.3% methane, and 1.2% propylene was observed at an ethanol conversion of 100.0% (Fig. 7B). When the reaction time was extended from 2 to 15 h, the remarkable selectivity of acetaldehyde was still obtained as the major product. Unfortunately, the slight reduction of acetaldehyde selectivity was obtained by decreasing from 73.8% to 61.6% when prolonging reaction time from 2 to 15 h. In addition, a low ethylene selectivity was obtained, approximately in the range of 10–14%. It might be due to the utilization of the Cu-NP active site while prolonging the reaction time. However, the Cu-BEA-IZT-red catalyst showed superior acetaldehyde production over the Cu-FAU-IWI-red catalyst, which might be due to the presence of a large number of metallic Cu active sites on Cu-BEA-IZT-red when compared to the lower amount of metallic Cu on the Cu-FAU-IWI-red catalyst.

Compared with the Cu-BEA-IWI-red catalyst, the Cu-BEA-IWI-red catalyst exhibits a similar catalytic activity to the Cu-FAU-IWI-red sample (Fig. 7C). An ethanol conversion above 95.0% was obtained within the first five hours of reaction time, which then gradually decreased to 89.1% at 15 h, which is similar to the Cu-FAU-IWI-red catalyst. It could be noted that the major product selectivity of 52.7% ethylene, 8.0% acetaldehyde, and 34.0% diethyl ether was obtained at a reaction time of 15 h. It should be important to mention that in this case, a very low amount of acetaldehyde can be produced with the main portion of ethylene. Moreover, the ethylene selectivity remarkably decreases, while the diethyl ether selectivity increases with an increase in reaction time.

To obtain insight into the catalytic activity, the L/B, Cu⁰/Cu^I ratios, acetaldehyde, and ethylene selectivity were plotted (Fig. S18). It was found that the Cu-BEA-IWI-red catalyst which had the lowest L/B (5.76) and Cu⁰/Cu^I (0.71) ratios exhibited the lowest acetaldehyde selectivity (7.99%) with relatively high ethylene selectivity (52.65%) after 15 h of TOS. Consequently, Cu-FAU-IWI-red which had a L/B ratio of 6.14 and a Cu⁰/Cu^I ratio of 1.82, exhibited an acetaldehyde selectivity of 18.00% and an ethylene selectivity of 64.20% after 15 h of TOS. Promisingly, Cu-BEA-IZT-red which had the highest L/B ratio of 26.64 and Cu0/CuI ratio of 3.37, exhibited the highest acetaldehyde selectivity of 64.70% and ethylene selectivity of 13.90% after 15 h of TOS. Furthermore, these observations can be related to the size of the Cu nanoparticle supported on the zeolite and low L/B and Cu⁰/Cu^I ratios were found with the catalyst with large Cu nanoparticles or clusters, eventually resulting in low acetaldehyde selectivity, where a Cu particle size of 7.48 ± 4.39 nm was obtained with Cu-BEA-IWI and 7.45 ± 3.20 nm was obtained with Cu-FAU-IWI (Fig. 2B and C). Interestingly, a small Cu particle size of 2.43 ± 0.58 nm, which was obtained with Cu-BEA-IZT, exhibited the highest L/B and Cu⁰/Cu^I ratios, resulting in the highest acetaldehyde selectivity. These observations indicate that metallic Cu (Cu⁰) plays a vital role in the bioethanol dehydrogenation reaction.

However, the presence of Brønsted acid sites might affect the shift to the side reaction to produce ethylene. The Brønsted acid site can directly promote the ethanol dehydration reaction to form ethylene. In addition, it can form ethylene through an alternative route via ethanol dimerization on a Brønsted acid site to form diethyl ether, which is then subsequently cracked on Brønsted acid sites and Lewis acid sites to form ethylene (Fig. S19).⁸¹ The alternative route of the side reaction might be facilitated with Cu-BEA-IWI-red catalysts. This is because Cu-BEA-IWI-red exhibits 34.02% diethyl ether selectivity after 15 h of TOS, and then this diethyl ether might be further cracked by the presence of a Lewis acid to form ethylene. Cu-FAU-IWI-red promotes 12.56% diethyl ether selectivity after 15 h of TOS; however, the Cu⁰/Cu^I ratio of Cu-FAU-IWI was >2-fold higher than that of Cu-BEA-IWI-red. Therefore, the direct acetaldehyde formation by the Lewis acid site and the side reaction were competitive. A more abundant metallic Cu active site can more proficiently promote acetaldehyde production. Hence, the acetaldehyde selectivity of Cu-FAU-IWI-red was higher than that of Cu-BEA-IWI-red. In the case of Cu-BEA-IZT-red, a diethyl ether selectivity of less than 10.00% was obtained within the first 10 h of TOS, and then it was promoted to 13.20% after 15 h of TOS. This might be due to the loss of the Cu active site used to produce acetaldehyde, resulting in an increase in diethyl ether formation by the side reaction. Indeed, this phenomenon indicates that acid properties synergize with the Cu nanoparticle size and oxidation state to promote the bioethanol dehydrogenation reaction to produce acetaldehyde.

Moreover, the characterization of spent catalysts (spent Cu-FAU-IWI-red, spent Cu-BEA-IZT-red, and spent Cu-BEA-IWI-red) was addressed. The coke formation was determined by thermogravimetric analysis (TGA) and the temperature programmed oxidation of oxygen (O₂-TPO) technique. The TGA analysis showed that the spent Cu-BEA-IZT-red contained carbonaceous deposition, as indicated by a weight loss of 13.9 wt% (Fig. S20 and Table S3). Consequently, the spent Cu-BEA-IWI-red exhibited a weight loss of 16.6 wt% and the spent Cu-FAU-IWI-red exhibited a weight loss of 19.5 wt%. In addition, the type of coke species could be analyzed by the temperature programmed oxidation of oxygen (O₂-TPO) technique (Fig. S21). The O₂-TPO profile exhibited two peaks, at approximately 400 °C and 500 °C, which can be assigned to soft and hard coke, respectively. Soft coke generally starts to combust at a temperature of around 200 to 400 °C, and hard coke starts to combust from 400 to 700 °C.^82,83 In addition, soft coke can be attributed to light carbonaceous species,^84,85 as well as confined oligomers, while hard coke can be attributed to more developed carbonaceous species.^86,87 The amount of each type of coke is calculated from the consumed oxygen amount and is shown in Table S3. The total amount of coke of spent Cu-FAU-IWI-red, spent Cu-BEA-IZT-red, and spent Cu-BEA-IWI-red was 2.60, 1.92, and 2.09 mmol g⁻¹, respectively. From these results, it might be useful to clarify that the decay of acetaldehyde selectivity might be affected by coke deposition and metal sintering, which could possibly result in pore blocking and the loss of the Cu active site, resulting in the reduction of acetaldehyde selectivity. Moreover, the physicochemical and morphological properties of the post-reaction catalyst were discussed. The results confirmed that the crystal structure and morphology of the zeolite support were not destroyed or collapsed (Fig. S22). The PXRD pattern of spent Cu-FAU-IWI-red was similar to those of FAU-Com and the reference FAU zeolite (Fig. S22A). Moreover, the PXRD patterns of spent Cu-BEA-IZT-red and spent Cu-BEA-IWI-red still preserved the characteristic diffraction of BEA topology. The PXRD patterns of the spent catalyst indicated that the crystal structure of the zeolite support was preserved during the bioethanol dehydrogenation reaction. In addition, the morphological structure confirmed that the zeolite support was stable and retained a similar morphological structure to the original shape (Fig. S22B and 1C), where the discrete crystal particles of the spent catalyst were observed. Therefore, from the overall characterization of spent catalysts, it can be clarified that the decline of acetaldehyde selectivity is affected by coke deposition and the Cu metal sintering phenomena.

2.6 Potential application of Cu-NP redispersion obtained from the spent Cu-FAU-IWI-red catalyst after bioethanol dehydration

To extend the feasibility of potential applications using the benefits of the solvent-free IZT process, we applied the spent Cu-FAU-IWI-red catalyst after the bioethanol dehydrogenation reaction as the starting source for the IZT process instead of using the fresh Cu-FAU-IWI as the starting source. The sample after the solvent-free IZT process using the spent Cu-FAU-IWI-red catalyst as the starting source was labelled as Re-Cu-BEA-IZT. The PXRD pattern confirms that the spent Cu-FAU-IWI-red catalyst has been successfully applied as the starting zeolite for the IZT process (Fig. 8A). The characteristic PXRD pattern of the recrystallized samples (Re-Cu-BEA-IZT) corresponding to BEA topology was obtained after the IZT process. The characteristic diffraction peaks at 2θ of 7.79° and 22.44° were obtained to confirm the formation of the BEA framework in the absence of the characteristic PXRD pattern of FAU zeolite. These observations confirm that the spent Cu-FAU-IWI-red is completely transformed into the recrystallized samples (Re-Cu-BEA-IZT). It is worth mentioning that the photograph of the spent Cu-FAU-IWI-red shows a brown–black color (Fig. 8B), while after the IZT process, the Re-Cu-BEA-IZT catalyst exhibits a blue color. This also implies that the Cu-NPs were redispersed, and the carbonaceous or coke-deposited compounds were completely removed after the IZT method. The amount of coke formation of the spent catalysts was analyzed using the thermogravimetric analysis (TGA) method, and the results are illustrated in Fig. S20. It can be seen that the coke deposition of 19.5% was obtained over the spent Cu-FAU-IWI-red catalyst. Moreover, the HAADF-STEM images reveal different Cu-NP sizes between spent Cu-FAU-IWI-red and Re-Cu-BEA-IZT catalysts. The Cu-NP size of the spent Cu-FAU-IWI catalyst is approximately 12.60 ± 6.15 nm (Fig. S23). After the IZT process, the redispersion of Cu-NPs was successfully performed. A Cu-NP size of 2.67 ± 0.52 nm was obtained with a Re-Cu-BEA-IZT sample. It has been observed that not only fresh parent Cu supported on zeolite can be used as the starting precursor for the IZT process, but also the spent catalyst can be used as the starting source for the solvent-free IZT process, eventually reproducing highly dispersed metals supported on recrystallized zeolite surfaces.

Fig. 8 (A) PXRD patterns, (B) photograph of the spent Cu-FAU-IWI-red catalyst, (C) photograph of the Re-Cu-BEA-IZT catalyst, (D) HAADF-STEM of the spent Cu-FAU-IWI catalyst, (E) HAADF-STEM of the Re-Cu-BEA-IZT catalyst, and (F) catalytic activity of Re-Cu-BEA-IZT.

To investigate the performance of Re-Cu-BEA-IZT, ethanol dehydrogenation was performed. Re-Cu-BEA-IZT was reduced at 450 °C for 30 min under a 10% H₂ atmosphere before testing the catalytic activity (Re-Cu-BEA-IZT-red). The catalytic activity of the Re-Cu-BEA-IZT-red catalyst is also illustrated in Fig. 8F. In the initial reaction (at 1 h), a remarkable predominant product selectivity of 86.3% acetaldehyde and 3.9% ethylene was observed at an ethanol conversion of 100.0%. Upon prolonging the reaction time to 15 h, an acetaldehyde selectivity of 87.7% was obtained. However, ethanol conversion decreased from 100.0% to 87.7% as TOS increased from 2 to 8 h. Obviously, Re-Cu-BEA-IZT-red exhibits a similar catalytic bioethanol conversion to Cu-BEA-IZT-red in the range of 85.0–100.0% with a constant acetaldehyde selectivity. This observation of the superior catalytic performance of the Re-Cu-BEA-IZT-red catalyst was highlighted. In addition, a comparison between our catalyst and the literature based on Cu-based catalysts for the ethanol dehydrogenation reaction to acetaldehyde in recent years has been performed, as shown in Table S4. It was found that acetaldehyde selectivity higher than 90.0% was achieved by using a Cu loading of higher than 10.0 wt%, while our catalyst had approximately 4.0 wt% Cu loading and can achieve up to 80.0% acetaldehyde selectivity. Therefore, this first example opened up a new perspective on catalyst preparation for sustainable bioethanol dehydrogenation via the solvent-free IZT process, which can redisperse sintered metals after a catalytic reaction.

In addition, prolonging the TOS up to 25 h using the Re-Cu-BEA-IZT-red catalyst was addressed (Fig. S24). The catalytic activity of the bioethanol dehydration reaction showed a slight decline in acetaldehyde selectivity over TOS. An acetaldehyde selectivity of higher than 80.0% was obtained within 20 h of TOS. After that, it decreased to less than 80.0% and an acetaldehyde selectivity of 77.1% was obtained at 25 h of TOS. In the case of bioethanol conversion, it was found that a bioethanol conversion of more than 80.0% was achieved in the first 15 h of TOS. Then, the bioethanol conversion dramatically decreased from 87.7% to 72.6% at TOS of 15 and 16 h, respectively. This could be affected by the loss of the Cu active site by coke deposition and Cu metal sintering, resulting in the shift to side reaction, which was found that the diethyl ether selectivity gradually increased after 15 h of TOS. However, this work demonstrates novel strategies for the redispersion of Cu-NP species deposited on zeolite surfaces via the solvent-free IZT approach for bioethanol dehydrogenation. Further study focusing on the active site and elimination of side reactions should hint at a suitable design process for the catalyst to achieve long-term durability.

3 Conclusions

In this work, small-sized Cu-NPs supported on BEA zeolite with high dispersion were successfully prepared via a solvent-free inter-zeolite transformation (IZT) process. The Cu-NP size was reduced from approximately 7.5 nm to 2.5 nm after the IZT process and encapsulated in the secondary pores of BEA zeolite.

The interaction between Cu nanoparticles and the zeolite support in Cu-BEA-IZT, prepared via the IZT process, was stronger than that in Cu-FAU-IWI and Cu-BEA-IWI, which were synthesized using the impregnation method. In addition, in situ TR-XANES showed that the reduced Cu-BEA-IZT (Cu-BEA-IZT-red) sample contained the highest metallic Cu content at 80.0%, while the reduced Cu-FAU-IWI (Cu-FAU-IWI-red) and the reduced Cu-BEA-IWI (Cu-BEA-IWI-red) contained 68.8% and 73.0% metallic Cu, respectively, during the in situ hydrogen pretreatment. The Cu-BEA-IZT-red sample showed a superior acetaldehyde selectivity of higher than 61.6%, while Cu-FAU-IWI-red and Cu-BEA-IWI-red showed a very low acetaldehyde selectivity of 18.0% and 8.0%, respectively. From the catalytic activity, it was found that superior catalytic acetaldehyde production was obtained with Cu-BEA-IZT-red, which has the largest number of metallic Cu active sites and highest L/B ratio. In addition, not only the freshly prepared Cu-FAU-IWI could be used as the starting zeolite for the solvent-free IZT process, but also the spent Cu-FAU-IWI-red could be surprisingly used as the starting precursor. The spent Cu-FAU-IWI-red was successfully applied as a starting precursor for the IZT process. After IZT, the Cu-NP size decreased from 12.60 ± 6.15 nm to 2.67 ± 0.52 nm for spent Cu-FAU-IWI-red and recrystallized Cu-BEA-IZT (Re-Cu-BEA-IZT) samples, respectively. Catalytic acetaldehyde production over Re-Cu-BEA-IZT-red showed an acetaldehyde selectivity of above 82.0% with a bioethanol conversion of higher than 87.7%. This work demonstrates novel strategies for the redispersion of Cu-NP species deposited on zeolite surfaces via the solvent-free IZT approach for bioethanol dehydrogenation.

Author contributions

Krissanapat Yomthong: conceptualization, performed all experiments, characterization, data collection, formal analysis, investigation, visualization, and writing of the manuscript. Ammarika Makdee: conceptualization, performed experiments, characterization, and investigations. Asadawut Soyphet: performed the experiment and characterization. Kachaporn Saenluang: performed the experiment and characterization. Narasiri Maineawklang: performed the experiment and the TR-XANES experiment. Somlak Ittisanronnachai: performed XPS experiment, analysis, and investigation. Wanwisa Limphirat: performed XAS and TR-XANES analysis and investigation. Pinit Kidkhunthod: XAS and TR-XANES support. Chularat Wattanakit: conceptualization, methodology, funding acquisition, resources, writing – review & editing of the manuscript, and supervision.

Conflicts of interest

There are no conflicts to declare.

Data availability

In addition, the data findings in this study can be accessed via the provided link below. https://drive.google.com/drive/folders/1yXKj0tmeGllH7XPPTfiBUyxyar9Fl72s?usp=sharing

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: experimental section, Scheme S1, Fig. S1–S24, and Tables S1–S4. See DOI: https://doi.org/10.1039/d5ta05755h.

Acknowledgements

This project is supported by the National Research Council of Thailand/Vidyasirimedhi Institute of Science and Technology (No.42A660307), the Thailand Science Research and Innovation (TSRI, FRB680014/0457), and the Program Management Unit for Human Resources & Institutional Development, Research and Innovation (grant number: B41G680026 (Global League)).

References

1. Q. Zhang, S. Gao and J. Yu, Chem. Rev., 2023, 123, 6039–6106.
2. M. Shamzhy, M. Opanasenko, P. Concepción and A. Martínez, Chem. Soc. Rev., 2019, 48, 1095–1149.
3. A. Martínez, M. A. Arribas and S. B. C. Pergher, Catal. Sci. Technol., 2016, 6, 2528–2542.
4. A. Gutiérrez, J. M. Arandes, P. Castaño, M. Olazar and J. Bilbao, Fuel, 2012, 94, 504–515.
5. C.-K. Park and S.-K. Ihm, Appl. Catal., A, 2000, 203, 201–209.
6. M. B. Gawande, A. Goswami, F. X. Felpin, T. Asefa, X. Huang, R. Silva, X. Zou, R. Zboril and R. S. Varma, Chem. Rev., 2016, 116, 3722–3811.
7. P. H. Finger, T. A. Osmari, N. M. Cabral, J. M. C. Bueno and J. M. R. Gallo, Catal. Today, 2021, 381, 26–33.
8. A. G. Sato, D. P. Volanti, I. C. De Freitas, E. Longo and J. M. C. Bueno, Catal. Commun., 2012, 26, 122–126.
9. P. C. P. Caldas, J. M. R. Gallo, A. Lopez-Castillo, D. Zanchet and J. M. C. Bueno, ACS Catal., 2017, 7, 2419–2424.
10. X. Zhang, J. X. Liu, B. Zijlstra, I. A. W. Filot, Z. Zhou, S. Sun and E. J. M. Hensen, Nano Energy, 2018, 43, 200–209.
11. Q. L. Tang, Q. J. Hong and Z. P. Liu, J. Catal., 2009, 263, 114–122.
12. H. Yue, Y. Zhao, S. Zhao, B. Wang, X. Ma and J. Gong, Nat. Commun., 2013, 4, 1859.
13. J. Ob-Eye, P. Praserthdam and B. Jongsomjit, Catalysts, 2019, 9, 66.
14. P. H. Finger, T. A. Osmari, M. S. Costa, J. M. C. Bueno and J. M. R. Gallo, Appl. Catal., A, 2020, 589, 117236.
15. I. C. Freitas, S. Damyanova, D. C. Oliveira, C. M. P. Marques and J. M. C. Bueno, J. Mol. Catal. A: Chem., 2014, 381, 26–37.
16. I. C. Freitas, J. M. R. Gallo, J. M. C. Bueno and C. M. P. Marques, Top. Catal., 2016, 59, 357–365.
17. A. G. Sato, D. P. Volanti, D. M. Meira, S. Damyanova, E. Longo and J. M. C. Bueno, J. Catal., 2013, 307, 1–17.
18. H. Wang, X.-K. Gu, X. Zheng, H. Pan, J. Zhu, S. Chen, L. Cao, W.-X. Li and J. Lu, Sci. Adv., 2019, 5, eaat6413.
19. Y. Zhong, Z. Sun, B. Y. Xia and Y. Su, Chem.–Eur. J., 2025, 31, e202500770.
20. T. W. van Deelen, C. Hernández Mejía and K. P. de Jong, Nat. Catal., 2019, 2, 955–970.
21. J. Wang, Y. Wang, C. Cai, Y. Liu, D. Wu, M. Wang, M. Li, X. Wei, M. Shao and M. Gu, Nano Lett., 2023, 23, 1897–1903.
22. Y. H. Huang, C. L. Hsueh, H. P. Cheng, L. C. Su and C. Y. Chen, J. Hazard. Mater., 2007, 144, 406–411.
23. J. Li, A. Ozden, M. Wan, Y. Hu, F. Li, Y. Wang, R. R. Zamani, D. Ren, Z. Wang, Y. Xu, D. H. Nam, J. Wicks, B. Chen, X. Wang, M. Luo, M. Graetzel, F. Che, E. H. Sargent and D. Sinton, Nat. Commun., 2021, 12, 2808.
24. C. Xu, Y. Shan, M. Bilal, B. Xu, L. Cao and Q. Huang, Chem. Eng. J., 2020, 395, 125093.
25. D. Hou, Z. Ding, P. Wang, M. Wang, Q. Xu, X. Wang, J. Guan, Y. Su and Y. Zhang, ACS Sustain. Chem. Eng., 2022, 10, 8094–8104.
26. Y. Wu, X. Zhang, F. Wang, Y. Zhai, X. Cui, G. Lv, T. Jiang and J. Hu, Ind. Eng. Chem. Res., 2021, 60, 8386–8395.
27. N. M. Latiff, X. Fu, D. K. Mohamed, A. Veksha, M. Handayani and G. Lisak, Carbon, 2020, 168, 245–253.
28. K. A. Tarach, M. Jabłońska, K. Pyra, M. Liebau, B. Reiprich, R. Gläser and K. Góra-Marek, Appl. Catal., B, 2021, 284, 119752.
29. M. Jabłońska, ChemCatChem, 2020, 12, 4490–4500.
30. P. Cao, L. Lin, H. Qi, R. Chen, Z. Wu, N. Li, T. Zhang and W. Luo, ACS Catal., 2021, 11, 10246–10256.
31. C. Paolucci, J. R. Di Iorio, W. F. Schneider and R. Gounder, Acc. Chem. Res., 2020, 53, 1881–1892.
32. X. Tang, J. Ye, L. Guo, T. Pu, L. Cheng, X.-M. Cao, Y. Guo, L. Wang, Y. Guo, W. Zhan and S. Dai, Adv. Mater., 2023, 35, 2208504.
33. J. Kim, E. Jang, Y. Jeong, H. Baik, S. J. Cho, C. Y. Kang, C. H. Kim and J. Choi, Chem. Eng. J., 2022, 430, 132552.
34. E. Jang, L. Choi, J. Kim, Y. Jeong, H. Baik, C. Y. Kang, C. H. Kim, K. Y. Lee and J. Choi, Appl. Catal., B, 2021, 287, 119951.
35. L. Singh, P. Rekha and S. Chand, Sep. Purif. Technol., 2016, 170, 321–336.
36. M. Richter, M. J. G. Fait, R. Eckelt, E. Schreier, M. Schneider, M. M. Pohl and R. Fricke, Appl. Catal., B, 2007, 73, 269–281.
37. H. Kim, E. Jang, Y. Jeong, J. Kim, C. Y. Kang, C. H. Kim, H. Baik, K. Y. Lee and J. Choi, Catal. Today, 2018, 314, 78–93.
38. W. B. Widayatno, G. Guan, J. Rizkiana, J. Yang, X. Hao, A. Tsutsumi and A. Abudula, Appl. Catal., B, 2016, 186, 166–172.
39. F. W. Chang, H. C. Yang, L. S. Roselin and W. Y. Kuo, Appl. Catal., A, 2006, 304, 30–39.
40. H. Wang, L. Wang and F. S. Xiao, ACS Cent. Sci., 2020, 6, 1685–1697.
41. M. Choi, Z. Wu and E. Iglesia, J. Am. Chem. Soc., 2010, 132, 9129–9137.
42. L. Wang, S. Xu, S. He and F. S. Xiao, Nano Today, 2018, 20, 74–83.
43. Y. Yan, Z. Zhang, S. M. Bak, S. Yao, X. Hu, Z. Shadike, C. L. Do-Thanh, F. Zhang, H. Chen, X. Lyu, K. Chen, Y. Zhu, X. Lu, P. Ouyang, J. Fu and S. Dai, ACS Catal., 2019, 9, 1923–1930.
44. Q. Sun, N. Wang, T. Zhang, R. Bai, A. Mayoral, P. Zhang, Q. Zhang, O. Terasaki and J. Yu, Angew. Chem., Int. Ed., 2019, 131, 18743–18749.
45. Q. Sun, B. W. J. Chen, N. Wang, Q. He, A. Chang, C. M. Yang, H. Asakura, T. Tanaka, M. J. Hülsey, C. H. Wang, J. Yu and N. Yan, Angew. Chem., Int. Ed., 2020, 59, 20183–20191.
46. Q. Sun, N. Wang, Q. Fan, L. Zeng, A. Mayoral, S. Miao, R. Yang, Z. Jiang, W. Zhou, J. Zhang, T. Zhang, J. Xu, P. Zhang, J. Cheng, D. C. Yang, R. Jia, L. Li, Q. Zhang, Y. Wang, O. Terasaki and J. Yu, Angew. Chem., Int. Ed., 2020, 59, 19450–19459.
47. M. J. Mendoza-Castro, E. De Oliveira-Jardim, N. T. Ramírez-Marquez, C. A. Trujillo, N. Linares and J. García-Martínez, J. Am. Chem. Soc., 2022, 144, 5163–5171.
48. P. Pornsetmetakul, F. J. A. G. Coumans, J. M. J. J. Heinrichs, H. Zhang, C. Wattanakit and E. J. M. Hensen, Chem.–Eur. J., 2024, 30, e202302931.
49. M. Liu, C. Miao and Z. Wu, Ind. Chem. Mater., 2024, 2, 57–84.
50. S. Goel, S. I. Zones and E. Iglesia, Chem. Mater., 2015, 27, 2056–2066.
51. S. Goel, S. I. Zones and E. Iglesia, J. Am. Chem. Soc., 2014, 136, 15280–15290.
52. S. Kweon, Y. Cho, D. Jo, H. K. Min and M. B. Park, Chem. Mater., 2023, 35, 4717–4730.
53. S. Kweon, Y. W. Kim, C. H. Shin, M. B. Park and H. K. Min, Chem. Eng. J., 2022, 431, 133364.
54. S. Kweon, J. H. Kim, H. Seo, D. W. Kim, J. S. Park, Y. Cho, S. Lee, J. Bae and M. B. Park, Appl. Catal., A, 2024, 670, 119542.
55. P. Iadrat, J. Jongthong, A. Prasertsab, S. Thanphrom, N. Toewiwat, S. Ittisanronnachai, T. Wongnate and C. Wattanakit, ACS Appl. Mater. Interfaces, 2023, 15, 42854–42867.
56. Ch. Baerlocher, D. Brouwer, B. Marler and L. B. McCusker, Database of Zeolite Structures, n.d., https://www.iza-structure.org/databases/, accessed May 14, 2025.
57. M. Limlamthong and A. C. K. Yip, Bioresour. Technol., 2020, 297, 122488.
58. H. Jon, K. Nakahata, B. Lu, Y. Oumi and T. Sano, Microporous Mesoporous Mater., 2006, 96, 72–78.
59. R. K. Parsapur and P. Selvam, Sci. Rep., 2018, 8, 16291.
60. Y. Zhou, Y. Li, X. Yuan, G. Hou, M. Chu, H. Kang, W. Qin and X. Wu, Sep. Purif. Technol., 2023, 325, 124729.
61. T. Ikuno, W. Chaikittisilp, Z. Liu, T. Iida, Y. Yanaba, T. Yoshikawa, S. Kohara, T. Wakihara and T. Okubo, J. Am. Chem. Soc., 2015, 137, 14533–14544.
62. P. Sun, C. Liu, H. Wang, Y. Liao, X. Li, Q. Liu, B. F. Sels and C. Wang, Angew. Chem., Int. Ed., 2023, 62, e202215737.
63. N. D. Hould, S. Kumar, M. Tsapatsis, V. Nikolakis and R. F. Lobo, Langmuir, 2010, 26, 1260–1270.
64. L. Liu, U. Díaz, R. Arenal, G. Agostini, P. Concepción and A. Corma, Nat. Mater., 2017, 16, 132–138.
65. L. Liu, R. Arenal, D. M. Meira and A. Corma, Chem. Commun., 2019, 55, 1607–1610.
66. K. K. Dey, A. Kumar, R. Shanker, A. Dhawan, M. Wan, R. R. Yadav and A. K. Srivastava, RSC Adv., 2012, 2, 1387–1403.
67. Y. Yue, B. Liu, N. Lv, T. Wang, X. Bi, H. Zhu, P. Yuan, Z. Bai, Q. Cui and X. Bao, ChemCatChem, 2019, 11, 4744–4754.
68. M. Chen, G. Sun, Y. Wang, D. Liang, C. Li, J. Wang and Q. Liu, Fuel, 2022, 314, 122733.
69. Z. A. Alothman, A review: Fundamental aspects of silicate mesoporous materials, Materials, 2012, 5, 2874–2902.
70. K. A. Cychosz, R. Guillet-Nicolas, J. García-Martínez and M. Thommes, Chem. Soc. Rev., 2017, 46, 389–414.
71. C. Schlumberger and M. Thommes, Adv. Mater. Interfaces, 2012, 8, 2002181.
72. P. Morales-Pacheco, F. Alvarez, L. Bucio and J. M. Domínguez, J. Phys. Chem. C, 2009, 113, 2247–2255.
73. N. Maineawklang, P. Iadrat, P. Pornsetmetakul, A. Prasertsab, P. Chaipornchalerm, S. Salakhum, S. Tantisriyanurak, C. Rodaum and C. Wattanakit, ChemNanoMat, 2024, 10, e202400100.
74. B. Pereda-Ayo, U. De La Torre, M. J. Illán-Gómez, A. Bueno-López and J. R. González-Velasco, Appl. Catal., B, 2014, 147, 420–428.
75. L. Lin, P. Cao, J. Pang, Z. Wang, Q. Jiang, Y. Su, R. Chen, Z. Wu, M. Zheng and W. Luo, J. Catal., 2022, 413, 565–574.
76. A. Yin, X. Guo, W. L. Dai and K. Fan, J. Phys. Chem. C, 2009, 113, 11003–11013.
77. X. Shi, Y. Lin, L. Huang, Z. Sun, Y. Yang, X. Zhou, E. Vovk, X. Liu, X. Huang, M. Sun, S. Wei and J. Lu, ACS Catal., 2020, 10, 3495–3504.
78. E. Borfecchia, K. A. Lomachenko, F. Giordanino, H. Falsig, P. Beato, A. V. Soldatov, S. Bordiga and C. Lamberti, Chem. Sci., 2015, 6, 548–563.
79. V. L. Sushkevich and J. A. Van Bokhoven, Chem. Commun., 2018, 54, 7447–7450.
80. M. Wen, Z. Liu, C. Liu and W. Zhuang, Chem.–Asian J., 2025, 20, e202400973.
81. L. Ouayloul, M. El Doukkali, M. Jiao, F. Dumeignil and I. Agirrezabal-Telleria, Green Chem., 2023, 25, 3644–3659.
82. I. M. S. Anekwe and Y. M. Isa, RSC Adv., 2025, 15, 3988–3999.
83. R. Chu, B. Yang, Y. Zhou, J. Wu, P. Li, M. Dai, X. Meng, X. Li, W. Li, G. Wu and C. Wang, J. Taiwan Inst. Chem. Eng., 2023, 145, 104819.
84. Y. Wang, N. Yan and Z. Chen, RSC Adv., 2024, 14, 22056–22062.
85. C. H. L. Tempelman and E. J. M. Hensen, Appl. Catal., B, 2015, 176–177, 731–739.
86. M. Díaz, E. Epelde, J. Valecillos, S. Izaddoust, A. T. Aguayo and J. Bilbao, Appl. Catal., B, 2021, 291, 120076.
87. P. Iadrat, K. Yomthong, C. Rodaum, P. Pornsetmetakul, A. Thivasasith, A. Prasertsab, X. Fan, T. Sooknoi and C. Wattanakit, Fuel, 2023, 338, 127208.

---