Rational Modulation of Pt d Electrons to Significantly Enhance the Catalytic Dehydrogenation Performance of Liquid Organic Hydrogen Carriers

Chao Sun; Tianzuo Wang; Ruijie Gao; Xiaoyang Liu; Kang Xue; Chengxiang Shi; Xiangwen Zhang; Lun Pan; Ji-Jun Zou

doi:10.1016/j.eng.2025.07.045

Engineering ›› 2026, Vol. 57 ›› Issue (2) :217 -226. DOI: 10.1016/j.eng.2025.07.045

Research

research-article

Rational Modulation of Pt d Electrons to Significantly Enhance the Catalytic Dehydrogenation Performance of Liquid Organic Hydrogen Carriers

Chao Sun ^a^,^b^,^d
, Tianzuo Wang ^a^,^b^,^d
, Ruijie Gao ^a^,^b^,^c^,^d
, Xiaoyang Liu ^d
, Kang Xue ^a^,^b^,^c^,^d
, Chengxiang Shi ^a^,^b^,^d
, Xiangwen Zhang ^a^,^b^,^c^,^d
, Lun Pan ^a^,^b^,^c^,^d^,^*
, Ji-Jun Zou ^a^,^b^,^c^,^d^,^*

Author information +

History +

PDF (2934KB)

Abstract

Understanding the intrinsic features of dehydrogenation reactions of liquid organic hydrogen carriers (LOHCs) is a formidable challenge due to the combined impact of electronic and geometric effects. Herein, we constructed a series of Pt/MO_x catalysts (CeO₂, MgO, ZrO₂, TiO₂, Al₂O₃, or SiO₂) with similar sizes of Pt (∼1.7 nm) to investigate the effects of Pt electron structures (tuned by electronic metal-support interactions) on the catalytic dehydrogenation of LOHCs. The results revealed a volcano-shaped correlation between Pt d electrons on different supports and the turnover frequency of catalytic dehydrogenation. Importantly, the Pt/MgO catalyst exhibited the highest dehydrogenation activity. With decreasing d electron content, Pt/MgO increases the bonding orbital dominance of Pt-C bonds and leads to stable adsorption of H6-monobenzyltoluene (MBT), which facilitates subsequent C-H bond scission. This study offers insight for the strategic development of high-efficiency dehydrogenation catalysts via d electron density modulation of Pt sites.

Graphical abstract

Keywords

Pt nanoparticles / Liquid organic hydrogen carriers / Dehydrogenation reaction / d electron

Cite this article

Download citation ▾

Chao Sun, Tianzuo Wang, Ruijie Gao, Xiaoyang Liu, Kang Xue, Chengxiang Shi, Xiangwen Zhang, Lun Pan, Ji-Jun Zou. Rational Modulation of Pt d Electrons to Significantly Enhance the Catalytic Dehydrogenation Performance of Liquid Organic Hydrogen Carriers. Engineering, 2026, 57(2): 217-226 DOI:10.1016/j.eng.2025.07.045

登录浏览全文

4963

注册一个新账户忘记密码

1. Introduction

Liquid organic hydrogen carriers (LOHCs) serve as adaptable media for sustainable hydrogen storage and transportation via the reversible hydrogenation and dehydrogenation of cyclic hydrocarbon compounds, making them highly adaptable for practical applications [[1], [2], [3]]. The reversible systems of perhydro-monobenzyltoluene/monobenzyltoluene (H12-MBT/H0-MBT) and perhydro-dibenzyltoluene/dibenzyltoluene (H18-DBT/H0-DBT) have been identified as promising, safe, effective, low-pollution, and affordable materials for large-scale hydrogen energy storage and release [[4], [5], [6]]. These materials are compatible with existing infrastructure and can be transported over long distances at ambient temperature.

The hydrogen storage and release processes of LOHCs involve fully hydrogenating unsaturated organic liquids that store hydrogen and subsequently release hydrogen via dehydrogenation. In contrast to hydrogenation, the dehydrogenation of saturated hydrocarbons is an endothermic process that demands high energy to activate C-H [[7], [8], [9]], which usually requires a high reaction temperature and may cause side reactions (such as cracking). As shown in Fig. 1, the hydrogen release processes of H12-MBT and H18-DBT are multistep reactions. Owing to their superior ability to activate C-H [10,11], platinum-based catalysts have garnered extensive research attention as the predominant materials employed to catalyze the release of hydrogen from H12-MBT [12,13] and H18-DBT [14,15]. Nevertheless, the relatively low rate of hydrogen release and high energy consumption remain the main obstacles to the widespread implementation of LOHCs.

To improve the catalytic activity and stability, researchers have attempted various methodologies to improve the surface properties of platinum (Pt), such as nonmetal atom modification (e.g., Pt-S) [16,17], the addition of secondary metal atoms to form alloys (e.g., Pt-Ru, Pt-Co, and PtFe_4.7) [[18], [19], [20]], metal oxide modification (e.g., MnO₂ [12,21] or WO_x [22]), and plasma treatment [23]. In addition to the modification of Pt active sites, the optimization of supports can also improve the dehydrogenation performance of the catalyst by modifying the Pt electron structure via metal-support interactions [[24], [25], [26], [27], [28], [29], [30], [31], [32]]. Nevertheless, the characteristics of the metals on different supports, including their particle size, shape, and bonding form, are highly intricate, which limits the rational design of dehydrogenation catalysts. Advancements in the precise control of the size and shape of colloidal nanoparticles (NPs) have offered a valuable approach to explore the effects of electron structure on catalytic activity [33,34].

Herein, we carefully fabricated a collection of Pt/MO_x catalysts with a uniform size of Pt (∼1.7 nm) and different types of supports (CeO₂, MgO, ZrO₂, TiO₂, γ-Al₂O₃, α-Al₂O₃, or SiO₂) to systematically investigate the effects of the electron structures of Pt sites (tuned by electronic metal-support interactions) on the dehydrogenation performance of LOHCs (using H12-MBT and H18-DBT as models). The results confirm a volcano-shaped relationship between the d electron density of the surface Pt site and the turnover frequency (TOF) of dehydrogenation. Specifically, the Pt/MgO catalyst demonstrated the greatest catalytic dehydrogenation activity compared with the other catalysts. Theoretical results indicate that the d electron density on Pt/MO_x affects the dominance of bonding orbitals in Pt-C bonds, H6-MBT adsorption, and C-H activation, which ultimately tune the catalytic dehydrogenation activity.

2. Experiments

2.1. Materials

H₂PtCl₆·6H₂O (Pt ≥ 37.5 wt%), ethylene glycol (≥ 99%), and absolute ethyl alcohol (99.8%) were purchased from Shanghai Titan Scientific Co., Ltd. (China). Poly(vinylpyrrolidone) (PVP, weight-average molecular weight = 58 000) was obtained from Shanghai Macklin Biochemical Co., Ltd. (China). NaOH (99.9%) was obtained from Shanghai Aladdin Biochemical Technology (China). HCl (36-38 wt%) was obtained from Tianjin Jiangtian Chemical Technology (China). To avoid the size effect of supports, the particle size range of all the selected supports was 20-50 nm, which included CeO₂ (20-50 nm, 99.5%, Shanghai Yuanye Bio-Technology Co., Ltd. (China)), MgO (98 wt%, 20 nm, Nanjing XFNANO Materials Tech Co., Ltd. (China)), ZrO₂ (99 wt%, 20-40 nm, Nanjing XFNANO Materials Tech Co., Ltd.), TiO₂ (P25, 20 nm), γ-Al₂O₃ and α-Al₂O₃ (99.9%, 20 nm, Shanghai D&B Biological Science and Technology Co., Ltd. (China)), and SiO₂ (99.5%, 30 nm, Shanghai Aladdin Biochemical Technology). H0-MBT (99 wt%) and H0-DBT (99 wt%) were purchased from Liaoning Zhenghe Chemical Co., Ltd. (China).

2.2. Fabrication of catalysts

2.2.1. Fabrication of Pt NPs

Pt NPs with an average particle size of 1.7 nm were produced via the ethylene glycol reduction technique (Fig. S1 in Appendix A) [33]. Typically, 1 g of NaOH was dissolved in 111 g of ethylene glycol with sonication for 30 min to formulate the reducing solution. H₂PtCl₆·6H₂O (1 g) was dissolved in the reducing solution to form an orange mixture, which was subsequently heated at 180 °C under nitrogen gas in a three-necked flask with magnetic stirring for 3 h. The as-prepared Pt NPs (8 mL of solution) were precipitated via the addition of 2 mL of 2 mol∙L⁻¹ HCl and subsequently dispersed in 50 mL of absolute ethyl alcohol containing 30 mg of PVP (labelled PVP-Pt NPs).

2.2.2. Fabrication of Pt/support catalysts

In a typical procedure (Fig. S1), 6 g of support (CeO₂, MgO, ZrO₂, TiO₂, γ-Al₂O₃, α-Al₂O₃, or SiO₂) was dispersed in a mixture of ultrapure water (200 mL) and absolute ethyl alcohol (150 mL). Then, a certain amount of PVP-Pt NPs (with a Pt loading of 0.5 wt%) dispersed in absolute ethyl alcohol was added dropwise into the above mixed solution. The resulting slurry was then stirred for 5 h at room temperature. Subsequently, the produced powders were collected, rinsed with absolute ethanol, and dried at 60 °C for 12 h. Ultimately, the dried powder was calcined in a muffle furnace under ambient air, heated to 450 °C at a constant rate of 5 °C∙min⁻¹, and subsequently held at this temperature for 4 h. As shown in Fig. S2 in Appendix A, PVP was removed via washing and calcination.

2.3. Characterizations

Powder X-ray diffraction (XRD) studies were performed using a D/Max-2500 diffractometer with a Cu Kα radiation source (40 kV and 40 mA; Rigaku Corporation, Japan). Transmission electron microscopy (TEM) analysis was carried out using a JEM-F200 field emission transmission electron microscope operating at 200 kV (JEOL Ltd., Japan). Inductively coupled plasma optical emission spectroscopy (ICP-OES) measurements were performed using an Optima 5300 DV apparatus (PerkinElmer, Inc., USA). X-ray absorption fine structure (XAFS) spectroscopy was performed with a TableXAFS-500A instrument (Anhui Specreation Instrument Technology Co., Ltd., China) equipped with a Mo anode X-ray source (1.2 kW). The Pt L_III edge was collected in fluorescence mode at 30 kV and 30 mA. Prior to the experiment, the sample was extracted from the vacuum storage container and positioned on the test platform while being shielded by nitrogen.

In situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) was used to analyze the CO adsorption of Pt/MO_x. The measurements were conducted in a reaction chamber equipped with a temperature control device. The catalyst was reduced via a mixture of 20% H₂/N₂ at a flow rate of 50 mL∙min⁻¹ at 350 °C for 3 h. The reduced sample was subjected to a N₂ flow (50 mL∙min⁻¹) at 350 °C for 1 h and subsequently cooled to 25 °C. After the background infrared spectrum was collected at 25 °C under N₂ flow (50 mL∙min⁻¹), CO molecules (5% CO/N₂, 10 mL∙min⁻¹) were allowed to adsorb to the pre-reduced catalyst samples for 10 min, followed by purging (N₂, 50 mL∙min⁻¹) to record the infrared spectra related to CO adsorption.

In situ X-ray photoelectron spectroscopy (XPS) measurements were conducted on an ESCALAB Xi⁺ XPS instrument with a monochromatized Al Kα source (Thermo Fisher Scientific, USA). The instrument radiation parameters were set to 14.5 kV and 150 W. The catalyst powders were compressed into thin layers, positioned on a pristine high-temperature resistant sample holder and subsequently placed into the sample preparation chamber. The Pt/MO_x underwent reduction using a gas mixture (20% H₂/N₂, 50 mL∙min⁻¹) at 350 °C for 3 h. Upon decreasing the temperature of the samples to ambient temperature under a flow of N₂ (50 mL∙min⁻¹), the samples underwent dynamic vacuum treatment; they were then transferred to an X-ray test chamber to obtain the electron spectra of Pt 4d and Pt 4f. All electron binding energies in the entire Pt spectrum were calibrated using the C 1s peak at 284.6 eV as a reference.

2.4. Density functional theory calculations

All calculations to determine the relationship between the Pt electronic density and the energy barriers of the dehydrogenation steps were carried out using the Vienna Ab initio Simulation Package (VASP) [35,36]. The Perdew-Burke-Ernzerhof (PBE) [37] functional and projector augmented wave (PAW) [38] methods were used to describe the electron exchange-correlation energy term and the interaction between the valence electrons and ionic cores, respectively. The kinetic energy cut-off was set to 520 eV, and the Brillouin zone was collected via a 5 × 5 × 1 k-point mesh. The geometry optimization was stopped when the force residue on the atom reached a value that was less than 0.02 eV∙Å⁻¹ and the energy difference was less than 10⁻⁶ eV. A p(4 × 4) supercell with a 3-layer slab for CeO₂(111) was modelled, whereas p(3 × 3) with layer slabs was used for the TiO₂(001) model. A p(4 × 4) supercell with 4-layer slabs and a p(4 × 4) supercell with 3-layer slabs were used for γ-Al₂O₃(001) and ZrO₂(111), respectively. The MgO and SiO₂ were modelled via a p(4 × 4) supercell for MgO(111) with 3-layer slabs and a p(4 × 4) supercell for SiO₂(001) with 4-layer slabs, respectively. The following typical method was applied to calculate the adsorption energy:

(1)${E}_{\text{ads}}={E}_{\text{total}}-\left({E}_{\text{ads}}+{E}_{\text{sur}}\right)$

where E_total is the total energy of the adsorbed surface, E_ads is the energy of the adsorbed surface, and E_sur is the energy of the pure surface. By empirical definition, the negative value of E_ads represents the energy released or relatively stable adsorption.

2.5. Dehydrogenation tests of H12-MBT and H18-DBT

The catalytic hydrogenation-saturation of monobenzyltoluene and dibenzyltoluene was conducted with stainless-steel reaction equipment using a Ni/Al₂O₃ catalyst, which produced H12-MBT (99 wt%) and H18-DBT (98 wt%), respectively. The results of the catalytic dehydrogenation tests of saturated compounds were evaluated in a stainless-steel tubular flow reactor (Fig. S3 in Appendix A). The as-prepared powders were compressed into granules and then sorted into 420-840 μm agglomerates. The agglomerates were physically mixed with quartz (purity > 99%, size range of 420-840 μm) to ensure the elimination of concentration and temperature gradients throughout the samples. Prior to the catalytic evaluation, the catalysts were activated at 350 °C for 3 h (10% H₂/N₂, 50 mL∙min⁻¹). The temperature of the reaction equipment was subsequently decreased to the desired reaction temperature (320 °C for H12-MBT; 330 °C for H18-DBT) in a N₂ flow (50 mL∙min⁻¹) and pressurized to 0.2 MPa. The reactant was introduced to the reactor via a high-pressure liquid-metering pump. The produced H₂ stream was separated from a stainless-steel gas-liquid separator, and its flow rate was subsequently quantified via an online mass flow meter. The liquid products were analyzed via a gas chromatograph (GC-2030, Shimadzu, Japan) equipped with an SH-Rtx-5 capillary column (30 m × 0.32 mm × 0.25 µm) and a flame ionization detector (FID).

The degree of dehydrogenation (DoD_M for H12-MBT, DoD_D for H18-DBT) was calculated based on the molar amounts of the liquid phase compositions and the theoretical molar amounts of H₂ that would be produced by the complete dehydrogenation of the reactants (F_in,theo-MBT and F_in,theo-DBT) (Eqs. (2), (3)).

(2)${\text{DoD}}_{\text{M}}=\frac{3{n}_{\mathrm{H}6-\mathrm{M}\mathrm{B}\mathrm{T}}+6{n}_{\text{H0-MBT}}}{{F}_{\mathrm{i}\mathrm{n},\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{o}-\mathrm{M}\mathrm{B}\mathrm{T}}}$

(3)${\text{DoD}}_{\text{D}}=\frac{3{n}_{\text{H12-DBT}}+6{n}_{\text{H6-DBT}}+9{n}_{\text{H0-DBT}}}{{F}_{\text{in,theo-DBT}}}$

where n_H6-MBT, n_H0-MBT, n_H12-DBT, n_H6-DBT, and n_H0-DBT represent the molar amounts of the corresponding liquid phase composition, respectively.

The TOF was calculated based on the amount of hydrogen gas produced per Pt site (Eq. (4)).

(4)$\text{TOF}=\frac{{n}_{{\text{H}}_{2}}}{{n}_{\text{act-Pt}}\times t}$

where n_H₂and n_act-Pt are the molar amounts of hydrogen production and active Pt sites, respectively; and t is the reaction time.

3. Results and discussion

3.1. Crystal structures of the Pt/support catalysts

As shown in the high-resolution transmission electron microscopy (HR-TEM) image (Fig. 2(a)), the as-prepared PVP-Pt NPs have a hexagonal shape with a crystal size (D_p,avg) of ∼1.7 nm and a Pt crystal (111) surface (characteristic lattice spacing of ∼2.3 Å, 1 Å = 10⁻¹⁰ m). As shown in Figs. 2(b)-(g), the deposition of the above Pt NPs on different supports, including CeO₂, MgO, ZrO₂, TiO₂, γ-Al₂O₃, and SiO₂, does not change the Pt crystal size (maintaining 1.6-1.8 nm). This particle size remained similar to that of Pt particles (maintaining 1.6-2.0 nm) calculated from CO chemisorption (Table S1 in Appendix A), and the supported Pt NPs also had a lattice spacing of 2.3 Å. In addition, the Pt/MO_x catalysts exhibit comparable Pt dispersion (65%-69%) over different metal oxide supports (Table S1). The minimal differences between the morphologies and structures of loaded Pt NPs and those of unsupported Pt NPs demonstrate the effective packing of Pt NPs, which was achieved without agglomeration or damage. The diffraction peaks in the XRD patterns (Fig. 2(h)) are the same as the characteristic peaks of the supports (Fig. S4 in Appendix A). However, diffraction peaks typically attributed to metallic Pt(1 1 1) were lacking, indicating the high dispersion of the Pt NPs. The inductively coupled plasma spectroscopy results verify that the loading amounts of Pt were similar (Table S2 in Appendix A).

3.2. Electronic structures of the Pt/support catalysts

To precisely characterize the ordering of the Pt 5d electron density in the Pt/MO_x catalysts, we employed an integrated approach combining in situ XPS, X-ray absorption spectroscopy, and in situ CO adsorption DRIFTS. The in situ XPS spectra of the prepared Pt-based catalysts were determined. To avoid peak overlap interference in the platinum binding energy analysis, the 4d orbital was not detected for the Pt/MgO catalysts, whereas the 4f orbital was not detected for Pt/Al₂O₃. As shown in Figs. 3(a) and (b), the binding energies of Pt/SiO₂ at 70.7 and 73.9 eV refer to Pt 4f_7/2 and Pt 4f_5/2, respectively [39], whereas 314.0 and 331.0 eV are attributed to Pt 4d_5/2 and Pt 4d_3/2, respectively [12]. Importantly, the characteristic peaks of Pt 4d and Pt 4f on different supports shift towards higher binding energies in the order of SiO₂ < α-Al₂O₃ ≈ γ-Al₂O₃ < TiO₂ < ZrO₂ < MgO < CeO₂, suggesting that the electron density of Pt NPs decreases in the sequence of interactions with different supports. Recently, Wang et al. [40] found that the Pt of Pt/TiO₂ is more electron deficient than that of Pt/MgO, which may be due to different reaction conditions (reduction atmosphere or temperature) [41,42]. Thus, normalized X-ray absorption near-edge structure (XANES) curves at the Pt L_III edge were obtained, in which the white-line intensity (a strong adsorption feature associated with bound electron transitions from 2p orbitals into unoccupied orbitals with d character) is indicative of 5d band occupancy [43,44]. As shown in Fig. 3(c), the white-line intensity gradually increases in the order of SiO₂ < α-Al₂O₃ ≈ γ-Al₂O₃ < TiO₂ < MgO < CeO₂, which indicates that the decrease in Pt 5d electrons in the sequence aligns with the XPS results.

CO-probe DRIFTS analysis was performed to further explore the electron density of the Pt NPs on the CeO₂, MgO, Al₂O₃, and SiO₂ supports (Fig. 3(d)). CO-adsorption peaks were not readily identified on the pure supports (Fig. S5 in Appendix A). As depicted in Fig. 3(d), the spectral profile can be deconvoluted into two distinct vibrational bands. The higher-wavenumber band corresponds to the on-top CO at the terraces of the Pt NPs, whereas the lower-wavenumber band originates from the on-top CO at the edges of the particles [40,45]. The similar peak shapes in the spectra indicate that the Pt NPs maintained a consistent geometric structure on different supports. When CO adsorbs to Pt surfaces, the lowest unoccupied molecular orbital (LUMO, i.e., 2π* antibonding orbital) of CO hybridizes with the occupied Pt 5d orbitals. This interaction induces electron transfer from Pt 5d to CO 2π*, known as the feedback effect [46,47], which weakens the C-O bond. Consequently, the vibrational frequency of adsorbed CO serves as a probe for the electronic density of Pt d orbitals. As shown in Fig. 3(d), Pt/SiO₂ has a CO-adsorption peak at 2069 cm⁻¹. The vibration frequencies gradually shift towards greater wavenumbers for adsorbed CO from Pt/SiO₂ (2069 cm⁻¹) to Pt/Al₂O₃ (2074 cm⁻¹), Pt/MgO (2079 cm⁻¹), and Pt/CeO₂ (2084 cm⁻¹), which confirms that the d orbitals of the Pt sites for the above catalysts became more electron deficient [48]. Owing to the negligible difference in the particle diameter and crystallographic state of the Pt NPs, the electron transfer capability of the Pt particles on various supports should be caused by the effect of the electronic metal-support interaction.

3.3. Effects of support type on catalytic dehydrogenation performance

The catalytic performances of the Pt/MO_x catalysts were tested for the H12-MBT and H18-DBT dehydrogenation reactions (Fig. 4, Figs. S6 and S7 in Appendix A). No dehydrogenation activity was detected for H12-MBT or H18-DBT over the pure supports (Fig. S8 in Appendix A). As shown in Figs. 4(a) and (b), the trends in the DoD_M and DoD_D values of the prepared Pt-based catalysts were similar. The Pt/MgO catalyst had the highest catalytic activity in terms of the degree of dehydrogenation (DoD_M of ∼77% and DoD_D of ∼83%), whereas Pt/SiO₂ had the lowest activity (DoD_M of ∼28% and DoD_D of ∼44%). After the six-hour activity test, the coke deposition amounts on the Pt/MgO catalyst were 1.9 wt% for H12-MBT dehydrogenation and 3.4 wt% for H18-DBT dehydrogenation, both of which are lower than those of the other catalysts (Table S3 in Appendix A). Previous studies have demonstrated that in the dehydrogenation of H12-MBT and H18-DBT, Pt-based catalysts predominantly utilize Al₂O₃ as the support material [12,23,49]. The Pt/Al₂O₃ catalyst exhibited unstable activity during dehydrogenation (Figs. 4(a) and (b), Fig. S9 in Appendix A). As shown in Fig. S10 in Appendix A, Pt/MgO exhibited no significant deactivation during the 30-hour long-term evaluation, demonstrating its high stability. This difference may be attributed to side reactions (e.g., cracking and cyclization) occurring with different catalysts during the dehydrogenation process, leading to varying degrees of carbon deposition (Table S3). However, no correlation was observed between the electron structure and either the deactivation rate (Fig. S11 in Appendix A) or the degree of carbon deposition (Fig. S12 in Appendix A).

Because the dehydrogenations of H12-MBT and H18-DBT are multistep reactions (Fig. 1), the degree of dehydrogenation is directly related to the production distribution. As shown in Figs. 4(c) and (d), the selectivity of reactant (H12-MBT or H18-DBT) exhibits a decline followed by an increase with the list of supports. Among all the catalysts, the Pt/MgO catalyst exhibited the lowest selectivity for H12-MBT or H18-DBT, indicating a relatively high conversion rate of the reactants. H6-MBT and H6-DBT, which contain the last saturated ring, become the main intermediates among all the catalysts. Proton nuclear magnetic resonance (¹H NMR, as shown in Fig. S13 in Appendix A) analysis revealed that monosubstituted side rings are more susceptible to dehydrogenation. The selectivity values of H6-MBT and H6-DBT remained consistent for all the Pt/support catalysts, averaging 40.1% ± 2.0% and 38.8% ± 3.0%, respectively. These results indicate a potential counterbalancing mechanism in which the rapid consumption of other compounds (H12-MBT or H12-DBT) compensated for the consumption of H6-MBT or H6-DBT. This effect may be due to the occupation of the active site by the saturated ring of the reactant (H6-MBT or H6-DBT), which restricts the dehydrogenation of other intermediates. The selectivity of the complete dehydrogenation products serves as an important indicator of both the dehydrogenation rates. According to the order of Pt d electron reduction when Pt is loaded onto SiO₂, α-Al₂O₃, γ-Al₂O₃, TiO₂, ZrO₂, MgO, and CeO₂, the selectivity of complete dehydrogenation products (H0-MBT or H0-DBT) exhibits a volcano-shaped trend, and Pt/MgO shows the maximum value (57% for H0-MBT or 58% for H0-DBT). Quantitative analysis of the liquid-phase products (Figs. S14 and S15 in Appendix A) revealed that the heavy constituents (C_>14 for H12-MBT dehydrogenation and C_>21 for H18-DBT dehydrogenation) constituted less than 1% of the total product distribution, a finding that is similar to previously reported results [50]. Dehydrocyclization and molecular cracking may be associated with the kinks in steps on the Pt surface [50] and the acidic sites of the catalyst [49,51], respectively, which did not appear to be correlated with Pt electron density (Fig. S16 in Appendix A). Considering the negligible differences in the particle diameter and dispersity of the Pt NPs, the variations in the dehydrogenation performance should be closely related to the properties of the Pt d electrons. We correlated the TOF values of dehydrogenation with different densities of Pt d electrons among the Pt/support materials (Figs. 5(a) and (b)) and used the binding energy difference (ΔBE, i.e., the Pt 4d or Pt 4f binding energy difference (Table S4 in Appendix A) between the Pt/support materials and Pt/SiO₂, as shown in Figs. 3(a) and (b)) to describe the electron structure. To mitigate interference from XPS peak convolution for the Pt/CeO₂, Pt/TiO₂, Pt/ZrO₂, and Pt/Al₂O₃ samples, the ΔBE was determined based on Pt 4d, whereas Pt/MgO employed Pt 4f as the reference. A larger ΔBE value indicates a more electron-deficient state at the Pt d orbitals. The TOF value increased as ΔBE increased, reached its maximum for the Pt/MgO catalyst, but decreased for Pt/CeO₂, which has the highest ΔBE. Previous experimental observations have provided clear evidence of the similar electron structures of the Pt/α-Al₂O₃ and Pt/γ-Al₂O₃ catalysts. Consequently, the approximate TOF values (Fig. 5) for the Pt/α-Al₂O₃ and Pt/γ-Al₂O₃ catalysts suggest that the dominant factor influencing dehydrogenation performance is the electronic effect rather than the crystallographic form when the shape change of the Pt NPs loaded on α-Al₂O₃ and γ-Al₂O₃ is excluded. The similar volcano-shaped plots for the dehydrogenation reactions of H12-MBT and H18-DBT reveal that the support type strongly affects the dehydrogenation by adjusting the d electron structure of the Pt NPs.

3.4. Mechanisms by which Pt d electrons affect the dehydrogenation reaction

A representative Pt₁₀ model on various supports (Fig. S17 in Appendix A) was constructed to explore the d electron structure of the Pt NPs via density functional theory (DFT) calculations. The interaction between Pt NPs and various supports results in varying levels of electron rearrangement. The results show that Pt NPs gradually transfer more electrons to the carrier in the order of SiO₂ < Al₂O₃ < TiO₂ < ZrO₂ < MgO < CeO₂, suggesting more deficient properties of the Pt d electrons in the sequence. This finding corresponds to the XPS, XANES, and DRIFTS data. Moreover, the observed linear association between the number of electron transfers and the accompanying alteration in ΔBE (Fig. 6(a)) further substantiate the calculation results.

During dehydrogenation, the conversion rate of the intermediate H6-MBT determines the concentration of complete dehydrogenation products, which serves as an important indicator of the activity of the catalyst. Hence, H6-MBT (Fig. S18 in Appendix A) served as the model substrate to analyze the effect of the electron structure on the activity. According to the volcano-type activity trend, Pt/MgO exhibited the optimal catalytic performance, whereas Pt/SiO₂ and Pt/CeO₂ represented the extremes in electron density (highest and lowest, respectively). Consequently, Pt/MgO, Pt/SiO₂, and Pt/CeO₂ were selected as model systems for the DFT investigations. After optimization, the γ-C (meta-C of the mono-substituted side ring) of H6-MBT adsorbed to the Pt sites formed a Pt-C bond (Fig. S19 in Appendix A). To elucidate the connection between the adsorption stability and the electron structure of Pt, the integrated crystal orbital Hamilton population (ICOHP) of Pt-C bonds was calculated. As shown in Fig. 6(a), a comparison between Pt NPs supported on SiO₂ and MgO reveals that the latter is more electron deficient, leading to an increased predominance of bonding orbitals in the Pt-C bond (Fig. 6(b)). Hence, the intensity of the Pt-C bond from H6-MBT adsorbed to the Pt site of the Pt/SiO₂ catalyst (−ICOHP = 0.25) is weaker than that on Pt/MgO (−ICOHP = 2.81). The Pt site in the Pt/CeO₂ catalyst, which transfers 1.12 e to the support, results in greater electron loss than the Pt/MgO catalyst does (0.98 e), leading to an insufficient electron supply for bonding orbitals within the Pt-C bond and a reduced intensity of the Pt-C bond (−ICOHP = 0.38).

DFT and kinetic studies have suggested that the first C-H bond cleavage is a kinetically relevant step of dehydrogenation [7,32]. As plotted in Fig. 6(c), the activation energy of the initial C-H bond in the reaction mechanism, specifically the transition from H6-MBT* to H5-MBT* and H* on the Pt/MgO, Pt/CeO₂, and Pt/SiO₂ catalysts, was investigated. The corresponding C-H activation energies were 0.866, 1.236, and 1.354 eV, respectively, indicating a gradual increase in the energy barrier necessary to break the C-H bond. This finding is consistent with the Pt-C stability sequence and the measured dehydrogenation performance. Therefore, the stable adsorption of H6-MBT on catalysts can promote the breaking of C-H bonds and improve dehydrogenation activity. This work confirms that the Pt d electron density is crucial for modulating the dehydrogenation activity (Fig. 6(d)). Owing to its relatively low d electron density, Pt/MgO has the highest dehydrogenation activity for H12-MBT and H18-DBT. Compared with d electron-rich Pt NPs on Pt/SiO₂, those on Pt/MgO promote a greater prevalence of bonding orbitals within the Pt-C bond and enhance the stabilization of H6-MBT adsorption, which is advantageous for subsequent C-H activation steps. As the d electron density further decreases (for Pt/CeO₂), an inadequate electron supply from Pt NPs to Pt-C bonding orbitals destabilizes H6-MBT adsorption, increasing the energy barrier for C-H activation and reducing dehydrogenation activity. Understanding the connections between the electron structure and performance will aid in the development and investigation of catalysts that are more effective and resilient in promoting dehydrogenation and similar processes. Future research should focus on the relationship between modifications in the electron structure and alterations in the surface structure caused by supports or promoters to maximize catalytic activity.

4. Conclusions

In this work, we constructed a range of Pt/MO_x catalysts with a controlled average size of 1.7 nm over different supports to investigate the Pt electron effect on the dehydrogenation performance of H12-MBT and H-DBT. Based on in situ XPS and dehydrogenation tests, the volcano-shaped ΔBE (representing d electron density)-TOF plot demonstrates that supports modify the dehydrogenation performance by altering the d electron density of the Pt NPs. Specifically, Pt/MgO exhibited the highest dehydrogenation activity of H12-MBT and H18-DBT among all the Pt/support catalysts. As the density of d electrons decreases, Pt NPs induce an increased predominance of bonding orbitals in the Pt-C bond, leading to adsorption stabilization of H6-MBT, which is beneficial for subsequent C-H activation. However, the insufficient electron supply from Pt NPs (Pt/CeO₂) to the bonding orbitals within the Pt-C bond results in the adsorption instability of H6-MBT. Thus, the energy barrier of C-H activation that must be overcome is greater, resulting in decreased dehydrogenation activity. This work provides a framework for the systematic development of highly efficient catalysts for dehydrogenation reactions via d electron density modulation of Pt sites.

CRediT authorship contribution statement

Chao Sun: Writing - original draft, Visualization, Validation, Methodology, Investigation, Data curation, Conceptualization. Tianzuo Wang: Writing - review & editing. Ruijie Gao: Writing - review & editing. Xiaoyang Liu: Writing - review & editing. Kang Xue: Writing - review & editing. Chengxiang Shi: Writing - review & editing. Xiangwen Zhang: Supervision, Resources. Lun Pan: Writing - review & editing, Supervision, Resources, Project administration. Ji-Jun Zou: Writing - review & editing, Validation, Supervision, Project administration, Funding acquisition.

Declaration of competing interest

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

Acknowledgments

The authors appreciate the support from the National Natural Science Foundation of China (U24A20547 and 22222808) and the Haihe Laboratory of Sustainable Chemical Transformations for financial support.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.eng.2025.07.045.

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]	Markiewicz M, Zhang YQ, Empl MT, Lykaki M, Thöming J, Steinberg P, et al. Hazard assessment of quinaldine-, alkylcarbazole-, benzene- and toluene- based liquid organic hydrogen carrier (LOHCs) systems. Energy Environ Sci 2019; 12(1):366-83.

[2]	Teichmann D, Stark K, Müller K, Zöttl G, Wasserscheid P, Arlt W. Energy storage in residential and commercial buildings via liquid organic hydrogen carriers (LOHC). Energy Environ Sci 2012; 5(10):9044-54.

[3]	Cho J, Kim B, Venkateshalu S, Chung DY, Lee K, Choi SI. Electrochemically activatable liquid organic hydrogen carriers and their applications. J Am Chem Soc 2023; 145(31):16951-65.

[4]	Gianotti E, Taillades-Jacquin M, Rozière J, Jones DJ. High-purity hydrogen generation via dehydrogenation of organic carriers: a review on the catalytic process. ACS Catal 2018; 8(5):4660-80.

[5]	Li J, Zhang J, Zhang S, Gao R, Zhang X, Pan L, et al. Intrinsic kinetics of benzyltoluene hydrogenation over a supported Ni catalyst for green hydrogen storage. Int J Hydrogen Energy 2024;49(Pt B):1586-96.

[6]	Geburtig D, Preuster P, Bösmann A, Müller K, Wasserscheid P. Chemical utilization of hydrogen from fluctuating energy sources—catalytic transfer hydrogenation from charged liquid organic hydrogen carrier systems. Int J Hydrogen Energy 2016; 41(2):1010-7.

[7]	Guo J, Peng M, Jia Z, Li C, Liu H, Zhang H, et al. Kinetic evidence of most abundant surface intermediates variation over Ptn and Ptp: few-atom Pt ensembles enable efficient catalytic cyclohexane dehydrogenation for hydrogen production-II. ACS Catal 2022; 12(12):7248-61.

[8]	Al-ShaikhAli AH, Jedidi A, Anjum DH, Cavallo L, Takanabe K. Kinetics on NiZn bimetallic catalysts for hydrogen evolution via selective dehydrogenation of methylcyclohexane to toluene. ACS Catal 2017; 7(3):1592-600.

[9]	Rao PC, Kim Y, Kim H, Son Y, Choi Y, Na K, et al. Methylbenzyl naphthalene: liquid organic hydrogen carrier for facile hydrogen storage and release. ACS Sustain Chem Eng 2023; 11(34):12656-66.

[10]	Yook H, Jeong K, Hwang J, Park JH, Han JW. Roles of bridge carbon and heteroatom during the dehydrogenation reaction of dicyclohexylmethane derivatives on Pd and Pt catalysts for liquid organic hydrogen carrier. Appl Catal B 2024; 342:123394.

[11]	Li X, Shen P, Han X, Wang Y, Zhu Y, Wu Z. Dehydrogenation mechanisms of liquid organic hydrogen carriers over Pt, Pd, Rh, and Ni surfaces: cyclohexane as a model compound. Appl Surf Sci 2021; 543:148769.

[12]	Jo Y, Kim D, Kim TW, Yoon D, Suh YW. Highly dispersed Pt-MnOx nanoclusters for the promoted activity of mesoporous Pt-MnOx-Al2O3 in dehydrogenation of perhydro-benzyltoluene. Appl Catal B 2023; 334:122848.

[13]	Kim TW, Jo Y, Jeong K, Yook H, Han JW, Jang JH, et al. Tuning the isomer composition is a key to overcome the performance limits of commercial benzyltoluene as liquid organic hydrogen carrier. J Energy Storage 2023; 60:106676.

[14]	Shi L, Zhou Y, Tan X, Qi S, Smith KJ, Yi C, et al. The effects of alumina morphology and Pt electron property on reversible hydrogenation and dehydrogenation of dibenzyltoluene as a liquid organic hydrogen carrier. Int J Hydrogen Energy 2022; 47(7):4704-15.

[15]	Brückner N, Obesser K, Bösmann A, Teichmann D, Arlt W, Dungs J, et al. Evaluation of industrially applied heat-transfer fluids as liquid organic hydrogen carrier systems. ChemSusChem 2014; 7(1):229-35.

[16]	Chen X, Gierlich CH, Schötz S, Blaumeiser D, Bauer T, Libuda J, et al. Hydrogen production based on liquid organic hydrogen carriers through sulfur doped platinum catalysts supported on TiO2. ACS Sustain Chem Eng 2021; 9 (19):6561-73.

[17]	Auer F, Blaumeiser D, Bauer T, Bösmann A, Szesni N, Libuda J, et al. Boosting the activity of hydrogen release from liquid organic hydrogen carrier systems by sulfur-additives to Pt on alumina catalysts. Cat Sci Technol 2019; 9 (13):3537-47.

[18]	Zhu T, Yang M, Chen X, Dong Y, Zhang Z, Cheng H. A highly active bifunctional Ru-Pd catalyst for hydrogenation and dehydrogenation of liquid organic hydrogen carriers. J Catal 2019; 378:382-91.

[19]	Alconada K, Barrio VL. Evaluation of bimetallic Pt-Co and Pt-Ni catalysts in LOHC dehydrogenation. Int J Hydrogen Energy 2024;51(Pt D):243-55.

[20]	He Z, Li K, Chen T, Feng Y, Villalobos-Portillo E, Marini C, et al. High-purity hydrogen production from dehydrogenation of methylcyclohexane catalyzed by zeolite-encapsulated subnanometer platinum-iron clusters. Nat Commun 2025; 16:92.

[21]	Nakano A, Manabe S, Higo T, Seki H, Nagatake S, Yabe T, et al. Effects of Mn addition on dehydrogenation of methylcyclohexane over Pt/Al2O3 catalyst. Appl Catal A 2017; 543:75-81.

[22]	Kim CH, Lee MW, Jang JS, Lee SH, Lee KY. Enhanced activity of a WOx- incorporated Pt/Al2O3 catalyst for the dehydrogenation of homocyclic LOHCs: effects of impregnation sequence on Pt-WOx interactions. Fuel 2022; 313:122654.

[23]	Shi L, Zhou Y, Qi S, Smith KJ, Tan X, Yan J, et al. Pt catalysts supported on H2 and O2 plasma-treated Al2O3 for hydrogenation and dehydrogenation of the liquid organic hydrogen carrier pair dibenzyltoluene and perhydrodibenzyltoluene. ACS Catal 2020; 10(18):10661-71.

[24]	Ahn CI, Kwak Y, Kim AR, Jang M, Badakhsh A, Cha J, et al. Dehydrogenation of homocyclic liquid organic hydrogen carriers (LOHCs) over Pt supported on an ordered pore structure of 3-D cubic mesoporous KIT-6 silica. Appl Catal B 2022; 307:121169.

[25]	Tuo Y, Meng Y, Chen C, Lin D, Feng X, Pan Y, et al. Partial positively charged Pt in Pt/MgAl2O 4 for enhanced dehydrogenation activity. Appl Catal B 2021; 288:119996.

[26]	Aakko-Saksa PT, Vehkamäki M, Kemell M, Keskiväli L, Simell P, Reinikainen M, et al. Hydrogen release from liquid organic hydrogen carriers catalysed by platinum on rutile-anatase structured titania. Chem Commun 2020; 56 (11):1657-60.

[27]	Meng H, Yang Y, Shen T, Yin Z, Zhang J, Yan H, et al. Highly efficient hydrogen production from dehydrogenation reaction of nitrogen heterocycles via Pd0-Pdd+ synergistic catalysis. ACS Catal 2023; 13(13): 9234-44.

[28]	Sattler A, Paccagnini M, Lanci MP, Miseo S, Kliewer CE. Platinum catalyzed C-H activation and the effect of metal-support interactions. ACS Catal 2020; 10 (1):710-20.

[29]	Liu Z, Lustemberg P, Gutiérrez RA, Carey JJ, Palomino RM, Vorokhta M, et al. In situ investigation of methane dry reforming on metal/ceria(111) surfaces: metal-support interactions and C-H bond activation at low temperature. Angew Chem Int Ed 2017; 56(42):13041-6.

[30]	Auer F, Hupfer A, Bösmann A, Szesni N, Wasserscheidpeter P. Influence of the nanoparticle size on hydrogen release and side product formation in liquid organic hydrogen carrier systems with supported platinum catalysts. Cat Sci Technol 2020; 10(19):6669-78.

[31]	Chen L, Verma P, Hou K, Qi Z, Zhang S, Liu YS, et al. Reversible dehydrogenation and rehydrogenation of cyclohexane and methylcyclohexane by single-site platinum catalyst. Nat Commun 2022; 13:1092.

[32]	Deng Y, Guo Y, Jia Z, Liu JC, Guo J, Cai X, et al. Few-atom Pt ensembles enable efficient catalytic cyclohexane dehydrogenation for hydrogen production. J Am Chem Soc 2022; 144(8):3535-42.

[33]	Rioux RM, Song H, Hoefelmeyer JD, Yang P, Somorjai GA. High-surface-area catalyst design: synthesis, characterization, and reaction studies of platinum nanoparticles in mesoporous SBA-15 silica. J Phys Chem B 2005; 109 (6):2192-202.

[34]	Ge S, Fan W, Tang X, Cui Y, Wang D, Gong XQ, et al. Revealing the size effect of ceria nanocube-supported platinum nanoparticles in complete propane oxidation. ACS Catal 2024; 14(4):2532-44.

[35]	Kresse G, Hafner J. Ab initio molecular dynamics for liquid metals. Phys Rev B 1993; 47(1):558-61.

[36]	Perdew JP, Burke K, Ernzerhof M. Generalized gradient approximation made simple. Phys Rev Lett 1996; 77(18):3865-8.

[37]	Hammer B, Hansen LB, Nørskov JK. Improved adsorption energetics within density-functional theory using revised Perdew-Burke-Ernzerhof functionals. Phys Rev B 1999; 59(11):7413-21.

[38]	Blöchl PE. Projector augmented-wave method. Phys Rev B 1994; 50 (24):17953-79.

[39]	Gao R, Wang J, Huang ZF, Zhang R, Wang W, Pan L, et al. Pt/Fe2O3 with Pt-Fe pair sites as a catalyst for oxygen reduction with ultralow Pt loading. Nat Energy 2021; 6(6):614-23.

[40]	Wang Z, Wang C, Mao S, Lu B, Chen Y, Zhang X, et al. Decoupling the electronic and geometric effects of Pt catalysts in selective hydrogenation reaction. Nat Commun 2022; 13:3561.

[41]	Tang X, Yu A, Yang Q, Yuan H, Wang Z, Xie J, et al. Significance of epitaxial growth of PtO 2 on rutile TiO2 for Pt/TiO2 catalysts. J Am Chem Soc 2024; 146 (6):3764-72.

[42]	Shu G, Lin Y, Wang S, Zhang S, Fan L, Wang C, et al. Dynamic metal-support interaction-activated sub-nanometer Pt clusters on FeOx supports for aqueous phase reforming and hydrogenolysis of glycerol. ACS Catal 2023; 13 (13):8423-36.

[43]	Jia Z, Qin X, Chen Y, Cai X, Gao Z, Peng M, et al. Fully-exposed Pt-Fe cluster for efficient preferential oxidation of CO towards hydrogen purification. Nat Commun 2022; 13:6798.

[44]	Malta G, Kondrat SA, Freakley SJ, Davies CJ, Lu L, Dawson S, et al. Identification of single-site gold catalysis in acetylene hydrochlorination. Science 2017; 355 (6332):1399-403.

[45]	Kusche M, Enzenberger F, Bajus S, Niedermeyer H, Bösmann A, Kaftan A, et al. Enhanced activity and selectivity in catalytic methanol steam reforming by basic alkali metal salt coatings. Angew Chem Int Ed 2013; 52(19):5028-32.

[46]	Yamagishi S, Fujimoto T, Inada Y, Orita H. Studies of CO adsorption on Pt(100), Pt(410), and Pt(110) surfaces using density functional theory. J Phys Chem B 2005; 109(18):8899-908.

[47]	Liu J, Hibbitts D, Iglesia E. Dense CO adlayers as enablers of CO hydrogenation turnovers on Ru surfaces. J Am Chem Soc 2017; 139(34):11789-802.

[48]	Borges LR, Lopez-Castillo A, Meira DM, Gallo JMR, Zanchet D, Bueno JMC. Effect of the Pt precursor and loading on the structural parameters and catalytic properties of Pt/Al2O3. ChemCatChem 2019; 11(13):3064-74.

[49]	Alconada K, Barrio VL. Enhancing perhydrobenzyltoluene dehydrogenation performance with Co, Mo and Mn metal oxides: a comparative study with Pt/ Al2O3 catalyst. Appl Catal B 2024; 357:124349.

[50]	Modisha P, Bessarabov D. Stress tolerance assessment of dibenzyltoluene-based liquid organic hydrogen carriers. Sustain Energy Fuels 2020; 4 (9):4662-70.

[51]	Zhao W, Yu H, Peng S, Liu W, Zhang W, Mei D. Mechanistic understanding of anthracene hydrocracking over HY zeolite encapsulated single-atom Pt catalysts. ACS Catal 2024; 14(11):8836-55.