Double-Ligand [Fe/PNP/PP₃] and Their Hybrids [Fe/SiO₂@PNP/PP₃] as Catalysts for H₂-Production from HCOOH

Abstract

Two types of iron-based catalysts, [Fe/SiO₂@^iProPNP/PP3] and [Fe/SiO₂@^tBuPNP/PP3], for the dehydrogenation of formic acid (FADH), were synthesized. These catalysts were developed using a double-ligand approach combining a PNP ligand and a PP3 ligand, demonstrating functionality without the need for additional cocatalysts or additives. Furthermore, hybrid catalysts [Fe/SiO₂@^iProPNP/PP3] and [Fe/SiO₂@^tBuPNP/PP3] were created by covalently grafting PNP ligands onto SiO₂ particles. The hybrid [Fe/SiO₂@^iProPNP/PP3] exhibited enhanced recyclability, with turnover numbers (TONs) exceeding 74,000. In situ ATR-FTIR and UV-Vis spectroscopies were used to monitor the structure and dynamics of the catalysts under catalytic conditions, revealing the formation of active catalysts through the involvement of all components: [Fe (metal)/PNP (first ligand)/PP3 (second ligand)/FA (substrate)], which are crucial to FADH catalysis. An Arrhenius study revealed that the hybrid [Fe/SiO₂@^iProPNP/PP3] had a lower activation energy (E_a = 42.5 kJ/mol) compared to its homogeneous counterpart (E_a = 48.2 kJ/mol), indicating superior catalytic performance. Conversely, [Fe/SiO₂@^tBuPNP/PP3] showed an increased activation energy (E_a = 48.3 kJ/mol) compared to its homogeneous form (E_a = 46.4 kJ/mol). This study discusses the differing roles of ^tBuPNP and ^iProPNP in catalyst configuration, highlighting the potential of double-ligand catalysts to enhance the performance and recyclability of PNP ligands in FADH, offering significant implications for the development of efficient and reusable catalytic systems.

1. Introduction

Due to its high energy per mass content (120 MJ kg⁻¹), H₂ is attractive for energy technologies [1]. However, its low volumetric energy density (0.0108 MJ L⁻¹) [2] mandates storage and transportation under high pressure (i.e., 100–700 bar) or as a liquid at −253 °C. This necessity makes H₂ storage and transportation a standalone task requiring further investigation [3]. Within this context, the scientific community is shifting toward in situ H₂ production, either through water splitting [4], ammonia cracking [5], or the dehydrogenation of organic molecules such as methanol, formaldehyde, and formic acid, which can serve as liquid organic hydrogen carriers (LOHCs) [6,7]. Among LOHCs, formic acid (FA) has gained considerable interest due to its high volumetric energy intensity (6.36 MJ L⁻¹), low toxicity, and very low flammability [8].

The pioneering works of Beller [9,10], Himeda [11,12], and Laurenczy [13,14,15] reported efficient H₂ production from FA catalyzed by systems based on metal complexes of iridium, ruthenium, or iron combined with organic ligands containing P-, N-, and O-heteroatoms. PNP pincer-type ligands, which often surround the metal and provide high electron density, maintain a reductive character and act as σ-electron donors and/or π-electron acceptors within the metal complex structure [16]. The involvement of phosphine moieties attempts to influence transition metal hydrides by controlling steric hindrance and M-H bond characteristics [17].

One of the first catalysts using Ir and a pincer PNP-type ligand was reported in 2011 by Tanaka et al., where the ligand bears a [2,6-(ECH₂)C₆H₃]− structure with E being iso-propyl (iPro) groups. This Ir(III) trihydride catalyst achieved highly efficient dehydrogenation of 5 mmol HCOOH and trimethylamine-base, providing a TON of 2000 and TOF of 120,000 h⁻¹ [18], at 80 °C in tert-butanol. The first PNP-type catalyst with a non-noble metal was [Fe(tBuPNP)(H)₂(CO)], which was reported in 2013 by Milstein et al. using 2,6-di(di-tert-butylphosphinomethyl)pyridine as the PNP ligand. This system was efficient and stable in dioxane with NEt₃ at 40 °C, dehydrogenating 1 mol of formic acid in 10 days and achieving a TON value of 100,000 and 416 h⁻¹ TOF [19].

In 2014, Pidko et al. demonstrated that the Ru-hydride-PNP complex (PNP)Ru(H)Cl(CO), with an aromatic PNP ligand with tert-butyl groups, was very active in the cycling catalytic operation of CO₂ to formate, where FA dehydrogenation provided TON = 326,500, TOF = 257,000 h⁻¹, with NHex₃ as a base at 90 °C in DMF [20]. In the same year, Bielinski et al. published a FeII(H)(iPrPNP)(CO) complex, achieving one of the highest values of TON (983,642) and TOF (196,728 h⁻¹) within one hour at 80 °C in dioxane in the presence of a Lewis acid, specifically LiBF₄. The use of LiBF₄ boosted the catalytic performance, as the same system without the additive yielded a TON of 182 in 48 h [21]. A system similar to the one used by Bielinski et al., with MnI instead of FeII in 1,4-dioxane at 80 °C, exhibited low performance for FA dehydrogenation, yielding a TON of only 190 after 14 h of operation. Interestingly, the use of LiBF₄ here had a negative impact, reducing the TON values from 190 to 19 for the same operation time [22].

A Ru-PN₃P catalyst, where the ligand contained a dearomatized pyridine moiety and an imine arm, with Ru coordinated by 1N- and 2P-atoms, was able to work under continuous HCOOH feed (14.5 mL for 150 h) yielding TON > 10⁶ in DMSO at 90 °C, with NEt₃ as an additive [23]. Additionally, the corresponding Mn(PN₃P)(CO)₂ catalyst published by the same group in 2023 achieved TON = 15,200 and 1416 mL of [H₂+CO₂] gases [24].

Based on a 2,6-diaminopyridine scaffold, the corresponding Fe(II)-hydrido-carbonyl catalyst performed quite well in the dehydrogenation of 0.01 mol FA in THF with a base, giving a TON value of 10,000 [25]. In 2019, Tondreau et al. combined a tBuPNNOP chelate ligand to form the (tBuPNNOP)Mn(CO)₂ catalyst, which, operating with 1 mL FA in chlorobenzene at 80 °C and NEt₃ as an additive, gave TOF = 8500 h⁻¹ [26]. In 2021, the group of Milstein introduced a Ru-9H-acridine pincer catalyst with extraordinary ability for the dehydrogenation of neat FA, with a total TON of 1,700,000 and demonstrated activity for continuous operation over 19 days, decomposing a total of 1.2 L of FA [27].

The hydrido complex of Ru with iProPNP, [Ru(H)(Cl)(CO)(HN{CH₂CH₂P(iPr)₂}₂)], was evaluated at different pH values for FA decomposition, providing the highest activity at pH = 4.5 with 1693 TON and 1912 h⁻¹ TOF within 3 h of operation [28]. By replacing the H-N group with Me-N, the system’s performance was enhanced, resulting in a TON of 6801 and a TOF of 2598 h⁻¹; however, the corresponding catalysts with Fe and Mn did not adopt this trend [28].

PNP-type ligands have also shown potential in CO₂ hydrogenation [29], methanol dehydrogenation [30], and reversible CO₂ hydrogenation-FA dehydrogenation [20,31], with metal hydride species demonstrating their significance.

Marketwise, the cost of H₂ production via in situ catalytic H₂ production from LOHCs is a key factor, in conjunction with the efficiency and safety of the process. A recent cost analysis of such systems [32] revealed that the ligand significantly contributes to the final cost [32]. Therefore, developing efficient and recyclable heterogeneous molecular catalysts based on low-cost ligands becomes mandatory.

So far, a brief review of the key progress in developing heterogeneous catalysts for FADH highlights that heterogeneous catalysts are a promising technology. For example, successful immobilization of the Ru/mTTPTS complex onto the MCM-41 surface [14] or immobilization of an IrCp* complex on a covalent triazine framework [33]. The Ru catalyst in [14] was active in 20 consecutive cycles with a TON of 71,000, while in [33], an IrCp*-bipyridine complex achieved 1,060,000 TONs under continuous FA feed for 40 h.

Our research group engineered two hybrid catalysts, RPPh₂@SiO₂ and polyRPhphos@SiO₂ [34], by grafting two phosphines, triphenylphosphine, and tris [2-(diphenyl-phosphine)ethyl]phosphine, onto the surface of SiO₂ particles. These systems exhibited totals of 176,000 and 65,000 TONs, respectively, under a continuous HCOOH feed [34]. In the same context as hybrid materials, a heterogeneous catalyst produced by grafting imidazole-based ligands onto the surface of SiO₂, which, with the incorporation of Fe^II and a polyphosphine, showed significant performance and reusability [35].

A key finding was that using PP₃ as an additional ligand in FeL and RuL catalysts [16] significantly boosts the FADH efficiency. This ‘Double-Ligand’ technology in FeL/PP₃ and RuL/PP₃ benefits from the high efficiency and stability of the catalyst.

The working hypothesis was that the combined use of hybrid materials and ‘Double-Ligand’ technology would result in a superior catalytic system for FA dehydrogenation (FADH). Within this context, in the present study, we chose two cheap and commercial PNP ligands: bis[(2-di-iso-propylphosphino)ethyl]amine (^iProPNP) and bis [2-(di-tert-butylphosphino)ethyl]amine (^tBuPNP), in order to engineer two double-ligand catalysts: [Fe/^iProPNP/PP3] and [Fe/^tBuPNP/PP3]. Based on these catalysts, we further developed heterogeneous systems [Fe/SiO₂@^iProPNP/PP3] and [Fe/SiO₂@^tBuPNP/PP3]. The results demonstrate the following: (i) the in situ formation of the Fe-catalysts [Fe/^iProPNP/PP3] and [Fe/^tBuPNP/PP3] provides efficient FADH catalysis without the need for additives; (ii) the heterogeneous catalysts [Fe/SiO₂@^iProPNP/PP3] and [Fe/SiO₂@^tBuPNP/PP3] exhibit high activity in FADH; (iii) among these, [Fe/SiO₂@^iProPNP/PP3] is the most efficient; (iv) an Arrhenius study reveals that the activation energy (Ea) of [Fe/SiO₂@^iProPNP/PP3] is 5.5 kJ/mol lower than that of the homogeneous [Fe/^iProPNP/PP3], indicating superior thermodynamic performance; and (v) [Fe/SiO₂@^iProPNP/PP3] is stable and reusable, achieving a significant turnover number (TON) of 74,450 within 12 h of continuous operation.

Thus, in this study, we aimed to address the limitations of current catalytic systems by developing a more efficient and straightforward approach. Traditional catalytic systems often require the synthesis of expensive and complex metal hydride structures before the catalytic procedure, which is both time-consuming and resource-intensive. These systems typically rely on pre-synthesized metal complexes with hydrides and necessitate the use of additives to boost performance. Our approach, however, synthesizes the catalyst in situ without the need for any performance-enhancing additives. This innovative method not only simplifies the synthesis process but also enhances the recyclability and efficiency of the catalysts in heterogeneous systems. By eliminating the challenges associated with pre-synthesized complexes and additives, our research provides a significant improvement in catalytic synthesis, addressing the existing research gap and offering a viable alternative for sustainable H₂ production.

2. Materials and Methods

Materials: Formic Acid (97.5/2.5H₂O[v/v]), RuCl₃x(H₂O) (ruthenium-III content 40.0–49.0%), Fe(BF₄)₂ × 6H₂O, and tris [2-(diphenylphosphino)ethyl]phosphine PP₃ (P(CH₂CH₂PPh₂)₃) (98% purity) were purchased from Sigma Aldrich and stored under argon. Propylene carbonate (PC), bis[(2-di-i-propylphosphino)ethyl]amine (^iProPNP), 10 wt.% in THF and bis [2-(di-tert-butylphosphino)ethyl]amine (^tBuPNP), 10 wt.% in hexane were obtained from Strem.

Characterization Techniques: Fourier Transform-Infrared spectra were recorded using an FT-IR Nicolet IS5 system equipped with an OMNIC software package 9.2.86 from 450 cm⁻¹ to 4000 cm⁻¹, with 2 cm⁻¹ resolution. Solid-state NMR spectra for ¹³C, ³¹P, and ²⁹Si were recorded in a 400 MHz magnet using cross-polarization (CP-MAS) at a spinning rate of 14,000 Hz with proton decoupling. The parameters used were as follows: acquisition time of 0.053 s, recycle delay of 0.5 s, contact time of 2 s, and 10,000 scans. Peak correction was performed using a standard reference and field adjustment (adamantane). Thermogravimetric analysis (TGA-DTA) was performed using a SETARAM TGA 92 analyzer, (SETARAM Instrumentation, Caluire-et-Cuire, France) with a heat rate of 10 °C/min from 25 °C to 800 °C and a flow rate of 20 mL/min for the oxygen carrier gas. E_h measurements were performed using an ORP electrochemical electrode combined with a Pt ring vs. Standard Hydrogen Electrode (SHE) [35] from Metrohm (Herisau, Switzerland), and the measurements were performed in situ during the catalytic procedure. Attenuated Total Reflection-FTIR (ATR-FTIR) experiments were recorded in situ in the region of 400–2000 cm⁻¹ by an Agilent spectrometer (Agilent Technologies, Santa Clara, CA, USA), including a ZnSe-attenuated total reflection accessory. UV-Vis spectra were recorded on a Hitachi spectrophotometer, U2900 (Hitachi High-Tech Corporation, Tokyo, Japan) operating in the range λ = 250–800 nm. Liquid samples were measured in a 3 mL quartz cuvette (1 cm optical path).

ATR-FTIR measurement procedure: The ATR-FTIR procedure was carried out in situ using approximately 5 mg of ground KBr as the solid matrix. Two drops of each liquid sample (PNPs, PC, and FA) or the catalytic solutions were then added and allowed to absorb for 5 min. After absorption, the samples were pressed with an ATR-FTIR diamond head and measured. Solid samples, such as PP3 and SiO₂, were measured directly without the addition of KBr.

Solution Potential (E_h) measurements: The reaction was carried out in a wide-neck glass tube to accommodate the redox-electrode while stirring. The amounts of reagents used and the adopted conditions were strictly similar to those for catalysis experiments. E_h was continuously recorded after each addition of any reagent. Amounts of reagents for Fe catalytic system: 5 mL PC with 2 mL formic acid, 7.5 μmol [Fe^II(BF₄)₂ × 6H₂O], different amounts of PNP and SiO₂@PNPs ligands, 7.5 μmol PP3.

Catalytic procedure: Catalytic reactions were carried out in a thermostatic double-wall reactor with the insertion of argon gas for the first 10 min and continuous stirring at 85 °C (±2 °C). The reactor was directly linked to a GC system (Shimadzu GC-2014 Gas Chromatograph with a Thermal Conductivity Detector, GC-TCD, Shimadzu Corporation, Kyoto, Japan) and a Carboxen-1000 column (Supelco, Bellefonte, PA, USA) for the analysis and identification of the produced gases. The total volume of the evolved gases was measured using a manual gas burette. In a typical catalytic experiment, 7.5 μmol Fe(BF₄)₂ × 6H₂O were dispersed in 5 mL of PC and 2 mL of formic acid, followed by the addition of different ratios of the PNP ligand. The reaction took place for 10 min and then 7.5 μmol of P(CH₂CH₂PPh₂)₃ was inserted into the mixture. Without the use of PP3 the dehydrogenation of formic acid did not occur, meaning that the presence of PP3 is necessary. In all catalytic experiments, a molar ratio of [metal:PP3] = [1:1] was used. For the continuous operation catalytic experiments, the addition of formic acid was 1 mL when the produced gases reached around 1.2 Lt (1 mL of formic acid produces 1.2 Lt of gases). All catalytic data presented in this study were obtained from the average of at least three sets of experiments, with a standard error of 5%. Recycling experiments: To recover the catalytic materials, the reaction mixture was centrifuged at 6000 rpm for 10 min. The materials were separated from the solution, washed with methanol three times, and then dried into a dry pistol at 80 °C under vacuum. In the recycling sets of experiments, the same procedure as for a typical catalytic test described above was adopted; however, instead of new amounts of metals and PNP-based materials, the recovered solid was added. Addition of NaBH₄: The NaBH₄ was added, where it is mentioned, after the generation of the catalytic system [metal/PNP/PP3] or [metal/SiO₂@PNPs/PP3] as an aqueous solution. In the case of NaBH₄ addition, to ensure that the released gases were derived from FA dehydrogenation, the gas volume was measured 10 min after this addition. TONs and TOFs are calculated using the equations in our previous publication by our group [16,36] and are shown in the Supporting Information as Equations (S1) and (S2).

• Synthesis and Characterization of the immobilized SiO₂@^iProPNP and SiO₂@^tBuPNP

Synthesis: In a 50 mL flask with 10 mL of methanol, 1.2 mmol of a PNP ligand and 1.1 mmol of 3-(chloropropyl)-trimethoxysilane were added and refluxed for 48 h. Then, 250 mg of dried SiO₂ particles were added to the solution, and the suspension was refluxed for 24 h. The obtained hybrid material was washed three times with methanol and was dried under a vacuum at 80 °C for 24 h (Scheme 1).

FT-IR characterization: In the FT-IR spectrum of SiO₂@^iProPNP (see Figure S1), the bands appeared at 1093 and 468 cm⁻¹ and were assigned to the stretching of the Si-O-Si bonds, while those at 3430 and 802 cm⁻¹ were assigned to the stretching of the surface Si-OH bonds of the silica support. The corresponding bands of SiO₂@^tBuPNP (see Figure S2) are shown at 1100, 470, 3434, and 803 cm⁻¹, respectively [22,37]. The bands at 2360 cm⁻¹ and 2341 cm⁻¹ (Figure S1) and 2361 cm⁻¹ and 2338 cm⁻¹ (Figure S2) are derived from atmospheric CO₂. The successful grafting of the PNP ligands onto the SiO₂ surface is proven by the appearance of the bands derived from vibrations of PNP scaffold, i.e., at 2920 cm⁻¹, 2850 cm⁻¹ (for SiO₂@^iProPNP) and 2921 cm⁻¹, 2851 cm⁻¹ (for SiO₂@^tBuPNP) assigned to the stretching of C-H bonds, at 1540 cm⁻¹ (SiO₂@^iProPNP) and 1539 cm⁻¹ (SiO₂@^tBuPNP) attributed to the stretching of the N-C bonds, at 1454 cm⁻¹ (for SiO₂@^iProPNP) and 1449 cm⁻¹ ( for SiO₂@^tBuPNP) due to the bending of P-C bonds, at 1230 cm⁻¹ (SiO₂@^iProPNP) and 1248 cm⁻¹ (SiO₂@^tBuPNP) to the bending of the N-C bonds and finally, at 1418 cm⁻¹, 1384 cm⁻¹ (for SiO₂@^iProPNP) and 1415 cm⁻¹, 1381 cm⁻¹ (for SiO₂@^tBuPNP) attributed to the stretching of P-CH₃ bonds [22,37].

TGA-DTA analysis: The loading of PNP ligands onto the surface of the SiO₂ particles was calculated from thermogravimetric analysis (see thermograms in Figures S3 and S4). From 250 °C to 500 °C a complete combustion of the carbonaceous backbone of the PNP ligands occurs, followed by two exothermic peaks at 300 and 500 °C. The calculated (%) loading based on the mass loss was found to be 9.5% (17 mmol ^iProPNP/g of material) for SiO₂@^iProPNP and 9.7% (22 mmol ^tBuPNP/gr of material) for SiO₂@^tBuPNP.

Solid-State NMR: Alongside FT-IR spectroscopy, an ss-NMR study was conducted by running ¹³C, ³¹P, and ²⁹Si spectra to further ensure the successful immobilization of the PNP ligands onto the surface of the SiO₂ particles.

¹³C spectrum of SiO₂@^iProPNP is shown in the Supplementary Material in Figure S5, where, for convenience, the structure of the grafted molecule is shown on the left side. The peak at 9.46 ppm corresponds to the methyl carbon atoms (g) of the iso-propyl groups, while those at 16.14 and 21.59 ppm are assigned to methylene C atoms (b) and (a), respectively, of the linker propyl-group. The sharp peak at 26.46 ppm corresponds to the methylene carbons (e) connected directly to P atoms, while that at 45.50 ppm to the carbons (f) of the iso-propyl groups, which are also connected directly to P atoms. Finally, the peaks observed at 50.41 and 59.46 ppm are attributed to the methylene carbons (d) and (c), respectively, which are both bonded with the N atom of the ^iProPNP ligand [22,38].

Accordingly, the ¹³C spectrum of SiO₂@^tBuPNP and the molecular structure of the grafted ^tBuPNP functionality are shown in Figure S6. The peak at 9.15 ppm corresponds to the methylene carbon atoms (e), which are connected directly to P atoms, while those at 16.21 and 30.04 ppm are assigned to methylene C atoms (b) and (a), respectively, of the linker propyl-group. The sharp peak at 26.32 ppm corresponds to the methyl carbons (g) of the tert-butyl groups, while the (f) carbons of the tert-butyl groups, which are connected to P atoms, appear at 45.53 ppm. Finally, the peaks observed at 49.47 and 59.41 ppm are attributed to the methylene carbons (d) and (c), respectively, which are both linked to the N atom of the ^tBuPNP ligand [22,38].

The ³¹P spectra of SiO₂@^iProPNP and SiO₂@^tBuPNP are presented in Figures S7 and S8, respectively. The intense and sharp peak at 41.68 ppm for SiO₂@^iProPNP (Figure S7) and at 71.41 ppm for SiO₂@^tBuPNP (Figure S8) were assigned to the two chemically equivalent phosphorus atoms of the PNP ligands. The broad and low to medium-intensity peaks appeared at 35.84 and 64.39 ppm in the spectrum of SiO₂@^iProPNP and at 47.52 and 41.61 ppm in the SiO₂@^tBuPNP spectrum are attributed to chemically non-equivalent phosphorus atoms of the PNP ligands; the observed chemical shifts could be a result of different interactions of the two phosphorus atoms due to spatial limitations derived from the grafting of the PNP functionalities on the silica support [22,38].

The ²⁹Si spectra of the hybrid materials SiO₂@^iProPNP and SiO₂@^tBuPNP are presented in Figures S9 and S10, respectively. Both show two sets of bands centered at −100 and −60 ppm, which correspond to Qⁿ and Tⁿ type of Si atoms. In the ²⁹Si spectrum of SiO₂@^iProPNP_, the intense peak at −102.0 ppm is assigned to Q³ atoms, the second highest peak at −110.8 ppm to Q⁴ species, and a shoulder appearing at −92.0 ppm corresponds to Q² Si atoms; these siloxane-type Si are derived from the silica support used for grafting [39,40]. The corresponding peaks of SiO₂@^tBuPNP rose at −101.3, −110.7 and −90.4 ppm, respectively. The Tⁿ species originated from Si atoms of organo-silane alkoxy precursor and indicates the degree of co-polymerization with the silica surface; the T³ bears 3 siloxane bonds and represents a species that has undergone full co-polymerization. In the ²⁹Si spectrum of SiO₂@^iProPNP_, the T² peak at −58.9 ppm is the highest of the Tⁿ set peaks, with T³ at −67.2 ppm being the second highest; the shoulder at −50.8 ppm is assigned to the T¹ species. The set of Tⁿ peaks for SiO₂@^tBuPNP shows a similar profile, and the corresponding peaks appear at −57.9, −66,4, and −49,8 ppm, respectively. These data further confirm that the grafting process was successful, resulting in a strong attachment of the PNP functionalities on the silica surface via the formation of 2 or 3 siloxane bonds [39,40].

3. Results

• Evaluation of the Solution Potential, E_h, of the Catalytic Process

Before assessing H₂ production via FADH, the solution potential profile of the catalytic systems was monitored. As shown recently, in situ monitoring of the solution potential E_h [41] offers a convenient and reliable tool to frame the physicochemical characteristics of the catalytic system under study. In this context, the E_h-profiles of homogeneous [Fe^II/^iProPNP/PP3], [Fe^II/^tBuPNP/PP3], and heterogeneous [SiO₂@^iProPNP/Fe^II/PP3], and [SiO₂@^tBuPNP/Fe^II/PP3] systems were monitored using the optimal [PNP:Fe] ratio [1:1:1]. The experimental Eh-data are presented in Figure 1. As seen, when all the reagents have been added, the solution potential Eh reaches a negative value of −49 mV to −67 mV, indicating the formation of an active environment for FA dehydrogenation [41].

Subsequently, spontaneous generation of bubbles was observed. It is important to note that, in the absence of PP3, the Eh of the catalytic reactions based on the [Fe^II/^iProPNP], [Fe^II/^tBuPNP], [SiO₂@^iProPNP/Fe^II], and [SiO₂@^tBuPNP/Fe^II] systems consistently remained at positive values (refer to data in Figure S11 in the Supplementary Information), and gas production was minimal.

• Spectroscopic study of the catalytic systems

UV-Vis Spectroscopy: The in situ formation of the active [Fe^II/^iProPNP/PP3] and [Fe^II/^tBuPNP/PP3] catalysts was confirmed using UV-Vis spectroscopy, as shown in Figure 2.

The UV-Vis spectra of [Fe^II/^iProPNP] and [Fe^II/^tBuPNP] present analogous profiles. The intense bands observed at 290 and 343 nm, as well as the low-intensity peaks at 410 and 470 nm, are attributed to the Fe²⁺ oxidative species and the Metal-to-Ligand Charge Transfer (MLCT) transitions due to the interaction of Fe²⁺ with the PNP ligands (d→π_PNP) [31]. In both systems, after the addition of the PP3 ligand, a band with a maximum at 500 nm is observed, which is assigned to the MLCT transition derived from the interaction of Fe²⁺ with PP3 (d→π_PP3) [16]. This clearly indicates the formation of tertiary [Fe^II/^iProPNP/PP3] and [Fe^II/^tBuPNP/PP3] systems during FADH catalysis. The medium-intensity band with a maximum at 390 nm, which also appeared after the addition of PP3 in both systems, indicates the generation of active Fe-H species directly involved in this type of catalysis and further supports the reactivity of the studied systems [42]. Thus, UV-Vis spectroscopy was crucial in understanding the electronic interactions between the Fe²⁺ ions and the ligands, as well as the evolution of the metal coordination environment during catalysis by identifying intermediate species.

ATR-FTIR: The in situ ATR-FTIR data under catalytic conditions for the homogeneous [Fe^II/^iProPNP/PP3] and [Fe^II/^tBuPNP/PP3], as well as for the heterogeneous [Fe^II/SiO₂@^iProPNP/PP3] and [Fe^II/SiO₂@^tBuPNP/PP3] systems, are presented in the Supplementary Material (see Figure S12). The corresponding spectra for the [Fe^II/^iProPNP], [Fe/^tBuPNP], [Fe^II/SiO₂@^iProPNP], and [Fe^II/SiO₂@^tBuPNP] systems without PP3 were also recorded. As a background study, the ATR-FTIR spectra of (i) the PC solvent with FA, (ii) PP3, and (iii) the hybrid materials SiO₂@^iProPNP and SiO₂@^tBuPNP were obtained and are shown in Figure S12.

The ATR-FTIR spectrum of FA in PC (Figure S12a) is characterized by bands at 1787 cm⁻¹, 1394 cm⁻¹, and 1195 cm⁻¹, corresponding to the stretching of the C=O bond, bending of the O-H bond, and stretching of the C-O bond, respectively. Additionally, the bands at 1115 cm⁻¹ and 1048 cm⁻¹ correspond to the stretching of the asymmetric C-O-C bond in PC. In Figure S12b, the bands at 1480 cm⁻¹, 735 cm⁻¹, and 692 cm⁻¹ correspond to the stretching of the C=C bonds in aromatic structures and the stretching of the P-C bonds, respectively, in the PP3 ligand [35,43]. The ATR-FTIR spectra of the hybrid materials SiO₂@^iProPNP and SiO₂@^tBuPNP (Figure S12c,d) are dominated by the bands at 1054 cm⁻¹ and 450 cm⁻¹, which correspond to the stretching of the Si-O-Si bonds, while the band at 795 cm⁻¹ refers to the stretching of the Si-OH bonds. More detailed information concerning the FT-IR characterization of these materials is provided in Figures S1 and S2.

Figure S13 presents the ATR-FTIR spectra of [Fe^II/^iProPNP] (Figure S13a), [Fe^II/^tBuPNP] (Figure S13b), [Fe^II/SiO₂@^iProPNP] (Figure S13c), and [Fe^II/SiO₂@^tBuPNP] (Figure S13d) in PC in the presence of FA. The bands at 510 cm⁻¹, 450 cm⁻¹, and 417 cm⁻¹ are assigned to the stretching of Fe-P and Fe-N bonds, respectively, confirming the formation of the Fe-PNP complexes in all cases. The peak at 1380 cm⁻¹ (Figure S13a,c) is attributed to the iso-propyl groups of ^iProPNP [44], and the peak at 1462 cm⁻¹ (Figure S13b,d) is attributed to the tert-butyl groups of the ^tBuPNP ligands [45]. The vibrations of the v(Si-O-Si) bonds from the SiO₂ matrix (Figure S12c,d) overlap with the strong bands from the stretching of the v(C-O-C) bond derived from the PC solvent (Figure S13c,d). Additionally, two other bands located at 1585–1581 cm⁻¹ and 1508–1513 cm⁻¹ in all spectra are attributed to formate vibrations, indicative of the type of bonding of the HCOO⁻ anion, i.e., η¹-O₂CH and η²-O₂CH, respectively [35,43]. These data provide direct evidence that the complex [Fe/PNP/FA] has been formed in all cases. However, it should be noted that in the absence of PP3, the catalytic data show zero catalytic FADH activity. This indicates that [Fe/PNP/FA] without PP3 fails to generate an appropriate reduced environment for FADH catalysis, as evidenced by the positive E_h values measured herein, as discussed above.

Figure 3 shows the ATR-FTIR spectra of the complete catalytic system (in the presence of PP3): [Fe^II/^iProPNP/PP3] (Figure 3a), [Fe^II/^tBuPNP/PP3] (Figure 3b), [Fe^II/SiO₂@^iProPNP/PP3] (Figure 3c), and [Fe^II/SiO₂@^tBuPNP/PP3] (Figure 3d) in PC in the presence of FA. Comparing this set of ATR-FTIR spectra with those in Figure 3, the difference arises from the addition of PP3, which results in a slight shift of the v(Fe-P) bond bands due to PP3 coordination to the Fe center. Specifically, these bands, initially at 521 cm⁻¹ and 449 cm⁻¹ for [Fe^II/^iProPNP] (Figure S13a) and at 508 cm⁻¹ and 456 cm⁻¹ for [Fe^II/^tBuPNP] (Figure S13b), are shifted to 520 cm⁻¹ and 451 cm⁻¹ (Figure 3a) and to 510 cm⁻¹ and 458 cm⁻¹ (Figure 3b). An analogous shift has been recorded for the heterogeneous systems: the v(Fe-P) vibrations initially at 508 cm⁻¹ and 455 cm⁻¹ in the [Fe^II/SiO₂@^iProPNP] (Figure S13c) and at 508 cm⁻¹ and 449 cm⁻¹ in the [Fe^II/SiO₂@^tBuPNP] (Figure S13d) are shifted to 510 cm⁻¹ and 458 cm⁻¹ in the [Fe^II/SiO₂@^iProPNP/PP3] (Figure 3c) and to 501 cm⁻¹ and 451 cm⁻¹ in the [Fe^II/SiO₂@^tBuPNP/PP3] (Figure 3d).

Ligation of PP3 to Fe also affects the Fe-N bond, likely leading to less strong coordination with the PNP molecules in general. The vibration of Fe-N, found at 417 cm⁻¹ in [Fe^II/^iProPNP] (Figure S13a) and [Fe^II/^tBuPNP] (Figure S13b), shifts to 420 cm⁻¹ and 419 cm⁻¹ in [Fe^II/^iProPNP/PP3] (Figure 3a) and [Fe^II/^tBuPNP/PP3] (Figure 3b), respectively. Similarly, this band shifts from 415 cm⁻¹ in [Fe^II/SiO₂@^iProPNP] (Figure S13c) and [Fe^II/SiO₂@^tBuPNP] (Figure S13d) to 426 cm⁻¹ and 418 cm⁻¹ in the spectra of [Fe^II/SiO₂@^iProPNP/PP3] (Figure 3c) and [Fe^II/SiO₂@^tBuPNP/PP3] (Figure 3d), respectively. These spectroscopic data provide further evidence that the entire homogeneous [Fe/PNP/PP3] and heterogeneous [Fe^II/SiO₂@PNP/PP3] systems are the active and functional units in FADH catalysis [35,43]. Additionally, in the spectra of Figure 3a,c, the band at 1380 cm⁻¹ is assigned to the iso-propyl groups of ^iProPNP molecules [44], and those at 1462 cm⁻¹ (Figure 3b,d)) to the tert-butyl groups of the ^tBuPNP ligands [45]. The band close to 1485 cm⁻¹ is attributed to the stretching of the C=C bond of the aromatic groups included in the PP3 molecule structure. The vibrations of the SiO₂ matrix, ν(Si-O-Si), are covered by the strong ν(C-O-C) stretching from the abundant PC solvent (Figure 3c,d)). The bands appearing at 1589–1593 cm⁻¹ and 1505–1510 cm⁻¹ in all spectra are attributed to formate anion vibrations, indicating monodentate and/or chelate coordination [35,43].

Overall, the ATR-FTIR spectra provide insights into the chemical bonding and structural changes occurring in the catalysts. The characteristic bands observed confirmed the successful coordination of the PNP and PP3 ligands with the Fe²⁺ center, thus the formation of the active catalysts as well as the involvement of the formic substrate.

• Standard catalytic H₂-production

Based on the data obtained from E_h mapping, UV-Vis, and ATR-FTIR data, a standard FA dehydrogenation experiment was conducted in propylene carbonate (PC) at 85 °C with 2 mL FA (48 mmol), catalyzed by Fe^II/^iProPNP/PP3, Fe^II/SiO₂@^iProPNP/PP3, Fe^II/^tBuPNP/PP3, or Fe^II/SiO₂@^tBuPNP/PP3. The catalytic performance, as demonstrated by H₂ production from FA, directly correlates with the electronic and structural properties of the catalysts observed in the UV-Vis and ATR-FTIR analyses. The volume of gases produced over time, as presented in Figure 4, confirmed the catalytic reactivity of the studied systems. All catalytic gaseous products were subjected to a quality check by GC-TCD, and a typical chromatograph is shown in Supporting Information Figure S15. The only detected products were H₂ and CO₂ without any trace of CO, which makes the present catalytic systems suitable for use in fuel cell technology.

More specifically, the homogeneous [Fe^II/^iProPNP/PP3] generated over 2300 mL of gases in 50 min, with a TON of 6406 and a TOF value of 7718 h⁻¹. [Fe/^tBuPNP/PP3] produced 2260 mL of gases in 60 min, resulting in a TON of 6242 h⁻¹. For comparison, the homogeneous [Fe^II/L_g/PP3], first reported by our research group [14], achieved a TOF value of 5208 h⁻¹, significantly lower than the [Fe^II/^tBuPNP/PP3] reported here. Interestingly, the heterogeneous [Fe^II/SiO₂@^iProPNP/PP3] proved to be more efficient than its homogeneous counterpart, achieving a TON of 6460 within 40 min and a TOF of 9787 h⁻¹, the highest among all systems studied herein. Finally, [Fe^II/SiO₂@^tBuPNP/PP3] achieved a TON of 6297 h⁻¹, slightly higher than the TON of the corresponding homogeneous [Fe^II/^tBuPNP/PP3] system.

• Continuous catalytic H₂-production

To further evaluate the catalytic systems [Fe^II/^iProPNP/PP3], [Fe^II/^tBuPNP/PP3], [Fe^II/SiO₂@^iProPNP/PP3], and [Fe^II/SiO₂@^tBuPNP/PP3] in FA dehydrogenation, catalytic experiments were run under a continuous feed of FA.

The catalysis proceeds under the same conditions and follows the protocol described above, using 2 mL of FA at the beginning of the reaction. However, when 1200 mL of gases are released, a volume approximately estimated to be produced from the decomposition of 1 mL of FA, an additional 1 mL of FA is added. Under this setup, the catalysis driven by the present systems continues for 4 h and consumes 8–10 mL of FA. The gas production from FA over time catalyzed by tertiary [Fe^II/^iProPNP/PP3], [Fe^II/^tBuPNP/PP3], [Fe^II/SiO₂@^iProPNP/PP3], and [Fe^II/SiO₂@^tBuPNP/PP3] is shown in Figure 5. For comparison, the efficiency of [Fe^II/PP3] under the same experimental conditions is detailed in Figure S14b in the Supplementary Material and represented by the gray lines in Figure 5a,b.

Under continuous operation, the homogeneous [Fe^II/^iProPNP/PP3] (Figure 5a, red line) and [Fe^II/^tBuPNP/PP3] (Figure 5b, green line) converted 9 and 10 mL of FA (see Table S1 of the Supplementary Material) and produced 8120 mL and 8770 mL of gases at rates of 45 and 53 mL/min, respectively. In terms of TONs, they achieved 22,135 and 23,907 with TOF values of 5533 h⁻¹ and 5976 h⁻¹, respectively. Both [Fe^II/^iProPNP/PP3] and [Fe^II/^tBuPNP/PP3] exhibited better performance than the reference homogeneous [Fe/PP3] system, which provided 7680 mL of gases from 9 mL of FA at a rate of 40 mL/min, resulting in a TON of 20,936 and a TOF value of 5234 h⁻¹. The enhanced reactivity of the tertiary systems compared to the binary [Fe/PP3], which does not include any PNP molecule as a second ligand, is due to the ‘double ligand’ beneficial effect [33]. This effect arises from the second σ-donor ligand, which further electronically enriches the metal center and enhances its catalytic behavior.

The heterogeneous [Fe^II/SiO₂@^tBuPNP/PP3] system (Figure 5b, orange line) showed slightly better catalytic performance than the corresponding homogeneous [Fe^II/^tBuPNP/PP3] system (Figure 5b, green line) during the first 2 h of catalysis. However, it showed a delay, ultimately providing 7860 mL of gases at an average rate of 44 mL/min, with a TON of 21,427 and a TOF value of 5356 h⁻¹. In contrast, under continuous FA feeding, the heterogeneous [Fe^II/SiO₂@^iProPNP/PP3] (Figure 5a, blue line) performed much better than the homogeneous [Fe^II/^iProPNP/PP3] (Figure 5a, red line) and was by far the best catalytic system studied herein. It produced 12,010 mL of gases within 220 min at a rate of 65 mL/min from the decomposition of 11 mL of FA, achieving a TON of 35,079 with a TOF of 9643 h⁻¹.

Indeed, during the continuous operation experiment, gas production ceased after an adequate amount of FA had been added. This could be because the FA used contained 2.5% water. This small amount of water accumulates in the catalytic mixture over time, causing the catalyst to slowly deactivate and lose its efficiency. The amount of active catalyst is 7.5 μmol, and even this small amount of water is sufficient to impact its performance. Molecular catalysts, like the ones we used, do not operate at maximum efficiency in the presence of water, leading to incomplete catalysis of the FA. In homogeneous catalytic systems, the suppressive impact of water is not easily overcome [36]. Nevertheless, as we demonstrate below, current heterogeneous systems offer an economical solution to mitigate this inhibitory effect. That is, when catalytic H₂ production ceased, the solid catalyst was centrifuged, rinsed, and reused, leading to continuous H₂ production from new quantities of added FA (Figure 6). This indicates that the drop in catalytic performance after the dehydrogenation of approximately 11 mL of FA was not due to irreversible catalyst damage but could be remedied through a simple washing and drying process.

• Recycling of the catalytic materials

The enhanced reactivity of [Fe^II/SiO₂@^iProPNP/PP3] under continuous catalytic operation and its heterogeneous nature prompted the recovery of the solid material from the catalysis medium for reuse in a second catalytic process with a continuous FA feed. The solid was recovered from the reaction mixture by centrifugation, washed with methanol (MeOH), and dried.

In the recycling experiments, the same protocol as for the continuous operation process was adopted; however, instead of new amounts of the Fe precursor and SiO₂@^iProPNP or SiO₂@^tBuPNP, the recovered solids were used. The [Fe^II/SiO₂@^tBuPNP] was not able to perform after the first use; however, the [Fe^II/SiO₂@^iProPNP] operated for a second time after the addition of a new dose of PP3, producing 6620 mL of gas from decomposing 8 mL of FA at an average rate of 36 mL/min (Figure 6, violet line), with a TON of 23,822 and a TOF value of 6438 h⁻¹ (Table S1). The [Fe^II/SiO₂@^iProPNP] catalyst was recovered for a second time, washed, dried, and reused in a catalytic process under continuous FA feed for a third time. Its performance in terms of gas production vs. time is shown in Figure 6 (cyan line). Although it remained active, the observed reactivity was significantly reduced (for details, see Table S1). After the third use, it was no longer able to release gas. This is mainly attributed to the reduction in the amount of solid catalyst because, for practical reasons, its recuperation from the reaction mixture via centrifugation is not fully successful, leading to a significant loss of the catalytic material. Despite this, the recycling experiments highlight the ability of the SiO₂@^iProPNP hybrid material to form stable and durable [Fe^II/SiO₂@^iProPNP] units capable of operating in FADH under continuous FA feed and to be recovered and reused in new processes. Within this framework, [Fe^II/SiO₂@^iProPNP], associated with PP3, operated three times in total, producing 21,890 mL of gas from the decomposition of 23 mL of FA and achieving a TON of 74,451.

As demonstrated hereafter, the high reactivity of [Fe^II/SiO₂@^iProPNP/PP3] has a thermodynamic basis, as proven by Arrhenius analysis.

• Arrhenius analysis

A thermodynamic study was conducted on the [Fe^II/^iProPNP/PP3], [Fe^II/^tBuPNP/PP3], [Fe^II/SiO₂@^iProPNP/PP3], and [Fe^II/SiO₂@^tBuPNP/PP3] systems. The gas production over time from FADH catalysis driven by each system at different temperatures (85 °C, 75 °C, 65 °C, and 55 °C) was measured and is presented in Figure 7. The corresponding TOF values were applied to the Arrhenius equation $\mathrm{l}\mathrm{n}\mathrm{T}\mathrm{O}\mathrm{F}=-{\displaystyle \frac{{\mathrm{E}}_{\mathrm{a}}}{\mathrm{R}}}\times \left({\displaystyle \frac{1}{\mathrm{T}}}\right)+\mathrm{c}$ (where E_a represents the activation energy in J/mol, T is the temperature in K, and R = 8.314 J/K·mol), to create the Arrhenius plots. Figure 7a,c shows the gas production over time at different temperatures for the [Fe^II/^iProPNP/PP3] and [Fe^II/SiO₂@^iProPNP/PP3] systems, respectively. The corresponding Arrhenius plots are shown in Figure 7b,d. Based on the linear Arrhenius fits, the activation energies for [Fe/^iProPNP/PP3] and [Fe^II/SiO₂@^iProPNP/PP3] were calculated to be E_a = 48.2 ± 0.1 kJ/mol and E_a = 42.5 ± 0.1 kJ/mol, respectively, elucidating the higher catalytic activity of the latter system.

The gas production vs. time from FADH at different temperatures catalyzed by [Fe^II/^tBuPNP/PP3] and [Fe^II/SiO₂@^tBuPNP/PP3] is shown in Figure 8a,c, respectively. Figure 8b,d present the corresponding Arrhenius plots, where the lines fit to the data parameterized by the Arrhenius equation. This analysis estimated the activation energy, which is found to be E_a = 46.4 ± 0.09 kJ/mol for [Fe^II/^tBuPNP/PP3] and E_a = 48.3 ± 0.08 kJ/mol for [Fe^II/SiO₂@^tBuPNP/PP3]. [Fe^II/^tBuPNP/PP3] exhibits a slightly lower E_a compared to that of [Fe^II/SiO₂@^tBuPNP/PP3], which could be attributed to the spatial restrictions of the immobilized ^tBuPNP with the bulky t-butyl groups on the silica surface.

The E_a values of the homogeneous [Fe^II/^iProPNP/PP3] and [Fe^II/^tBuPNP/PP3], as well as the heterogeneous [Fe^II/SiO₂@^iProPNP/PP3] and [Fe^II/SiO₂@^tBuPNP/PP3], found herein, are within the range of 40–60 kJ/mol, similar to those reported for analogous catalytic systems active in FADH catalysis [28,34]. However, it is highlighted that the low E_a of the heterogeneous [Fe^II/SiO₂@^iProPNP/PP3], which is 42.5 ± 0.1 kJ/mol, is the lowest among the catalysts studied herein. This lower E_a explains the higher catalytic activity of [Fe^II/SiO₂@^iProPNP/PP3] on a physicochemical basis.

• Comparison with the literature

Table 1. Comparison of [Fe^II/^iProPNP/PP3] and [Fe^II/^tBuPNP/PP3] with PNP-based catalytic complexes from pertinent literature.

Catalyst	Solvent	Additive	TON	TOF (h−1) Working Time (h)	Reference
Ir complex with [2,6-(ECH2)C6H3]−, E = iso-propyl (iPro) groups (homogeneous)	tert-butanol	NEt3	2000	120,000 0.017	[18]
[Fe(tBuPNP)(H)2(CO)] with 2,6-di(di-tert-butylphosphinomethyl) pyridine (homogeneous)	1,4-dioxane	NEt3	100,000	416 240	[19]
(PNP)Ru(H)Cl(CO) (homogeneous)	DMF	NEt3	326,500	257,000 1.28	[20]
FeII(H)(iPrPNP)(CO) (homogeneous)	1,4-dioxane	LiBF4	983,642	196,728 9.5	[21]
Mn(H)(iPrPNP)(CO) (homogeneous)	1,4-dioxane	LiBF4	190	13.5 14	[22]
Ru(H)(Cl)(CO)(HN{CH2CH2P(iPr)2}2), (iProPNP) (homogeneous)	Triglyme and H2O	KOH	5263	5263 1	[28]
Fe/iProPNP/PP3 (homogeneous in situ)	Propylene carbonate	-	22,135	5533 4	This work
Fe/tBuPNP/PP3 (homogeneous in situ)	Propylene carbonate	-	23,907	5976 4	This work
Fe/SiO2@iProPNP/PP3, (heterogeneous in situ)	Propylene carbonate	-	74,451	- 12	This work
Fe/SiO2@tBuPNP/PP3, (heterogeneous in situ)	Propylene carbonate	-	21,427	5356 4	This work

presents the activities of molecular metal catalysts involving ^iProPNP and ^tBuPNP ligands for FADH, discussed in terms of TONs, TOFs, operational time, and the need for additives. Currently, the PNP ligands used in FADH catalysis include ^iProPNP, which are associated with Ir [18], Ru [28], Fe [19], and Mn [22] to form the corresponding metal complexes. In all these cases, the active catalysts are {hydrido}- and/or {carbonylo}-{^iProPNP}-metal complexes, which require additives to boost their activity [19,28]. Some of these catalysts have achieved high TON values along with long working times of up to 9.5 h and high TOF [21]. To date, no references in the literature have been found on the use of ^tBuPNP as a ligand in FADH catalysis.

In this research, Fe-catalysts are formed in situ by the association of [Fe/^iProPNP/PP3] and [Fe/^tBuPNP/PP3]. Both catalysts are immediately active in FADH catalysis without the need for an additive or additional ligation with hydrides and/or CO. They operate for 4 h and achieve TON values of 22,100 and 23,900, respectively. Covalent grafting of ^iProPNP and ^tBuPNP on a silica surface results in hybrids SiO₂@^iProPNP and SiO₂@^tBuPNP, which are used as modified scaffolds for catalysis. Their association with Fe and PP3 forms in situ the heterogeneous [Fe/SiO₂@^iProPNP/PP3] and [Fe/SiO₂@^tBuPNP/PP3], which are active in FADH catalysis. Additionally, the [Fe/SiO₂@^iProPNP/PP3] catalyst is recyclable and demonstrates significant performance, achieving 74,450 TONs within 12 h of continuous operation.

4. Conclusions

This work explores the use of two PNP ligands, ^iProPNP and ^tBuPNP, in double-ligand Fe-catalysts for the dehydrogenation of FA. A convenient in situ formation of the [Fe^II/^iProPNP/PP3] and [Fe^II/^tBuPNP/PP3] catalysts is presented, which does not require additives or additional ligation with hydrides and/or CO for efficient catalysis. The synthesis of SiO₂@^iProPNP and SiO₂@^tBuPNP materials via covalent immobilization on a silica surface has been achieved and demonstrated by advanced spectroscopic techniques. Accordingly, the in situ generation of [Fe^II/SiO₂@^iProPNP/PP3] and [Fe^II/SiO₂@^tBuPNP/PP3] in catalysis has been proven, as well as their efficiency. Among the catalysts presented here, [Fe^II/SiO₂@^iProPNP/PP3] exhibited remarkable catalytic behavior; it was recyclable, operating three times under a continuous feed of FA, producing a total of 10,945 H₂ gas and achieving a TON of 74,451. This reactivity has a thermodynamic basis, as proven by Arrhenius analysis, due to a low activation energy (E_a) of 42.5 ± 0.1 kJ/mol, which is the lowest among the others measured in this study. The development of efficient and recyclable catalysts can potentially lower the costs associated with H₂ production and storage, facilitating broader adoption in various industrial applications. Moreover, our catalysts’ performance under mild conditions aligns with the industry’s need for scalable and cost-effective solutions. By addressing the current challenges in H₂ production and offering viable alternatives, our work contributes to the advancement of new energy technologies and supports the transition to a more sustainable energy future. The commercial application of these catalysts could enhance H₂’s role as a clean energy carrier, promoting environmental benefits and industrial innovation.