www.eurjic.org Unveiling Solvent-Dependent Divergent Hydrogen Production Pathways during the Dehydrogenation of Formic Acid Using N,N 0-Iminopyridine Ruthenium(II) Complexes Thabiso Mphuti, Rotondwa Mphephu, Moegamat Joseph, and Andrew J. Swarts* The preparation of a panel of cationic, half-sandwich iminopyri- dine ruthenium(II) complexes of the type [(L1)RuCl(p-cymene)]Cl (C1–C11) is reported, where L= substituted N-phenyl-1-(2- pyridinyl)methanimine ligand derivates (L1–L9) and their effi- ciency in the catalytic dehydrogenation of formic acid (FA). The activity could be correlated with the nature of the substituent on the imine nitrogen and the solubility of the complexes in water, with C1 exhibiting the highest initial turnover frequency (TOF) of 281 hr�1 at 90 °C. Kinetic and mechanistic investigations are undertaken in dimethyl sulfoxide (DMSO) and H2O, revealing that the proton source in the hydrogen production step is sol- vent-dependent. FA is found to be responsible for H2 formation in DMSO, and H3Oþ ions are involved in generating H2 in water. Complex C1 is more stable in water, as it maintains efficient gas evolution for 16 cycles without any deactivation, reaching a turn- over number (TON) of 13 791. Furthermore, C1 is active, although with reduced activity (TOF= 43 hr�1), for over 34 h when oper- ated under continuous FA addition conditions. 1. Introduction Efforts to implement a “hydrogen economy” have often been impeded by a lack of suitable storage systems for hydrogen gas. Its low volumetric density (0.08988 g L�1 at 1 atm) makes it difficult to store and transport using existing infrastructure for fuels.[1] An alternative approach to attain this objective is the intro- duction of liquid organic hydrogen carriers (LOHCs) as potential H2 energy vectors. LOHCs such as formic acid (FA, HCOOH) can undergo reversible hydrogenation and dehydrogenation cycles, which enable hydrogen storage and release.[2] FA is a liquid under ambient conditions, enabling easy storage, handling, and transpor- tation. It has a moderately high hydrogen content (4.4 wt%) and a high energy density (1.77 kWh L�1).[3] The dehydrogenation of FA to generate H2 gas and carbon dioxide using a homogeneous complex was first described by Coffey more than half a century ago.[4] The use of heterogeneous[5,6] and homogeneous catalyst systems[7,8] toward FA dehydrogenation (FADH) is well docu- mented in the literature. Although heterogeneous catalysts offer excellent recovery, they do not provide the superior catalytic activ- ity achieved by homogeneous catalysts. Since the initial seminal reports,[9,10] several state-of-the-art complexes based on iron (I), iridium (II, III), and ruthenium (Ru, IV) have been reported for the catalytic dehydrogenation of FA. Hazari and Schneider reported an iron complex, I, based on an earth-abundant metal, generating turnover number (TON) and turnover frequency (TOF) of 980 000 and 200 000, respectively.[11] Precious metal- containing II, III, and IV could generate TONs of up to 3 000 000, allowing access to industrially competitive FADH catalysts (Figure 1).[7,12,13] These exceptional catalyst systems have led to a surge in the number of investigations seeking to access high- performance FADH catalysts (Figure 1). However, most of them require the inclusion of organic solvents and sophisticated ligand scaffolds for efficient gas evolution, while exhibiting poor solubility and activity toward FA dehydrogenation in water as a solvent.[14–16] To date, only a few ruthenium-based complexes that can function inwater have been reported (Figure 1).[17–19] Huang and co-workers reported the dehydrogenation of FA in water using a Ru complex, V, bearing an N,N 0-diimidazole ligand in the absence of organic additives.[17] A TOF of 12 000 hr�1 and a TON of 350 000 at 90 °C were obtained from an aqueous solution of FA and sodium for- mate (HCOONa). In addition, they demonstrated the production of high-pressure H2 and CO2 up to 24.0MPa (3480 psi). Singh and coworkers reported H2 production from the dehydrogenation of FA over pyridinemethanol-based Ru complexes, VI, in water.[18] The complexes exhibited high stability in water and could be recycled up to seven times with a total TON of 6050. He and coworkers reported the dehydrogenation of aqueous FA by a T. Mphuti, R. Mphephu, M. Joseph, A. J. Swarts Molecular Sciences Institute School of Chemistry University of the Witwatersrand PO Wits 2050 Johannesburg, South Africa E-mail: andrew.swarts@wits.ac.za M. Joseph Department of Chemistry and Polymer Science Stellenbosch University Private Bag 1, Matieland, 7601 Stellenbosch, South Africa Supporting information for this article is available on the WWW under https:// doi.org/10.1002/ejic.202500212 © 2025 The Author(s). European Journal of Inorganic Chemistry published by Wiley-VCH GmbH. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. Eur. J. Inorg. Chem. 2025, 28, e202500212 (1 of 10) © 2025 The Author(s). European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202500212 http://www.eurjic.org https://orcid.org/0000-0002-7942-8557 mailto:andrew.swarts@wits.ac.za http://creativecommons.org/licenses/by-nc-nd/4.0/ http://doi.org/10.1002/ejic.202500212 http://crossmark.crossref.org/dialog/?doi=10.1002%2Fejic.202500212&domain=pdf&date_stamp=2025-07-09 ruthenium complex ligated by a diamino-bis(1H-1,2,4-triazole) ligand, VII.[19] Using 0.01mol% of catalyst and 20mmol of FA at 90 °C, a maximum TOF of 570 hr�1 was reached. Herein, we report a panel of water-soluble cationic half- sandwich ruthenium(II) complexes C1–C11 bearing iminopyri- dine ligands as active catalyst precursors for H2 production from FA dehydrogenation in dimethyl sulfoxide (DMSO) and water. Our investigation, utilizing spectroscopic, kinetic, and thermodynamic studies, provides insight into the role the solvent plays in regu- lating the proton source during hydrogen production. We also demonstrate the reusability and long-term performance of the most active catalyst in our panel. 2. Results and Discussion 2.1. Synthesis and Characterization As shown in Scheme 1, the prepared iminopyridine ligands (L1–L9) react with the [Ru(p-cymene)Cl2]2 precursor to afford cationic half-sandwich iminopyridine ruthenium(II) complexes (C1–C11) inmoderate-to-good yields (73%–87%). These complexes, except for C5, and C6, are air- andmoisture-stable yellow solids and display good solubility in polar solvents such as MeOH, H2O, and DMSO, but poor solubility in hexane and ethers. The1H nuclear magnetic resonance (NMR) spectra of the complexes C1–C11 dis- play a singlet for the imine proton in the region δ= 8.49–9.09 ppm, which is considerably downfield compared to the imine proton of the corresponding free ligand, δ= 8.31–8.61 ppm (Section S3, Supporting Information). Additionally, the1H NMR resonances cor- responding to the coordinated p-cymene ring for C1–C11 appear in the expected region, δ= 5.32–6.31 ppm.[20] In the fourier transform infrared (FT-IR) spectra for complexes C1–C11, the imine (νC=N) band appears in the low wavenumber region, 1611–1626 cm�1 compared to the free imine ligands, 1627–1649 cm�1 (Section S3, Supporting Information). The electrospray ionization mass spec- trometry (ESI-MS) spectra of complexes C1–C11 recorded in the positive ion mode showed prominent isotope clusters correspond- ing to the [M]þ fragment ions (Section S3, Supporting Information). Suitable single crystals of C1 and C11 were obtained from the slow Figure 1. Benchmark metal-based and water-soluble ruthenium-based catalysts for FA dehydrogenation. Scheme 1. Synthesis of cationic half-sandwich iminopyridine Ru(II) complexes C1–C11. Eur. J. Inorg. Chem. 2025, 28, e202500212 (2 of 10) © 2025 The Author(s). European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202500212 10990682c, 2025, 24, D ow nloaded from https://chem istry-europe.onlinelibrary.w iley.com /doi/10.1002/ejic.202500212 by South A frican M edical R esearch, W iley O nline L ibrary on [27/10/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense http://doi.org/10.1002/ejic.202500212 diffusion of diethyl ether in methanolic solutions. The molecular structure of C1 is illustrated in Figure 2, while that of C11 is pro- vided in Figure S81, Supporting Information. Both complexes crys- tallized in the triclinic crystal system with the P-1 space group. The complexes adopt a three-legged piano stool arrangement in which the Ru(II) metal center has a pseudo-octahedral molecular geome- try (Figure 2). The coordination sphere around ruthenium is com- pleted by the iminopyridine ligand, which coordinates to the Ru metal center through the N atoms of the imine and the pyridine ring, p-cymene ring at the tip of the stool, and a Cl atom at the bottom of the stool. Suitable single crystals of C1 and C11 were obtained from the slow diffusion of diethyl ether in methanolic sol- utions. The molecular structure of C1 is illustrated in Figure 1, while that of C11 is provided in Figure S81, Supporting Information. Both complexes crystallized in the triclinic crystal system with the P-1 space group. The complexes adopt a three-legged piano stool arrangement in which the Ru(II) metal center has a pseudo- octahedral molecular geometry (Figure 2). The coordination sphere around ruthenium is completed by the iminopyridine ligand, which coordinates to the Ru metal center through the N atoms of the imine and the pyridine ring, p-cymene ring at the tip of the stool, and a Cl atom at the bottom of the stool. The Ru─Npy bond lengths, i.e., Ru(1)-N(1), are 2.0961(15) Å and 2.083(2) Å for C1 and C11, respectively. These bond lengths are consistent with the sp2-hybridized Npy and are within the range observed for related ruthenium complexes.[21] The increased rigidity in the complexes is reflected in the N(2)─Ru(1)─N(1) bond angles of 76.43(6)° and 76.75(9)° for C1 and C11, respectively, which deviate from the ideal 90° bond angles for octahedral complexes. The planar nature, with slight distortion, of the iminopyridine ligands in the complexes is indicated by the Ru(1)─N(1)─ C(1)─C(2) torsion angles of 171.98(4)° and 175.1(2)° for C1 and C11, respectively. Key crystallographic details and selected bond parameters are summarized in Table S1, S2, Supporting Information. 2.2. FA Dehydrogenation No gas evolution was observed when the reaction was conducted without a catalyst or in the presence of the dichloro(p-cymene) ruthenium(II) dimer (Table 1, entry 1). H2 gas formation was con- firmed by1H NMR as a singlet at δ= 4.61 ppm in DMSO-d6 (Figure S82, Supporting Information). The observed chemical shift for H2 gas is in accordance with the literature precedent.[22] The production of CO2 with negligible CO formation and the absence of H2O were confirmed by gas chromatography coupled with thermal conductivity detector analyses, with He as carrier gas (Figure S83, Supporting Information). For complexes C1–C4, the imine nitrogen is bonded to aromatic substituents bearing different alkyl moieties. The catalysis performed with C1, bearing the unsubstituted phenyl ring, yielded the highest amount of CO2 þ H2, i.e., 236mL, reaching TONs of 483 after 120 minutes (Table 1, entry 2). A drastic decrease in activity was observed when varying the alkyl substitution from C1 to C2 (TON= 130) and C3 (TON= 51) owing to the introduction of successively bulkier methyl and di-isopropyl substituents on the aromatic ring, which hinder FA coordination to the Ru metal center (Table 1, entries 3 and 4). Complex C4 containing mesityl substituents on the aromatic ring displayed similar catalytic activity as the 2,6-dimethyl-substituted version, C2 (Table 1, entries 3 and 5). Figure 2. Molecular structure of C1, illustrating the two independent molecules in the asymmetric unit cell, with atomic numbering illustrated with 30% probability ellipsoids. All hydrogen atoms are omitted for clarity. Eur. J. Inorg. Chem. 2025, 28, e202500212 (3 of 10) © 2025 The Author(s). European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202500212 10990682c, 2025, 24, D ow nloaded from https://chem istry-europe.onlinelibrary.w iley.com /doi/10.1002/ejic.202500212 by South A frican M edical R esearch, W iley O nline L ibrary on [27/10/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense http://doi.org/10.1002/ejic.202500212 For complexes C5–C7, the imine nitrogen is bound to a range of aliphatic substituents. Complex C5 bearing a benzyl substituent exhibited good catalytic activity, i.e., TON= 444 compared to C6 (TON= 7) and C7 (TON= 73) bearing cyclohexyl and propyl sub- stituents, respectively (Table 1, entries 6–8). The high activity of C5 is attributed to the flexibility of the benzyl ring, which enables easy substrate coordination to the metal center. Complexes C8 and C9 possess a methyl and a trifluoromethyl substituent on the α carbon of the pyridine ring, respectively. In both cases, introducing these substituents led to decreased catalytic activity, i.e., TON= 238 for C8 and TON= 294 for C9, compared to the analogous unsubstituted complex C1 with TON of 483 (Table 1, entries 9 and 10). C9 exhibits slightly higher catalytic performance than C8 due to the electron-withdrawing trifluoromethyl substit- uent that increases the electrophilicity of the Ru metal center. The nature and amount of formate cation exerted a crucial role in facilitating the rate of FA dehydrogenation over C1. A reac- tivity trend K (TON= 499) > Na (TON= 483) > Li (TON= 332) was established when the nature of the alkali formate cation was varied (Figure S84, Supporting Information). This behavior is attributed to the relatively large size of Kþ compared to that of Naþ and Liþ, which enables easy interaction with the Cl� ligand of C1 according to the hard-soft acid-base principle to afford the catalytically active Ru-hydride species. Results indicated that TONs increased with an increase in the amount of potassium ormate (HCOOK) from 0.2 to 1.0 equivalent with respect to FA. Adding excess HCOOK, i.e., 1.4 equivalents, led to the dehydro- genation of formate (HCOO�) upon full consumption of available FA. TONs of 521 are achieved within 50min for the dehydroge- nation of 0.2 mL of FA using 0.6 eq of HCOOK over C1 under opti- mized reaction conditions (Figure S85, Supporting Information). Solvent-dependent FA dehydrogenation rates were deduced from performing the reaction in various solvent media. Interestingly, C1was soluble and effective in catalyzing FA dehydro- genation in water reaching TONs= 208 after 50min (Figure S86, Supporting Information). Halving and/or doubling the initial solvent volume (2 mL) led to decreased catalytic activities due to dilution effects associated with high reaction volumes and sol- ubility effects accompanied by low reaction volumes, respectively (Figure S87, Supporting Information). Varying the nature of the ruthenium complex counterion from the less sterically demand- ing Cl in C1 to more sterically demanding BPh4 and PF6 counter- ions in C10 and C11, respectively, did not influence either the rate or the amount of FA dehydrogenated (Figure S88, Supporting Information). 2.3. Mechanistic Studies in DMSO The activation energy (EA) was obtained by measuring the initial rates of FA consumption from 70 to 100 °C (Figure S89, Supporting Information). The estimated apparent EA from the Arrhenius plot is 84 kJ/mol (Figure 3b), within the range of FA dehydrogenation reactions catalyzed by other noble metals.[23] The activation parameters ΔH╪= 81 kJmol�1 and ΔS╪= �134 J Kmol�1 were calculated from the Eyring plot based on the first 60min of FA consumption (Figure S90, Supporting Information). The highly negative ΔS╪ value points toward an asso- ciative rate-determining step, i.e., the interaction of a reaction inter- mediate with a molecule of FA.[24] The double-logarithmic plots of the initial rates of FA dehydrogenation (2.5M, 0.2 mL) with the concentration of C1 (0.025–0.1mol%) illustrate the first-order dependence between 0.025 and 0.075mol% (order of 1.30), followed by saturation kinetics at higher concentrations up to 0.2mol% (Figure S91–S93, Supporting Information).[23] Further, the reaction order of 0.963 for the dehydrogenation of varying con- centrations of FA over 0.2mol% of C1 at 90 °C evidenced instant coordination of a molecule of FA or formate ions to the Ru metal center to afford [HCOO-Ru] species, i.e., no preequilibrium exists between C1 and FA (Figure S94–S96, Supporting Information). The involvement of Ru–hydride species in FA dehydrogena- tion was evidenced by the presence of multiple1H NMR signals in the region δ=�11 to �14.5 ppm, indicative of Ru–hydride species from the reaction of C1 with 1 equivalent of HCOOK in DMSO-d6 at 90 °C (Figure S97, Supporting Information). The presence of several hydride species is likely due to competing equilibria under these reaction conditions. Next, the catalyst-to- formate ratio was increased to catalytically relevant ratios, i.e., 1:50. Under these reaction conditions, an intense 1H NMR signal was observed at δ=�11 ppm as the major Ru-hydride species during the reaction of C1 with HCOOK in DMSO-d6 at 90 °C (Figure S98, Supporting Information). This species was identified as the Ru–monohydride species according to literature prece- dent.[17,25] Further, kinetic isotope effect (KIE) studies were con- ducted to gain deeper insight into the catalytic process of FA dehydrogenation over C1 (Table 2). Triethylamine (NEt3) was employed as the additive to avoid deuterium scrambling. Replacing HCOOH with formic-d acid (DCOOH) had a minimal effect on the reaction rate with KIE= 1.2 (Table 2, entry 2). Employing formic acid-d2 (DCOOD) as the substrate led to a more pronounced decrease in the reaction rate with KIE= 2.1 (Table 2, entry 3). The KIE effect with formic acid-d (HCOOD) was found to be 1.4, which in comparison to DCOOH suggests a more Table 1. Initial screening of iminopyridine Ru(II) complexes C1–C9. Entrya) Catalyst CO2þ H2 [mL] TON TOF [hr�1] Conversion [%] 1 [Ru(p-cymene)Cl2]2 – – – – 2 C1 236 483 281 91 3 C2 64 130 46 25 4 C3 25 51 21 10 5 C4 72 147 54 28 6 C5 217 444 191 84 7 C6 4 7 5 2 8 C7 36 73 27 14 9 C8 117 238 85 45 10 C9 144 294 123 56 a)Conditions: FA (5 mmol, 0.2 mL), C1–C9 (0.2 mol%, 0.01 mmol), HCOONa (1 mmol), DMSO (2 mL), temperature (90 °C), and total reaction volume (2.2 mL). TOF estimated based on gas evolution at 90 mins. TONs and conversion determined at 120 mins. Each reaction was repeated twice with an error of less than 5%. Maximum theoretical volume (CO2 þ H2): 259 mL. Eur. J. Inorg. Chem. 2025, 28, e202500212 (4 of 10) © 2025 The Author(s). European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202500212 10990682c, 2025, 24, D ow nloaded from https://chem istry-europe.onlinelibrary.w iley.com /doi/10.1002/ejic.202500212 by South A frican M edical R esearch, W iley O nline L ibrary on [27/10/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense http://doi.org/10.1002/ejic.202500212 pronounced effect of protonation and/or O─H bond cleavage. The KIE data, in combination with the highly negative entropy of activation and first-order dependence in [FA], suggests that the rate-determining step likely involves the cleavage of the O─H bond and/or the protonation of the Ru─hydride intermedi- ate. The observed trend corroborates well with the literature pre- cedent for reactions catalyzed in DMSO.[13] 2.4. Mechanistic Studies in Water The activation energy (EA) was obtained by measuring the initial rates of FA consumption from 70 to 100 °C (Figure S99, Supporting Information). The estimated apparent EA from the Arrhenius plot is 108 kJ mol�1 (Figure 4b), within the range of FA dehydrogenation reactions catalyzed by noble metals.[23] The obtained EA is �24 kJ mol�1 higher than the EA determined from the reactions performed in DMSO. From the Eyring analyses (Figure S100, Supporting Information), the activation parameters ΔH╪= 105 kJ mol�1 and ΔS╪=�75 J K mol�1 were calculated. As in DMSO, the rate-determining step for the reac- tion performed in H2O is associative as indicated by the negative ΔS╪ value. A linear dependence (reaction order= 0.924) was deduced from the double-logarithmic plot of the initial rates of FA consumption with C1 concentration (0.025–0.1 mol%) and points toward the generation of the monomeric species of C1 as the most active catalytic species and the absence of dimeric species during the catalytic dehydrogenation of FA (Figure S101–S103, Supporting Information). From the double- logarithmic plot of the initial rates of FA consumption with the concentration of FA (0.05–0.2 mL), a positive reaction order of 0.306 was obtained (Figure S104–S106, Supporting Information). Interestingly and in contrast to observations in DMSO, perform- ing the catalytic reaction in water leads to the existence of a pre- equilibrium between the C1 and FA as well as the interaction of FA or formate ions with the Ru metal center. Usually, half-order reaction kinetics with respect to FA concentration indicate equi- molar interaction of the FA substrate with the Ru complex.[18,23] This difference in Ru–FA interactions is also reflected in the more positive ΔS╪ value in water (ΔS╪=�134 J K mol�1 in DMSO) and implies divergent species participating in the hydrogen- formation step. Our kinetic and thermodynamic data suggest that FA is involved in the hydrogen formation step in DMSO,[13] whereas hydronium (H3Oþ) ions are involved when generating H2 in water.[17] KIE studies were conducted to better understand the role of formate activation versus hydrogen evolution on the rate of FA dehydrogenation (Table 3). The observed trend in KIE measure- ments indicated that both d2-FA (DCOOD, KIE 1.9) and HCOOD (KIE 1.9) were more influential than D2O (KIE 1.1) on the reaction rate for FA dehydrogenation over C1 under the reaction condi- tions in Table 3. Crucially, replacing HCOOH with DCOOH led to a more pronounced KIE of 6.0 in H2O. These results, in combination with kinetic and thermodynamic data, infer that the rate- determining step involves C–H bond cleavage during decarbox- ylation of the Ru-formato species to yield the Ru–hydride species, rather than the proton (Hþ)-assisted hydrogen gas evolution.[26,27] The dehydrogenation of FA in the presence of DCOOD and/or D2O may involve a complex combination of reaction kinetics, equilibria, and isotope exchange processes, leading to the slightly low KIE values shown in Table 3. Nonetheless, the obtained data follows literature precedent. For instance, Himeda and coworkers reported a KIE of 1.6 when D2O replaced H2O and a KIE of 2.3 when DCOOD replaced HCOOH.[28] Our analysis of the mechanistic features in DMSO and water allows us to propose Table 2. KIE in the dehydrogenation of FA over catalyst C1 in DMSO. Entrya) Catalyst Substrate Solvent TOF KIE 1 C1 HCOOH DMSO 115 – 2 C1 DCOOH DMSO 100 1.2 3 C1 DCOOD DMSO 54 2.1 4 C1 HCOOD DMSO 85 1.4 a)Conditions: FA (1.33 mmol, 0.05 mL), C10 (0.2 mol%, 0.00266 mmol), NEt3 (1 mmol), DMSO (1 mL), and temperature (80 °C). TOF measured at 30mins. KIE= TOFHCOOH/TOFisotopologue). (a) (b) Figure 3. a) Influence of reaction temperature on the activity of C1 in FA dehydrogenation. Conditions: FA (5 mmol, 0.2 mL), C1 (0.2 mol%, 0.01 mmol), HCOOK (3 mmol), DMSO (2 mL), temperature (as indicated), and total reaction volume (2.2 mL). b) Arrhenius plot based on FA consumption in the first 60 min of the reaction using C1. Eur. J. Inorg. Chem. 2025, 28, e202500212 (5 of 10) © 2025 The Author(s). European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202500212 10990682c, 2025, 24, D ow nloaded from https://chem istry-europe.onlinelibrary.w iley.com /doi/10.1002/ejic.202500212 by South A frican M edical R esearch, W iley O nline L ibrary on [27/10/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense http://doi.org/10.1002/ejic.202500212 divergent hydrogen production pathways. In DMSO, hydrogen production proceeds with FA as the hydrogen source, which is preceded by Ru-hydride formation from the Ru-formato, with the rate-determining step likely being O─H bond cleavage during the protonation step [Scheme 2, pathway a)]. In contrast, in water, hydrogen production proceeds with hydronium ions as the hydrogen source, with decarboxylation to generate Ru-hydride as the rate-determining step [Scheme 2, pathway b)]. Both path- ways share Ru-hydride as the resting state. 2.5. Optimized Catalytic Performance To determine the reusability and long-term stability of C1 in DMSO, fresh batches of 0.1 mL FA were added to the reaction solution in 30-minute intervals, i.e., time taken to consume the initial amount of FA added into the system. Under the optimized reaction conditions shown in Figure 5, the total volume of gas generated decreased with each FA refill (Figure 5a). During the reaction, TOFs of about 1636 hr�1 were attained at the end of the initial addition and halved to 837 hr�1 at the end of the sixth refill, with estimated TONs of 3000 (Figure S107, Supporting Information). This drastic reduction in activity can be attributed to possible catalyst deactivation due to pH changes associated with the consumption of the substrate and further drastic pH changes during FA refills. To further explore the robustness of catalyst C1, FA was continually introduced using an automatic syringe pump at the rate of its consumption, i.e., 0.003 mL min�1. The catalytic rate decreased with time, with TOFs of 540 hr�1 reached in the first hour of the reaction decreasing to 199 hr�1 after 5 h. Figure 5b indicates the plateau in the generated TONs as a function of time with maximum TONs of 1009 after 5 h under the optimized reaction conditions. These results further reinforce the pH sensitivity of C1 when DMSO is employed as a solvent medium. Similarly, the reusability and long-term stability of C1 for FA dehydrogenation in water were evaluated. C1 was subjected to repeated 0.1 mL FA injections every 80 min under the reaction conditions shown in Figure 6. To our delight, C1 could dehydro- genate 0.1 mL batches of FA for seven cycles without losing activ- ity, achieving TONs= 7872 at the end of the seventh cycle. TOFs of 682 hr�1 were obtained after complete dehydrogenation of the initial FA addition and amounted to 844 hr�1 at the end of the seventh refill of FA, illustrating the retained activity of C1 with each FA injection. This sustained catalytic activity can be attrib- uted to the ability of C1 to withstand rigorous pH changes asso- ciated with the addition and/or consumption of the FA substrate in a water solvent medium. Under these reaction conditions, sustained activity with TONs 13 791 could be obtained within 21 h, before deactivation was observed. Further, the suitability of C1 for long-term applications was evaluated by injecting FA at a constant rate (0.00125mLmin�1) equal to its consumption in the reaction. Under the optimized reaction conditions shown in Figure S108, Supporting Information, catalyst C1 was able, although with diminished catalytic activity (TOF= 43 hr�1), to cat- alyze the dehydrogenation of FA for over 34 h. These results fur- ther reinforce the increased stability of the catalytic species of C1 in water. The decreased catalytic activity under these conditions is likely due to the reaction not proceeding at the desired pH range, i.e., highly basic conditions due to the excess HCOOK in the reaction solution. (a) (b) Figure 4. a) Influence of reaction temperature on the activity of C1 in FA dehydrogenation. Conditions: FA (5 mmol, 0.2 mL), C1 (0.2 mol%, 0.01 mmol), HCOOK (3 mmol), H2O (2 mL), temperature (as indicated), and total reaction volume (2.2 mL). b) Arrhenius plot based on FA consumption in the first 60 min of the reaction using C1. Table 3. KIE in the dehydrogenation of FA over catalyst C1 in water. Entrya) Catalyst Substrate Solvent TOF KIE 1 C1 HCOOH H2O 375 – 2 C1 HCOOH D2O 329 1.1 3 C1 DCOOD H2O 231 1.6 4 C1 DCOOD D2O 202 1.9 5 C1 HCOOD H2O 196 1.9 6 C1 HCOOD D2O 110 3.4 7 C1 DCOOH H2O 63 6.0 8 C1 DCOOH D2O 98 3.8 a)Conditions: substrate (1.33 mmol, 0.05 mL), C10 (0.2 mol%, 0.00266 mmol), NEt3 (1 mmol), solvent (1 mL), and temperature (80 °C). TOF measured at 40 min. KIE = TOFHCOOH/TOFisotopologue). Eur. J. Inorg. Chem. 2025, 28, e202500212 (6 of 10) © 2025 The Author(s). European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202500212 10990682c, 2025, 24, D ow nloaded from https://chem istry-europe.onlinelibrary.w iley.com /doi/10.1002/ejic.202500212 by South A frican M edical R esearch, W iley O nline L ibrary on [27/10/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense http://doi.org/10.1002/ejic.202500212 2.6. Comparative Analysis with Reported Catalytic Systems In this section, the performance of Ru catalyst C1 toward FA dehy- drogenation is compared with reported ruthenium complexes operated under similar reaction conditions. Table S3 and Figure S109, Supporting Information, summarize the catalytic performance of the existing literature work and compare it to the work presented in this study. As described previously, com- pounds I–IV (Figure 1) represent benchmark catalysts, which out- perform most catalysts, based on both earth-abundant and platinum group metals (PGMs). In addition, Huang and coworkers reported a highly efficient Ru complex VIII that achieved maxi- mum TONs of 1 100 000 in DMSO after 150 h, attributable to enhanced stabilization by the tridentate noninnocent PNP ligand framework (Table S3, Supporting Information, entry 1).[29] The best-performing Ru-catalyst in water was reported by Huang and co-workers, V, reaching TONs of 350 000 and TOFs of 12 000 hr�1 (Figure 1,V).[17] Catalyst C1 was found to outperform several reported Ru- and other metal-containing systems. For instance, Singh and coworkers reported a Ru complex, VI, which attained TONs of 6050 and TOFs of 1548 in water (Figure 1, VI).[18] In another report, Singh and coworkers illustrated a Ru catalyst system, IX, that reached TONs of 2248 in 15min with reasonably high TOFs of 940 hr�1 (Table S3, Supporting Information, entry 2).[30] C1 outperformed our recently reported Ru-catalysts bearing 1) pyridyl-formamidine ligands, X, and 2) pyridyl- pyrazolyl ligands, XI, which in the presence of formate could dehydrogenate FA with TONs and TOFs of 7892 and 225 hr�1 Scheme 2. Divergent hydrogen production pathways, for the dehydrogenation of FA using C1 in a) DMSO and b) water. Eur. J. Inorg. Chem. 2025, 28, e202500212 (7 of 10) © 2025 The Author(s). European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202500212 10990682c, 2025, 24, D ow nloaded from https://chem istry-europe.onlinelibrary.w iley.com /doi/10.1002/ejic.202500212 by South A frican M edical R esearch, W iley O nline L ibrary on [27/10/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense http://doi.org/10.1002/ejic.202500212 and 3169 and 336 hr�1, respectively (Table S3, Supporting Information, entries 3 and 4).[31,32] The better performance of C1may be attributable to both its faster activation and enhanced stability under operating conditions. A π-coordinated phenoxy- ligated Ru-dimer, XII, was active for FA dehydrogenation at 90 °C, generating TONs of 173 and TOFinitial of 202 h�1 in DMSO (Table S3, Supporting Information, entry 5).[33] The lower performance of XII may be ascribed to its sensitivity to water, leading to catalyst inhibition. The [Ru(triphos)(MeCN)3](OTf )2 system, XIII, was reported to achieve full conversion in an hour, at low catalyst loadings with N,N-dimethyloctanamine as addi- tive, achieving TONs and TOFs of 10 000 and 1000 h�1, respec- tively (Table S3, Supporting Information, entry 6), with slower activation kinetics translating into TONs and TOFs of 1000 and 100 h�1 after 180min.[34] The PCP-Ni(II)-hydride, XIV, dehydrogen- ated FA in the presence of triethylamine at 80 °C, with TONs of 626 (Table S3, Supporting Information, entry 7) and was found to generate product gas with higher CO2 content than expected, attributed to propylene carbonate cleavage as a side reaction.[35] A related (PNP)Ir(III)-trihydride complex, XV, catalyzed the dehy- drogenation of FA in acidic water, with pH-dependent activity. The best activity displayed a TOFmax of 1880 h�1 at pH 2.8, with catalyst deactivation observed at lower pH (Table S3, Supporting Information, entry 8).[36] Compound XVI, a P4-tetradentate tetraphos-ligated Fe(II) tetrafluoroborate complex, dehydrogen- ated FA under base-free conditions in propylene carbonate at 60 °C (Table S3, Supporting Information, entry 9).[37] TONs of 6061 were obtained after 6 h of reaction, with TOFinitial of 1737 h�1. Finally, the [BrMn(L)(CO)3] complex, (L= 2-(4,5-dihydro-1H- imidazol-2-yl)pyridine), XVII, reported by Beller dehydrogenated FA under continuous-flow conditions, with TON of 5476 after 45 h and TOF up to 192 h�1 after 3 h.[38] Our catalyst system pro- vides efficient dehydrogenation activity, combined with ease-of- synthesis, modularity in terms of electronic and steric properties, as well as sustained activity in water. These are distinguishing fea- tures which can be exploited in future catalyst design strategies. (a) (b) Figure 5. a) Stacked plots of the gas evolution upon repetitive FA addition with t= 0mins for each addition. b) Influence of continuous FA injections on the rate of FA dehydrogenation. Conditions: FA (2.5 mmol, 0.1 mL), C1 (0.1 mol%, 0.0025 mmol), HCOOK (3 mmol), DMSO (2 mL), temperature (90 °C), and total reaction volume (2.1 mL). (a) (b) Figure 6. a) Stacked plots of the gas evolution upon repetitive FA addition with t= 0mins for each addition. b) Influence of repetitive FA injections on the rate of FA dehydrogenation. Conditions: FA (2.5 mmol, 0.1 mL), C1 (0.1 mol%, 0.0025 mmol), HCOOK (3 mmol), H2O (2 mL), temperature (90 °C), and total reaction volume (2.1 mL). Eur. J. Inorg. Chem. 2025, 28, e202500212 (8 of 10) © 2025 The Author(s). European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202500212 10990682c, 2025, 24, D ow nloaded from https://chem istry-europe.onlinelibrary.w iley.com /doi/10.1002/ejic.202500212 by South A frican M edical R esearch, W iley O nline L ibrary on [27/10/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense http://doi.org/10.1002/ejic.202500212 3. Conclusion A series of iminopyridine ruthenium complexes, C1–C11, were evaluated for FA dehydrogenation, where the catalytic activity was closely linked to the nature of the substituent on the imine nitrogen of the ligands. Catalyst C1 was found to be most effec- tive, reaching TONs of 483 after 2 h. The activity of C1 toward FA dehydrogenation was solvent-dependent with EA= 81 kJ mol�1 in DMSO and EA= 108 kJ mol�1 in water. The main Ru-hydride species responsible for FA dehydrogenation was observed at δ=�11 ppm using1H NMR experiments. KIE experiments identi- fied the protonation of Ru-hydride species and the decarboxyl- ation of Ru-formato species as the turnover-limiting steps in DMSO and water, respectively. The hydrogen source was found to be solvent dependent, with FA involved in the hydrogen for- mation step in DMSO, whereas hydronium (H3Oþ) ions are involved when generating H2 in water. Complex C1 was observed to be more stable in water than DMSO, achieving TONs of 13 791 with loss of activity commencing after 16 cycles. Under continuous FA addition, C1 was active, although with reduced performance (TOF= 43 hr�1), for over 34 h in water. 4. Experimental Section General Considerations All reactions involving air- and/or moisture-sensitive compounds were performed under purified dry nitrogen using normal Schlenk techni- ques. All reagents were obtained from Sigma-Aldrich and used as received. Solvents were purchased from Promark Chemicals and freshly dried before use. IR spectra were recorded as neat oils or solids on a Bruker Tensor 27 FT-IR spectrometer fitted with an attenuated total reflection (ATR) accessory. NMR spectra were recorded on a Bruker Avance NMR spectrometer (1H at 400MHz,13C at 101MHz). ESI-MS anal- yses were recorded on a Thermo Scientific Dionex Ultimate 3000 UHPLC coupled to a Bruker Compact QqTOF HRMS. Elemental analysis was recorded using a Thermo Scientific Flash 2000 Series CHNS-O Analyzer, at the University of Johannesburg, as a service. All the ligands, except for L9, have previously been reported and were prepared according to the reported literature methods, as described in the Supplementary Information. Representative synthesis of C1 is provided below. The remainder is described in the Supplementary Information. General Considerations: [(L1)RuCl(p-Cymene)]Cl, C1 Ligand L1 (47.8 mg, 0.26 mmol) was added to a stirred solution of [RuCl2(p-cymene)]2 (80.4 mg, 0.13 mmol) in MeOH (10mL), and the resulting solution was reacted for 24 h at room temperature under a N2 atmosphere. After the reaction, the solution was concentrated to 1mL, redissolved in DCM (10mL), and layered with excess diethyl ether to afford the product as a yellow crystalline solid (145.1mg, 87%). 1H NMR (400 MHz, DMSO-d6): δ (in ppm)= 9.69 (d, J= 5.7 Hz, 1H), 9.08 (s, 1H), 8.37–8.32 (m, 2H), 7.88–7.81 (m, 3H), 7.63 (m, 3H), 6.12 (d, J= 6.2 Hz, 1H), 5.82 (d, J= 6.2 Hz, 1H), 5.64 (m, 2H), 2.15 (s, 3H), 0.98 (d, J= 7.1 Hz, 6H).13C NMR (101 MHz, DMSO-d6): δ (in ppm)= 167.99, 156.19, 154.58, 151.71, 139.88, 130.19, 129.72, 129.50, 128.91, 122.53, 105.05, 103.31, 86.59, 86.10, 85.11, 84.99, 30.45, 21.69, 21.60, 18.27. FT-IR (ATR, neat): 1612.64 cm�1 (νC=N), 1587.62 cm�1 (νpyr-C=N). HRMS (ESI) m/z calcd for C22H24ClN2Ruþ: 453.00 [M]þ; found: 453.0669. Anal. calcd. for C22H24Cl2N2Ru: C 54.10; H 4.95; N 5.74. Found: C 54.03; H 5.16; N 5.63. General Considerations: Procedure for Catalytic FA Dehydrogenation A stock solution of the complex in DMSO or H2O was prepared. From the stock solution, the desired quantity of complex was added to a Schlenk flask followed by the addition of HCOOM (where M= Li, Na, K). The flask was sealed with a rubber septum, and the atmo- sphere was exchanged for N2. The reaction solution was stirred at 800 rpm for 5 min at room temperature. Then, FA was added, and the mixture was stirred for an additional minute. Lastly, the reaction mixture was heated at the desired temperature using a preheated oil bath and the gas evolution was monitored using an inverted gradu- ated measuring cylinder. For the continuous FA addition experiment, a similar approach was used, and an NE-300 Just Infusion syringe pump was used to inject FA from a Hamilton syringe fitted with a long metallic needle directly after the immersion of the reaction flask in the oil bath. X-Ray Crystal Structure Determination Single-crystal X-ray diffraction data were collected on a Bruker APEX-II CCD diffractometer. The crystal was kept at 173.00 K during data col- lection. Using Olex2,[39] the structure was solved with the SHELXT[40] structure solution program using intrinsic phasing and refined with the SHELXL[41] refinement package using least-squares minimization. All nonhydrogen atoms were refined anisotropically, while hydrogen atoms were placed in calculated positions using riding models. High- resolution molecular diagrams were generated with POV-ray.[42] Deposition Numbers 2 423 518 (for C1) and 2 423 519 (for C2), contain the supplementary crystallographic data for this paper. These data are provided free of charge by the joint Cambridge Crystallographic Data Centre and Fachinformationszentrum Karlsruhe Access Structures service. Acknowledgements The authors gratefully thank the School of Chemistry, University of the Witwatersrand, for infrastructure and equipment support, Prof. Manuel Fernandes (Wits University) for his assistance with SCXRD analyses, and Sasol South Africa for its financial support to TIM and RM. A.J.S acknowledges funding support of the National Research Foundation (ZA), with grant numbers 137758, 127291, and CPRR240312208672. Conflict of Interest The authors declare no conflict of interest. Author Contributions Thabiso Mphuti: investigation (lead); methodology (lead); writing—original draft (lead). Rotondwa Mphephu: investigation (supporting); methodology (supporting); writing—original draft (supporting). Moegamat Joseph: formal analysis (supporting); writing—review and editing (supporting). Andrew J. Swarts: funding acquisition (lead); project administration (lead); resources (lead); supervision (lead); writing—review and editing (lead). Eur. J. Inorg. Chem. 2025, 28, e202500212 (9 of 10) © 2025 The Author(s). European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202500212 10990682c, 2025, 24, D ow nloaded from https://chem istry-europe.onlinelibrary.w iley.com /doi/10.1002/ejic.202500212 by South A frican M edical R esearch, W iley O nline L ibrary on [27/10/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://www.ccdc.cam.ac.UK/services/structures?id=doi:10.1002/ejic.202500212 http://www.ccdc.cam.ac.UK/structures http://www.ccdc.cam.ac.UK/structures http://doi.org/10.1002/ejic.202500212 Data Availability Statement The data that support the findings of this study are available in the supplementary material of this article. Keywords: dehydrogenation · hydrogen · iminopyridines · ruthenium [1] M. R. Usman, Renew. Sustain. Energy Rev. 2022, 167, 112743. [2] P. M. Modisha, C. N. M. Ouma, R. Garidzirai, P. Wasserscheid, D. Bessarabov, Energy Fuels. 2019, 33, 2778. [3] D. Mellmann, P. Sponholz, H. Junge, M. Beller, Chem. Soc. Rev. 2016, 45, 3954. [4] R. S. Coffey, Chem. Commun. London 1967, 923. [5] H. Jeon, Y.-H. Chung, Appl. Catal. B 2017, 210, 212. [6] J. L. Santos, C. Megías-Sayago, S. Ivanova, M. A. Centew, J. A. Odriozola, Chem. Eng. J. 2021, 420, 127641. [7] S. Kar, M. Rauch, G. Leitus, Y. Ben-David, D. Milstein, Nat. Catal. 2021, 4, 193. [8] N. H. Anderson, J. Boncella, A. M. Tondreau, Chem. Eur. J. 2019, 25, 10557. [9] B. Loges, A. Boddien, H. Junge, M. Beller, Angew. Chem., Int. Ed. 2008, 47, 3962. [10] C. Fellay, J. Dyson, G. Laurenczy, Angew. Chem. Int. Ed. 2008, 47, 3966. [11] E. A. Bielinski, P. O. Lagaditis, Y. Zhang, B. Q. Mercado, C. Würtele, W. H. Bernskoetter, N. Hazari, S. Schneider, J. Am. Chem. Soc. 2014, 136, 10234. [12] Z. Wang, S.-M. Lu, J. Li, J. Wang, C. Li, Chem. - Eur. J. 2015, 21, 12592. [13] N. Lentz, M. Albrecht, ACS Catal. 2022, 12, 12627. [14] K. Sordakis, C. Tang, L. K. Vogt, H. Junge, P. J. Dyson, M. Beller, G. Laurenczy, Chem. Rev. 2018, 118, 372. [15] A. Wang, P. He, J. Wu, N. Chen, C. Pan, E. Shi, H. Jia, T. Hu, K. He, Q. Cai, R. Shen, Energy Fuels 2023, 37, 17075. [16] I. Dutta, S. Chatterjee, H. Cheng, R. K. Parsapur, Z. Liu, Z. Li, E. Ye, H. Kawanami, J. S. E. Low, Z. Lai, X. J. Loh, K.-W. Huang, Adv. Energy Mat. 2022, 12, 2103799. [17] C. Guan, D.-D. Zhang, Y. Pan, M. Iguchi, M. J. Ajitha, J. Hu, H. Li, C. Yao, M.- H. Huang, S. Min, J. Zheng, Y. Himeda, H. Kawanami, K.-W. Huang, Inorg. Chem. 2017, 56, 438. [18] S. Patra, S. K. Singh, Inorg. Chem. 2020, 59, 4234. [19] S. Ge, J.-H. Liu, L. Gong, X.-F. Mo, C. Liu, X.-Y. Yi, P. He, Eur. J. Inorg. Chem. 2024, 27, e202300509. [20] C. Binnani, R. K. Rai, D. Tyagi, S. M. Mobin, S. K. Singh, Eur. J. Inorg. Chem. 2018, 2018, 1435. [21] P. A. Scattergood, U. Khushnood, A. Tarig, D. J. Cooke, C. R. Rice, P. I. P. Elliott, Inorg. Chem. 2016, 55, 7787. [22] G. R. Fulmer, A. J. M. Miller, N. H. Sherden, H. E. Gottlieb, A. Nudelman, B. M. Stoltz, J. E. Bercaw, K. I. Goldberg, Organometallics 2010, 29, 2176. [23] A. Boddien, D. Mellmann, F. Gärtner, R. Jackstell, H. Junge, P. J. Dyson, G. Laurenczy, R. Ludwig, M. Beller, Science 2011, 333, 1733. [24] S. Upadhyay, Chemical Kinetics and Reaction Dynamics, Springer, Netherlands 2006, pp. 46–54. [25] C. Fellay, N. Yan, P. J. Dyson, G. Laurenczy, Chem. Eur. J. 2009, 15, 3752. [26] Z. J. Wang, S. M. Lu, J. Li, J. J. Wang, C. Li, Chem. Eur. J. 2015, 21, 12592. [27] W. H. Wang, J. F. Hull, J. T. Muckerman, E. Fujita, T. Hirose, Y. Himeda, Chem. Eur. J. 2012, 18, 9397. [28] W. H. Wang, S. Xu, Y. Manaka, Y. Suna, H. Kambayashi, J. T. Muckerman, E. Fujita, Y. Himeda, ChemSusChem 2014, 7, 1976. [29] Y. Pan, C. L. Pan, Y. Zhang, H. Li, S. Min, X. Guo, B. Zheng, H. Chen, A. Anders, Z. Lai, J. Zheng, K.-W. Huang, Chem. Asian J. 2016, 11, 1357. [30] S. Patra, M. K. Awasthi, R. K. Rai, H. Deka, S. M. Mobin, S. K. Singh, Eur. J. Inorg. Chem. 2019, 2019, 1046. [31] J. M. Edor, M. C. Joseph, J. H. L. Jordaan, H. C. M. Vosloo, A. J. Swarts, Appl. Organomet. Chem. 2025, 39, e70016. [32] R. Mphephu, M. C. Joseph, A. J. Swarts, Eur. J. Inorg. Chem. 2025, 28, e202500063. [33] R. Verron, E. Puig, P. Sutra, A. Igau, C. Fischmeister, ACS Catal. 2023, 13, 5787. [34] I. Mellone, M. Peruzzini, L. Rossi, D. Mellmann, H. Junge, M. Beller, L. Gonsalvi, Dalton Trans. 2013, 42, 2495. [35] S. Enthaler, A. Brück, A. Kammer, H. Junge, E. Irran, S. Gülak, ChemCatChem 2015, 7, 65. [36] R. Tanaka, M. Yamashita, K. Nozaki, J. Am. Chem. Soc. 2009, 131, 14168. [37] F. Bertini, I. Mellone, A. Ienco, M. Peruzzini, L. Gonsalvi, ACS Catal. 2015, 5, 1254. [38] A. Léval, A. Agapova, C. Steinlechner, E. Alberico, H. Junge, M. Beller, Green Chem. 2020, 22, 913. [39] O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard, H. Puschmann, J. Appl. Cryst. 2009, 42, 339. [40] G. M. Sheldrick, Acta Cryst. 2015, A71, 3. [41] G. M. Sheldrick, Acta Cryst. 2015, C71, 3. [42] POV-Ray, Version 3.7, Williamstown, Australia, Persistence of vision ray. Manuscript received: April 21, 2025 Revised manuscript received: June 12, 2025 Version of record online: June 18, 2025 Eur. J. Inorg. Chem. 2025, 28, e202500212 (10 of 10) © 2025 The Author(s). European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Research Article doi.org/10.1002/ejic.202500212 10990682c, 2025, 24, D ow nloaded from https://chem istry-europe.onlinelibrary.w iley.com /doi/10.1002/ejic.202500212 by South A frican M edical R esearch, W iley O nline L ibrary on [27/10/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense http://doi.org/10.1002/ejic.202500212 Unveiling Solvent-Dependent Divergent Hydrogen Production Pathways during the Dehydrogenation of Formic Acid Using N,N′-Iminopyridine Ruthenium(II) Complexes 1. Introduction 2. Results and Discussion 2.1. Synthesis and Characterization 2.2. FA Dehydrogenation 2.3. Mechanistic Studies in DMSO 2.4. Mechanistic Studies in Water 2.5. Optimized Catalytic Performance 2.6. Comparative Analysis with Reported Catalytic Systems 3. Conclusion 4. Experimental Section Outline placeholder General Considerations General Considerations: [(L1)RuCl(p-Cymene)]Cl, C1 General Considerations: Procedure for Catalytic FA Dehydrogenation X-Ray Crystal Structure Determination