Dual Activation of H2 and CO2 by a Pincer‐Type Ni–Zn Heterobimetallic Complex Authors: Krishnendu Dey, Matthias Borowski, Gregor Schnakenburg, Connie C Lu Published: 2026 ABSTRACT ABSTRACTVarious Ni‐based metalloenzymes employ synergistic heterometallic cooperativity where a Ni center and a second metal ion cofacilitate small‐molecule conversion. Inspired by these biological systems, synthetic model complexes have been developed that exploit bimetallic cooperativity to achieve analogous reactivity. In this work, we report a bimetallic Ni−Zn complex comprising a pincer‐type diamido‐diaminodiphosphine ligand, in which a weakly Lewis acidic Zn(II) center stabilizes a T‐shaped Ni(0) site. Subsequent reactivity with H2 led to the isolation of a rare Ni dihydride complex, which dynamically interconverts between Zn(μ‐H)Ni(H) and trans‐Ni(H)2 motifs via rotational rocking about the P−Ni−P axis. The Ni dihydride complex is a competent alkene hydrogenation catalyst (5 mol% loading, 70°C, 2 bar H2) for terminal olefins. Also, the Ni dihydride inserts CO2 to give a rare Ni hydrido formate. This product further reacts with a second equivalent of CO2 through a carboxylate shift mechanism that unveils a nucleophilic Zn(II) amide group that inserts CO2 to give carbamate. Collectively, our experimental and computational results highlight the role of Zn(II) as a weakly Lewis acidic partner, which promotes the dual activation of H2 and CO2 by a single Ni center in the absence of any external base.Keywords: CO2 hydrogenation, dihydride, heterobimetallic, Lewis acid, nickel, zinc Introduction Several nickel‐containing metalloenzymes, such as [NiFe]‐hydrogenase, [NiFe]‐carbon monoxide dehydrogenase (CODH), and acetyl coenzyme A synthase, feature active sites with a redox‐active Ni center in close proximity to a second metal center (Figure 1a) [1, 2, 3, 4, 5]. Motivated to understand and to mimic the activity of these metalloenzymes, many Ni‐based heterobimetallic systems have been developed to exploit bimetallic cooperativity in the activation of H2 or CO2 [6, 7, 8, 9, 10, 11, 12, 13, 14]. H2 evolution from weak acids at low overpotentials and the reverse reaction, H2 oxidation, have been successfully catalyzed with Ni–Fe synthetic model complexes (Figure 1b, left) [8, 11]. This cooperativity concept has also been extended to Ni‐main group and Ni‐lanthanide catalysts that exploit direct metal‐metal cooperativity to activate H2 for olefin and alkyne (semi)hydrogenation [15, 16, 17, 18, 19, 20, 21, 22, 23]. (a) Enzymatic active site of [NiFe]‐hydrogenase [1, 2] and [NiFe]‐carbon monoxide dehydrogenase (CODH) [3]. (b) Catalytic model systems for H+ and CO2 reduction [10, 11]. (c) Model complexes for stoichiometric CO2 reduction [12, 14]. (d) Ni−Zn heterobimetallic complex for dual activation of H2 and CO2 (this work). Moving to CO2, only one heterobimetallic Ni‐based catalyst system for CO2 reduction has been reported to date [10, 24, 25, 26]. Namely, the NiFe catalyst (Figure 1b, right) in combination with Ru(bpy)3 as a photosensitizer generates CO in high selectivity under irradiation [10]. Another Ni–Ga catalyst was shown to hydrogenate CO2 to formate, but a high pressure of a H2/CO2 gas mixture and a strong stoichiometric base were needed for high turnovers [27]. Of biological relevance, two NiFe heterobimetallic systems have demonstrated stoichiometric CO2 reduction to CO via bridging carboxylate intermediates that are akin to one observed for the C‐cluster in [NiFe]‐CODH [28] (Figure 1c) [12, 14]. Considering the widespread role of Zn(II) as a Lewis acid in enzymes, we began to investigate the Ni‐Zn pairing (Figure 1d). While Ni–Zn(II) heterobimetallic complexes are already known, the reported reactivity has been largely limited to H2 activation [29, 30, 31, 32, 33]. Using a new ligand framework, [((o‐NCH2PtBu2‐C6H4)NMeCH2)2]2− (abbreviated as L 2−), a Ni–Zn complex was isolated featuring a T‐shaped Ni(0) center bound to Zn(II). The reactivity of the Ni–Zn system with small molecules (H2, CO, and CO2) was investigated, and the products were characterized by X‐ray crystallography, variable‐temperature NMR spectroscopy, and IR spectroscopy. Density functional theory (DFT) calculations were also performed to provide further insight into the reaction mechanisms. A highlight of this work is the formation of a Ni hydrido formate complex from the dual activation of H2 and CO2 without any external base. While H2 activation by heterobimetallic Ni systems has been established, the subsequent reactivity of the resulting Ni hydrides is largely limited to olefin hydrogenation, with no reports of base‐free CO2/carbonyl reactivity thus far. Indeed, among the late first‐row transition metals, well‐defined metal dihydrides that can directly insert CO2 are only reported for Fe [34, 35, 36, 37, 38, 39, 40, 41, 42]. After formate formation, another CO2 equivalent is inserted into a Zn−Namide bond. The DFT‐calculated mechanism reveals that this insertion is triggered by a change in the binding mode of the formate ligand. The latter is reminiscent of the carboxylate‐shift mechanism in metalloenzymatic active sites. The Zn(II) co‐active site plays an important functional role and enables the unique reactivity of this Ni–Zn system. Results and Discussion Synthesis and Characterization of Ni–Zn Bimetallic Compound Entry into the Ni–Zn heterobimetallic system began with the salt metathesis of the lithiated ligand Li2 L (1) and ZnCl2 to obtain the neutral Zn metalloligand 2 as white powder in good yield (78%, see Supporting Information Experimental Section). Consistent with C 2 symmetry in solution, the NMR spectra of 2 displayed a single 31P{1H} NMR signal (28.8 ppm) and two sets of diastereotopic methylene protons, one for CH 2PtBu2 and one for the ethylene backbone of the ligand (Figures S28, S29). Next, 2 was reacted with Ni(COD)2, where COD is 1,5‐cyclooctadiene, to yield the bimetallic Ni–Zn complex (3), which was isolated as a dark‐red solid (65% yield, Figure 2a). Like 2, complex 3 displays a C 2 symmetric solution‐state structure, as evidenced by the single 31P{1H} NMR signal (61 ppm) and the two sets of diastereotopic methylene protons (Figures S35, S36). The UV‐Vis spectrum of 3 shows three overlapping bands in the visible region from 440 to 560 nm (Figure S93). (a) Synthesis of the Ni–Zn bimetallic complex 3. (b) Molecular structure of 3 shown at 50% thermal ellipsoid probability. One of two independent molecules is shown. H atoms have been omitted for clarity. The average Ni−Zn bond length (Å) is shown in red. (c) QTAIM molecular graphs of 3 showing key bond critical points (bcps, in red) and ring critical points (rcps, in orange). Single crystals of 3 were grown from vapor diffusion of hexane into a concentrated benzene solution [43]. The solid‐state structure of 3 (Figure 2b, see Supporting Information X‐ray Crystallographic Section, Table S1) revealed a T‐shaped Ni(0) geometry (∠P–Ni–P = 176.92(4)°). Only a handful of T‐shaped Ni(0) examples are known, but so far, they all bear one tetrylene‐based σ‐acceptor in combination with phosphine and/or N‐heterocyclic carbene donors [20, 44, 45, 46]. The geometry around Zn in 3 is distorted trigonal bipyramidal with a τ5 value of 0.6 [47, 48]. Of interest, the Ni–Zn bond distance of 2.6145(6) Å may signify a weak bonding interaction as it is slightly longer than the sum of the two metals’ covalent radii at 2.46 Å (Figure 2b) [49]. A quantum theory of atoms in molecules (QTAIM) analysis was conducted on 3 to evaluate whether the Ni and Zn are in a direct bonding interaction [50, 51]. One bond critical point (bcp) was located between Ni and Zn, with significant electron density, ρ(r) = 0.038 e·Bohr−3 and a positive Laplacian (∇2ρ) of 0.046 e·Bohr−5 that indicates a polar Ni–Zn bond (Figure 2c) [52]. A frontier molecular orbital (MO) analysis of 3 essentially shows five doubly‐occupied Ni 3d orbitals, which is consistent with Ni(0) (Figure S109). Dihydrogen Activation Addition of 1.5 bar of H2 to 3 at room temperature (rt) in THF resulted in a color change from dark red to light orange in 12 h and a shift of the 31P signal to 95.3 ppm. Of note, the product 4, which was isolated in 86% yield, is stable to vacuum at rt and does not revert back to 3 (Scheme 1). The 1H NMR spectrum of 4 at rt showed a characteristic triplet at −10.19 ppm (J P−H = 40.8 Hz), that integrated to two protons (Figure S43). Such a peak is consistent with the formation of a terminal dihydride species, LZnNi(H)2. To verify the presence of dihydride ligands, we carried out T 1 measurements between 193 and 303 K (Figure S96). The T 1(min) relaxation time of 523 ms at 233 K is fully consistent with dihydride ligands and excludes the possibility of an intact H2 ligand (Table S4) [53]. Reaction of Ni–Zn 3 with H2. Exposure of 3 to D2 in THF solution generated 4−D in 12 h, which was subsequently isolated as an orange solid powder in 90% yield. Characteristic 2H resonances for the Ni dideuteride complex were observed at −10 ppm as a triplet (J P−D = 6.6 Hz, Figure S50). The ATR‐IR spectrum of 4 showed a prominent absorption at 1642 cm−1 that corresponds to the Ni–H stretch, which for 4–D shifts to 1182 cm−1 (predicted νNi−D = 1171 cm−1 using Hooke's law, Figures S85−S88). A single crystal of 4, grown from vapor diffusion of hexane into a concentrated THF solution, was investigated by X‐ray diffraction (Table S2). The Ni−Zn distance of 2.6690(6) Å in 4 is elongated by 0.055 Å compared to that in 3 and indicates the absence of any metal‐metal interaction (Figure 3a). In support, QTAIM predicted no bcps between Ni and Zn (Figure 3b). Both hydride atoms were successfully located in the electron difference map, and the refined structure is consistent with LZn(μ‐H)Ni(H), where one hydride is terminal at Ni and the second hydride bridges between Ni and Zn. The HNi(μ‐H)Zn connectivity was also predicted by DFT (see below), and the natural population analysis (NPA) charges on the H atoms (−0.38, −0.24) further support their hydridic nature (Figure 3c). Moreover, bcps were located along the Ni−Hterminal (ρ(r) = 0.13 e·Bohr−3), Ni−Hbridging (0.11 e·Bohr−3), and Zn−Hbridging (0.06 e·Bohr−3) bonds. (a) Molecular structure of 4 shown at 50% thermal ellipsoid probability except for H atoms. H atoms of the pincer‐type ligand have been omitted for clarity. (b) QTAIM molecular graphs of 4 showing key bcps (in red) and rcps (in orange). (c) Charges obtained by natural population analysis (NPA) for the first coordination shell of 4. The seeming discrepancy between the solution‐ and solid‐state structures was resolved by variable‐temperature NMR studies (see below), which demonstrate that 4 is fluxional and can interconvert between trans‐Ni(H)2 and Zn(μ‐H)Ni(H). The latter motif had been previously proposed in other Ni−E heterobimetallic systems, [MesDPBPh]Ni and (MesPAlP)Ni, which also react with H2 to form dihydride species featuring the HNi(μ‐H)E core [15, 16, 54]. However, these literature samples, which are generated in situ under an H2 atmosphere, release H2 readily under vacuum, and hence, their solid‐state structures have not been reported. Complex 4 is distinct in that it is stable to H2 loss, and to the best of our knowledge, this is the first structural characterization of the HNi(μ‐H)E core that arises from H2 activation. Of note, 4 is a competent hydrogenation catalyst for terminal olefins such as styrene and its para‐substituted derivatives, as well as 1‐hexene and 1‐octene (5 mol% loading, 70°C, 2 bar H2; Table S7). In comparison to other base‐metal hydrogenation catalysts, both the activity and the substrate scope of 4 are limited [55, 56]. The trans‐Ni(H)2 structural motif with two terminal hydrides at Ni, which is accessible in solution, has not been previously invoked to the best of our knowledge. A search of the Cambridge Structural Database (CSD) [57] of Ni(H)2 yielded no hits for the two terminal hydrides, with the majority of examples bearing two bridging hydrides, Ni(μ‐H)2Ni, in a Ni2H2 diamond core and one containing Al(μ‐H)2Ni [58, 59, 60, 61, 62, 63, 64, 65]. However, two structures are noteworthy for further comparison: a Ni cis‐dihydride and a hexagonal Ni trihydride, which are supported by disilylene and Mg(nacnac) ligands, respectively [66, 67]. In the first case, the Ni cis‐dihydride bears a “significant bonding interaction” to the silylene ligand (Ni−H—Si), as evidenced by the 1 J Si−H coupling of 44.2 Hz, which is intermediate between that of a classical Ni silyl hydride (11 Hz) [68, 69] and that of a nonclassical Ni hydrosilane σ‐complex (111 Hz) [66, 70, 71, 72, 73, 74, 75]. In the second case, a more delocalized Ni(μ‐H)Mg bonding was invoked, with Mg–H bond lengths of 2.03(3) to 2.04(3) Å, based on a neutron diffraction study. We postulate that the lability of the bridging hydride in 4 is due to the weak Lewis acidity of the Zn(II) metalloligand. In support, the fluoride ion affinity (FIA) for 4 in the gas phase was calculated to be 233 kJ/mol, which is on par with other weak Lewis acids, including Si(OH)4 (FIA 223) and PPh5 (226) [76]. VT‐NMR and DFT Study of the Stereodynamics of 4 The stereodynamics of 4 was investigated using variable temperature (VT) 1H NMR spectroscopy in the temperature range of 193 to 303 K (Figure 4). At 193 K, two broad peaks were observed at −8.65 and −11.71 ppm, corresponding to the two distinct hydrides. Typically, in the HNi(μ‐H)E systems, the more upfield signal corresponds to the terminal hydride [16]. Upon raising the temperature, the two hydride peaks coalesce at 218 K and further sharpen into a triplet at −10.19 ppm above 263 K. Standard activation parameters of the exchange process were determined through Eyring analysis of the temperature‐dependent exchange rate constants (k/Hz), which were obtained by gNMR full lineshape analysis of the hydride signals (Figure S98 and Table S5 and S6) [77]. Stacked 1H VT‐NMR plot of 4 in THF‐d 8 from 193 to 303 K in the region from −7 to −13 ppm to show the fluxionality of the two hydride resonances. (For additional NMR spectra, see Figures S94, S95). The standard activation enthalpy (ΔH ‡ 218 K) and the standard free activation energy (ΔG ‡ 218 K) for the hydride exchange in 4 are 33.4(1.0) and 37.5(0.6) kJ mol−1, respectively. The standard activation entropy for the hydride exchange in 4 is negative and relatively small, ΔS ‡ 218 K = −18.6(4.1) J (mol K)−1. Hence, the enthalpic contribution to the energy barrier is more significant, which is consistent with a mechanism that involves Zn−H bond cleavage. In support of the small entropic barrier, the 1H NMR shifts of the ligand backbone are nearly unchanged from 193 to 303 K (Figure S94). The reaction profile of the H2 activation for 3 was investigated by DFT calculations at the r2Scan D4/def2‐TZVP level of theory [78, 79]. A transition state (4TS) was located at an activation energy barrier of 124.4 kJ mol−1 and is consistent with a concerted activation of H2 across the Ni−Zn bond (Figure 5). 4TS is a late transition state, wherein the Ni−H, Ni−(μ‐H) and Zn−(μ‐H) bond lengths are within 0.1 Å of those found in the product 4. The H−H distance is significantly lengthened to 1.948 Å versus 0.74 Å in free H2, indicating that the H−H bond is completely cleaved [53]. The notable differences in 4TS compared to 4 are the presence of a Ni−Zn bond (2.48 Å) in 4TS, and the acute H−Ni−H and P−Ni−P bond angles of 79.5 and 156.0 deg, respectively. (top) DFT‐calculated reaction pathways for cooperative H2 activation by 3 and the self‐exchange of hydrides in 4. Values in the parentheses refer to the computed free energy differences in kJ mol−1. (bottom) DFT structures of key intermediates and transition states. Ligand H atoms and the methyl groups of the tBu substituents have been omitted for clarity. Next, two possible mechanisms were investigated for the self‐exchange of the dihydrides in 4 (from 4A to 4B). The first pathway involves H2 reductive elimination to form the nonclassical Ni−H2 adduct intermediate, 4int , which is endergonic by 51.5 kJ mol−1. The transition state (4ATS1 ) was located at 156.9 kJ mol−1 above 4, and resembles a cis‐dihydride Ni species. As an aside, 4ATS1 is energetically similar to 4TS, making Ni‐mediated H2 activation (4int ) also viable. The second pathway is a direct exchange reaction via a C 2 symmetric transition state, 4TS_sym, with a square‐planar trans‐dihydride Ni motif. The low activation energy barrier of 41.3 kJ mol−1 makes this the energetically preferred pathway, and involves the cleavage of the Zn−(μ‐H) bond and partial rotation about the P−Ni−P axis. Of note, the calculated activation energy barrier closely matches that determined experimentally. Reaction of Ni−Zn Bimetallic With CO Since 4 does not release H2 spontaneously to give 3, we investigated the reaction of 4 with CO to force the reductive elimination of H2. Gratifyingly, the addition of 1 bar of CO to 4 in benzene resulted in the disappearance of the 1H hydride resonances with a concomitant color change from light orange to light yellow within 5 min, while the 31P peak shifted to 68.5 ppm. The ATR‐IR spectra of the newly formed complex 5 show two CO bands at 1926 cm−1 and 1985 cm−1, which we assign to a Ni bis(carbonyl) species (Figures 6 and S89). Formation of Ni–Zn bis‐ and monocarbonyl complexes (5 and 6, respectively) via H2 reductive elimination from 4. (inset) Molecular structure of 6 shown at 50% thermal ellipsoid probability. H atoms have been omitted for clarity. The mono carbonyl complex 6 (νCO = 1900 cm−1) was accessed in quantitative yield by reacting 5 with one equiv of 3, resulting in a rare release of a CO ligand from a Ni(0) center (Figures 6 and S90) [80]. Alternatively, 6 can be independently prepared by careful addition of one equiv of CO to either 3 or 4. Single‐crystal X‐ray diffraction analysis confirmed the structure of 6 with a distorted trigonal planar Ni and a long Ni∙∙∙Zn distance of 3.80 Å, indicating no interaction (Tables S2, S3). Reactivity of the Dihydride Complex 4 With CO2 Addition of 1 bar of CO2 to a THF‐d 8 solution of 4 at rt resulted in an immediate color change from orange to light yellow with a new 31P{1H} peak at 54.7 ppm. In addition to the ligand signals, two new 1H peaks were observed at 8.1 (singlet) and −27.6 ppm (triplet, 2 J P‐H = 67 Hz), which would be consistent with a formate H and a terminal Ni−H, respectively (Figure 7a). Further support for this new product 7 was evidenced through isotope‐labeling studies. Exposure of 4−D to CO2 led to the expected 2H signals for both the formate D and the Ni–D resonances (Figure 7b). Reacting 4 with 13CO2 resulted in a splitting of the formate H resonance into a doublet (1 J C−H = 192.5 Hz, Figure 7c). Partial 1H/2H NMR spectra of reactions between 4 or 4−D and 12CO2 or 13CO2 (1 bar) at t ∼5 min. (a) 1H spectrum of 4 and CO2, (b) 2H spectrum of 4−D and CO2, and (c) 1H spectrum of 4 and 13CO2. For full NMR spectra, see Figures S68, S76, and S78. The new product 7 exhibits two carboxylate bands in the IR spectrum in C6D6. The antisymmetric (ν a(COO)) mode at 1625 cm−1 shifts to 1585 cm−1 for the 13C‐labelled analogue, 713C (Figures S91, S92). The less intense symmetric (ν s(COO)) mode was identified in the difference IR spectrum and appears at 1313 cm−1 for 7, and 1290 cm−1 for 713C . For metal carboxylate complexes, the energy difference of these two stretches (Δ = ν a − ν s) informs on the nature of the binding of the carboxylate [81]. The Δ of 312 cm−1 is most consistent with a κ1‐monodentate formate (228−470 cm−1) and falls outside the expected values for bridging (138−190 cm−1) and κ2‐bidentate (42−190 cm−1) binding modes. Two stable isomers of 7 were predicted by DFT, 7a and 7b, of which the latter is more stable by a mere 2.7 kJ mol−1 (Figure 9). While 7a is a Ni(μ‐H)Zn with a monodentate κ 1 O‐formate bound to Ni, 7b is a bridging Ni(μ‐OCHO‐κ 1 O,κ 1 O')Zn with a terminal Ni hydride. The DFT‐calculated Δ values are 365 and 260 cm−1, respectively. The observed Δ is essentially the same as their average value, suggesting that 7a and 7b may coexist in rapid equilibrium. Of note, the Δ for 7b is higher than expected for bridging carboxylate ligands. However, this discrepancy can be attributed to the asymmetric bridging mode in the predicted structure, with a longer Zn–O distance of 2.20 Å and a shorter Ni−O distance of 1.93 Å. DFT‐calculated reaction pathways for the insertion of CO2 into 4. Ligand H atoms and the methyl groups of the tBu substituents have been omitted for clarity. Values in the parentheses refer to the computed free energy differences in kJ mol−1. Complex 7 is unstable to vacuum and reverts slowly to 4 under an N2 atmosphere (Figure 8). Due to the reversibility of CO2 insertion, clean cycling between 4 and 7 was effected through repeated exposures to vacuum and CO2 (Figures S100, S101). Complex 7 reacts further with CO2 (t½ ∼ 12 h under 1 bar CO2) to form a new product 8, which was crystallized from a concentrated benzene solution under CO2 (Figure 8). The molecular structure revealed a second CO2 insertion into the Zn–Namide bond, forming a new bridging carbamate ligand between Zn and Ni. Simultaneously, the formate ligand has migrated fully to Zn as a bidentate ligand. This is reminiscent of the carboxylic shift mechanism observed in metalloenzymes, where a carboxylate ligand can switch from binding one metal ion to two in bridging fashion, and vice versa [82]. Reversible reactivity between 4 and CO2 (1 bar) to yield 7 (t ∼ 5 min) and 8 (t ∼ 1 day). (inset) Molecular structure of 8 shown at 50% thermal ellipsoid probability. H atoms (except Ni−H and formyl H) and lattice solvent molecules have been omitted for clarity. The DFT‐calculated mechanism of double CO2 insertion into 4 to generate 8 is also shown in Figure 9. The first CO2 insertion into the Ni−H bond in 4 proceeds via a low activation energy barrier of 45.1 kJ mol−1, via a 4‐centered late‐stage transition‐state, with the following bond distances in the 4‐membered ring: Ni–H, 1.77; C–O, 1.26; Ni–O, 2.20; and C–H, 1.20 Å, to give 7a. The exchange process between the two isomers 7a and 7b is essentially a toggling of the bridging ligand from H to formate. This proceeds through 7TS1 , where neither H nor formate is bridging, with an energy barrier of 53.6 kJ mol−1. The second molecule of CO2, which formally inserts into the Zn−Namide bond, undergoes nucleophilic attack by the amide upon its approach. A high‐energy transition state 8TS1 (ΔG‡ = 82.7 kJ mol−1) was located, where the Zn−Namide bond has elongated to 2.52 Å with partial formation of the Namide—CO2 bond at 1.65 Å. We postulate that the elongation of the Zn−Namide bond is necessary to create a sufficiently nucleophilic N to attack CO2. In support, the other Zn−Namide bond remains short at 2.05 Å. Moreover, the bridging formate ligand forms a stronger interaction with Zn at 2.01 Å, which likely compensates for the weakening of one Zn−Namide interaction. This is an interesting case of the carboxylate shift mechanism. A classical carboxylate shift mechanism in methane monooxygenases involves a change from bridging mode to a single‐metal binding, so as to open a coordination site for oxygen binding at one of the iron centers [82, 83, 84, 85]. In the present case, the carboxylate shift from single‐metal to bridging weakens the coordination of an existing metal‐ligand bond, that is, Zn−Namide, so as to enable further reactivity with a substrate, that is, nucleophilic attack of the amide on an external CO2. The immediate product 8int thus features a new carbamate motif that is κ 1 O‐bound to Zn. Next, 8int goes through a very low energy transition state 8TS2 (ΔG‡ = 4.1 kJ mol−1) to give the final product 8. 8TS2 is interesting as one of the oxygen atoms of the carbamate bridges between the two metal centers and completes a 6‐membered cyclic transition state along with the bridging formate ligand. From 8int to 8, the carbamate ligand changes coordination modes from κ 1 O to Zn, to κ 1 O‐bridging between Zn and Ni, and finally to a bridging κ 1 O,κ 1 O'‐carbamate. This is accompanied by a reverse formate shift from bridging to bidentate only to Zn, as well as a cleavage of a Zn−Namine bond. Next, we briefly investigated the potential for the Ni−Zn system to catalyze CO2 reduction. As the release of formic acid from 7 was significantly uphill by +104.9 kJ/mol (Figure S115), different bases were added to drive formate release. Complex 4 was heated in the presence of 4 equiv base under 1 bar of H2:CO2. While partial formate release was detected (Figure S103), no catalytic turnover was observed. Replacing H2 with HBPin did generate CO2‐reduced products, HCO2BPin and CO; but the HCO2BPin turnovers were quite limited at ∼23, and CO deactivated the catalyst as the monocarbonyl 6 [86]. While catalytic CO2 hydrogenation was not achieved, the Ni−Zn system has accomplished the first step of CO2 hydrogenation to formate without needing additives or stoichiometric bases. Conclusion In summary, a T‐shaped Ni(0) complex supported by a Lewis acidic Zn(II) center (3) reacted with H2 to form a stable Ni dihydride species (4), which is rare. In the solid state, the Zn(μ‐H)Ni(H) motif was observed, but in solution at rt, the two hydrides are equivalent in the 1H NMR spectrum, consistent with a trans‐Ni(H)2 motif. The ability to toggle easily between these two motifs is attributed to the weak Lewis acidity of the Zn(II) ion in 4, which is supported by QTAIM calculations, where the Zn−(μ‐H) bond critical point has a lower electron density (0.06 e·Bohr−3) compared to the Ni−(μ‐H) one (0.11 e·Bohr−3). Moreover, the barrier between the Zn(μ‐H)Ni(H) and trans‐Ni(H)2 motifs is predicted by DFT to be only 41.3 kJ mol−1. The Ni hydrido formate complex (7) is, to the best of our knowledge, the first Ni example that forms from the simple addition of H2 and CO2 without an external base. This insertion of CO2 into an H2‐derived Ni dihydride is a desirable transformation because it could represent a key step in Ni‐mediated CO2 hydrogenation that only uses H2 and CO2. However, additional strategic designs are needed to enable formic acid release in this system to allow for efficient catalysis. Finally, 7 reacted with another CO2 molecule to give the dual CO2 activation product, 8, which features a bridging κ 1 O,κ 1 O'‐carbamate ligand between Ni and Zn. The DFT‐predicted mechanism reveals the importance of a carboxylate shift mechanism to create a sufficiently nucleophilic Zn−Namide that directly inserts CO2 to generate a carbamate group. In all these transformations, the Zn metalloligand plays a key role. Compared to other Ni‐Lewis acid pairings in the literature, the weaker Lewis acidity of Zn(II) enables lability of the bridging ligands in Ni(μ‐L)Zn, where L is hydride in 4 or formate in 7. This fluxionality allows for a rocking rotation about the P−Ni−P vector that toggles between bridging and terminal binding modes. The fluxional binding of the formate ligand to Zn(II) is also key to the reaction of 7 and CO2 to give 8. A combined study with X‐ray structures and IR analysis supported by DFT calculations showed three different binding modes of formate: κ 1 O on Ni, bridging κ 1 O,κ 1 O' between Ni and Zn, and terminal κ2 O,O' on Zn. As a whole, this work shows a new variation of metal‐metal cooperativity to promote Ni‐mediated H2/CO2 activation, which can inform the design of future Ni‐based catalysts for atom‐efficient CO2 hydrogenation. Author Contributions Krishnendu Dey: writing – original draft, writing – review and editing, conceptualization, investigation, visualization. Matthias Borowski: investigation. Gregor Schnakenburg: writing – review and editing, supervision, formal analysis, visualization, data curation. Connie C. Lu: writing – original draft, writing – review and editing, supervision, conceptualization, project administration, resources, data curation. Conflicts of Interest The authors declare no conflicts of interest. Supporting information The authors have cited additional references within the Supporting Information. Supporting File 1: anie73052‐sup‐0001‐SuppMat.pdf. Supporting File 2: anie73052‐sup‐0002‐Data.zip. Acknowledgments The experimental work was supported by the Deutsche Forschungsgemeinschaft (DFG), Grant No. 521217569. Start‐up funding from the University of Bonn is also acknowledged. The authors thank Dr. Senada Nozinovic, Charlotte Rödde, Nevena Petrova, and Dr. Jürgen Tirrée for experimental assistance and Tom Bergmann for proofing of the experimental section. Open access funding enabled and organized by Projekt DEAL. Contributor Information Gregor Schnakenburg, Email: gschnake@uni-bonn.de. Connie C. Lu, Email: clu@uni-bonn.de. Data Availability Statement The data that supports the findings of this study are available in the supplementary material of this article. References Associated Data Supplementary Materials The authors have cited additional references within the Supporting Information. Supporting File 1: anie73052‐sup‐0001‐SuppMat.pdf. Supporting File 2: anie73052‐sup‐0002‐Data.zip. Data Availability Statement The data that supports the findings of this study are available in the supplementary material of this article.