Distinct Reactivity Modes of a Copper Hydride Enabled by an Intramolecular Lewis Acid

Abstract

We disclose a 1,4,7-triazacyclononane (TACN) ligand featuring an appended boron Lewis acid. Metalation with Cu(I) affords a series of tetrahedral complexes including a boron capped cuprous hydride. We demonstrate distinct reactivity modes as a function of chemical oxidation: hydride transfer to CO₂ in the copper(I) state and oxidant-induced H₂ evolution as well as alkyne reduction.

Graphical Abstract

INTRODUCTION

The most prominent strategy for tuning the reactivity of metal sites within the synthetic community is primary-sphere ligand tuning;¹ less emphasis has been placed on reactivity changes imparted by modifying the secondary coordination sphere.² Extensive analyses of primary sphere structure/function relationships have provided a blueprint for how metal geometry influences substrate activation. For many substrates, metals in pseudo-tetrahedral geometries provide appropriate metal-substrate orbital overlap to stabilize reactive intermediates.^3–5

1,4,7-Triazacyclononane (TACN) ligands are facially binding chelates for metal complexes that are traditionally reported in 5- or 6-coordinate geometries,⁶ and have been widely employed within the bioinorganic, coordination chemistry, and catalysis fields.^7–9 Building upon work by Wieghardt and others,¹⁰ TACN ligands containing sterically-encumbering tertiary-alkyl groups have been shown to stabilize transition metal complexes in pseudotetrahedral geometries (Figure 1, bottom).¹¹ This bulky subset of TACN ligands has been used to access complexes that stabilize reactive oxygen species.^12–13 For example, Scarborough and coworkers reported the formation of a room temperature stable dicopper(II) μ-η²:η²-peroxo complex that is competent for catalytic aerobic oxidation of 3,5-di-tertbutylcatechol at and above 25 °C.¹² More recently, Ray and coworkers reported a highly reactive Fe(IV)-oxo supported by tri-tert-butyl TACN that is an electronic and functional model of taurine dioxygenase.¹³ An alternative avenue to tune reactivity in TACN manifolds is to incorporate secondary-sphere acidic groups: a modification that has received little attention.¹⁴

Figure 1.

Top: Representative copper hydride complexes in the Cu(I) state (top left) and oxidized states (top right). Bottom: TACN ligand with bulky tert-butyl groups provide access to stable pseudo-tetrahedral complexes (bottom left) and extension to include an appended boron Lewis acid (this work, bottom right).

Secondary-sphere interactions with metal-coordinated substrates are widely used by most metalloenzymes to facilitate small molecule activation/functionalization.^15–16 These interactions can enhance the activation of typically-inert molecules by stabilizing otherwise unstable reaction intermediates, thereby providing accessible reaction pathways. Although routinely exploited in biology, the use of secondary sphere interactions has yet to be widely employed in the field of synthetic chemistry. Synthetic ligand architectures containing appended acidic groups may be used to clarify the molecular-level details that inform on mechanism and ultimately enhance function. Hydrogen-bonding interactions are the most common appended groups that have been used to stabilize high-valent intermediates (i.e. metal-oxos):^{2, 17–20} such intermediates are not generally susceptible to protonation by weak Brønsted acids (hydrogen bond donors). However, these types of secondary sphere donors are less compatible with hydridic metal hydrides because of competitive protonation (H₂ elimination).²¹ Ligands containing appended Lewis acidic groups overcome this incompatibility^22–27 and are ideally suited to facilitate small molecule binding/reduction.

Our group has investigated cooperative binding modes and subsequent reactivity enabled by Lewis acidic boron-appended ligands.^28–33 A key design principle we uncovered within the context of otherwise unstable metal hydrides was that moderately acidic Lewis acids provide an optimal balance between stability and reactivity.³⁴ Leveling of ligand donor properties through M–H–BR₃ interactions provided enhanced stability, without the loss of metal-hydride reactivity (e.g. reductive elimination). We found that there are clear changes to M-H redox properties that are uniquely accessible to Lewis-acid capped metal hydrides. We hypothesized that these principles could be extended to other metals and potentially unveil access to new reactivity modes.

One class of metal hydrides that could benefit from the addition of an appended Lewis acidic group is copper hydrides. Considerable work in the last decade has focused on regio- and enantioselective functionalization of organic substrates including olefins,^35–36 alkynes,³⁷ imines,³⁸ ketones,^39–40 and α,β-unsaturated carbonyls facilitated by discrete as well as proposed intermediate Cu(I) hydrides.^41–42 One precedented pathway through which Cu(I) hydride complexes react with alkyne substrates is through insertion into the Cu-H bond to form a Cu(I) alkenyl complex.^42–45 Importantly, we note that this pathway is only precedented at the Cu(I) state. Cu(I) hydride complexes with exogenous borane interactions are also known,^46–50 although oxidation studies of these types of complexes have not been explored. In contrast to well-defined reactivity at the Cu(I) state, reactivity pathways accessible to oxidized Cu-H adducts are not clear. The Norton group investigated oxidation of Stryker’s reagent-type clusters and found that electron transfer can be mechanistically relevant for net hydride transfer reactions (Figure 1, top),⁵¹ although this concept has not been extended to monomeric or dimeric Cu(I) hydride complexes. A rare example of an oxidized Cu-H-Cu fragment reported by the Warren group demonstrated phenylacetylene reduction to styrene.⁵² In this case, the authors formulate the oxidized adduct as a Cu^1.5-H-Cu^1.5. To clarify the requirement of the second Cu atom for oxidized Cu-H reactivity, we hypothesized that a monometallic Cu-H stabilized by a borane Lewis acid might provide similar access to oxidized Cu-H adducts and thus we targeted reactivity modes of both cuprous and oxidized Cu-H adducts.

RESULTS AND DISCUSSION

We targeted a sterically-encumbered heteroalkylated TACN ligand that features two tert-butyl groups in addition to a modifiable auxiliary, by adapting a similar literature protocol originally used for the synthesis of 1,4,7-tri-tert-butyl-1,4,7-triazacyclononane.¹¹ Ring closure of N,N’-(ethane-1,2-diyl)bis(N-(tert-butyl)-2-chloroacetamide) with allylamine followed by LiAlH₄ reduction provided 1-allyl-4,7-di-tert-butyl-1,4,7-triazacyclononane (^allylTACN^tBu) as a colorless oil in 16% yield over 3 steps.

We chose to employ copper(I) for metalation/reactivity studies because it favors low coordination numbers and also participates in single electron redox processes.⁵³ Metalation with [Cu(MeCN)₄][PF₆] or CuI in THF solvent afforded the cationic or neutral complexes [(^allylTACN^tBu)Cu(MeCN)][PF₆] or (^allylTACN^tBu)CuI, respectively(1-PF₆ and 1-I; Figure 2). Single crystal X-ray diffraction (XRD) experiments were employed to confirm connectivity, and importantly, each molecular structure confirmed that the allylic moiety is not interacting with copper and is available for post-metalation hydroboration.

Figure 2.

Synthetic approaches to generate complexes 2-X. Molecular structures of 1-X and 2-X displayed with 50% probability ellipsoids. H-atoms not attached to allylic moieties and any minor disordered moieties are omitted and the 9-BBN substituents are displayed in wireframe for clarity.

Complexes 1 were subjected to hydroboration experiments to install an appended Lewis acid. Treating 1-I with one equiv. of 9-borabicyclo[3.3.1]nonane (9-BBN) in THF at room temperature for 16 h afforded the anti-Markovnikov hydroboration product (^BBNTACN^tBu)CuI (2-I) in 55% yield, assessed by ¹H NMR spectroscopy. We attribute the moderate yield to competitive decomposition pathways: in addition to 2-I, we observed an insoluble grey precipitate. To overcome this hurdle, we undertook an alternate synthetic approach (Figure 2). Treating a C₆H₆ solution of ^allylTACN^tBu with 9-BBN (50 °C, 90 min) resulted in successful in situ anti-Markovnikov hydroboration of the ligand which, when followed by metalation with CuI or Cu₂(OTf)₂(C₆H₆) afforded (^BBNTACN^tBu)Cu(I) (2-I) or (^BBNTACN^tBu)Cu(OTf) (2-OTf) as a white powders in 67% and 90% yield respectively over the two steps. In contrast to the post-metallation hydroboration route, the pre-metallation hydroboration route provided the desired borane-appended Cu(I) complexes in high yield.⁵⁴

To ascertain the molecular structures of the Lewis acid appended complexes, we carried out single crystal XRD studies on complexes 2. Data refinement revealed each molecule displays a pseudo-tetrahedral coordination environment (τ₄: 2-I = 0.66; 2-OTf = 0.71)⁵⁵ around copper that is similar to that of other mononuclear Cu(I) TACN complexes.^{11, 56–57} Each structure contains a planar, three-coordinate borane (ΣB_α: 2-I = 359.6(3)°; 2-OTf = 359.4(4)°) that is far removed from any Lewis basic sites (>3.6 Å), consistent with an unquenched Lewis acid in each case.

Voltammetry experiments suggest the X-type ligand in 2-OTf is labile in comparison to 2-I. The voltammogram (0.2 M [Bu₄N][PF₆], THF, 200 mV/s) of 2-OTf displays an oxidative event at E_pa = 0.38 V (vs. Cp₂Fe/Cp₂Fe⁺) that shifts to E_pa = 0.16 V upon titration with [Bu₄N][OTf] (~2, 4 , and 23 equiv); Δ = −0.22 V). This potential, assessed by square wave voltammetry, observed in the presence of [Bu₄N][OTf] (0.09 V vs. Cp₂Fe/Cp₂Fe⁺) is similar to that measured for 2-I (−0.03 V), consistent with similar electronic environments in the absence of ligand exchange.

The lability of the X-type ligand in 2-OTf is further borne out by reactivity studies. To baseline the electronics of the new chelating ligand, ^BBNTACN^tBu, we treated each complex 2 with CO. Although substitution of the iodide in 2-I did not occur, the trifluoromethanesulfonate ligand in 2-OTf was readily displaced to afford [(^BBNTACN^tBu)Cu(CO)][OTf] (2-CO) as a vacuum stable white solid.⁵⁸ The IR spectrum (KBr) of 2-CO displays a sharp ν(CO) stretch at 2069.5 cm⁻¹ that is at comparable energy to related cationic Cu(I)-CO species chelated by triamine-type ligands^59–60 suggesting that Lewis acid incorporation in ^BBNTACN^tBu has negligible effects on its ligand field properties and does not interact with CO.

The absence of a Lewis acid/CO interaction in 2-CO may indicate either a geometric or basicity mismatch with the appended Lewis acid. Previous work from our group showed that cooperative binding can be enhanced by modifying the appended Lewis acid tether length as well as the basicity of the substrate.⁶¹ To provide a single atom ligand with less directionality requirements that we anticipated would better accommodate a metal-ligand-interaction, we targeted a copper hydride. Treating a freshly thawed toluene solution of 2-OTf with KBHEt₃ afforded a new hydride-containing complex, (^BBNTACN^tBu)CuH (3). The formation of 3 is reversible: treating 3 with [Ph₂NH₂][OTf] cleanly regenerates 2-OTf with extrusion of H₂ (Figure 3). IR spectroscopy of 3 revealed a strong absorption (KBr) at 1831 cm⁻¹ that is sensitive to deuteride incorporation (Δ = 195 cm⁻¹) and assigned as the ν_(Cu-H-B) stretch. This absorption is at lower energy than most reported Cu-H-B interactions, notably from scorpionate complexes (~2350–2250 cm⁻¹),^46–48 and is more comparable to two-coordinate (IMes)Cu-HBPh₃ (ν_(Cu-H-B) = 1693 cm⁻¹).⁴⁹ The ¹H NMR spectrum of 3 (C₆D₆) displays a broad resonance at −1.79 ppm that disappeared in the (^BBNTACN^tBu)CuD isotopologue. ¹¹B NMR spectroscopy reveals a broad singlet at −6.44 ppm, indicative of a 4-coordinate, tetrahedral boron environment. Together, these data suggest a copper-hydride that interacts with the Lewis acidic trialkylborane.⁶²

Figure 3.

A) Synthesis and reversibility of 3. B) Molecular structure and Wiberg bond indices of 3 as well as of limiting bonding descriptions of Cu-H and B-H. The structure of 3 is displayed with 50% probability ellipsoids. H-atoms not attached to copper are omitted and the 9-BBN is displayed in wireframe for clarity. C) Voltammetry comparison between 3 (top, red) and 2-I (bottom, blue).

To further clarify the binding mode in 3, we analyzed single, X-ray quality crystals that were obtained by layering a C₆H₆ solution of 3 with hexamethyldisiloxane at room temperature (Figure 3). Data refinement from the XRD experiment revealed a pseudo-tetrahedral (τ₄ = 0.67) copper-hydride that is capped by the appended trialkylborane (ΣB_α = 322.7(8)°). The position of the hydride atom in 3 was refined and displayed a Cu-H and B-H distance of 1.560(18) and 1.315(19) Å, respectively (Cu-H-B = 125.77°; Cu-B = 2.5617(15) Å). While the Cu-H distance is similar to that observed in (NHC)CuH-BR₃ species,^49–50 the B-H distance in 3 is considerably longer than for other reported Cu-H-BR₃ examples.^63–65 The primary coordination sphere of the Cu(I)-TACN in 3 is very similar to 2-OTf and 2-I (τ₄ within 0.04).

Two limiting descriptions of complex 3 can be formulated where it is viewed as either a 1) copper-hydride capped by a trialkylborane, or 2) borohydride anion coordinated to the copper center. While the X-ray and spectroscopic data provide a structural description of the Cu–H–BR₃ interaction, we undertook a DFT study of 3 (B3LYP-D3/SVP (PCM = THF)) to provide a clearer electronic description. Wiberg bond analyses revealed a Cu-H bond index of 0.13 and a B-H bond index of 0.79. We compared these values to both limiting bonding descriptions: a terminal Cu–H with no borane interaction ((^nBuTACN^tBu)CuH, 3’; 0.65) and R₃B–H (0.94) (see Figure 3B). These data suggest that 3 is best described as a borohydride anion interacting with copper, with some copper hydride resonance contribution. This depiction has also been invoked in related compounds,^{50, 66} and we note that similar (NHC)CuH-BR₃ complexes can be induced to exhibit Cu-hydride-type reactivity (carbonyl hydrosilylation) in select organometallic transformations.⁴⁹ Importantly, our analysis of 3 suggests that the Cu-H-BR₃ moiety contains both Cu-H and B-H character.

The electronic effect that the appended trialkylborane imparts on the Cu–H–BR₃ unit of 3 was probed by electrochemical analysis. Although Stryker’s reagent-type clusters, [(Ar₃P)CuH]₆, are commonly employed as reducing agents (−1.00 – −1.20 V),^{51, 67} Cu-H motifs have rarely been subjected to electrochemical analysis, perhaps due to limited stability. In contrast, isolated samples of 3 are indefinitely stable in the solid state at −35 °C and have solution (C₆D₆ or THF) half-lives of >2 weeks at room temperature.⁶⁸ Voltammetry experiments of 3 were performed in THF (0.2 M [Bu₄N][PF₆]) and revealed an irreversible oxidation event at −0.13 V vs. Cp₂Fe/Cp₂Fe⁺. The difference in the measured potential of 2-I vs 3 (Δ = 100 mV) is small. Redox potentials between hydride and halide ligands are typically quite different; for instance, in isostructural Rh^III-X (X = Cl, H) complexes, the two redox potentials vary by >800 mV.⁶⁹ We attribute the similar redox potentials for 2-I and 3 to the appended trialkylborane, which we propose syphons electron density from the hydrido-ligand and thus levels the overall redox-potential of the system (redox leveling).

To ascertain the requirement of the boron Lewis acid to provide access to TACN-ligated cuprous hydrides, we attempted to synthesize a Lewis acid-free variant of 3. We designed a new ligand, 1-n-butyl-4,7-di-tert-butyl-1,4,7-triazacyclononane (^nBuTACN^tBu), where the −(CH₂)₃BBN substituent was replaced by an inert n-butyl group. Metalation with Cu₂(OTf)₂(C₆H₆) proceeded smoothly, affording (^nBuTACN^tBu)Cu(OTf) (2’-OTf; see SI for full characterization). Unfortunately, attempts to form an analogous copper hydride complex (3’) were unsuccessful. Reactions with 1 equiv. KHBEt₃ and 2’-OTf under the same conditions used to prepare 3 yielded free-ligand and an insoluble black precipitate, and neither (^nBuTACN^tBu)CuH nor (^nBuTACN^tBu)Cu(HBEt₃) were observed, which highlights the importance of the appended Lewis acid in 3 to stabilize the hydrido ligand.

We used DFT analyses to provide further insight into the electronic structure of 2-I vs. 3 as well as their Lewis acid-free counterparts, 2’-I and 3’.⁷⁰ The HOMOs of 2-I and 2’-I were calculated to be very similar, separated by 0.10 eV (−4.53 and −4.43 eV respectively). In comparison to complexes 2, the corresponding hydride 3 is stabilized by 0.37 eV (−4.90 eV) while 3’ is destabilized by 0.65 eV (−3.78 eV). These results are consistent with the ability of the appended Lewis acidic borane to provide substantial stabilization to the otherwise unstable Cu-H unit (Table S18 and Figure S83).

Related boron-capped Cu-H systems have been shown to reduce C=O bonds,^{49, 71–73} through hydride transfer. To evaluate the ability of 3 to similarly deliver a hydride, we introduced CO₂ to 3 (Figure 4A). Treating a C₆D₆ solution of 3 with CO₂ (20 psig) at room temperature resulted in the precipitation of a new copper-containing species (4) as a white solid. ¹H NMR spectroscopy (THF) is consistent with reduction of CO₂ to CO₂H⁻, indicated by a diagnostic resonance at 8.21 ppm assigned as the formate C-H bond.⁷⁴ An XRD experiment revealed that 4 is a dimeric species, [(^BBNTACN^tBu)Cu(O₂CH)]₂, containing two pseudotetrahedral copper centers (τ₄= 0.70, 0.67), where the pyramidalized boron Lewis acid (ΣB_α = 322.14(18)°) from one molecule interacts (B-O = 1.609(4) Å) with the copper-bound formate (Cu-O = 1.9537(17) Å) of an adjacent molecule. We propose that the dimeric form is favored for this substrate due to a geometrical mismatch for intramolecular binding of this substrate through the Cu and borane Lewis acid, which we note is distinct from the hydride and consistent with recently described criteria for cooperative binding.³³

Figure 4.

A: CO₂ reduction by 3 and formation of 4. The molecular structure of 4 is displayed with 50% probability ellipsoids. H-atoms not connected to the formate moieties are omitted and 9-BBN substituents are displayed in wireframe for clarity. Cutout: free energy landscape for monomer/dimer formation of the Cu-OCHO-BR₃. B: 1) chemical oxidation of 3 and formation of H₂, 2) formation of H₂ via delivery of an H-atom to Cu(I) via acid/reductant combination, and 3) control reaction. C: 1) Cu(I) no reaction with phenylacetylene, 2) redirection of reactivity via oxidation to facilitate substrate reduction and 3) control reactions.

To further investigate the preference of 4 to form a dimer, rather than a monomer, we examined the energetic landscape for monomer/dimer equilibra by DFT. The dimeric (Cu-OCHO-BR₃)₂ complex is calculated to be more stable than its monomeric counterpart (ΔG=−11.49 kcal/mol; Figure 4A, cutout). This result clarifies the thermodynamic driving force to form the dimer, and may indicate that a different geometry or appended Lewis acid tether length may be required for the stabilization of a monomeric formate complex. Analysis of the bonding within formate was consistent with XRD data, with calculated Wiberg bond indices indicating slightly more double bond character in the copper-bound C-O (1.44) than in the boron-bound C-O (1.30). This bonding depiction is comparable to that observed in systems employing strong exogenous Lewis acids.^75–77 Importantly, compound 4 demonstrates that the same electronic structure can be attained by using intramolecular Lewis acids that are considerably weaker.⁷⁸

The availability of single electron redox chemistry to 3 provides a strategy to modify the reactivity of a copper hydride. Further analysis of the electrochemistry of 3 indicated that additional chemical reactivity pathways occur following oxidation. After scanning more positive than −0.13 V, the voltammogram of 3 featured a fully reversible oxidation event at +0.39 V (vs Cp₂Fe/Cp₂Fe⁺), identified as [(^BBNTACN^tBu)Cu]⁺. (Note: this redox event was observed in our analysis of 2-OTf) This result suggests that oxidation of 3 results in loss of an H-atom. To investigate the products of oxidation, we treated 3 with a chemical oxidant in a sealed vessel. When 1 equiv. ferrocenium trifluoromethanesulfonate ([Cp₂Fe][OTf]) was added to a THF solution of 3 in a sealed J-Young NMR tube, we observed 0.42 (±0.03) equiv of H₂ (Figure 4B-1), in addition to 2-OTf. To clarify the pathway by which H₂ forms, we repeated this reaction using the deuteride analogue of 3 in protio THF solvent. Using these reagents, only D₂ was formed (no HD or H₂). The stoichiometry of H₂ formation and the deuterium labeling results are consistent with a bimolecular process where H₂ is derived solely from the copper-(boro)hydride.

The requirement of the appended Lewis acid to stabilize the Cu-H in 3 suggested that the Lewis acidic group might also enable unique reactivity. We undertook a series of control reactions to examine the extent to which H₂ generation could be accessed from different synthetic routes, and influenced by the appended Lewis acid. Mayer and coworkers have tabulated the H-atom BDFE values of mixtures of acids and reductants that can be used to deliver H• equivalents.⁷⁹ We selected a combination of pyridinium tetrakis[3,5-bis(trifluoromethyl)phenyl]borate ([pyr-H][BArF₂₄]) with decamethylferrocene (Cp*₂Fe), which has a BDFE of 58.0 kcal/mol in acetonitrile and is thus a weak H-atom donor.⁷⁹ While the combination of [pyr-H][BArF₂₄] and Cp*₂Fe alone does not generate H₂, addition of [pyr-H][BArF₂₄] and Cp*₂Fe to a C₆D₆ solution of 2-OTf resulted in the formation of H₂ along with the reformation of 2-OTf (Figure 4B-2).⁸⁰ The observation of H₂ formation suggests that this reaction may share a common intermediate with that of chemical oxidation of 3: a transient Cu(II)-H. To evaluate the role(s) of the appended Lewis acid, we examined reactivity employing 2’-OTf, which does not contain a Lewis acid, and found that H₂ does not form under analogous conditions (Figure 4B-3).

Finally, we sought to circumvent the H₂ formation pathway by intercepting the product of chemical oxidation of 3 with a sterically-accessible alkyne substrate. We found that there is no reaction between 3 and 10 equiv. phenylacetylene (Figure 4C-1). In contrast, when the reaction was repeated in the presence of 1.2 equiv. [Cp₂Fe][OTf] oxidant, the reduction product (styrene) formed, albeit in low yield (Figure 4C-2).⁸¹

We examined two sets of control reactions to ascertain the requirements of each component of 3 to form styrene: the Lewis acid and copper ion. To evaluate the role(s) of the appended Lewis acid, we examined analogous reactivity using mixtures of 2’-OTf and KHBEt₃ in place of 3, providing analogue of 3 with an exogenous borane. Solutions of 2’-OTf, KHBEt₃, phenylacetylene, and [Cp₂Fe][OTf] were sequentially layered and frozen in a J-Young NMR tube to prevent unwanted side reactions between individual reagents. Subsequent thawing and mixing of the reagents revealed only H₂, and no styrene (Figure 4C-3). Similarly, to evaluate the role(s) of the Cu ion, when a solution of KHBEt₃ was treated under analogous conditions, only H₂ formed.^82–83 Combined, these results confirm that in 3, both the Cu and the appended borane are crucial cooperative components for phenylacetylene reduction and we propose this is due to unique stabilization imparted to an otherwise unstable copper hydride.

Although an oxidant was added to 3 in the above reactions, we did not observe solution colors that are characteristic of Cu(II) at the end of reactions. We further investigated this observation using EPR and NMR spectroscopy to assess the oxidation state of Cu. Addition of 0.9 equiv. [Cp₂Fe][OTf] to a mixture of 3 and phenylacetylene at −78 °C generated a transient EPR signal assigned as Cu(II) which decayed over time, and no paramagnetic signals were observed at the end of the reaction (2h at room temperature; see SI:Figure S75–S76). These results are consistent with ¹H NMR spectra obtained after completion of the reaction, which indicated 2-OTf formed in high yield. The results of both experiments provide key insight into the reactivity of 3: addition of an oxidant to the Cu(I)-H oxidizes at the Cu center, which induces reduction of phenylacetylene and reforms Cu(I). This reactivity is distinct from previous Cu(I)-H complexes that react directly with alkynes and represents an unusual pathway because addition of an oxidant is required for alkyne reduction with 3.

CONCLUSION

In conclusion, we have presented a tetrahedral-enforcing tridentate ligand scaffold containing an appended borane Lewis acid and metalation with Cu(I). The appended Lewis acid enables the isolation of an otherwise unstable Cu(I) hydride that exhibits distinct reactivity modes as a function of chemical oxidation: hydride transfer to CO₂ in the copper(I) state and oxidant-induced H₂ evolution as well as alkyne reduction. Access to these divergent reactivity modes is attributed to an intramolecular appended Lewis acid. Ongoing work is focused on clarifying the mechanistic details as well as extending these concepts to promote selective reduction of more inert small molecules.