Skip to main content
NCBI home page
Chemical Science logoLink to Chemical Science
. 2026 Oct 1. Online ahead of print. doi: 10.1039/d6sc01654e

Force-modulated and strain-dependent stereoselectivity of copper-catalyzed reductive coupling

Chenghao Duan a,†, Xujun Zheng a,†, Jiaqi Zhu b, Rosemary L Buhrman a, Stephen L Craig a,✉, Ross A Widenhoefer a,✉
PMCID: PMC13650620  PMID: 42857459

Abstract

Force coupled to the dihedral angle of a macrocyclic biaryl bisphosphine ligand modulates the stereoselectivity of the copper-catalyzed reductive coupling of 2-vinyl pyridine with acetophenone to form 1-phenyl-2-(pyridin-2-yl)butan-2-ol (1) as a mixture of the (2S,3S)/(2R,3R) (1a) and (2R,3S)/(2S,3R) (1b) diastereomers. Across a ∼280 pN range of extension forces, the diastereomeric ratio (1a : 1b) decreases from 2.7 : 1 to 1.7 : 1 and the enantiomeric ratios of 1a and 1b decrease from 5.60 to 2.37 and 7.23 to 1.93, respectively. DFT calculations support a model for the observed force-dependent stereoselectivity of reductive coupling in which the competing transition states are differentially destabilized by the applied tensile force. The degree of destabilization corresponds to the magnitude of the nuclear motions along the coupled vector of applied force that occur on moving from the ground state to the transition state. The stereoselectivity of reductive coupling is likewise modulated in the bulk by compressing elastomeric polyacrylate gels containing crosslinked [(R)-Biphep]Cu active sites. The diastereomeric ratio (1a : 1b) decreases from 1.9 to 1.7 : 1 and the enantiomeric ratios of 1a and 1b decrease from 2.88 to 2.58 and 2.77 to 2.48, respectively, when the support is compressed by 50% in a single dimension. Stereoselectivity returns to basal levels when the polymer is relaxed. Control experiments employing polymers containing mechanically-decoupled [(R)-Biphep]Cu active sites support the mechanical origin of the strain dependent stereoselectivity of reductive coupling.

Force applied to the ligand of a biaryl bisphosphine copper complex modulates the stereoselectivity of Cu-catalyzed reductive coupling in both small molecule systems and in elastomeric gels containing crosslinked [(R)-Biphep]Cu active sites.graphic file with name d6sc01654e-ga.webp

Introduction

There has been longstanding interest in the heterogenization of single-site transition metal catalysts via covalent attachment to polymeric organic supports as a means to simplify catalyst recycling and recovery while retaining the desirable reactivity and selectivity of homogeneous catalysts.1 More recently, porous organic polymers (POPs) have emerged as particularly attractive supports for single-site catalysts owing to their high stability, large surface area and pore size, and desirable swelling behavior with the potential for high catalyst loadings and diffusion rates that rival homogeneous systems.2 In addition to the nature of the metal and ligand comprising the active site, the polymeric support plays a significant role in modulating the selectivity of a supported catalyst.1 For example, polymer rigidity and ligand density can affect selectivity by biasing the geometry and/or coordination state of the active catalyst, while polymer porosity, channel size, and main-chain composition can modulate selectivity via size- and polarity-selective diffusion effects. However, optimization of these features typically requires iterative modification of the polymeric support including variations in main-chain composition, extent of polymerization, and crosslink and ligand density.

Extrinsic mechanical force represents an intriguing but largely untapped means to modulate the selectivity of polymer-supported single-site catalysts in a controlled manner. Macroscopic mechanical forces have been shown to bias the outcomes of a number of stoichiometric organic transformations,3 including outcomes that are not realized under thermal conditions.4 Conversely, mechanochemical manipulation of polymer-supported single-site catalysts has been largely restricted to the activation of latent catalysts via metal–ligand bond cleavage followed by thermally-controlled reactivity of the resulting metal fragment.5,6 Distinct from force-induced M–L bond cleavage is the use of extrinsic mechanical force to modulate the reactivity of an embedded, intact transition metal catalyst. Because polymers can be stretched and deformed both incrementally and reversibly, this approach offers the potential to generate a continuous range of force-dependent catalyst states, each possessing distinct reactivity, which we term “multi-state mechanocatalysis” (MMC). In a singular example, we have recently shown that extrinsic mechanical force generated through uniaxial extension or compression modulates the enantioselectivity of the hydrogenation of dehydroamino acids catalyzed by a rhodium bisphosphine complex embedded in a flexible polymer, in one case leading to a two-fold increase in the enantiomeric ratio when the support was compressed by 73% in a single dimension (Fig. 1).7

Fig. 1. Strain-dependent enantioselectivity of hydrogenation employing a supported rhodium catalyst.

Fig. 1

Open in a new tab

In addition to the development of advanced polymeric supports with improved force transduction and mass transfer properties, we have sought to expand the scope of MMC from the lone example of rhodium-catalyzed hydrogenation to include a more diverse palette of catalytic transformations including carbon–carbon bond forming processes. Toward this objective, we were drawn to the reductive coupling of alkenylazaarenes with π-electrophiles catalyzed by copper bisphosphine complexes.8,9 These coupling reactions proceed efficiently under mild conditions to form two contiguous sp3-stereocenters. Importantly, the stereoselectivity of these transformations is sensitive to the nature of the bisphosphine supporting ligand,8,9 suggesting that selectivity might be similarly modulated by force-dependent ligand perturbations. Here we show that mechanical compression of elastic polyacrylate networks containing crosslinked, catalytically active copper bisphosphine groups modulates the stereoselectivity of the reductive coupling of 2-vinyl pyridine with acetophenone. Homogeneous, single-site copper complexes containing macrocyclic bisphosphine ligands are employed to quantify the effects of force on the stereoselectivity of reductive coupling and to probe the differential effects of force on the diastereomeric reaction pathways.

Results and discussion

Force-modulated selectivity of reductive coupling

From the larger family of copper-catalyzed reductive coupling processes,8 we focused on the reductive coupling of 2-vinyl pyridine with acetophenone to form 2-phenyl-3-(pyridin-2-yl)butan-2-ol (1).9a To quantify the effect of force on the stereoselectivity of copper-catalyzed reductive coupling, we employed molecular copper complexes bearing the known macrocyclic force probe ligands (FPLs) (R)-E(m,n) and (R)-Z(m,n) with untethered (R)-MeOBiphep serving as a force-free benchmark (Fig. 2).10–13 These FPLs comprise a 1,1′-biindane photoswitch tethered to the 6,6′ oxygen atoms of an (R)-Biphep moiety. The Z and E isomers of the 1,1′-biindane apply a compressive or stretching force, respectively, to the oxygen atoms of the bisphosphine ligand and the magnitude of the restoring force, i.e. the force applied to the 6,6′ oxygen atoms perpendicular to the axis of rotation, is modulated by the tether length. The restoring force for each ligand has been quantified by DFT and within the family of ligands employed here, forces ranging from −65 pN (compressive) to +228 pN (extension) are realized without perturbation of the electron-donor abilities of the ligands.10–13 Computational evaluation of related systems has determined that the effective tension on the free ligand within a given macrocycle is similar to that felt by the metal-coordinated ligand;13 similarly, the compressive force applied by the biaryl motif to the stiff stilbene changes by only 7–15% upon metal coordination.14 Because the effective forces are relatively insensitive to small changes in coordination geometry, we interpret the observed mechanochemical effects in terms of a constant force applied over the course of the reaction. We emphasize that this assumption is not strictly true – the effective tension does fluctuate in response to small geometric changes on going between reactant, intermediate, and transition states – but it is a reasonable approximation for the system. Previous work has also shown that the effect of force applied intramolecularly and quantified in this way is indistinguishable to that applied in a strained polymer,15 and we have previously employed these FPLs to quantify the force effects of a range of stoichiometric and catalytic transformations.10–14,16

Fig. 2. Molecular force probe ligands (R)-Z(m,n) and (R)-E(m,n) and (R)-MeOBiphep.

Fig. 2

Open in a new tab

The reactions of 2-vinylpyridine with acetophenone and PhSiH3 catalyzed by a mixture of Cu(OAc)2 and bisphosphine ligand in each case proceeded to ≥95% conversion to form 1 as a mixture of the (2S,3S)/(2R,3R) (1a) and (2R,3S)/(2S,3R) (1b) diastereomers, both enriched in the 3S enantiomer, and without formation of detectable amounts of C Created by potrace 1.16, written by Peter Selinger 2001-2019 C hydrogenation product 2-ethylpyridine (Table 1). The diastereomeric ratio for each reaction was determined by 1H NMR analysis of the crude reaction mixture, whereas the enantiomeric ratios of both 1a and 1b were determined by chromatographic separation of the diastereomers followed by HPLC equipped with chiral stationary phase (Table 1); the absolute configurations of the four stereoisomers of 1 were unambiguously assigned via derivatization and X-ray crystallography (see SI). The diastereomeric ratio (1a : 1b) of reductive coupling decreases from 2.7 : 1 to 1.7 : 1 with increasing forces of extension from −65 (compression) to +228 (extension) pN. Similarly, the enantiomeric ratio of 1a decreases from 5.60 to 2.37 and the enantiomeric ratio of 1b decreases from 7.23 to 1.93 across the same range of extension forces (Table 1). Although higher enantioselectivities (≥99% ee) for the copper-catalyzed coupling of 2-vinylpyridine with acetophenone have been realized with chiral ferrocene ligands,9 the modest enantioselectivity realized with (R)-MeOBiphep is advantageous for our purposes, as it renders potential MMC effects more easily detectable.

Table 1. Reductive coupling of 2-vinylpyridine (0.20 M) with acetophenone (0.21 M) and phenylsilane (0.22 M) catalyzed by a mixture of Cu(OAc)2 (10 mM) and (R)-bisphosphine ligand (10 mM) in toluene at 25 °C for 15 h to form a mixture of diastereomers 1a and 1b depicting the major 3S enantiomers.

graphic file with name d6sc01654e-u1.jpg
(R)-P–P Force (pN) Convn (%) er(1a)a er(1b)a 1a : 1bb
Z(2,2) −65 95 5.60 ± 0.15 7.23 ± 0.14 2.7 ± 0.09
Z(3,3) −3 >99 5.31 ± 0.04 6.72 ± 0.11 2.6 ± 0.10
MeOBiphep 0 >99 5.14 ± 0.10 6.12 ± 0.05 2.3 ± 0.13
E(3,3) 130 95 3.29 ± 0.12 2.95 ± 0.07 2.1 ± 0.04
E(2,3) 228 >99 2.37 ± 0.06 1.93 ± 0.06 1.7 ± 0.06
Open in a new tab
a

er(1a) = (2S,3S)-1a:(2R,3R)-1a and er(1b) = (2R,3S)-1b:(2S,3R)-1b, both determined by HPLC equipped with chiral stationary phase.

b

Diastereomeric ratio determined by 1H NMR analysis of the crude reaction mixture. Values are reported as the mean of three replicate measurements; uncertainties correspond to the standard deviation of the mean (n = 3).

Because the difference in activation energies (ΔEa) of competing reaction pathways is given by ΔEa = ln(k1/k2)RT, the linear dependence of the natural logarithms of er(1a), er(1b), and dr on force establish the linear relationship between the ΔEa of er(1a), er(1b), and dr, respectively, and force (Fig. 3). The simplest model of mechanochemical coupling assumes a reaction that is unperturbed in the reactant and transition states, but in which the change in geometry along the coupled vector of applied force, quantified by a term Δx, leads to a change in activation energy (here, of the reaction at the coordinated metal center) that is equal to the change in the mechanical potential energy of the component of the system that is applying the force (here, the part of the macrocycle applying force to the biaryl oxygen atoms). The change in distance Δx in this scenario is due primarily to motion of the oxygen atoms, the change in activation energy is −FΔx, and er depends exponentially on force. Therefore, the semilog relationship between relative rate and force is consistent with a picture in which the observed force-dependent stereoselectivity of copper-catalyzed reductive coupling arises predominantly through the mechanochemical coupling of force to the nuclear motions (O⋯O in the present case) as the complexes progress from the ground state to the stereochemically-determining transition states. Ground state perturbations cannot be ruled out as a contributor to the force/selectivity behavior, but the trends we observe do not immediately suggest them as the major contributor. This interpretation is consistent with previous work on other organometallic reactions employing these ligands.17 Linear relationships between Ea and force have been previously established for the oxidative addition of bromobenzene to Pd(0) complexes and for C(sp2)–C(sp2) reductive elimination from Pt(ii) and Ni(ii) diaryl complexes employing these same macrocyclic FPLs.12,13,16 The relative rigidity of the biaryl bisphosphine ligand likely contributes to the force-rate relationships by limiting the geometric deformation of the reactant and relevant transition states.14

Fig. 3. Plots of ln[er(1a)] (Δ), ln[er(1b)] (□), and ln(dr) (◯Δ) versus restoring force for the copper-catalyzed reductive coupling of 2-vinyl pyridine and acetophenone, where y = –(4.9 ± 0.4)x × 10−3 + (1.77 ± 0.05) pN−1, y = –(3.1 ± 0.3)x × 10−3 + (1.60 ± 0.03) pN−1, and y = –(1.5 ± 0.2)x × 10−3 + (0.90 ± 0.03) pN−1, respectively. Data are shown as mean ± s.d., with n = 3.

Fig. 3

Open in a new tab

Further analysis of the effect of force on the isomeric distribution of 1 reveal that the relative abundance of the 2R,3S isomer (i.e. the major enantiomer of 1b) is largely independent of force and constitutes 25 ± 1% of the isomeric mixture at all applied forces (Fig. 4). The most significant changes in the isomeric distribution as a function of force are observed for the minor enantiomers of 1a (2R,3R) and 1b (2S,3R); the relative abundance of the former isomer nearly doubles (11 to 19%) with increasing force of extension from −65 to 228 pN, while the relative abundance of the latter isomer increases four-fold (3.3 to 13%) over the same range of forces (Fig. 4). The increasing abundance of these 3R stereoisomers with increasing extension force is offset by the decreasing abundance of the 2S,3S isomer (i.e. the major enantiomer of 1a), which decreases from 62 to 44% with increasing force of extension from −65 to 228 pN (Fig. 4).

Fig. 4. Plots of the relative abundance of each of the four stereoisomers of 1 formed in the reductive coupling of 2-vinyl pyridine with acetophenone catalyzed by copper bisphosphine complexes versus restoring force.

Fig. 4

Open in a new tab

Lam has proposed a stereochemical model for the copper-catalyzed reductive coupling of vinyl arenes with ketones that invokes stereochemical-determining nucleophilic addition of an (E)-azaallyl copper intermediate (INT) to the ketone through a cyclic six-membered Zimmerman–Traxler transition state.9 This transition state model is similarly consistent with our observations and accurately predicts formation of 1a as the major diastereomer via transition states TS(2S,3S) and TS(2R,3R) in which the C3 methyl group and C2 phenyl group occupy pseudo-equatorial positions (Scheme 1). Minor diastereomer 1b is presumably formed via transition states TS(2R,3S) and TS(2R,3R) in which the C2 and C3 methyl groups occupy pseudo-equatorial positions (Scheme 1). Alternative transition states for the formation of 1b derived from a (Z)-azaallyl copper intermediate in which the C2 and C3 methyl groups occupy pseudo-axial positions appear unlikely as this pathway would require formation of diastereomer 1b with the opposite sense of enantiomeric induction relative to 1a (Scheme S3).

Scheme 1. (a) Stereochemical model for the formation of the stereoisomers of 1 via copper-catalyzed reductive coupling of 2-vinyl pyridine with acetophenone. (b) Calculated structures with ligand backbone omitted for clarity; relative free energies (kcal mol−1) in parentheses.

Scheme 1

Open in a new tab

Computational analysis

To validate the stereochemical model depicted in Scheme 1a and gain insight into the force-dependent stereoselectivity of reductive coupling, we performed DFT calculations on the key C–C bond forming, stereochemical-determining step between the copper azaallyl species derived from 2-vinylpyridine and acetophenone. DFT was performed at the (M06/6-311+G(d,p)-SDD(Cu)//B3LYP/6-31G(d)-SDD(Cu) level of theory18 with quasi-harmonic corrections19 applied to all structures. In accord with Lam's proposed mechanism, the three-coordinate, N-bound (E)-azaallyl copper complex INT was located as a stable intermediate (Scheme 1b). Importantly, both of the corresponding, diastereomeric η3-azaallyl complexes collapsed to INT upon attempted minimization, consistent with INT as a common intermediate in the formation of the four stereoisomers of 1. From the minimized structure of INT, acetophenone was brought into proximity in one of the four possible orientations. For each of these orientations, we were able to locate the corresponding transition state [TS(2S,3S), TS(2R,3R), TS(2R,3S), and TS(2S,3R)] residing 4.1–7.6 kcal mol−1 above INT (Scheme 1b). All four transition state structures adopt a distorted chair-like orientation possessing either a pseudo-equatorial phenyl (TS(2S,3S) and TS(2R,3R)) or methyl (TS(2R,3S) and TS(2S,3R)) group, in accord with our expectations derived from Lam's stereochemical model (Scheme 1).

The calculated free energies of the four diastereomeric transition states correctly predict the sense of the experimentally observed enantioselectivities: TS(2S,3S) is favored over TS(2R,3R) (ΔΔG‡ = 2.45 kcal mol−1) and TS(2R,3S) is favored over TS(2S,3R) (ΔΔG‡ = 3.42 kcal mol−1), consistent with the major enantiomers (2S,3S) and (2R,3S) for 1a and 1b, respectively. However, the corresponding Boltzmann-predicted ratios [er(1a) ≈ 63, er(1b) ≈ 321] substantially overestimate the experimentally observed values (er = 5.14 and 6.12, respectively). We suggest this overestimation may reflect, at least in part, a breakdown of the fast-pre-equilibrium assumption underlying Curtin–Hammett-type selectivity predictions, given the unusually low computed barriers for these transition states (ΔG‡ = 4.1–7.6 kcal mol−1). According to Eyring theory, the lowest barrier corresponds to a rate of k ≈ 6 × 109 s−1 at 298 K, comparable to the reorientation rate of the incoming ketone within the solvent cage, estimated via the Stokes–Einstein–Debye relation (kreorient ≈ 1 × 1010 s−1 using the viscosity of toluene).20 Because these two rates differ by only a factor of 2–3, the assumption that all four competing approach geometries equilibrate rapidly relative to bond formation cannot be taken completely; under such conditions, the observed product ratio is expected to be diluted toward a more statistical distribution relative to the pure free-energy-based (Curtin–Hammett) prediction, consistent with the direction of the discrepancy observed here.

The linear dependence of ln[er(1a)], ln[er(1b)], and ln(dr) on applied force (Fig. 4) suggests that the force-dependent stereoselectivity of reductive coupling stems from the extent of nuclear motion along the coupled vector of applied force that occur on moving from INT to the respective transition states. To evaluate this hypothesis, we analyzed the P-M-P bite angle (β), P⋯P distance (d(P⋯P)), and Biphep dihedral angle (θ) of transition state structures TS(2S,3S), TS(2R,3R), TS(2R,3S), and TS(2S,3R) relative to INT (Table 2). We focus on these largely interdependent structural parameters because they serve as effective reporters of the nuclear motions in the first coordination sphere of copper that are most strongly coupled to a tensile force applied perpendicular to the Biphep rotation axis. With a single exception, the bite angle, P–P distance, and dihedral angle for all four transition state structures are diminished relative to those of INT (Table 2), suggesting that applied tensile force effectively opposes the natural nuclear motions as INT evolves to each of the respective transition states. Therefore, these data suggest that applied tensile forces destabilizes all four transition states relative to the force-free reaction, with the extent of destabilization increasing with the magnitude of the geometric perturbation relative to INT. Although caution must be applied considering the errors associated with these small geometric perturbations, evaluation of these data supports this interpretation. Specifically, the values for β, d(P⋯P), and θ for the four transition states decrease in the order TS(2S,3R) > TS(2R,3R) > TS(2R,3S) > TS(2S,3S) (Table 2), which parallels the order of the increase in the relative abundance of the corresponding stereoisomers of 1 formed in the copper-catalysed reaction of 2-vinyl pyridine with acetophenone with increasing tensile force (Fig. 4).

Table 2. Calculated structural parameters for INT, TS(2S,3S), TS(2R,3R), TS(2R,3S), and TS(2S,3R).

Structure β (°) d (P⋯P) (Å) θ (°)
INT 105.27 3.673 80.87
TS(2S,3R) 102.43 3.651 81.26
TS(2R,3R) 102.25 3.638 80.29
TS(2R,3S) 101.23 3.633 80.11
TS(2S,3S) 100.51 3.629 79.86
Open in a new tab

Catalysis on elastomeric supports

Having established that the stereoselectivity of copper-catalyzed reductive coupling depends on force, we sought to test whether these force-coupled changes in the selectivity could be realized in the bulk employing elastomeric supported catalysts. To this end, we employed the double network (DN) polyacrylate gel DN1 that is similar to the support employed for rhodium-catalyzed enantioselective hydrogenation.7,21 Gel DN1 comprises a poly(ethyl acrylate) (PEA) first network containing the mechanically-coupled (R)-Biphep crosslinker P1 (2.5 mol%) and an interpenetrating second poly(methyl acrylate) network containing 1,4-butanediol diacrylate (BDA) crosslinkers (0.005 mol%) with total ligand concentration calculated to be 4.6 × 10−2 mmol g−1 (Fig. 5). In a similar manner, we synthesized the analogous control network DN2 in which the bisphosphine ligand P2 is covalently linked to the main chain in the first network through a single oxygen atom on the phosphine backbone and therefore decoupled from tension (Fig. 5). Double networks DN1 and DN2 were immersed in methanol solutions of Cu(OAc)2 (2 equiv./P–P) for ≥3 days, washed successively with methanol and toluene, and dried to give the copper-loaded gels DN1·Cu and DN2·Cu, respectively. UV-Vis absorption spectra of DN1·Cu displays a distinct absorption band at 296 nm that is similar to the absorption spectrum of the complex generated in situ from an equimolar mixture of (R)-MeOBiphep and Cu(OAc)2 (25 µM) in toluene, but which is absent in copper-free DN1, confirming the formation of intact (P–P)Cu groups in DN1·Cu (Fig. 6). ICP-MS analysis of the analogous rhodium-loaded DN gel synthesized through a similar procedure revealed that the support contained approximately 3.2 × 10−2 mmol Rh per g polymer, indicating that ∼70% of the P–P moieties (4.6 × 10−2 mmol gram−1) were occupied with rhodium, and we anticipate a similar extent of copper loading in DN1·Cu and DN2·Cu.

Fig. 5. Schematic representation of the synthesis and characteristics of double network polymers DN1·Cu and DN2·Cu for MMC.

Fig. 5

Open in a new tab

Fig. 6. UV-vis absorption spectra of DN1·Cu (Δ) and an incubated solution of (R)-MeOBiphep and Cu(OAc)2 (25 µM) in toluene (◯).

Fig. 6

Open in a new tab

We analyzed the stereoselectivity of the reductive coupling of 2-vinyl pyridine with acetophenone as a function of strain employing supported catalyst DN1·Cu under conditions that approximated those used in the small molecule system. The catalytically active gel DN1·Cu (∼63 mg, 4.2 µmol Cu) was mounted between parallel stainless-steel plates, compressed manually to the desired strain state, and fully immersed in a toluene solution of 2-vinylpyridine, acetophenone, and PhSiH3. The surrounding solution was stirred at room temperature for 7 days and then analyzed by 1H NMR spectroscopy and HPLC. The supported catalyst was then washed with toluene and recycled for subsequent runs (Table 3, entries 1–4, Fig. 7). As strain (ε) increases from ε = 0 to 30 to 50%, the 1a : 1b ratio decreases from 1.9 to 1.8 to 1.7, the enantiomeric ratio of 1a decreases from 2.88 to 2.69, to 2.58, and the enantiomeric ratio of 1b decreases from 2.77 to 2.64, to 2.48 (Table 3, entries 1–3, Fig. 7). Importantly, each of these values return to their basal level [dr = 1.9, er(1a) = 2.87, er(1b) = 2.72] when DN1·Cu is relaxed in the fourth cycle (Table 3, entry 4).

Table 3. Effect of compressive strain on the reductive coupling of 2-vinylpyridine (0.20 M) with acetophenone (0.22 M) and phenylsilane (0.24 M) catalyzed by DN1·Cu or DN2·Cu (∼63 mg, 4.2 µmol Cu) in toluene at 25 °C for 7 days.

Entry catalysta Strainb (%) Convn. (%) er(1a) er(1b) 1a : 1b
1 DN1·Cu 0 >99 2.88 ± 0.03 2.77 ± 0.05 1.9 ± 0.02
2 DN1·Cu 30 90 2.69 ± 0.02 2.64 ± 0.02 1.8 ± 0.06
3 DN1·Cu 50 60 2.58 ± 0.04 2.48 ± 0.04 1.7 ± 0.03
4 DN1·Cu 0 >99 2.87 ± 0.03 2.72 ± 0.04 1.9 ± 0.02
5 DN2·Cu 0 >99 3.20 ± 0.05 4.25 ± 0.06 1.8 ± 0.03
6 DN2·Cu 60 95 3.18 ± 0.06 4.28 ± 0.03 1.8 ± 0.05
Open in a new tab
a

Single samples of DN1·Cu (entries 1–4) and DN2·Cu (entries 5 and 6) were employed in these experiments.

b

Strain (ɛ) of the compressed network was calculated as ε = (Li − Lf)/Li, where Li is the initial (ɛ = 0) distance between the plates and Lf is the distance under compression. Values are reported as the mean of three replicate measurements; uncertainties correspond to the standard deviation of the mean (n = 3).

Fig. 7. Plots of er 1a (Δ), er 1b (□), and dr (◯) versus compressive strain for the reductive coupling of 2-vinyl pyridine with acetophenone catalyzed by DN1·Cu. Data are shown as mean ± s.d., with n = 3.

Fig. 7

Open in a new tab

To confirm that the strain-dependent stereoselectivity of the reductive coupling of 2-vinyl pyridine with acetophenone catalyzed by DN1·Cu was due to coupled force, we evaluated the strain-dependent stereoselectivity of reductive coupling catalyzed by the mechanically decoupled catalyst DN2·Cu under the same reaction conditions employed for catalysis with DN1·Cu. Importantly, neither the diastereomeric ratio nor the enantiomeric ratios of 1a or 1b were significantly (<1%) perturbed when DN2·Cu was compressed to ε = 60% (Table 3, entries 5 and 6), as compared to the >10% change in these values for reductive coupling catalyzed by DN1·Cu at ε = 50%. These observations support the mechanical origin of the strain-dependent modulation of selectivity observed for reductive coupling catalyzed by DN1·Cu.

Several observations regarding the reductive coupling of 2-vinylpyridine with acetophenone on supported copper bisphosphine catalysts are worthy of comment. The rate of reductive coupling catalyzed by DN1·Cu appears to decrease with increasing compression as judged by the decrease in conversion from >99 to 90 to 60% as the applied strain increases from ε = 0 to 30 to 50% in subsequent runs (Table 3, entries 1–3), although the origin of this effect remains unclear. Catalyst deactivation or leaching can be safely ruled out as conversion returns to its nascent value (>99%) when DN1·Cu is relaxed in the fourth reaction cycle (Table 3, entry 4). Although we anticipate a decrease in the rate of mass transport into the network with increasing compression owing to the concomitant decrease in the accessible surface area and increasing lateral distance of the gel,22 diminished mass transport alone does not appear to account for the observed rate decrease owing to the absence of similar behavior in the case of DN2·Cu (Table 3, entries 5 and 6). We likewise cannot fully rule out a decrease in the overall rate of reductive coupling with the increasing tension generated at the individual sites through macroscopic compression; however, we observed no obvious dependence of the rate of reductive coupling on restoring force for reactions catalyzed by homogeneous (FPL)Cu complexes (Table 1).

We similarly note that the enantioselectivity of reductive coupling catalyzed by mechanically unstrained (ε = 0%) DN1·Cu is lower than that of reductive coupling catalyzed by mechanically unstrained DN2·Cu (Table 3, entries 1 and 5). This behavior can be safely attributed to the latent tension in DN1·Cu generated via solvent swelling that is absent in the mechanically-decoupled supported catalyst DN2·Cu. Indeed, the tension generated by solvent swelling in cross-linked poly(methyl methacrylate) is sufficient to cleave covalent bonds23 and we have shown that differential three-dimensional swelling of a supported rhodium/poly(methyl methacrylate) catalyst enhances the enantioselectivity of the hydrogenation of methyl 2-acetamidoacrylate (2) more efficiently than does uniaxial mechanical deformation, when the effect is characterized as a function of the extension in a single, primary dimension.24

In addition to the differential reactivity of the unstrained DNs, the stereoselectivity of reductive coupling catalyzed by either DN1·Cu or DN2·Cu at ε = 0% is significantly lower than is the stereoselectivity of reductive coupling catalyzed by the force-free homogeneous catalyst generated from Cu(OAc)2 and (R)-MeOBiphep (Table 1, entry 3; Table 3, entries 1 and 5). We considered that the lower stereoselectivity of reductive coupling on supported catalysts vis-a-vis homogeneous catalysts might be due to either (1) medium effects owing to the more polar, carbonyl-rich environment in the interior of the supported catalysts relative to the homogeneous medium (i.e. toluene) and/or (2) catalysis away from the chiral P–P binding site resulting from off-site binding of copper within the network. Supporting the former rationale, the reductive coupling of 2-vinylpyridine with acetophenone catalyzed by Cu(OAc)2/(R)-MeOBiphep in a homogeneous mixture of polymethylacrylate (PMA) and toluene (v/v = 0.017 : 1) (designed to mimic the medium in the interior of the toluene-swollen gels) is both slower and less stereoselective [dr = 1.4, er(1a) = 3.56, er(1b) = 2.81; Table 4, entry 1] than is catalysis by Cu(OAc)2/(R)-MeOBiphep in toluene (Table 1, entry 3). Arguing against catalysis away from the P–P binding site, reaction of 2-vinylpyridine with acetophenone and a catalytic amount of Cu(OAc)2 in PMA/toluene (v/v = 0.017) in the absence of (R)-MeOBiphep led to no detectable consumption of starting materials after 15 h at 25 °C (Table 4, entry 2), strongly suggesting that only phosphine-bound copper centers are active within the gel. The lower stereoselectivity of reductive coupling in PMA/toluene mixtures relative to neat toluene is consistent with the proposed cyclic transition state involving a Lewis acid/Lewis base interaction between copper and the acetophenone oxygen atom (Scheme 1). We posit that this cyclic transition state may be disrupted by bonding of copper to a carbonyl group in the PMA matrix leading to carbon–carbon bond formation via open, and presumably less selective, transition states.

Table 4. Reaction of 2-vinylpyridine (0.20 M), acetophenone (0.22 M), and phenylsilane (0.24 M) catalyzed by Cu(OAc)2 (10 mM) with and without (R)-MeOBiphep in a homogeneous mixture of toluene and polymethylacrylate (PMA) (v/v = 0.017).

graphic file with name d6sc01654e-u2.jpg
P–P (mM) Convn (%) er(1a) er(1b) 1a : 1b
10 80 3.56 2.81 1.4 : 1
0 0 — — —
Open in a new tab

Lastly, we note that the stereoselectivity of reductive coupling catalyzed by DN1·Cu is ∼8 times less sensitive to force than is the enantioselectivity of the hydrogenation of methyl 2-acetamidoacrylate (2) employing an analogous supported rhodium bisphosphine catalyst (Fig. 1).7 Specifically, the dr, er(1a), and er(1b) of reductive coupling catalyzed by DN1·Cu all decreased by ∼10% over 50% compressive strain, whereas the er of the hydrogenation of 2 employing the supported rhodium catalyst increased by ∼80% (2.12 to 3.76) over the same range of compressive strains at the same temperature.7 The greater strain sensitivity of hydrogenation catalyzed by the supported Rh catalyst can be attributed in part to the greater force sensitivity of Rh-catalyzed hydrogenation relative to copper-catalyzed reductive coupling. For example, the er of methyl acetylalaninate formed via hydrogenation of 2 catalyzed by homogeneous (FPL)Rh+ complexes increases by a factor of 11 over a ∼280 pN range of extension forces,7 whereas the er of 1b (i.e. the most sensitive measurable) formed via reductive coupling of 2-vinylpyridine with acetophenone catalyzed by homogeneous (FPL)Cu complexes decreases by a factor of 3.7 across the same range of forces. However, the ∼3-fold greater force sensitivity of Rh-catalyzed hydrogenation in solution relative to reductive coupling does not fully account for the ∼8-fold greater strain sensitivity of hydrogenation relative to reductive coupling on supported catalysts. Because we ruled out catalyst leaching and/or catalysis away from P–P binding site for reductive coupling catalyzed by DN1·Cu, the lower strain/selectivity dependence of copper-catalyzed reductive coupling relative to rhodium-catalyzed hydrogenation is likely tied to medium effects in the supported copper catalyst. That is, the more polar environment in the interior of DN1·Cu relative to neat toluene diminishes both the stereoselectivity and the force sensitivity of stereoselectivity of reductive coupling catalyzed by DN1·Cu. This observation is also consistent with the carbonyl-rich environment within the network promoting catalysis via open, and presumably less force-sensitive, transition states relative to the cyclic transition states that predominate in the non-polar medium employed in the homogeneous system. The enantioselectivity of Rh-catalyzed hydrogenation, in contrast, is largely insensitive to medium effects.25

Conclusions

We have shown that the stereoselectivity of the reductive coupling of 2-vinylpyridine with acetophenone catalyzed by molecular copper bisphosphine complexes is modulated by force applied to the ligand backbone. Extension forces of ∼280 pN decrease the enantiomeric excesses of major diastereomer 1a from 70% to 40% ee and minor diastereomer 1b from 76% to 32% ee and decrease the diastereoselectivity from 73% to 63%. DFT calculations support a model for the observed force-dependent stereoselectivity of reductive coupling in which the competing transition states are differentially destabilized by the applied tensile force. The degree of destabilization is determined by the extent to which the nuclear motions that occur on going from the ground state to the transition state are opposed by the applied tensile force. Copper-catalyzed reductive coupling represents a rare example of a metal-catalyzed reaction for which force-dependent stereoselectivity has been documented and the lone example in which stereoselectivity decreases with increasing forces of extension.7,10,11

We have likewise shown that the stereoselectivity of reductive coupling catalyzed by copper bisphosphine groups embedded as tension-bearing crosslinkers in a polyacrylate network is modulated by extrinsic mechanical force. Although the effects of strain on selectivity are small (∼10%) and lead to a decrease in selectivity, copper-catalyzed reductive coupling represents only the second documented example of multi-state mechanocatalysis (MMC).7,24 The modest force/selectivity response of reductive coupling on supported copper catalysts relative to the corresponding homogeneous catalysts was traced in part to the polar, carbonyl-rich environment within the network that presumably disrupts the cyclic, six-membered transition state for C–C bond formation that predominates in non-polar medium. In any event, these observations encourage the development of a more diverse family of elastomeric supports for MMC of varied microstructure and polarity.

Despite the modest force/sensitivity response of reductive coupling on supported copper catalysts, the results described here nevertheless encourage the future development of MMC, specifically the identification of polymer supports able to transduce and focus even greater force free from detrimental medium effects. For example, extrapolation of the linear relationship between ln(isomer ratio) and force established for reductive coupling catalyzed by homogeneous (FPL)Cu complexes suggests that if forces of 1 nN could be transmitted uniformly to active sites within a supported copper catalyst in the absence of medium effects, the 2S,3R stereoisomer of 1, which constitutes only ∼3% of the isomeric mixture under force-free conditions, would represent the major stereoisomer of reductive coupling, formed with 1.5 : 1 dr and 92% ee.

Although the polymer supports described here and in our related work are specifically tailored to transduce extrinsic mechanical force to the active site, it appears plausible that many extant supported catalysts, particularly those in which the metal is chelated to multiple main-chain phosphine or pyridine moieties,1 may possess undocumented mechanochemical properties. Furthermore, because (1) solvent swelling generates latent tension within a network and (2) the extent of swelling is solvent dependent,26 it is similarly plausible that mechanochemical effects contribute to the medium effects observed in supported catalysis.1

Experimental

Reductive coupling catalyzed by molecular (P–P)Cu complexes

A solution of 2-vinylpyridine (41 mg, 0.39 mmol), acetophenone (52 mg, 0.43 mmol), Cu(OAc)2 (3.6 mg, 0.020 mmol), and (R)-MeOBiphep (12 mg, 0.020 mmol) in toluene (2 mL) was stirred at 4 °C for 15 min. Phenylsilane (59 µL, 0.48 mmol, 1.2 eq.) was then added dropwise via syringe over 1 min to the resulting solution. The reaction mixture was stirred at 4 °C for 30 min and then room temperature for 15 h. Silica gel (200 mg) was added to the reaction mixture and the resulting suspension was stirred for 15 min, filtered through a short plug of silica gel eluting with EtOAc (10 mL), and concentrated under vacuum. The diastereomeric ratio was determined via1H NMR analysis of the resulting residue by integrating the pyridyl aromatic resonances at δ 8.36 (1a) and δ 8.56 (1b) and the C3 methine resonances at δ 3.28 (1a) and δ 3.15 (1b). The residue was then chromatographed (SiO2; 10–15% EtOAc/hexane) to give 1a (22 mg, 36% yield) as colorless oil and 1b (7.4 mg, 27% yield) as white solid. The 1H NMR spectra of 1a and 1b were consistent with published spectra.9 The enantiomeric purity of 1a (67.4% ee) was determined by HPLC equipped with a Chiralpak AD-H column (98 : 2 hexane:i-PrOH 0.8 mL min−1, 254 nm, 25 °C); tr (minor) = 12.2 min, tr (major) = 13.4 min (Fig. S6). The enantiomeric purity of 1b (71.9% ee) was determined by HPLC equipped with a Chiralcel OD-H column (98 : 2 hexane:i-PrOH, 0.8 mL min−1, 280 nm, 25 °C); tr (major) = 9.8 min, tr (minor) = 11.5 min (Figure S12). The copper-catalyzed reductive coupling of 2-vinylpyridine with acetophenone employing force probe ligands Z(2,2), Z(3,3), E(2,3), and E(3,3) was performed employing analogous procedures. All reported er and dr values represent the mean of three replicate measurements (±s.d., n = 3) (Fig. S7–S10, S13–S16).

Reductive coupling catalyzed by DN1·Cu and DN2·Cu

Supported catalyst DN1·Cu of dimensions ∼5 mm × ∼5 mm × ∼1.1 mm (∼63 mg, ∼4.2 µmol Cu) was secured without additional compression between the plates of a compressing device, built in house, comprising two 21 mm-diameter stainless steel plates attached to one another via four machine screws which control the separation of the plates (Fig. S4 and S5). The secured gel and device were then fully immersed in a solution of 2-vinylpyridine (42 mg, 0.40 mmol) and acetophenone (53 mg, 0.44 mmol) in toluene (2 mL) and the solution was stirred at 4 °C for 15 min. Phenylsilane (59 µL, 0.48 mmol) was then added dropwise to the solution over 1 min and the resulting solution was stirred at 4 °C for 30 min and room temperature for 96 h. Silica gel was added to the reaction mixture and the resulting suspension was stirred for 15 min and filtered through a short plug of silica gel eluting with EtOAc (5 mL). The filtrate was concentrated under vacuum and the resulting residue was chromatographed (10% EtOAc/hexane) to give 1a (19 mg, 32%) as a colorless oil and 1b (7.5 mg, 24%) as white solid (Table S3). The diastereomeric ratio and enantiomeric ratios of 1a and 1b were determined as described above employing molecular catalysts. All er and dr values are means of three replicate measurements (±s.d., n = 3).

Employing analogous procedures, the reductive coupling of 2-vinylpyridine and acetophenone catalyzed DN1·Cu was performed at ε = 30%, 50% and 0% employing the same sample of DN1·Cu, which was rinsed three times with toluene between experiments (Table 3). For each experiment, the desired strain (ɛ) of the DN1·Cu gel was achieved by tightening the machine screws connecting the two stainless steel plates to the targeted gel thickness (Fig. S4 and S5), which was measured with a set of calipers. The strain (ε) of the compressed network was calculated as ε = (Li − Lf)/Li, where Li is the initial (ɛ = 0) distance between the plates and Lf is the distance under compression. The reductive coupling of 2-vinylpyridine with acetophenone catalyzed by DN2·Cu as a function of uniaxial compression was performed employing a procedure similar to that used for the reductive coupling catalyzed by DN1·Cu (Table 3).

Computational methods

DFT calculations were performed using Gaussian 16 (ref. 27) with the SMD solvation model for toluene.18b Stationary points were characterized by frequency analysis, and thermochemical corrections at 298.15 K were evaluated using GoodVibes with quasi-harmonic treatments.19 Transition states were located by relaxed potential-energy-surface scans followed by transition-state optimization. Additional computational details are provided in the SI.

Author contributions

S. L. C. and R. A. W. conceived the project. C. D. and X. Z. designed and carried out the experiments and characterized the products. J. Z. performed computational analysis. R. L. B. performed the data collection for some complexes. C. D., J. Z., S. L. C., and R. A. W. analysed data and prepared the manuscript.

Conflicts of interest

There are no conflicts to declare.

Supplementary Material

SC-OLF-D6SC01654E-s001

Acknowledgments

This work was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences program under Award Number DE-SC0022146. Computational support was generously provided by the Duke Computing Cluster (DCC).

Data availability

CCDC 2528321 (rac-1b), 2528326 ((2R,3R,R)-S1) and 2528327 ((2R,3S,R)-S2) contain the supplementary crystallographic data for this paper.28a–c

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: stereochemical analyses of 1a and 1b, synthesis of supported catalysts DN1·Cu and DN2·Cu; computational analysis; scans of HPLC traces and NMR spectra. See DOI: https://doi.org/10.1039/d6sc01654e.

Notes and references

  1. (a) Chauvin Y. Commereuc D. Dawans F. Prog. Polym. Sci. 1977;5:95–226. doi: 10.1016/0079-6700(77)90001-6. [DOI] [Google Scholar]; (b) Haag R. Roller S. Top. Curr. Chem. 2004;242:1–42. doi: 10.1007/b96865. [DOI] [PubMed] [Google Scholar]; (c) Clapham B. Reger T. S. Janda K. D. Tetrahedron. 2001;57:4637–4662. doi: 10.1016/S0040-4020(01)00298-8. [DOI] [Google Scholar]; (d) Madhavan N. Jones C. W. Weck M. Acc. Chem. Res. 2008;41:1153–1165. doi: 10.1021/ar800081y. [DOI] [PubMed] [Google Scholar]; (e) Lu J. Toy P. H. Chem. Rev. 2009;109:815–838. doi: 10.1021/cr8004444. [DOI] [PubMed] [Google Scholar]; (f) Kann N. Molecules. 2010;15:6306–6331. doi: 10.3390/molecules15096306. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. (a) Sun Q. Dai Z. Liu X. Sheng N. Deng F. Meng X. Xiao F.-S. J. Am. Chem. Soc. 2015;137:5204–5209. doi: 10.1021/jacs.5b02122. [DOI] [PubMed] [Google Scholar]; (b) Zhang Y. Riduan S. N. Chem. Soc. Rev. 2012;41:2083–2094. doi: 10.1039/C1CS15227K. [DOI] [PubMed] [Google Scholar]; (c) Kaur P. Hupp J. T. Nguyen S. T. ACS Catal. 2011;1:819–835. doi: 10.1021/cs200131g. [DOI] [Google Scholar]; (d) Kramer S. Bennedsen N. R. Kegnæs S. ACS Catal. 2018;8:6961–6982. doi: 10.1021/acscatal.8b01167. [DOI] [Google Scholar]; (e) Sun Q. Dai Z. Meng X. Xiao F.-S. Chem. Soc. Rev. 2015;44:6018–6034. doi: 10.1039/C5CS00198F. [DOI] [PubMed] [Google Scholar]
  3. (a) Li J. Nagamani C. Moore J. S. Acc. Chem. Res. 2015;48:2181–2190. doi: 10.1021/acs.accounts.5b00184. [DOI] [PubMed] [Google Scholar]; (b) O'Neill R. T. Boulatov R. Nat. Rev. Chem. 2021;5:148–167. doi: 10.1038/s41570-020-00249-y. [DOI] [PubMed] [Google Scholar]; (c) Hu H. Ma Z. Jia X. Mater. Chem. Front. 2020;4:3115–3129. doi: 10.1039/D0QM00435A. [DOI] [Google Scholar]; (d) Lloyd E. M. Vakil J. R. Yao Y. Sottos N. R. Craig S. L. J. Am. Chem. Soc. 2023;145:751–768. doi: 10.1021/jacs.2c09623. [DOI] [PubMed] [Google Scholar]
  4. (a) Wang L. Zheng X. Kouznetsova T. B. Yen T. Ouchi T. Brown C. L. Craig S. L. J. Am. Chem. Soc. 2022;144:22865–22869. doi: 10.1021/jacs.2c10878. [DOI] [PubMed] [Google Scholar]; (b) Brown C. L. Bowser B. H. Meisner J. Kouznetsova T. B. Seritan S. Martinez T. J. Craig S. L. J. Am. Chem. Soc. 2021;143:3846–3855. doi: 10.1021/jacs.0c12088. [DOI] [PubMed] [Google Scholar]; (c) Lenhardt J. M. Ong M. T. Choe R. Evenhuis C. R. Martinez T. J. Craig S. L. Science. 2010;329:1057–1060. doi: 10.1126/science.1193412. [DOI] [PubMed] [Google Scholar]
  5. (a) Piermattei A. Karthikeyan S. Sijbesma R. P. Nat. Chem. 2009;1:133–137. doi: 10.1038/nchem.167. [DOI] [PubMed] [Google Scholar]; (b) Jakobs R. T. M. Sijbesma R. P. Organometallics. 2012;31:2476–2481. doi: 10.1021/om300161z. [DOI] [Google Scholar]; (c) Funtan S. Michael P. Binder W. H. Biomimetics. 2019;4:24. doi: 10.3390/biomimetics4010024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. (a) Michael P. Binder W. H. Angew. Chem., Int. Ed. 2015;54:13918–13922. doi: 10.1002/anie.201505678. [DOI] [PubMed] [Google Scholar]; (b) Jakobs R. T. M. Ma S. Sijbesma R. P. ACS Macro Lett. 2013;2:613–616. doi: 10.1021/mz400201c. [DOI] [PubMed] [Google Scholar]; (c) Biewend M. Michael P. Binder W. H. Soft Matter. 2020;16:1137–1141. doi: 10.1039/C9SM02185J. [DOI] [PubMed] [Google Scholar]; (d) Döhler D. Michael P. Binder W. H. Acc. Chem. Res. 2017;50:2610–2620. doi: 10.1021/acs.accounts.7b00371. [DOI] [PubMed] [Google Scholar]
  7. Zheng X. Chiou C.-Y. O'Neill R. T. Duan C. Yu Y. Malek J. Rivera Jr. N. A. Boulatov R. Craig S. L. Widenhoefer R. A. Nat. Synth. 2025;4:1319–1328. doi: 10.1038/s44160-025-00839-y. [DOI] [Google Scholar]
  8. (a) Shao X. Li K. Malcolmson S. J. J. Am. Chem. Soc. 2018;140:7083–7087. doi: 10.1021/jacs.8b04750. [DOI] [PMC free article] [PubMed] [Google Scholar]; (b) Choi B. Saxena A. Smith J. J. Churchill G. H. Lam H. W. Synlett. 2015;26:350–354. [Google Scholar]; (c) Liu Z.-C. Yue W.-J. Yin L. J. Org. Chem. 2022;87:399–405. doi: 10.1021/acs.joc.1c02426. [DOI] [PubMed] [Google Scholar]; (d) Suzuki H. Yoneoka K. Kondo S. Matsuda T. Chem. – Eur. J. 2022;28:e202104273. doi: 10.1002/chem.202104273. [DOI] [PubMed] [Google Scholar]; (e) Smith J. J. Best D. Lam H. W. Chem. Commun. 2016;52:3770–3772. doi: 10.1039/C6CC00603E. [DOI] [PubMed] [Google Scholar]; (f) Wang S.-Q. Liu Z.-C. Yue W.-J. Yin L. Angew. Chem., Int. Ed. 2021;60:4604–4608. doi: 10.1002/anie.202013207. [DOI] [PubMed] [Google Scholar]; (g) Best D. Lam H. W. J. Org. Chem. 2014;79:831–845. doi: 10.1021/jo402414k. [DOI] [PubMed] [Google Scholar]
  9. (a) Saxena A. Choi B. Lam H. W. J. Am. Chem. Soc. 2012;134:8428–8431. doi: 10.1021/ja3036916. [DOI] [PubMed] [Google Scholar]; (b) Bandar J. S. Ascic E. Buchwald S. L. J. Am. Chem. Soc. 2016;138:5821–5824. doi: 10.1021/jacs.6b03086. [DOI] [PMC free article] [PubMed] [Google Scholar]; (c) Shi S.-L. Wong Z. L. Buchwald S. L. Nature. 2016;532:353–356. doi: 10.1038/nature17191. [DOI] [PMC free article] [PubMed] [Google Scholar]; (d) Li C. Shin K. Liu R. Y. Buchwald S. L. Angew. Chem., Int. Ed. 2019;58:17074–17080. doi: 10.1002/anie.201911008. [DOI] [PMC free article] [PubMed] [Google Scholar]; (e) Liu R. Y. Zhou Y. Yang Y. Buchwald S. L. J. Am. Chem. Soc. 2019;141:2251–2256. doi: 10.1021/jacs.8b13907. [DOI] [PMC free article] [PubMed] [Google Scholar]; (f) Liu R. Y. Buchwald S. L. Acc. Chem. Res. 2020;53:1229–1243. doi: 10.1021/acs.accounts.0c00164. [DOI] [PMC free article] [PubMed] [Google Scholar]; (g) Dong Y. Schuppe A. W. Mai B. K. Liu P. Buchwald S. L. J. Am. Chem. Soc. 2022;144:5985–5995. doi: 10.1021/jacs.2c00734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Yu Y. Zheng X. Duan C. Craig S. L. Widenhoefer R. A. ACS Catal. 2022;12:13941–13950. doi: 10.1021/acscatal.2c04534. [DOI] [Google Scholar]
  11. Kean Z. S. Akbulatov S. Tian Y. Widenhoefer R. A. Boulatov R. Craig S. L. Angew. Chem., Int. Ed. 2014;53:14508–14511. doi: 10.1002/anie.201407494. [DOI] [PubMed] [Google Scholar]
  12. Wang L. Yu Y. Razgoniaev A. O. Johnson P. N. Wang C. Tian Y. Boulatov R. Craig S. L. Widenhoefer R. A. J. Am. Chem. Soc. 2020;142:17714–17720. doi: 10.1021/jacs.0c08506. [DOI] [PubMed] [Google Scholar]
  13. Yu Y. Wang C. Wang L. Sun C. L. Boulatov R. Widenhoefer R. A. Craig S. L. Chem. Sci. 2021;12:11130–11137. doi: 10.1039/D1SC03182A. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Yu Y. O'Neill R. T. Boulatov R. Widenhoefer R. A. Craig S. L. Nat. Commun. 2023;14:5074. doi: 10.1038/s41467-023-40842-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Akbulatov S. Tian Y. Boulatov R. J. Am. Chem. Soc. 2012;134:7620–7623. doi: 10.1021/ja301928d. [DOI] [PubMed] [Google Scholar]
  16. (a) Duan C. Malek J. C. Craig S. L. Widenhoefer R. A. ChemCatChem. 2024;16:e202301479. doi: 10.1002/cctc.202301479. [DOI] [Google Scholar]; (b) Duan C. Zheng X. Craig S. L. Widenhoefer R. A. Organometallics. 2023;42:1918–1926. doi: 10.1021/acs.organomet.3c00168. [DOI] [Google Scholar]
  17. (a) Ducrot E. Chen Y. Bulters M. Sijbesma R. P. Creton C. Science. 2014;344:186–189. doi: 10.1126/science.1248494. [DOI] [PubMed] [Google Scholar]; (b) Millereau P. Ducrot E. Clough J. M. Wiseman M. E. Brown H. R. Sijbesma R. P. Creton C. Proc. Natl. Acad. Sci. U.S.A. 2018;115:9110–9115. doi: 10.1073/pnas.1807750115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. (a) Becke A. D. J. Chem. Phys. 1993;98:5648–5652. doi: 10.1063/1.464913. [DOI] [Google Scholar]; (b) Lee C. Yang W. Parr R. G. Phys. Rev. B. 1988;37:785–789. doi: 10.1103/PhysRevB.37.785. [DOI] [PubMed] [Google Scholar]; (c) Hehre W. J., Radom L., Schleyer P. v. R. and Pople J. A., Ab Initio Molecular Orbital Theory, Wiley, New York, 1986 [Google Scholar]; (d) Figgen D. Rauhut G. Dolg M. Stoll H. Chem. Phys. 2005;311:227–244. doi: 10.1016/j.chemphys.2004.10.005. [DOI] [Google Scholar]; (e) Marenich A. V. Cramer C. J. Truhlar D. G. J. Phys. Chem. B. 2009;113:6378–6396. doi: 10.1021/jp810292n. [DOI] [PubMed] [Google Scholar]; (f) Zhao Y. Truhlar D. G. Theor. Chem. Acc. 2008;120:215–241. [Google Scholar]
  19. (a) Luchini G. Alegre-Requena J. V. Funes-Ardoiz I. Paton R. S. F1000Research. 2020;9:291. [Google Scholar]; (b) Ribeiro R. F. Marenich A. V. Cramer C. J. Truhlar D. G. J. Phys. Chem. B. 2011;115:14556–14562. doi: 10.1021/jp205508z. [DOI] [PubMed] [Google Scholar]; (c) Li Y. Gomes J. Sharada S. M. Bell A. T. Head-Gordon M. J. Phys. Chem. C. 2015;119:1840–1850. doi: 10.1021/jp509921r. [DOI] [Google Scholar]
  20. (a) Seeman J. I. Chem. Rev. 1983;83:83–134. doi: 10.1021/cr00054a001. [DOI] [Google Scholar]; (b) Debye P., Polar Molecules, Chemical Catalog Company, New York, 1929 [Google Scholar]; (c) CRC Handbook of Chemistry and Physics, ed. L. R. McEwen, CRC Press, Boca Raton, FL, 2026, “Viscosity of Toluene at 298 K” [Google Scholar]
  21. (a) Matsuda T. Kawakami R. Namba R. Nakajima T. Gong J. P. Science. 2019;363:504–508. doi: 10.1126/science.aau9533. [DOI] [PubMed] [Google Scholar]; (b) Wang Z. Zheng X. Ouchi T. Kouznetsova T. B. Beech H. K. Av-ron S. Matsuda T. Bowser B. H. Wang S. Johnson J. A. Kalow J. A. Olsen B. D. Gong J. P. Rubinstein M. Craig S. L. Science. 2021;374:193–196. doi: 10.1126/science.abg2689. [DOI] [PubMed] [Google Scholar]
  22. (a) Schmidt J. Kundu D. S. Blechert S. Thomas A. Chem. Commun. 2014;50:3347–3349. doi: 10.1039/C3CC49220F. [DOI] [PubMed] [Google Scholar]; (b) Liu H. Lin Y. Ma Z. Chin. J. Catal. 2016;37:73–82. doi: 10.1016/S1872-2067(15)60951-2. [DOI] [Google Scholar]
  23. (a) Clough J. M. van der Gucht J. Sijbesma R. P. Macromolecules. 2017;50:2043–2053. doi: 10.1021/acs.macromol.6b02540. [DOI] [PMC free article] [PubMed] [Google Scholar]; (b) Lee C. K. Diesendruck C. E. Lu E. Pickett A. N. May P. A. Moore J. S. Braun P. V. Macromolecules. 2014;47:2690–2694. doi: 10.1021/ma500195h. [DOI] [Google Scholar]; (c) Metze F. K. Sant S. Meng Z. Klok H.-A. Kaur K. Langmuir. 2023;39:3546–3557. doi: 10.1021/acs.langmuir.2c02801. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Zheng X. Duan C. Widenhoefer R. A. Craig S. L. J. Am. Chem. Soc. 2025;147:31085–31090. doi: 10.1021/jacs.5c09442. [DOI] [PubMed] [Google Scholar]
  25. Etayo P. Vidal-Ferran A. Chem. Soc. Rev. 2013;42:728–754. doi: 10.1039/C2CS35410A. [DOI] [PubMed] [Google Scholar]
  26. Watabe T. Otsuka H. Angew. Chem., Int. Ed. 2023;62:e20221646. doi: 10.1002/anie.202216469. [DOI] [PubMed] [Google Scholar]
  27. Frisch M. J., Trucks G. W., Schlegel H. B., Scuseria G. E., Robb M. A., Cheeseman J. R., Scalmani G., Barone V., Petersson G. A., Nakatsuji H., Li X., Caricato M., Marenich A. V., Bloino J., Janesko B. G., Gomperts R., Mennucci B., Hratchian H. P., Ortiz J. V., Izmaylov A. F., Sonnenberg J. L., Williams-Young D., Ding F., Lipparini F., Egidi F., Goings J., Peng B., Petrone A., Henderson T., Ranasinghe D., Zakrzewski V. G., Gao J., Rega N., Zheng G., Liang W., Hada M., Ehara M., Toyota K., Fukuda R., Hasegawa J., Ishida M., Nakajima T., Honda Y., Kitao O., Nakai H., Vreven T., Throssell K., Montgomery Jr. J. A., Peralta J. E., Ogliaro F., Bearpark M. J., Heyd J. J., Brothers E. N., Kudin K. N., Staroverov V. N., Keith T. A., Kobayashi R., Normand J., Raghavachari K., Rendell A. P., Burant J. C., Iyengar S. S., Tomasi J., Cossi M., Millam J. M., Klene M., Adamo C., Cammi R., Ochterski J. W., Martin R. L., Morokuma K., Farkas O., Foresman J. B. and Fox D. J., Gaussian 16, Revision A.03, Gaussian, Inc., Wallingford CT, 2016 [Google Scholar]
  28. (a) CCDC 2528321: Experimental Crystal Structure Determination, 2026, 10.5517/ccdc.csd.cc2qvxr8 [DOI]; (b) CCDC 2528326: Experimental Crystal Structure Determination, 2026, 10.5517/ccdc.csd.cc2qvxxf [DOI]; (c) CCDC 2528327: Experimental Crystal Structure Determination, 2026, 10.5517/ccdc.csd.cc2qvxyg [DOI]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Citations

  1. (a) CCDC 2528321: Experimental Crystal Structure Determination, 2026, 10.5517/ccdc.csd.cc2qvxr8 [DOI]
  2. (b) CCDC 2528326: Experimental Crystal Structure Determination, 2026, 10.5517/ccdc.csd.cc2qvxxf [DOI]
  3. (c) CCDC 2528327: Experimental Crystal Structure Determination, 2026, 10.5517/ccdc.csd.cc2qvxyg [DOI]

Supplementary Materials

SC-OLF-D6SC01654E-s001
SC-OLF-D6SC01654E-s001.pdf (6.8MB, pdf)

Data Availability Statement

CCDC 2528321 (rac-1b), 2528326 ((2R,3R,R)-S1) and 2528327 ((2R,3S,R)-S2) contain the supplementary crystallographic data for this paper.28a–c

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: stereochemical analyses of 1a and 1b, synthesis of supported catalysts DN1·Cu and DN2·Cu; computational analysis; scans of HPLC traces and NMR spectra. See DOI: https://doi.org/10.1039/d6sc01654e.

Articles from Chemical Science are provided here courtesy of Royal Society of Chemistry

RESOURCES