Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

A boron-transfer mechanism mediating the thermally induced revival of frustrated carbene–borane pairs from their shelf-stable adducts


Chemists have designed strategies that trigger the conformational isomerization of molecules in response to external stimuli, which can be further applied to regulate the complexation between Lewis acids and bases. We have recently developed a system in which frustrated carbene–borane pairs are revived from shelf-stable but external-stimuli-responsive carbene–borane adducts comprised of N-phosphine-oxide-substituted imidazolylidenes (PoxIms) and triarylboranes. Herein, we report the detailed mechanism on this revival process. A thermally induced borane-transfer process from the carbene carbon atom to the N-phosphinoyl oxygen atom initiates the transformation of the carbene–borane adduct. Subsequent conformational isomerization via the rotation of the N-phosphinoyl group in PoxIm moieties eventually leads to the revival of frustrated carbene–borane pairs that can cleave H2. We believe that this work illustrates an essential role of dynamic conformational isomerization in the regulation of the reactivity of external-stimuli-responsive Lewis acid-base adducts that contain multifunctional substituents.


There have been many recent developments in the chemistry of frustrated Lewis pairs (FLPs) that have been of note, for example, the activation of H2 mediated by main-group elements1,2,3,4,5,6,7,8,9. In general, FLPs are transient and not shelf-stable species making their isolation challenging. Meanwhile, chemists have developed strategies that trigger the conformational isomerization of molecules in response to external-stimuli10,11,12,13. These strategies can also be used to generate transient FLP species from classical Lewis adducts (CLAs) that act like their shelf-stable precursors14,15,16,17,18,19,20,21,22,23,24,25,26. In 2015, we demonstrated a strategy to generate FLPs from shelf-stable CLAs (PoxIm·B1 in Fig. 1) that are comprised of N-phosphine-oxide-substituted imidazolylidenes (PoxIms; 1) and B(C6F5)3 (B1). Here, the revival of the FLP from the CLA is closely controlled by a thermally induced conformational isomerization of the N-phosphinoyl moiety17,27,28,29,30. In 2018, Stephan et al. reported a system to control the generation of FLPs from CLAs via a light-induced E/Z isomerization of (C6F5)2B((p-Tol)S)C = CCH(tBu)31. Nevertheless, such FLP revival systems, including external-stimuli-responsive conformational isomerizations, are still underdeveloped. Thus, clarifying the relationship between external-stimuli-responsive conformational isomerizations and the interconversion that occurs between frustrated and quenched Lewis pairs is of great importance. This would allow a significant expansion of different strategies to design and apply FLP species19.

Fig. 1: Revival of FLPs from PoxIm·B1 adducts, induced by thermally responsive molecular motions.

a A previously proposed mechanism. b The updated mechanism proposed based on the results of this work.

In our system that uses PoxIms, the revival mechanism has not been fully explained. A tentative mechanism in which a B(C6F5)3 moiety is repelled by the N-phosphinoyl group via a thermally induced isomerization from the syn to anti conformation had been proposed. In this case, the syn/anti conformation refers to the relative orientation of the carbene carbon atom and the N-phosphinoyl oxygen atom with respect to the N‒P bond (Fig. 1a)17. Herein, we report the results of a combined experimental and theoretical mechanistic study that demonstrates the key role of a transfer step where the triarylborane (BAr3) unit on the carbene carbon atom moves to the N-phosphinoyl oxygen atom (Fig. 1b). In this study, PoxIms with 2,6-iPr2-C6H3, 2,4,6-Me3-C6H2, and 3,5-tBu2-C6H3 groups were studied and are herein referred to as 1a, 1b, and 1c, respectively.

Results and discussion

Effects of Lewis acidity

To explore the impact of the Lewis acidity of BAr3 on the formation and reactivity of the carbene–borane adducts, the reaction between 1a and B(p-HC6F4)3 (B2) was undertaken (Fig. 2a). Full consumption of 1a was confirmed after 20 min, resulting in the formation of two CLAs, i.e., 2aB2, which contains a N-phosphinoyl oxygen–boron bond, and 3aB2, which contains a carbene–boron bond, in 61% and 29% yield, respectively. Previously, we have reported that, even at ‒30 °C, 2aB1 could be converted to 3aB1 and that full identification of 2aB1 could therefore be achieved using NMR analysis conducted at ‒90 °C17. In the present case, 2aB2 exhibited a longer life-time at room temperature than 2aB1, which enabled us to prepare single crystals of 2aB2 by recrystallization from the reaction mixture at ‒30 °C. The molecular structure of 2aB2 was unambiguously confirmed using single-crystal X-ray diffraction (SC-XRD) analysis. A set of (Ra) and (Sa) atropisomers of 2aB2 was identified in the asymmetric unit of the single crystal. The molecular structure of (Ra)-2aB2 is shown in Fig. 2b and demonstrates a rare example of complexation-induced N‒P axial chirality29. As the reaction progressed, 2aB2 was converted to 3aB2 and 4a; 2aB2 was fully consumed within 6 h to afford these compounds in 75% and 25% yield, respectively. It should be noted that 4a is likely furnished via the migration of the N-phosphinoyl group from the nitrogen atom to the carbene carbon atom. However, in the absence of B2, this migration only proceeded to 9% at 100 °C, even after 25 h30. The formation of 4a was therefore promoted by the enhancement of the electrophilicity of the P center via the coordination of the N-phosphinoyl moiety to B2. Regeneration of B2 was observed along with the production of 4a. The molecular structure of 3aB2 was also confirmed by SC-XRD analysis (Fig. 2c). Comparison of the structural parameters between the solid-state structures of 3aB2 and 3aB1 shows their similarity. For example, the C1‒B distances in 3aB2 and in 3aB1 are 1.710(3) Å and 1.696(3) Å, respectively. The interatomic distance of 3.257(3) Å between the O and B atoms in 3aB2 suggests the absence of a specific interaction between these atoms, similar to that in 3aB1 (3.234(3) Å).

Fig. 2: Reaction between 1a and B(p-HC6F4)3 (B2).

a The reaction was monitored by NMR spectroscopy and the product yields were estimated based on 31P NMR analyses. b Molecular structure of (Ra)-2aB2 with thermal ellipsoids at 30% probability; H atoms and solvated C7H8 molecules are omitted for clarity. Selected bond lengths [Å] and angles [°]: O‒B 1.556(2), N1‒P 1.707(2), P‒O 1.513(1); P-O-B 165.2(1), C1-N1-P-O 128.0(1). c Molecular structure of 3aB2 with thermal ellipsoids at 30% probability; H atoms are omitted for clarity. For comparison with 3aB1 (cf. ref. 17), the carbene‒boron bond lengths and interatomic distances between oxygen and boron atoms are shown; C1-N1-P-O: 15.3(2)°.

Thermolysis of 3aB2 at 60 °C for 3 h resulted in the generation of 4a and B2 in 77% and 73% yield, respectively, with concomitant formation of [1a‒H][HO(B2)2] in 4% yield (conversion of 3aB2 = 81%; Fig. 3a). Although 2aB2 was not observed via NMR analysis of this reaction at 60 °C, the formation of 4a and B2 indicates the in situ regeneration of 2aB2 (vide supra). The formation of [1a‒H][HO(B2)2] can be rationalized in terms of a reaction between contaminated H2O and the FLP species regenerated from 3aB2 via 2aB2. The regeneration of the FLP species from 3aB2 was then clearly confirmed by treating 3aB2 with H2 (5 atm) at 22 °C, resulting in the formation of [1a‒H][H‒B2] (5aB2) in 19% yield with concomitant formation of [1a‒H][HO(B2)2] (8%) and 1a (6%) (Fig. 3b). Under identical conditions, no reaction occurred when 3aB1 was used17. At 60 °C, 5aB2 was generated in 90% yield after 3 h, which is almost comparable with the production of 5aB1 (89%) from 3aB1. Thus, the lower Lewis acidity of B2 relative to B1 allowed a more facile revival of the FLP species from 3aB2 than from 3aB1. However, the lower Lewis acidity did not affect the progress of the heterolytic cleavage of H2 by FLPs at 60 °C.

Fig. 3: Reactivity of carbene‒borane adducts 3aBn (n = 1, 2).

a Thermolysis of 3aB2 monitored via NMR spectroscopy. Product yields were calculated based on 31P and 19F NMR analyses. b Reaction between 3aBn and H2. Product yields were calculated based on 19F and 31P analysis. a[1a‒H][HO(B2)2] and 1a were also observed in 8% and 6% yield, respectively. b[1a‒H][HO(B2)2] was also observed in 7% yield. cResults obtained using 3aB1 are reproduced from ref. 17.

Kinetic studies

To gain further insight into the reaction mechanism, the initial rate constants for the generation of 5aB1, kint [10‒5 s‒1], from the reaction between 3aB1 and H2 in 1,2-dichloroethane-d4 (DCE-d4) at 60 °C were estimated by varying the H2 pressure from 0.5 to 5.0 atm (Fig. 4a). It should be noted here that when H2 was pressurized at 5.0 atm, an excess of H2 (ca. 0.3 mmol) with respect to 3aB1 (0.010 mmol) was added to the pressure-tight NMR tube. The concentration of H2 clearly influenced the progress of the reaction, suggesting that the heterolytic cleavage of H2 by the FLP species is involved in the rate-determining events. Next, the reaction between 3aB1 and H2 at 5.0 atm of pressure was monitored in DCE-d4 whilst the temperature was varied from 50 to 80 °C (Supplementary Figure 27). Pseudo-first order rate constants, kobs [10‒5 s‒1], of 2.95(2), 11.2(8), 46.4(4) and 183(2) were estimated for the reactions at 50, 60, 70, and 80 °C, respectively. Thus, the activation energy and pre-exponential factor obtained from the plot based on the Arrhenius equation, lnkobs = ‒(Ea/R)(1/T) + lnA, are Ea = 31.2 [kcal mol‒1] and A = 3.3(36) × 1016 [s‒1] (Fig. 4b). Given the close relation between Ea and ΔH, the values obtained for Ea suggest that the formation of 5aB1 via the reaction between 3aB1 and H2 only occurs at temperatures higher than 25 °C32.

Fig. 4: Kinetic studies for the reaction between 3aB1 and H2.

a Plot of the H2 pressure, P [atm], as a function of the initial reaction rate constants, kint [10‒5 s‒1]. b Plot of 1/T [10‒3 K‒1] as a function of lnkobs [s‒1]. The kobs values are the pseudo-first order rate constants for the formation of 5aB1 obtained from the reaction of 1a (2.0 × 10‒2 M in DCE-d4) and H2 (5 atm).

Based on the results presented here and those previously reported17, the reaction between the carbene‒borane adducts and H2 to give [PoxIm‒H][H‒BAr3] likely proceeds via the heterolytic cleavage of H2 by the FLP species that are formed following the regeneration of the N-phosphinoyl oxygen‒borane adducts. These steps are expected to be the rate-determining events because the concentration of H2 (Fig. 4a), the steric bulk of the N-aryl group17 and the Lewis acidity of the BAr3 moiety (Fig. 3b) influence the reaction rates and/or the temperature required to initiate the reaction between the carbene‒borane adducts and H2.

Theoretical studies

Density-functional theory (DFT) calculations were carried out at the ωB97X-D/6-311G(d,p), PCM (DCE)//ωB97X-D/6-31G(d,p) for H2 and 6-31G(d) for all other atoms level of theory (Fig. 5a). The relative Gibbs free energies with respect to [1a + B1] (0.0 kcal·mol‒1) are shown. During the transformation of 3aB1 (‒17.2 kcal·mol‒1) to 2aB1 (‒9.8 kcal·mol‒1), both of which were experimentally confirmed, the formation of an intermediate 2a′B1 (‒7.7 kcal·mol‒1) was predicted via a C-to-O transfer of B1 in 3aB1. This distinctive boron-transfer process takes place via saddle point TS1a (+7.3 kcal·mol‒1), while the potential energy surface around TS1a is very flat (Supplementary Figure 34 for details). The subsequent rotation of the N-phosphinoyl moiety via TS2a (+7.5 kcal·mol‒1) affords 2aB1. Next, the dissociation of the B‒O bond occurs to regenerate [1a + B1]. The optimized molecular structures of TS1a and 2a′B1 are shown in Fig. 5c. In TS1a, the interatomic distances C1···B and O···B are 4.24 and 3.34 Å, respectively, while B1 adopts a planar geometry. Thus, B1 is dissociated from both the carbene carbon and phosphinoyl oxygen atoms in TS1a, while the formation of the O‒B bond (1.59 Å) is confirmed in 2a′B1. Based on the quantum theory of atoms in molecule (AIM) method, neither bond paths nor bond critical points were confirmed between the B and C1/O atoms in TS1a (Supplementary Figure 37)33,34. This AIM analysis demonstrates that several non-covalent interactions, including π–π and H···F interactions, exist between the 1a and B1 moieties to stabilize TS1a.

Fig. 5: Theoretical studies.

The relative Gibbs energies [kcal mol‒1] are shown with respect to each [1+B1], calculated at the ωB97X-D/6-311G(d,p), PCM (DCE)//ωB97X-D/6-31G(d,p) (for H2) and 6-31G(d) (for all other atoms) level of theory (298.15 K, 1 atm). a Proposed mechanism for the regeneration of [1+B1] from the carbene‒borane complexes 3aB13cB1. b Proposed mechanism for the heterolytic cleavage of H2, enabled by the phosphinoyl oxygen and B(C6F5)3 moieties (left) or by the carbene and B(C6F5)3 moieties (right). c DFT-optimized molecular structures for TS1a and 2a′B1. d DFT-optimized molecular structures for TS4a and TS6a.

Two plausible mechanisms were evaluated for the FLP-mediated cleavage of H2 on the basis that the Lewis-basic center reacts with H2 via cooperation with B1 (Fig. 5b). One possibility is that the carbene carbon atom works as a Lewis base (path I; the right path in Fig. 5b)1,2,3,4,5,6,7,8,9,35,36, while the other is that the N-phosphinoyl oxygen functions as a Lewis base (path II; the left path in Fig. 5b)37. In path I, the heterolytic cleavage of H2 takes places via TS4a (+11.4 kcal·mol‒1), which arises from the insertion of H2 into the reaction field around the carbene carbon and boron atoms in FLP-1aB1, affording 5aB1 (‒34.8 kcal·mol‒1), a species more thermodynamically stable than 3aB1. In the optimized structure of TS4a (Fig. 5d), the dissociation of the H1–H2 bond (H1···H2 = 0.84 Å) occurs with the partial formation of the H2–C1/H1–B bonds (H2···C1 = 1.83 Å/H1···B = 1.49 Å). Based on these results, the overall path from 3aB1 to 5aB1 via FLP-1aB1 is substantially exothermic (ΔG° =‒17.6 kcal·mol‒1) and includes an overall activation energy barrier of +28.6 kcal·mol‒1 required to overcome TS4a. In path II, which takes place via TS5a (a transition state for the insertion of H2 into the O‒P bond) and TS6a (a transition state for the cleavage of H2 between the O and P atoms), a higher activation energy barrier of +32.7 kcal·mol‒1 is predicted to yield intermediate 8aB1, which contains a P=O‒H+ and B‒H species. It should be noted that the potential energy of the optimized TS6a (‒3633.288355 hartree) is almost identical to that of the optimized 7aB1 (‒3633.288363 hartree), which causes the reversed Gibbs energy levels as shown in Fig. 5b after the Gibbs energy correction and implementation of solvent effect. Therefore, the discussion on the activation energy barrier to overcome TS6a from 7aB1 should be not essential. The subsequent transfer of H+ from the N-phosphinoyl oxygen atom to the carbene carbon atom furnishes 5aB1, although the details of this process remain unclear at this point. The molecular structure of TS6a shows that the cleavage of the H1–H2 bond (H1···H2 = 0.85 Å) by the N-phosphinoyl oxygen and boron atoms occurs in a cooperative fashion (Fig. 5d). Given the experimental and theoretical results reported here, we conclude that path I is the more likely one.

The impact of the N-aryl substituents on the activation energy barriers for the regeneration of [1 + B1] was evaluated using calculations on 3bB1, which contains an N-2,4,6-Me3-C6H2 group, as well as 3cB1, which contains an N-3,5-tBu2-C6H3 group. This afforded ΔG values of +28.3 and +32.8 kcal·mol‒1 for 3bB1 and 3cB1, respectively (Fig. 5a). These results are consistent with the experimental observations, i.e., that 3aB13cB1 did not react in the presence or absence of H2 under ambient conditions under the applied conditions. Furthermore, these results might rationalize the fact that temperature to induce the reaction between these CLAs and H2 increases in the order 3aB1 (60 °C) < 3bB1 (80 °C) < 3cB1 (120 °C)17.


In summary, the reaction mechanism for the revival of frustrated carbene–borane pairs from external-stimuli-responsive classical Lewis adducts (CLAs), comprised of N-phosphine-oxide-substituted imidazolylidene (PoxIm) and triarylboranes (BAr3), is reported based on a combination of experimental and theoretical studies. Remarkably, a transfer of the borane moiety from the carbene carbon atom to the N-phosphinoyl oxygen atom was identified as a key step in the heterolytic cleavage of H2 by the regenerated FLP species. The optimized transition-state structure for this borane-transfer process was confirmed to include no bonding interactions between the carbene carbon/phosphinoyl oxygen and boron atoms, albeit that it is stabilized by intermolecular non-covalent interactions between the PoxIm and BAr3 moieties. The heterolytic cleavage of H2 takes place via the cooperation of the carbene carbon and the boron atoms, and exhibits a lower overall activation energy barrier than that of the path in which a combination of the N-phosphinoyl oxygen and boron atom mediates the H2 cleavage. These results demonstrate the essential role of dynamic conformational isomerization in the regulation of the reactivity of shelf-stable but external-stimuli-responsive Lewis acid-base adducts by multifunctional Lewis bases.


Synthesis of 3aB2

PoxIm 1a (154.8 mg, 0.40 mmol) and B(p-HC6F4)3 (B2) (183.4 mg, 0.40 mmol) were mixed in toluene (10 mL) at room temperature to furnish the yellow solution. Stirring this mixture for 4 h resulted into the precipitation of a white solid that was collected via removal of the supernatant solution. The obtained solid was washed with hexane (5 mL) and dried in vacuo to afford 3aB2 as a white solid (230.2 mg, 0.27 mmol, 68%). A single crystal suitable for X-ray diffraction analysis was prepared by recrystallization from CH2Cl2/hexane at ‒30 °C.

Synthesis of 5aB2

A solution of 3aB2 (51.6 mg, 0.06 mmol) in CH2Cl2 (3 mL) was transferred into an autoclave reactor, which was then pressurized with H2 (5 atm). Subsequently, the reaction mixture was stirred at 60 °C for 4 h, before the solvent was removed in vacuo to give 5aB2 as a white solid (51.8 mg, 0.06 mmol, >99%). A single crystal suitable for X-ray diffraction analysis was prepared by recrystallization from THF/hexane at ‒30 °C.

Reaction between 1a and B2 giving 2aB2

A solution of 1a (7.4 mg, 0.02 mmol) and B2 (9.3 mg, 0.02 mmol) in CD2Cl2 (0.5 mL) was prepared at ‒30 °C and then transferred into a J. Young NMR tube. The quantitative formation of 2aB2 was confirmed at ‒90 °C by 1H, 13C, 19F, and 31P NMR analysis (Supplementary Figs. 58). A single crystal suitable for X-ray diffraction analysis was prepared by recrystallization from toluene/hexane at ‒30 °C.

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request. Details for the preparation of materials, monitoring the reactions (Supplementary Figs. 530), and theoretical analysis/discussion are provided in the Supplementary Information. Compound characterization data including ESI-MS, elementary analysis, and SC-XRD data are also available in the Supplementary Information. NMR spectra for the isolated compounds are provided in the Supplementary Information. Results of the AIM analysis for TS1a are provided in Supplementary Fig. 37, as well as Supplementary Data 1. Metrical data for the solid-state structures are available from the Cambridge Crystallographic Data Centre (CCDC) under reference numbers CCDC2072358 (3aB2), 2072359 (5aB2), 2072360 (2aB2), and 2072638 ([1a–H][HO(B2)2]) (Supplementary Data 25). These data can be obtained free of charge from the CCDC via


  1. 1.

    Stephan, D. W. & Erker, G. Frustrated Lewis pairs: metal-free hydrogen activation and more. Angew. Chem. Int. Ed. 49, 46–76 (2010).

    CAS  Article  Google Scholar 

  2. 2.

    Paradies, J. Frustrated Lewis pair catalyzed hydrogenations. Synlett 24, 777–780 (2013).

    CAS  Article  Google Scholar 

  3. 3.

    Feng, X. & Du, H. Metal-free asymmetric hydrogenation and hydrosilylation catalyzed by frustrated Lewis pairs. Tetrahedron Lett. 55, 6959–6964 (2014).

    CAS  Article  Google Scholar 

  4. 4.

    Stephan, D. W. & Erker, G. Frustrated Lewis pair chemistry: development and perspectives. Angew. Chem. Int. Ed. 54, 6400–6441 (2015).

    CAS  Article  Google Scholar 

  5. 5.

    Weicker, S. A. & Stephan, D. W. Main group Lewis acids in frustrated Lewis pair chemistry: beyond electrophilic boranes. Bull. Chem. Soc. Jpn. 88, 1003–1016 (2015).

    CAS  Article  Google Scholar 

  6. 6.

    Scott, D. J., Fuchter, M. J. & Ashley, A. E. Designing effective ‘frustrated Lewis pair’ hydrogenation catalysts. Chem. Soc. Rev. 46, 5689–5700 (2017).

    CAS  Article  Google Scholar 

  7. 7.

    Fasano, V. & Ingleson, M. J. Recent advances in water-tolerance in frustrated Lewis pair chemistry. Synthesis 50, 1783–1795 (2018).

    CAS  Article  Google Scholar 

  8. 8.

    Jupp, A. R. & Stephan, D. W. New directions for frustrated Lewis pair chemistry. Trends Chem. 1, 35–48 (2019).

    CAS  Article  Google Scholar 

  9. 9.

    Hoshimoto, Y. & Ogoshi, S. Triarylborane-catalyzed reductive N-alkylation of amines: a perspective. ACS Catal. 9, 5439–5444 (2019).

    CAS  Article  Google Scholar 

  10. 10.

    Pan, T. & Liu, J. Catalysts encapsulated in molecular machines. ChemPhysChem 17, 1752–1758 (2016).

    CAS  Article  Google Scholar 

  11. 11.

    Cheng, C. & Stoddart, J. F. Wholly synthetic molecular machines. ChemPhysChem. 17, 1780–1793 (2016).

    CAS  Article  Google Scholar 

  12. 12.

    Kassem, S. et al. Artificial molecular motors. Chem. Soc. Rev. 46, 2592–2621 (2017).

    CAS  Article  Google Scholar 

  13. 13.

    Findlay, J. A. & Crowley, J. D. Functional nanomachines: recent advances in synthetic molecular machinery. Tetrahedron Lett. 59, 334–346 (2018).

    CAS  Article  Google Scholar 

  14. 14.

    Rokob, T. A., Hamza, A., Stirling, A. & Pápai, I. On the mechanism of B(C6F5)3-catalyzed direct hydrogenation of imines: inherent and thermally induced frustration. J. Am. Chem. Soc. 131, 2029–2036 (2009).

    CAS  Article  Google Scholar 

  15. 15.

    Jiang, C., Blacque, O., Fox, T. & Berke, H. Heterolytic cleavage of H2 by frustrated B/N Lewis pairs. Organometallics 30, 2117–2124 (2011).

    CAS  Article  Google Scholar 

  16. 16.

    Herrington, T. J. et al. Bypassing a highly unstable frustrated Lewis pair: dihydrogen cleavage by a thermally robust silylium-phosphine adduct. Chem. Commun. 50, 12753 (2014).

    CAS  Article  Google Scholar 

  17. 17.

    Hoshimoto, Y., Kinoshita, T., Ohashi, M. & Ogoshi, S. A strategy to control the reactivation of frustrated Lewis pairs from shelf-stable carbene borane complexes. Angew. Chem. Int. Ed. 54, 11666–11671 (2015).

    CAS  Article  Google Scholar 

  18. 18.

    Wu, L., Chitnis, S. S., Jiao, H., Annibale, V. T. & Manners, I. Non-metal-catalyzed heterodehydrocoupling of phosphines and hydrosilanes: mechanistic studies of B(C6F5)3-mediated formation of P–Si bonds. J. Am. Chem. Soc. 139, 16780–16790 (2017).

    CAS  Article  Google Scholar 

  19. 19.

    Han, Y., Zhang, S., He, J. & Zhang, Y. Switchable C–H silylation of indoles catalyzed by a thermally induced frustrated Lewis pair. ACS Catal. 8, 8765–8773 (2018).

    CAS  Article  Google Scholar 

  20. 20.

    Wang, X., Kehr, G., Daniliuc, C. G. & Erker, G. Internal adduct formation of active intramolecular C4-bridged frustrated phosphane/borane Lewis pairs. J. Am. Chem. Soc. 136, 3293–3303 (2014).

    CAS  Article  Google Scholar 

  21. 21.

    Holtrichter-Rößmann, T. et al. Generation of weakly bound Al–N Lewis pairs by hydroalumination of ynamines and the activation of small molecules: phenylethyne and dicyclohexylcarbodiimide. Organometallics 31, 3272–3283 (2012).

    Article  Google Scholar 

  22. 22.

    Mömming, C. M., Kehr, G., Wibbeling, B., Fröhlich, R. & Erker, G. Addition reactions to the intramolecular mesityl2P-CH2-CH2-B(C6F5)2 frustrated Lewis pair. Dalton Trans. 39, 7556–7564 (2010).

    Article  Google Scholar 

  23. 23.

    Spies, P. et al. Metal-free dihydrogen activation chemistry: structural and dynamic features of intramolecular P/B pairs. Dalton Trans. 1534–1541 (2009).

  24. 24.

    Spies, P. et al. Rapid intramolecular heterolytic dihydrogen activation by a four-membered heterocyclic phosphane-borane adduct. Chem. Commun. 5072–5074 (2007).

  25. 25.

    Kolychev, E. L. et al. Reactivity of a frustrated Lewis pair and small-molecule activation by an isolable arduengo carbene–B{3,5-(CF3)2C6H3}3 complex. Chem. -Eur. J. 18, 16938–16946 (2012).

    CAS  Article  Google Scholar 

  26. 26.

    Geier, S. J. & Stephan, D. W. Lutidine/B(C6F5)3: at the boundary of classical and frustrated Lewis pair reactivity. J. Am. Chem. Soc. 131, 3476–3477 (2009).

    CAS  Article  Google Scholar 

  27. 27.

    Hazra, S., Hoshimoto, Y. & Ogoshi, S. N-Phosphine oxide-substituted imidazolylidenes (poxims): multifunctional multipurpose carbenes. Chem. -Eur. J. 23, 15238–15243 (2017).

    CAS  Article  Google Scholar 

  28. 28.

    Hoshimoto, Y. & Ogoshi, S. Development of metal complexes equipped with structurally flexible carbenes. Bull. Chem. Soc. Jpn. 94, 327–338 (2021).

    CAS  Article  Google Scholar 

  29. 29.

    Asada, T., Hoshimoto, Y., Kawakita, T., Kinoshita, T. & Ogoshi, S. Axial chirality around N–P bonds induced by complexation between E(C6F5)3 (E = B, Al) and an N-phosphine oxide-substituted imidazolinylidene: a key intermediate in the catalytic phosphinoylation of CO2. J. Org. Chem. 85, 14333–14341 (2020).

    CAS  Article  Google Scholar 

  30. 30.

    Asada, T., Hoshimoto, Y. & Ogoshi, S. Rotation-triggered transmetalation on a heterobimetallic Cu/Al N-phosphine-oxide-substituted imidazolylidene complex. J. Am. Chem. Soc. 142, 9772–9784 (2020).

    CAS  Article  Google Scholar 

  31. 31.

    Fan, L., Jupp, A. R. & Stephan, D. W. Remote stereochemistry of a frustrated Lewis pair provides thermal and photochemical control of reactivity. J. Am. Chem. Soc. 140, 8119–8123 (2018).

    CAS  Article  Google Scholar 

  32. 32.

    Anslyn, E. V. & Dougherty, D. A. Modern Physical Organic Chemistry (University Science Books, 2006).

  33. 33.

    Bader, R. F. W. Ed. Atoms in Molecules. A Quantum Theory (Cambridge University Press, 1991).

  34. 34.

    Bader, R. F. W. & Stephens, M. E. Spatial localization of the electronic pair and number distributions in molecules. J. Am. Chem. Soc. 97, 7391–7399 (1975).

    CAS  Article  Google Scholar 

  35. 35.

    Chase, P. A. & Stephan, D. W. Hydrogen and amine activation by a frustrated Lewis pair of a bulky N-heterocyclic carbene and B(C6F5)3. Angew. Chem. Int. Ed. 47, 7433–7437 (2008).

    CAS  Article  Google Scholar 

  36. 36.

    Holschumacher, D., Bannenberg, T., Hrib, C. G., Jones, P. G. & Tamm, M. Heterolytic dihydrogen activation by a frustrated carbene–borane Lewis pair. Angew. Chem. Int. Ed. 47, 7428–7432 (2008).

    CAS  Article  Google Scholar 

  37. 37.

    Stepen, A. J., Bursch, M., Grimme, S., Stephan, D. W. & Paradies, J. Electrophilic phosphonium cation-mediated phosphane oxide reduction using oxalyl chloride and hydrogen. Angew. Chem. Int. Ed. 57, 15253–15256 (2018).

    CAS  Article  Google Scholar 

Download references


This work was supported by Grants-in-Aid for Young Scientists (JSPS KAKENHI grants JP15K17824 and JP18K14219), Grants-in Aid for Scientific Research on Innovative Areas “Stimuli-responsive Chemical Species (JSPS KAKENHI grant JP15H00943)” and “Precisely Designed Catalysts with Customized Scaffolding (JSPS KAKENHI grants JP15H05803 and 15H05805),” and the Environment Research and Technology Development Fund (No. 1RF-2101) of the Environmental Restoration and Conservation Agency of Japan. Y.H. expresses his special thanks to Prof. Dr. T. Sasamori (University of Tsukuba) for valuable suggestions on this work. T.K. expresses his special thanks to a Grant-in-Aid for JSPS Fellows. M.R. and J.H. appreciate support from the Integrated Research Consortium on Chemical Sciences. A part of the computational work was performed at RCCS in Okazaki, Japan. A part of the study was supported by the Cooperative Research Program of Institute for Catalysis, Hokkaido University. S.O. would like to dedicate this work to Prof. Dr. Christian Bruneau.

Author information




This work was conceptualized and directed by Y.H. with the support of J.H. and S.O. M.S. and T.K. performed the experiments, while Y.H., M.S. and T.K. analyzed these data. M.O., M.R. and J.H. performed the computational analysis. The manuscript was prepared by Y.H. supported by discussions with all the other authors.

Corresponding authors

Correspondence to Yoichi Hoshimoto, Jun-ya Hasegawa or Sensuke Ogoshi.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Peer review information Communications Chemistry thanks the anonymous reviewers for their contribution to the peer review of this work. Peer reviewer reports are available.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Hoshimoto, Y., Sakuraba, M., Kinoshita, T. et al. A boron-transfer mechanism mediating the thermally induced revival of frustrated carbene–borane pairs from their shelf-stable adducts. Commun Chem 4, 137 (2021).

Download citation


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing