DOI: 10.1002/chem.201500805
Communication
& Frustrated Lewis Pairs
Autoinduced Catalysis and Inverse Equilibrium Isotope Effect in the Frustrated Lewis Pair Catalyzed Hydrogenation of Imines Sebastian Tussing,[a] Lutz Greb,[a] Sergej Tamke,[a] Birgitta Schirmer,[b] Claudia Muhle-Goll,[c] Burkhard Luy,[c, d] and Jan Paradies*[a] Abstract: The frustrated Lewis pair (FLP)-catalyzed hydrogenation and deuteration of N-benzylidene-tert-butylamine (2) was kinetically investigated by using the three boranes B(C6F5)3 (1), B(2,4,6-F3-C6H2)3 (4), and B(2,6-F2C6H3)3 (5) and the free activation energies for the H2 activation by FLP were determined. Reactions catalyzed by the weaker Lewis acids 4 and 5 displayed autoinductive catalysis arising from a higher free activation energy (2 kcal mol¢1) for the H2 activation by the imine compared to the amine. Surprisingly, the imine reduction using D2 proceeded with higher rates. This phenomenon is unprecedented for FLP and resulted from a primary inverse equilibrium isotope effect.
The discovery of the metal-free activation of molecular hydrogen by sterically encumbered Lewis pairs sparked new interest in hydrogenation reactions.[1] The outstanding performance of the frustrated Lewis pair (FLP) concept[2] rests on the striking simplicity and the possibility to vary one component, for example the Lewis acid, while the Lewis base remains unchanged. Following this notion, distinct reactivity can be observed when the most popular Lewis acid in FLP B(C6F5)3 (1) was exchanged by weaker Lewis acids giving rise to functional group tolerance.[3] Chiral boranes even facilitated asymmetric hydrogenations of heterocycles,[4] silyl enol ethers,[5] and imines.[6] Similarly, the change of the Lewis base allowed the development of important reactions such as the first metal-free saturation of
olefins[7] using molecular hydrogen (H2). It was shown that the capability to heterolytically split H2 is strongly correlated with the basicity of the applied phosphines.[7b] An interesting situation occurs when the basicity of the Lewis base changes over the course of the reaction, hence leading to a change in the reactivity of the FLP towards H2. An example for such a fascinating reaction is the FLP-mediated hydrogenation of imines,[8] in which both the imine and the hydrogenation product, the amine, are viable Lewis bases for H2 activation. It has been speculated that this reaction might show the feasibility of autoinduced catalysis,[8a, 9] however, we are not aware of any experimental confirmation. The mechanism of the FLP-catalyzed hydrogenation of imines was proposed based on analogies to imine hydrosilylation[10] as well as on quantum-mechanical studies,[9, 11] but has not yet been supported by kinetic experiments. Herein we report that the autoinduced FLP-mediated hydrogenation of imines is only operative for weaker Lewis acids. Furthermore, an unusual inverse equilibrium (deuterium) isotope effect (IEIE)[12] was observed for the first time in FLP chemistry. This effect overcompensated the primary and secondary isotope effects[13] resulting in an overall rate increase. Now, a conclusive mechanistic picture of the most frequently used application of FLP is provided. According to the proposed catalytic cycle, the H2 activation is achieved by the FLP consisting of the borane 1 as Lewis acid and the imine 2 (Scheme 1, simple catalytic cycle), or with the
[a] Dipl.-Chem. S. Tussing, Dr. L. Greb, Dipl.-Chem. S. Tamke, Prof. Dr. J. Paradies Institute of Organic Chemistry, University of Paderborn Warburger Straße 100, 33098 Paderborn (Germany) E-mail: [email protected] [b] Dr. B. Schirmer Center for Multiscale Theory and Computation (CMTC) und Organisch-Chemisches Institut Westflische Wilhelms-Universitt Corrensstrasse 40, 48149 Mìnster (Germany) [c] Priv.-Doz. C. Muhle-Goll, Prof. Dr. B. Luy Institute of Organic Chemistry Karlsruhe Institute of Technology (KIT) Fritz-Haber-Weg 6, 76131 Karlsruhe (Germany) [d] Prof. Dr. B. Luy Department Magnetic Resonance of the Institute for Biological Interfaces Karlsruhe Institute of Technology (KIT), P.O. Box 3640, 76021 (Germany) Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201500805. Chem. Eur. J. 2015, 21, 8056 – 8059
Scheme 1. Simple and autoinduced catalytic cycle for the FLP-mediated imine hydrogenation; Lewis acids including their Lewis acidities.[3, 15]
8056
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication hydrogenation product 3 (autoinduced catalytic cycle) as the acting Lewis base (Scheme 1). It can be assumed that this mechanism is the basis of the autoinduced catalysis because both the starting material and the product are involved in the H2 activation.[14] Quantum-mechanical calculations supported the H2 activation by the FLP comprising borane 1 and imine 2 or amine 3 as the step with highest energy barrier making the heterolytic H2-splitting rate-determining for the overall process (Scheme 1).[9, 11] The transition state energies for those activations were calculated both to be 16.5 kcal mol¢1,[9] although the amine should be a more Figure 2. Eyring plots for the FLP-mediated hydrogenation of 2 (left) and deuterohydropotent Lewis base for the H2 activation than the imi- genation (right) in the presence of the three boranes. ne.[7a, b] Consequently, the H2-splitting by 1 is unselective with regard to the Lewis base resulting in the absence of the characteristic kinetic features of autoinduced cata general autocatalytic reaction.[18] The graphical analysis yieldalysis. ed the values for both rate constants, k1 and k2 for the imineThe reactivities of the three different boranes B(C6F5)3 (1), and the amine-mediated FLP-catalyzed hydrogenation of 2 B(2,4,6-F3-C6F3)3 (4), and B(2,6-F2-C6H3)3 (5) were studied in the and allowed the determination of the free activation paramehydrogenation of 2 at 70 8C (Figure 1). The reaction with ters. The results of the kinetic analysis are summarized in Figure 2 and in Table 1 (see Supporting Information for details). The reaction catalyzed by B(C6F5)3 (1) displayed a 1.5–1.9 fold rate increase for k2 (see Figure 2 a) resulting in an apparent first-order type reaction rate. Consequently, the free activation enthalpies for both cycles are very similar ((21.4 13.6) kcal mol¢1 and (20.1 23.7) kcal mol¢1, Table 1, for detailed error treatment see Ref. [19] and Supporting Information) and their absolute values are within an acceptable agreement with the previously calculated ones (lit. 16.5 kcal mol¢1).[9] This situation changed significantly when the weaker Lewis acids 4 and 5 were used. The rate constants (k2) for the product-accelerated Figure 1. Time versus yield plot for the FLP-catalyzed hydrogenation of Ncycle were about one order of magnitude larger than for the benzylidene-tert-butylamine (2) (5 mol % borane, CD2Cl2, 0.4 m, 70 8C; yields imine-catalyzed cycle (k1) (Figure 2 a). The comparison of the were determined by 1H NMR spectroscopy using hexamethylbenzene as infree reaction enthalpies of activation obtained from k1 and k2 & & ^ ternal standard; B(C6F5)3 (1); ~ B(2,4,6-F3-C6F3)3 (4); B(2,6-F2-C6H3)3 (5); show that the imine-mediated H2 activation by the less LewisB(2,6-F2-C6H3)3 (5) and 5 mol % benzyl-tert-butylamine (3)). acidic boranes is 1.6 kcal mol¢1 and 1.8 kcal mol¢1 higher compared to the amine-mediated H2-splitting, thus triggering the B(C6F5)3 qualitatively proceeded with highest rates exhibiting observed autoinduced catalysis (Table 1). The increased negathe characteristic curve shape for a second-order type reactive activation entropies for the boranes 4 and 5 compared to tion.[16] In contrast, the reaction profiles of the hydrogenations 1 may result from the formation of hydrogen-bonded comcatalyzed by the two weaker Lewis-acidic boranes 4 and 5 displexes in the transition state of the imine-reduction as proplayed sigmoidal curve shapes. This implies a significant inposed by Privalov.[11] Experimental evidence for such interaccrease in rate during the course of the reaction. Addition of 5 mol % of the reaction product 3 at the beginning of the reaction using B(2,6-F2-C6H3)3 (5) as Table 1. Activation parameters for the FLP-catalyzed hydrogenation of 2 (values in parentheses correspond to the reaction with D2). Lewis acid resulted in the same dramatic rate increase, which is a strong indication for autoinduced for k2 (autoinduced catalytic cycle) for k1 (standard catalytic cycle) catalysis. Kinetic data for the FLP-catalyzed hydrogeDH[b] DS[c] DG[a,b] DH[b] DS[c] DG[a,b] nation of 2 using the three boranes were acquired at [d] 1 21.4 13.9 15.6 7.1 ¢19.1 21.1 20.1 23.7 19.6 14.2 ¢5.8 42.5.0 three different temperatures and were analyzed ac(21.0 10.5) (12.9 5.4) (¢25.7 16.0) cording to the following criteria: i) the H2 and borane 4 24.1 3.5 8.5 1.9 ¢48.0 5.1 22.5 2.5 6.9 1.3 ¢49.7 3.7 (24.1 5.6) (6.0 3.0) (¢55.1 8.2) (22.4 3.4) (5.2 1.8) (¢52.0 4.9) concentration in solution is constant,[17] and ii) the ¢51.7 3.4 5 24.5 4.0 6.5 2.1 ¢55.1 5.9 22.7 2.3 5.8 1.2 shift of the equilibrium of the FLP 3/4 and 3/5 to(24.4 4.8) (5.8 2.6) (¢56.7 7.1) (22.5 2.5) (5.7 1.3) (¢51.3 3.7) wards its H2-activation products is slow (vide infra). [a] Calculated for 343 K. [b] In kcal mol¢1. [c] In cal mol¢1 K. [d] The reaction with D2 was These requirements allow a quantitative analysis of fitted to a first-order rate law, hence k2 could not be determined. the collected data conforming with the rate laws for Chem. Eur. J. 2015, 21, 8056 – 8059
www.chemeurj.org
8057
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication tion is supplied by NOESY NMR experiments, which revealed a distinct aggregation of the [BnNH2tBu] cation with the [HB(2,6-F2C6H3)3] anion via intermolecular dihydrogen bonding (see the Supporting Information).[3a] Hydrogenations are strongly affected by deuterium isotope effects. Surprisingly, the deuterohydrogenation of 2 occurred not only with higher rates with all boranes but also the reaction order changed for borane 1 from second-order to firstorder (compare Figure 2 right). Accordingly, the primary deuterium isotope effects decelerate the autoinduced catalytic cycle because an additional deuterium transfer is required to generate the activated iminium [2-D] for the nucleophilic attack. The two boranes 4 and 5 retained second-order kinetics with a strong inverse deuterium isotope effect for k2 (at 70 8C for 4: k1(H)/k1(D) = 0.90; k2(H)/k2(D) = 0.67; for 5: k1(H)/k1(D) = 0.87; k2(H)/k2(D) = 0.71). The Arrhenius analysis of the temperaturedependent equilibrium constants for both the H2 and D2 activation by the FLPs 3/4 and 3/5 provided negative values for the standard reaction enthalpy (see Scheme 2).[20] However,
Scheme 2. Equilibrium of the H2 and D2 activation by benzyl-tert-butylamine (3) and boranes 4 and 5.
DHo of the reaction of the FLP 3/4 or 3/5 with D2 is more negative than with H2 resulting in a substantial primary inverse equilibrium deuterium isotope effect (IEIE) KH/KD. This is the first time that an inverse isotope effect has been reported for FLP-catalyzed hydrogenations. The standard free reaction energies for the FLP-mediated H2 activation were calculated from DH8 and DS8 at 298 K amounting to (¢0.64 0.65) kcal mol¢1 and (0.16 0.26) kcal mol¢1 (for D2 : (¢1.33 0.09) kcal mol¢1 and (¢1.06 1.12) kcal mol¢1) for 3/4 and 3/5, respectively.[21] The enthalpy-driven higher concentrations of [D-4]¢ and [D-5]¢ account for the higher rate of the deuterohydrogenation of 2 and counterbalanced the primary and secondary deuterium isotope effects along the autocatalytic reaction pathway. Furthermore, the reaction was analyzed by DFT methods at a B2PLYP-D3(BJ)/def2-TZVP//TPSS-D3(BJ)/def2-TZVP level[22] as implemented in the TURBOMOLE package[23] including thermal and solvent corrections.[24] The results are summarized in Table 2. The activation of H2 by 1 and the imine 2 is slightly exergonic at 25 8C and almost ergoneutral at 70 8C (Table 2, [Eq. (1)]). For the two other boranes this reaction is endergonic (7–21 kcal mol¢1) at 25 8C and 70 8C. The hydride transfer becomes more exergonic with decreasing Lewis acidity of the borane (1: 100 %, 4: 70 %, 5: 56 %) due to the higher nucleophilicity of the corresponding hydridoborate anion [Eq. (2)].[25] The H2 activation by the amine is for all boranes exergonic but becomes less negative for the weaker Lewis-acidic boranes 4 and 5 (Table 2, [Eq. (3)]). As a result the corresponding ammoChem. Eur. J. 2015, 21, 8056 – 8059
www.chemeurj.org
Table 2. Free reaction energies in dichloromethane at 298 K (343 K) (in kcal mol¢1 1.5 kcal mol¢1).[a] Reaction
Eq. B(C6F5)3 (1)
(1) ¢1.7 (0.45) 2 + H2 + BArF3 ! [2-H][H-BArF3] [2-H][H-BArF3]!3 + BArF3 (2) ¢9.6 (¢11.2) (3) ¢16.9 3 + H2 + BArF3 ! (¢14.6) [3-H][H-BArF3] (4) 1.6 (3.4) 3 + BArF3 ![3·BArF3]
B(2,4,6-F3-C6H2)3 B(2,6-F2-C6H3)3 (5) (4) 7.4 (9.7)
9.8 (21.1)
¢18.8 (¢20.4) ¢7.0 (¢4.6) 7.2 (9.0)
¢21.2 (¢31.8) ¢5.9 (¢3.5) 7.5 (9.4)
[a] DFT calculations with B2PLYP-D3/def2-TZVP, thermal and solvent corrections; see Supporting Information for details.
nium hydridoborate salts are thermodynamically less stable, which is supported by the observation of the reversible H2 activation for boranes 4 and 5. The large negative value for B(C6F5)3 (¢16.9 kcal mol¢1) clearly indicates that the back reaction under H2 liberation is unfavored, which is in close agreement with our NMR results (vide supra). The endergonicity of the Lewis adduct formation (Table 2, [Eq. (4)]) strongly supports our observation that only the FLPs (3/4 and 3/5) or their H2-activation products were detected by NMR spectroscopy. It can be concluded that the FLP-catalyzed hydrogenations of imines proceed via the simple and autoinduced catalytic cycles which are switched by the strength of the Lewis acid. For FLP comprising weaker Lewis-acidic boranes and imines, DG for the H2 activation was 2 kcal mol¢1 higher compared to the H2 activation with the stronger donor. The unusual inverse isotope effect in the deuterohydrogenation of the imine was caused by the higher exothermicity of the D2-activation product, thus counterbalancing the primary and secondary isotope effects. In the future this detailed picture of the FLP-catalyzed imine hydrogenation will enable the development of even more sophisticated FLP and related catalyzed reactions.
Acknowledgements The Landesgraduiertenfçrderung of the State of Baden-Wìrttemberg and the German Science Foundation (DFG) are acknowledged for a Ph.D. grant to S.T. and for a Heisenberg-fellowship to J.P. B.L. acknowledges funding by DFG (Pro2NMR facility) and the HGF programme BIFTM. Dr. Rainer Kerssebaum (Bruker BioSpin GmbH) is thanked for 19F NMR experiments and Prof. Matthias Olzmann (KIT) is thanked for valuable discussions. Keywords: autoinduced catalysis · frustrated Lewis pairs · hydrogen activation · inverse equilibrium isotope effect · kinetics
8058
[1] a) G. C. Welch, R. R. S. Juan, J. D. Masuda, D. W. Stephan, Science 2006, 314, 1124 – 1126; b) G. C. Welch, D. W. Stephan, J. Am. Chem. Soc. 2007, 129, 1880 – 1881; c) D. W. Stephan, G. Erker, Angew. Chem. Int. Ed. 2010, 49, 46 – 76; Angew. Chem. 2010, 122, 50 – 81; d) J. Paradies, Synlett 2013, 777 – 780; e) T. Mahdi, D. W. Stephan, J. Am. Chem. Soc. 2014, 136, Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication
[2] [3]
[4] [5] [6]
[7]
[8]
[9] [10]
[11] [12]
[13]
[14]
[15]
15809 – 15812; f) D. J. Scott, M. J. Fuchter, A. E. Ashley, J. Am. Chem. Soc. 2014, 136, 15813 – 15816. a) D. W. Stephan, Org. Biomol. Chem. 2008, 6, 1535 – 1539; b) D. W. Stephan, Acc. Chem. Res. 2014, 48, 306-316. a) L. Greb, C. G. Daniliuc, K. Bergander, J. Paradies, Angew. Chem. Int. Ed. 2013, 52, 5876 – 5879; Angew. Chem. 2013, 125, 5989 – 5992; b) J. A. Nicasio, S. Steinberg, B. Ines, M. Alcarazo, Chem. Eur. J. 2013, 19, 11016 – 11020. Z. Zhang, H. Du, Angew. Chem. Int. Ed. 2015, 54, 623 – 626; Angew. Chem. 2015, 127, 623 – 626. a) X. Ren, G. Li, S. Wei, H. Du, Org. Let. 2015; b) S. Wei, H. Du, J. Am. Chem. Soc. 2014, 136, 12261 – 12264. a) D. J. Chen, Y. T. Wang, J. Klankermayer, Angew. Chem. Int. Ed. 2010, 49, 9475 – 9478; Angew. Chem. 2010, 122, 9665 – 9668; b) Y. B. Liu, H. F. Du, J. Am. Chem. Soc. 2013, 135, 6810 – 6813; c) M. Lindqvist, K. Borre, K. Axenov, B. Kûtai, M. Nieger, M. Leskel, I. Ppai, T. Repo, J. Am. Chem. Soc. 2015, 137, 4038 – 4041. a) L. Greb, P. Ona-Burgos, B. Schirmer, S. Grimme, D. W. Stephan, J. Paradies, Angew. Chem. Int. Ed. 2012, 51, 10164 – 10168; Angew. Chem. 2012, 124, 10311 – 10315; b) L. Greb, S. Tussing, B. Schirmer, P. Ona-Burgos, K. Kaupmees, M. Lokov, I. Leito, S. Grimme, J. Paradies, Chem. Sci. 2013, 4, 2788 – 2796; c) J. Paradies, Angew. Chem. Int. Ed. 2014, 53, 3552 – 3557; Angew. Chem. 2014, 126, 3624 – 3629. a) D. J. Chen, J. Klankermayer, Chem. Commun. 2008, 2130 – 2131; b) P. A. Chase, G. C. Welch, T. Jurca, D. W. Stephan, Angew. Chem. Int. Ed. 2007, 46, 8050 – 8053; Angew. Chem. 2007, 119, 8196 – 8199; c) D. W. Stephan, S. Greenberg, T. W. Graham, P. Chase, J. J. Hastie, S. J. Geier, J. M. Farrell, C. C. Brown, Z. M. Heiden, G. C. Welch, M. Ullrich, Inorg. Chem. 2011, 50, 12338 – 12348. T. A. Rokob, A. Hamza, A. Stirling, I. Papai, J. Am. Chem. Soc. 2009, 131, 2029 – 2036. a) J. Hermeke, M. Mewald, M. Oestreich, J. Am. Chem. Soc. 2013, 135, 17537 – 17546; b) J. M. Blackwell, E. R. Sonmor, T. Scoccitti, W. E. Piers, Org. Lett. 2000, 2, 3921 – 3923. T. Privalov, Eur. J. Inorg. Chem. 2009, 2229 – 2237. a) J. Bigeleisen, M. W. Lee, F. Mandel, Annu. Rev. Phys. Chem. 1973, 24, 407 – 440; b) O. W. Howarth, C. H. Mcateer, P. Moore, G. E. Morris, J. Chem. Commun. 1982, 745 – 747; c) K. E. Janak, G. Parkin, J. Am. Chem. Soc. 2003, 125, 13219 – 13224; d) K. E. Janak, G. Parkin, J. Am. Chem. Soc. 2003, 125, 6889 – 6891. a) F. H. Westheimer, Chem. Rev. 1961, 61, 265 – 273; b) H. Simon, D. Palm, Angew. Chem. Int. Ed. Engl. 1966, 5, 920 – 933; Angew. Chem. 1966, 78, 993 – 1007. a) A. Hadzovic, D. Song, C. M. MacLaughlin, R. H. Morris, Organometallics 2007, 26, 5987 – 5999; b) M. Zimmer-De Iuliis, R. H. Morris, J. Am. Chem. Soc. 2009, 131, 11263 – 11269. D. Chakraborty, A. Rodriguez, E. Y. X. Chen, Macromolecules 2003, 36, 5470 – 5481.
Chem. Eur. J. 2015, 21, 8056 – 8059
www.chemeurj.org
[16] The curve could not be fitted to a pure second order kinetic and was therefore analyzed according to the autocatalytic model. [17] H2 is supplied in large excess from the headspace. [18] F. Mata-Perez, J. F. Perez-Benito, J. Chem. Educ. 1987, 64, 925. [19] Absolute errors were calculated by propagation of uncertainty, rate constants and errors are given as arithmetic means of five individual samples. The large error for B(C6F5)3 is a result of the high reaction rate at the three temperatures. Consequently the acquisition of sufficient data points under our standardized conditions is problematic and more importantly the reaction time is not large in comparison to the time of NMR spectra acquisition. This perquisite is met for the boranes 4 and 5 resulting in lower errors. [20] Our NMR experiments revealed that the boranes 4 and 5 did not form the Lewis adduct with the amine 3 (1:1 mixture) and subsequent pressurization of these samples with H2 and D2 (4 bar) resulted in the equilibration of the FLPs 3/4 and 3/5 with their H2- and D2-activation products [3-H][H-4], [3-H][H-5], [3-D][D-4] and [3-D][D-5], respectively. This finding is in stark contrast to B(C6F5)3, which readily forms the Lewis adduct. see Ref. [21]. [21] The similar analysis could not be performed with B(C6F5)3 since solely the H2-activation product [BnNH2tBu][HB(C6F5)3] is observed implying that the equilibrium lies far on the right-hand side. see P. A. Chase, T. Jurca, D. W. Stephan, Chem. Commun. 2008, 1701 – 1703. [22] a) J. M. Tao, J. P. Perdew, V. N. Staroverov, G. E. Scuseria, Phys. Rev. Lett. 2003, 91; b) S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010, 132, 154104; c) S. Grimme, S. Ehrlich, L. Goerigk, J. Comput. Chem. 2011, 32, 1456 – 1465; d) S. Grimme, J. Chem. Phys. 2006, 124, 034108; e) T. Schwabe, S. Grimme, Phys. Chem. Chem. Phys. 2006, 8, 4398 – 4401; f) F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297 – 3305. [23] TURBOMOLE, version 6.4 and 6.5, R. Ahlrichs et al., Universitt Karlsruhe 2012 and 2013; http://www.turbomole.com. [24] a) F. Eckert, A. Klamt, COSMOtherm, Version C2.1, Release 01.11; COSMOlogic GmbH & Co. KG, Leverkusen, Germany, 2010; b) S. Grimme, Chem. Eur. J. 2012, 18, 9955 – 9964. [25] a) H. Mayr, M. Patz, Angew. Chem. Int. Ed. Engl. 1994, 33, 938 – 957; Angew. Chem. 1994, 106, 990 – 1010; b) H. Mayr, A. R. Ofial, J. Phys. Org. Chem. 2008, 21, 584 – 595; c) D. Richter, H. Mayr, Angew. Chem. Int. Ed. 2009, 48, 1958 – 1961; Angew. Chem. 2009, 121, 1992 – 1995; d) D. Richter, Y. Tan, A. Antipova, X. Q. Zhu, H. Mayr, Chem. Asian J. 2009, 4, 1824 – 1829 ; e) M. Horn, L. H. Schappele, G. Lang-Wittkowski, H. Mayr, A. R. Ofial, Chem. Eur. J. 2012, 18, 249 – 263; f) D. J. Morrison, W. E. Piers, Org. Lett. 2003, 5, 2857 – 2860.
Received: February 27, 2015 Published online on April 15, 2015
8059
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication
& Frustrated Lewis Pairs
Autoinduced Catalysis and Inverse Equilibrium Isotope Effect in the Frustrated Lewis Pair Catalyzed Hydrogenation of Imines Sebastian Tussing,[a] Lutz Greb,[a] Sergej Tamke,[a] Birgitta Schirmer,[b] Claudia Muhle-Goll,[c] Burkhard Luy,[c, d] and Jan Paradies*[a] Abstract: The frustrated Lewis pair (FLP)-catalyzed hydrogenation and deuteration of N-benzylidene-tert-butylamine (2) was kinetically investigated by using the three boranes B(C6F5)3 (1), B(2,4,6-F3-C6H2)3 (4), and B(2,6-F2C6H3)3 (5) and the free activation energies for the H2 activation by FLP were determined. Reactions catalyzed by the weaker Lewis acids 4 and 5 displayed autoinductive catalysis arising from a higher free activation energy (2 kcal mol¢1) for the H2 activation by the imine compared to the amine. Surprisingly, the imine reduction using D2 proceeded with higher rates. This phenomenon is unprecedented for FLP and resulted from a primary inverse equilibrium isotope effect.
The discovery of the metal-free activation of molecular hydrogen by sterically encumbered Lewis pairs sparked new interest in hydrogenation reactions.[1] The outstanding performance of the frustrated Lewis pair (FLP) concept[2] rests on the striking simplicity and the possibility to vary one component, for example the Lewis acid, while the Lewis base remains unchanged. Following this notion, distinct reactivity can be observed when the most popular Lewis acid in FLP B(C6F5)3 (1) was exchanged by weaker Lewis acids giving rise to functional group tolerance.[3] Chiral boranes even facilitated asymmetric hydrogenations of heterocycles,[4] silyl enol ethers,[5] and imines.[6] Similarly, the change of the Lewis base allowed the development of important reactions such as the first metal-free saturation of
olefins[7] using molecular hydrogen (H2). It was shown that the capability to heterolytically split H2 is strongly correlated with the basicity of the applied phosphines.[7b] An interesting situation occurs when the basicity of the Lewis base changes over the course of the reaction, hence leading to a change in the reactivity of the FLP towards H2. An example for such a fascinating reaction is the FLP-mediated hydrogenation of imines,[8] in which both the imine and the hydrogenation product, the amine, are viable Lewis bases for H2 activation. It has been speculated that this reaction might show the feasibility of autoinduced catalysis,[8a, 9] however, we are not aware of any experimental confirmation. The mechanism of the FLP-catalyzed hydrogenation of imines was proposed based on analogies to imine hydrosilylation[10] as well as on quantum-mechanical studies,[9, 11] but has not yet been supported by kinetic experiments. Herein we report that the autoinduced FLP-mediated hydrogenation of imines is only operative for weaker Lewis acids. Furthermore, an unusual inverse equilibrium (deuterium) isotope effect (IEIE)[12] was observed for the first time in FLP chemistry. This effect overcompensated the primary and secondary isotope effects[13] resulting in an overall rate increase. Now, a conclusive mechanistic picture of the most frequently used application of FLP is provided. According to the proposed catalytic cycle, the H2 activation is achieved by the FLP consisting of the borane 1 as Lewis acid and the imine 2 (Scheme 1, simple catalytic cycle), or with the
[a] Dipl.-Chem. S. Tussing, Dr. L. Greb, Dipl.-Chem. S. Tamke, Prof. Dr. J. Paradies Institute of Organic Chemistry, University of Paderborn Warburger Straße 100, 33098 Paderborn (Germany) E-mail: [email protected] [b] Dr. B. Schirmer Center for Multiscale Theory and Computation (CMTC) und Organisch-Chemisches Institut Westflische Wilhelms-Universitt Corrensstrasse 40, 48149 Mìnster (Germany) [c] Priv.-Doz. C. Muhle-Goll, Prof. Dr. B. Luy Institute of Organic Chemistry Karlsruhe Institute of Technology (KIT) Fritz-Haber-Weg 6, 76131 Karlsruhe (Germany) [d] Prof. Dr. B. Luy Department Magnetic Resonance of the Institute for Biological Interfaces Karlsruhe Institute of Technology (KIT), P.O. Box 3640, 76021 (Germany) Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201500805. Chem. Eur. J. 2015, 21, 8056 – 8059
Scheme 1. Simple and autoinduced catalytic cycle for the FLP-mediated imine hydrogenation; Lewis acids including their Lewis acidities.[3, 15]
8056
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication hydrogenation product 3 (autoinduced catalytic cycle) as the acting Lewis base (Scheme 1). It can be assumed that this mechanism is the basis of the autoinduced catalysis because both the starting material and the product are involved in the H2 activation.[14] Quantum-mechanical calculations supported the H2 activation by the FLP comprising borane 1 and imine 2 or amine 3 as the step with highest energy barrier making the heterolytic H2-splitting rate-determining for the overall process (Scheme 1).[9, 11] The transition state energies for those activations were calculated both to be 16.5 kcal mol¢1,[9] although the amine should be a more Figure 2. Eyring plots for the FLP-mediated hydrogenation of 2 (left) and deuterohydropotent Lewis base for the H2 activation than the imi- genation (right) in the presence of the three boranes. ne.[7a, b] Consequently, the H2-splitting by 1 is unselective with regard to the Lewis base resulting in the absence of the characteristic kinetic features of autoinduced cata general autocatalytic reaction.[18] The graphical analysis yieldalysis. ed the values for both rate constants, k1 and k2 for the imineThe reactivities of the three different boranes B(C6F5)3 (1), and the amine-mediated FLP-catalyzed hydrogenation of 2 B(2,4,6-F3-C6F3)3 (4), and B(2,6-F2-C6H3)3 (5) were studied in the and allowed the determination of the free activation paramehydrogenation of 2 at 70 8C (Figure 1). The reaction with ters. The results of the kinetic analysis are summarized in Figure 2 and in Table 1 (see Supporting Information for details). The reaction catalyzed by B(C6F5)3 (1) displayed a 1.5–1.9 fold rate increase for k2 (see Figure 2 a) resulting in an apparent first-order type reaction rate. Consequently, the free activation enthalpies for both cycles are very similar ((21.4 13.6) kcal mol¢1 and (20.1 23.7) kcal mol¢1, Table 1, for detailed error treatment see Ref. [19] and Supporting Information) and their absolute values are within an acceptable agreement with the previously calculated ones (lit. 16.5 kcal mol¢1).[9] This situation changed significantly when the weaker Lewis acids 4 and 5 were used. The rate constants (k2) for the product-accelerated Figure 1. Time versus yield plot for the FLP-catalyzed hydrogenation of Ncycle were about one order of magnitude larger than for the benzylidene-tert-butylamine (2) (5 mol % borane, CD2Cl2, 0.4 m, 70 8C; yields imine-catalyzed cycle (k1) (Figure 2 a). The comparison of the were determined by 1H NMR spectroscopy using hexamethylbenzene as infree reaction enthalpies of activation obtained from k1 and k2 & & ^ ternal standard; B(C6F5)3 (1); ~ B(2,4,6-F3-C6F3)3 (4); B(2,6-F2-C6H3)3 (5); show that the imine-mediated H2 activation by the less LewisB(2,6-F2-C6H3)3 (5) and 5 mol % benzyl-tert-butylamine (3)). acidic boranes is 1.6 kcal mol¢1 and 1.8 kcal mol¢1 higher compared to the amine-mediated H2-splitting, thus triggering the B(C6F5)3 qualitatively proceeded with highest rates exhibiting observed autoinduced catalysis (Table 1). The increased negathe characteristic curve shape for a second-order type reactive activation entropies for the boranes 4 and 5 compared to tion.[16] In contrast, the reaction profiles of the hydrogenations 1 may result from the formation of hydrogen-bonded comcatalyzed by the two weaker Lewis-acidic boranes 4 and 5 displexes in the transition state of the imine-reduction as proplayed sigmoidal curve shapes. This implies a significant inposed by Privalov.[11] Experimental evidence for such interaccrease in rate during the course of the reaction. Addition of 5 mol % of the reaction product 3 at the beginning of the reaction using B(2,6-F2-C6H3)3 (5) as Table 1. Activation parameters for the FLP-catalyzed hydrogenation of 2 (values in parentheses correspond to the reaction with D2). Lewis acid resulted in the same dramatic rate increase, which is a strong indication for autoinduced for k2 (autoinduced catalytic cycle) for k1 (standard catalytic cycle) catalysis. Kinetic data for the FLP-catalyzed hydrogeDH[b] DS[c] DG[a,b] DH[b] DS[c] DG[a,b] nation of 2 using the three boranes were acquired at [d] 1 21.4 13.9 15.6 7.1 ¢19.1 21.1 20.1 23.7 19.6 14.2 ¢5.8 42.5.0 three different temperatures and were analyzed ac(21.0 10.5) (12.9 5.4) (¢25.7 16.0) cording to the following criteria: i) the H2 and borane 4 24.1 3.5 8.5 1.9 ¢48.0 5.1 22.5 2.5 6.9 1.3 ¢49.7 3.7 (24.1 5.6) (6.0 3.0) (¢55.1 8.2) (22.4 3.4) (5.2 1.8) (¢52.0 4.9) concentration in solution is constant,[17] and ii) the ¢51.7 3.4 5 24.5 4.0 6.5 2.1 ¢55.1 5.9 22.7 2.3 5.8 1.2 shift of the equilibrium of the FLP 3/4 and 3/5 to(24.4 4.8) (5.8 2.6) (¢56.7 7.1) (22.5 2.5) (5.7 1.3) (¢51.3 3.7) wards its H2-activation products is slow (vide infra). [a] Calculated for 343 K. [b] In kcal mol¢1. [c] In cal mol¢1 K. [d] The reaction with D2 was These requirements allow a quantitative analysis of fitted to a first-order rate law, hence k2 could not be determined. the collected data conforming with the rate laws for Chem. Eur. J. 2015, 21, 8056 – 8059
www.chemeurj.org
8057
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication tion is supplied by NOESY NMR experiments, which revealed a distinct aggregation of the [BnNH2tBu] cation with the [HB(2,6-F2C6H3)3] anion via intermolecular dihydrogen bonding (see the Supporting Information).[3a] Hydrogenations are strongly affected by deuterium isotope effects. Surprisingly, the deuterohydrogenation of 2 occurred not only with higher rates with all boranes but also the reaction order changed for borane 1 from second-order to firstorder (compare Figure 2 right). Accordingly, the primary deuterium isotope effects decelerate the autoinduced catalytic cycle because an additional deuterium transfer is required to generate the activated iminium [2-D] for the nucleophilic attack. The two boranes 4 and 5 retained second-order kinetics with a strong inverse deuterium isotope effect for k2 (at 70 8C for 4: k1(H)/k1(D) = 0.90; k2(H)/k2(D) = 0.67; for 5: k1(H)/k1(D) = 0.87; k2(H)/k2(D) = 0.71). The Arrhenius analysis of the temperaturedependent equilibrium constants for both the H2 and D2 activation by the FLPs 3/4 and 3/5 provided negative values for the standard reaction enthalpy (see Scheme 2).[20] However,
Scheme 2. Equilibrium of the H2 and D2 activation by benzyl-tert-butylamine (3) and boranes 4 and 5.
DHo of the reaction of the FLP 3/4 or 3/5 with D2 is more negative than with H2 resulting in a substantial primary inverse equilibrium deuterium isotope effect (IEIE) KH/KD. This is the first time that an inverse isotope effect has been reported for FLP-catalyzed hydrogenations. The standard free reaction energies for the FLP-mediated H2 activation were calculated from DH8 and DS8 at 298 K amounting to (¢0.64 0.65) kcal mol¢1 and (0.16 0.26) kcal mol¢1 (for D2 : (¢1.33 0.09) kcal mol¢1 and (¢1.06 1.12) kcal mol¢1) for 3/4 and 3/5, respectively.[21] The enthalpy-driven higher concentrations of [D-4]¢ and [D-5]¢ account for the higher rate of the deuterohydrogenation of 2 and counterbalanced the primary and secondary deuterium isotope effects along the autocatalytic reaction pathway. Furthermore, the reaction was analyzed by DFT methods at a B2PLYP-D3(BJ)/def2-TZVP//TPSS-D3(BJ)/def2-TZVP level[22] as implemented in the TURBOMOLE package[23] including thermal and solvent corrections.[24] The results are summarized in Table 2. The activation of H2 by 1 and the imine 2 is slightly exergonic at 25 8C and almost ergoneutral at 70 8C (Table 2, [Eq. (1)]). For the two other boranes this reaction is endergonic (7–21 kcal mol¢1) at 25 8C and 70 8C. The hydride transfer becomes more exergonic with decreasing Lewis acidity of the borane (1: 100 %, 4: 70 %, 5: 56 %) due to the higher nucleophilicity of the corresponding hydridoborate anion [Eq. (2)].[25] The H2 activation by the amine is for all boranes exergonic but becomes less negative for the weaker Lewis-acidic boranes 4 and 5 (Table 2, [Eq. (3)]). As a result the corresponding ammoChem. Eur. J. 2015, 21, 8056 – 8059
www.chemeurj.org
Table 2. Free reaction energies in dichloromethane at 298 K (343 K) (in kcal mol¢1 1.5 kcal mol¢1).[a] Reaction
Eq. B(C6F5)3 (1)
(1) ¢1.7 (0.45) 2 + H2 + BArF3 ! [2-H][H-BArF3] [2-H][H-BArF3]!3 + BArF3 (2) ¢9.6 (¢11.2) (3) ¢16.9 3 + H2 + BArF3 ! (¢14.6) [3-H][H-BArF3] (4) 1.6 (3.4) 3 + BArF3 ![3·BArF3]
B(2,4,6-F3-C6H2)3 B(2,6-F2-C6H3)3 (5) (4) 7.4 (9.7)
9.8 (21.1)
¢18.8 (¢20.4) ¢7.0 (¢4.6) 7.2 (9.0)
¢21.2 (¢31.8) ¢5.9 (¢3.5) 7.5 (9.4)
[a] DFT calculations with B2PLYP-D3/def2-TZVP, thermal and solvent corrections; see Supporting Information for details.
nium hydridoborate salts are thermodynamically less stable, which is supported by the observation of the reversible H2 activation for boranes 4 and 5. The large negative value for B(C6F5)3 (¢16.9 kcal mol¢1) clearly indicates that the back reaction under H2 liberation is unfavored, which is in close agreement with our NMR results (vide supra). The endergonicity of the Lewis adduct formation (Table 2, [Eq. (4)]) strongly supports our observation that only the FLPs (3/4 and 3/5) or their H2-activation products were detected by NMR spectroscopy. It can be concluded that the FLP-catalyzed hydrogenations of imines proceed via the simple and autoinduced catalytic cycles which are switched by the strength of the Lewis acid. For FLP comprising weaker Lewis-acidic boranes and imines, DG for the H2 activation was 2 kcal mol¢1 higher compared to the H2 activation with the stronger donor. The unusual inverse isotope effect in the deuterohydrogenation of the imine was caused by the higher exothermicity of the D2-activation product, thus counterbalancing the primary and secondary isotope effects. In the future this detailed picture of the FLP-catalyzed imine hydrogenation will enable the development of even more sophisticated FLP and related catalyzed reactions.
Acknowledgements The Landesgraduiertenfçrderung of the State of Baden-Wìrttemberg and the German Science Foundation (DFG) are acknowledged for a Ph.D. grant to S.T. and for a Heisenberg-fellowship to J.P. B.L. acknowledges funding by DFG (Pro2NMR facility) and the HGF programme BIFTM. Dr. Rainer Kerssebaum (Bruker BioSpin GmbH) is thanked for 19F NMR experiments and Prof. Matthias Olzmann (KIT) is thanked for valuable discussions. Keywords: autoinduced catalysis · frustrated Lewis pairs · hydrogen activation · inverse equilibrium isotope effect · kinetics
8058
[1] a) G. C. Welch, R. R. S. Juan, J. D. Masuda, D. W. Stephan, Science 2006, 314, 1124 – 1126; b) G. C. Welch, D. W. Stephan, J. Am. Chem. Soc. 2007, 129, 1880 – 1881; c) D. W. Stephan, G. Erker, Angew. Chem. Int. Ed. 2010, 49, 46 – 76; Angew. Chem. 2010, 122, 50 – 81; d) J. Paradies, Synlett 2013, 777 – 780; e) T. Mahdi, D. W. Stephan, J. Am. Chem. Soc. 2014, 136, Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication
[2] [3]
[4] [5] [6]
[7]
[8]
[9] [10]
[11] [12]
[13]
[14]
[15]
15809 – 15812; f) D. J. Scott, M. J. Fuchter, A. E. Ashley, J. Am. Chem. Soc. 2014, 136, 15813 – 15816. a) D. W. Stephan, Org. Biomol. Chem. 2008, 6, 1535 – 1539; b) D. W. Stephan, Acc. Chem. Res. 2014, 48, 306-316. a) L. Greb, C. G. Daniliuc, K. Bergander, J. Paradies, Angew. Chem. Int. Ed. 2013, 52, 5876 – 5879; Angew. Chem. 2013, 125, 5989 – 5992; b) J. A. Nicasio, S. Steinberg, B. Ines, M. Alcarazo, Chem. Eur. J. 2013, 19, 11016 – 11020. Z. Zhang, H. Du, Angew. Chem. Int. Ed. 2015, 54, 623 – 626; Angew. Chem. 2015, 127, 623 – 626. a) X. Ren, G. Li, S. Wei, H. Du, Org. Let. 2015; b) S. Wei, H. Du, J. Am. Chem. Soc. 2014, 136, 12261 – 12264. a) D. J. Chen, Y. T. Wang, J. Klankermayer, Angew. Chem. Int. Ed. 2010, 49, 9475 – 9478; Angew. Chem. 2010, 122, 9665 – 9668; b) Y. B. Liu, H. F. Du, J. Am. Chem. Soc. 2013, 135, 6810 – 6813; c) M. Lindqvist, K. Borre, K. Axenov, B. Kûtai, M. Nieger, M. Leskel, I. Ppai, T. Repo, J. Am. Chem. Soc. 2015, 137, 4038 – 4041. a) L. Greb, P. Ona-Burgos, B. Schirmer, S. Grimme, D. W. Stephan, J. Paradies, Angew. Chem. Int. Ed. 2012, 51, 10164 – 10168; Angew. Chem. 2012, 124, 10311 – 10315; b) L. Greb, S. Tussing, B. Schirmer, P. Ona-Burgos, K. Kaupmees, M. Lokov, I. Leito, S. Grimme, J. Paradies, Chem. Sci. 2013, 4, 2788 – 2796; c) J. Paradies, Angew. Chem. Int. Ed. 2014, 53, 3552 – 3557; Angew. Chem. 2014, 126, 3624 – 3629. a) D. J. Chen, J. Klankermayer, Chem. Commun. 2008, 2130 – 2131; b) P. A. Chase, G. C. Welch, T. Jurca, D. W. Stephan, Angew. Chem. Int. Ed. 2007, 46, 8050 – 8053; Angew. Chem. 2007, 119, 8196 – 8199; c) D. W. Stephan, S. Greenberg, T. W. Graham, P. Chase, J. J. Hastie, S. J. Geier, J. M. Farrell, C. C. Brown, Z. M. Heiden, G. C. Welch, M. Ullrich, Inorg. Chem. 2011, 50, 12338 – 12348. T. A. Rokob, A. Hamza, A. Stirling, I. Papai, J. Am. Chem. Soc. 2009, 131, 2029 – 2036. a) J. Hermeke, M. Mewald, M. Oestreich, J. Am. Chem. Soc. 2013, 135, 17537 – 17546; b) J. M. Blackwell, E. R. Sonmor, T. Scoccitti, W. E. Piers, Org. Lett. 2000, 2, 3921 – 3923. T. Privalov, Eur. J. Inorg. Chem. 2009, 2229 – 2237. a) J. Bigeleisen, M. W. Lee, F. Mandel, Annu. Rev. Phys. Chem. 1973, 24, 407 – 440; b) O. W. Howarth, C. H. Mcateer, P. Moore, G. E. Morris, J. Chem. Commun. 1982, 745 – 747; c) K. E. Janak, G. Parkin, J. Am. Chem. Soc. 2003, 125, 13219 – 13224; d) K. E. Janak, G. Parkin, J. Am. Chem. Soc. 2003, 125, 6889 – 6891. a) F. H. Westheimer, Chem. Rev. 1961, 61, 265 – 273; b) H. Simon, D. Palm, Angew. Chem. Int. Ed. Engl. 1966, 5, 920 – 933; Angew. Chem. 1966, 78, 993 – 1007. a) A. Hadzovic, D. Song, C. M. MacLaughlin, R. H. Morris, Organometallics 2007, 26, 5987 – 5999; b) M. Zimmer-De Iuliis, R. H. Morris, J. Am. Chem. Soc. 2009, 131, 11263 – 11269. D. Chakraborty, A. Rodriguez, E. Y. X. Chen, Macromolecules 2003, 36, 5470 – 5481.
Chem. Eur. J. 2015, 21, 8056 – 8059
www.chemeurj.org
[16] The curve could not be fitted to a pure second order kinetic and was therefore analyzed according to the autocatalytic model. [17] H2 is supplied in large excess from the headspace. [18] F. Mata-Perez, J. F. Perez-Benito, J. Chem. Educ. 1987, 64, 925. [19] Absolute errors were calculated by propagation of uncertainty, rate constants and errors are given as arithmetic means of five individual samples. The large error for B(C6F5)3 is a result of the high reaction rate at the three temperatures. Consequently the acquisition of sufficient data points under our standardized conditions is problematic and more importantly the reaction time is not large in comparison to the time of NMR spectra acquisition. This perquisite is met for the boranes 4 and 5 resulting in lower errors. [20] Our NMR experiments revealed that the boranes 4 and 5 did not form the Lewis adduct with the amine 3 (1:1 mixture) and subsequent pressurization of these samples with H2 and D2 (4 bar) resulted in the equilibration of the FLPs 3/4 and 3/5 with their H2- and D2-activation products [3-H][H-4], [3-H][H-5], [3-D][D-4] and [3-D][D-5], respectively. This finding is in stark contrast to B(C6F5)3, which readily forms the Lewis adduct. see Ref. [21]. [21] The similar analysis could not be performed with B(C6F5)3 since solely the H2-activation product [BnNH2tBu][HB(C6F5)3] is observed implying that the equilibrium lies far on the right-hand side. see P. A. Chase, T. Jurca, D. W. Stephan, Chem. Commun. 2008, 1701 – 1703. [22] a) J. M. Tao, J. P. Perdew, V. N. Staroverov, G. E. Scuseria, Phys. Rev. Lett. 2003, 91; b) S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010, 132, 154104; c) S. Grimme, S. Ehrlich, L. Goerigk, J. Comput. Chem. 2011, 32, 1456 – 1465; d) S. Grimme, J. Chem. Phys. 2006, 124, 034108; e) T. Schwabe, S. Grimme, Phys. Chem. Chem. Phys. 2006, 8, 4398 – 4401; f) F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297 – 3305. [23] TURBOMOLE, version 6.4 and 6.5, R. Ahlrichs et al., Universitt Karlsruhe 2012 and 2013; http://www.turbomole.com. [24] a) F. Eckert, A. Klamt, COSMOtherm, Version C2.1, Release 01.11; COSMOlogic GmbH & Co. KG, Leverkusen, Germany, 2010; b) S. Grimme, Chem. Eur. J. 2012, 18, 9955 – 9964. [25] a) H. Mayr, M. Patz, Angew. Chem. Int. Ed. Engl. 1994, 33, 938 – 957; Angew. Chem. 1994, 106, 990 – 1010; b) H. Mayr, A. R. Ofial, J. Phys. Org. Chem. 2008, 21, 584 – 595; c) D. Richter, H. Mayr, Angew. Chem. Int. Ed. 2009, 48, 1958 – 1961; Angew. Chem. 2009, 121, 1992 – 1995; d) D. Richter, Y. Tan, A. Antipova, X. Q. Zhu, H. Mayr, Chem. Asian J. 2009, 4, 1824 – 1829 ; e) M. Horn, L. H. Schappele, G. Lang-Wittkowski, H. Mayr, A. R. Ofial, Chem. Eur. J. 2012, 18, 249 – 263; f) D. J. Morrison, W. E. Piers, Org. Lett. 2003, 5, 2857 – 2860.
Received: February 27, 2015 Published online on April 15, 2015
8059
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
