Metal-free transfer hydrogenation of olefins via dehydrocoupling catalysis Manuel Pérez, Christopher B. Caputo, Roman Dobrovetsky, and Douglas W. Stephan1 Department of Chemistry, University of Toronto, Toronto, ON, Canada M5S 3H6 Edited by Richard Eisenberg, University of Rochester, Rochester, NY, and approved June 12, 2014 (received for review April 23, 2014)
fluorophosphonium
A
wide range of commodity chemicals, petrochemicals, pharmaceuticals, materials, and foods depend on industrial-scale hydrogenation reactions. Catalysts for this process are based on both heterogeneous transition metals materials and homogeneous transition metal complexes. Indeed, it was the discovery of Sabatier (1) in the early 20th century that revealed the utility of amorphous Ni and other metals in hydrogenation processes. Following the dawn of organometallic chemistry in the 1960s, homogeneous catalysts were developed principally based on precious metals (2–4). Although subsequent modifications and optimizations have led to numerous highly selective and efficient catalyst technologies, concerns over cost, natural abundance, and toxicity have prompted efforts to uncover catalysts based on “earth-abundant” elements. The principal targets in these efforts have been the first-row transition metals. Indeed, the groups of Chirik (5, 6), Morris (7), Beller (8), and others (9) have developed remarkably active and selective catalysts based on Fe. Strategies to develop reduction methods using organic or main-group–based reagents have also been explored. For example, use of Hantzsch esters (10), Birch reduction of arenes (11), or the use of B or Al hydrides are effective, albeit stoichiometric reductions. More recently, the advent of “frustrated Lewis pairs” (FLPs) have led to the discovery that combinations of sterically encumbered donors and acceptors act as catalysts for the reductions of imines, enamines, silylenol ethers, olefins, and alkynes (12). To uncover hydrogenation strategies using earth-abundant elements, we have focused our efforts on Lewis acidic phosphorus species. Although Gudat (13), Burford and coworker (14), Yoshifuji (15), and Bertrand and coworker (16), among others, have reported phosphenium cations that demonstrate Lewis acidity, Radosevich and coworkers (17) have exploited the reaction of the unique planar P(III) species with ammoniaborane to give a P(V)H2 derivative that effects the subsequent reduction of diazobenzene. Although the acidity of P(V) has been exploited previously in ylide reagents (18), Diels–Alder reactions catalysis (19), and addition reactions to polar unsaturates (20), Gabbaï and coworkers (21) have more recently exploited the acceptor capabilities of phosphonium cations, in fluoride sensor strategies. In our own efforts, we have recently reported the synthesis of highly electrophilic phosphonium www.pnas.org/cgi/doi/10.1073/pnas.1407484111
cations (EPCs) (22). The species [(C6F5)3PF][B(C6F5)4] 1 proved to be an effective catalyst for the hydrodefluorination of fluoroalkanes in the presence of silane (22). We subsequently showed that 1 could effectively catalyze the hydrosilylation of olefins and alkynes (23). In this report, we first exploit the Lewis acidity of 1 to catalyze the dehydrocoupling of silane with a variety of species including amines, thiols, phenols, and carboxylic acids, liberating H2. We then show that these intermolecular and intramolecular dehydrocoupling reactions can be used for transfer hydrogenation of olefins, thus providing an unprecedented route to metal-free hydrogenation of olefins. Results and Discussion Initially, 1.5 mol% of 1, Ph2NH, and Et3SiH were allowed to react for 10 h. This afforded near complete conversion to Ph2NSiEt3 (Table 1, entry 1). Interestingly, when Et3SiH was replaced by ClMe2SiH, the reaction required only 1 h for complete conversion (Table 1, entry 2). In contrast, reactions with more sterically encumbered silanes such as Ph3SiH or PhMe2SiH resulted in an increased reaction time (Table 1, entries 3, 4), whereas the bulkier silane, iPr3SiH (Table 1, entry 5) precluded the reaction. para-Methyl substituted anilines showed a slower reaction time (Table 1, entries 6–8). iPr2NH and aniline (Table 1, entries 9, 10) led to catalyst degradation and thus no catalytic activity. Compound 1 was also very effective in the catalytic dehydrocoupling of thiophenols (Table 1, entries 11–14) and phenols (Table 1, entries 16–21) with Et3SiH. In general, these reactions were complete in 1 h with the exception of C6F5SH and C6F5OH, which required a longer time for completion (Table 1, entries 15 and 20, respectively). In this prior case (C6F5SH), heating at 100 °C accelerated the reaction, resulting in complete conversion in 3 h. In the case when para-methoxy phenol was reacted with the excess of Et3SiH, both dehydrocoupling of Significance For more than a century, hydrogenation has been limited to the use of transition metal-based catalysts. With the emerging focus on the chemistry of earth-abundant elements, the 21st century has seen a renaissance in main-group chemistry. In this work, an electrophilic phosphonium cation is shown to act as main-group catalyst effecting the dehydrocoupling of silane and amines, phenols, thiols, and carboxylic acids with the concurrent release of H2. In addition, performing the reactions in the presence of olefins, dehydrocoupling occurs with simultaneous hydrogenation of the olefin. This chemistry provides an unprecedented avenue to metal-free transfer hydrogenation catalysis of olefins. Author contributions: M.P., C.B.C., R.D., and D.W.S. designed research; M.P., C.B.C., and R.D. performed research; M.P., C.B.C., R.D., and D.W.S. analyzed data; and M.P., C.B.C., R.D., and D.W.S. wrote the paper. The authors declare no conflict of interest. This article is a PNAS Direct Submission. 1
To whom correspondence should be addressed. E-mail: [email protected].
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. 1073/pnas.1407484111/-/DCSupplemental.
PNAS Early Edition | 1 of 5
CHEMISTRY
A major advance in main-group chemistry in recent years has been the emergence of the reactivity of main-group species that mimics that of transition metal complexes. In this report, the Lewis acidic phosphonium salt [(C6F5)3PF][B(C6F5)4] 1 is shown to catalyze the dehydrocoupling of silanes with amines, thiols, phenols, and carboxylic acids to form the Si-E bond (E = N, S, O) with the liberation of H2 (21 examples). This catalysis, when performed in the presence of a series of olefins, yields the concurrent formation of the products of dehydrocoupling and transfer hydrogenation of the olefin (30 examples). This reactivity provides a strategy for metalfree catalysis of olefin hydrogenations. The mechanisms for both catalytic reactions are proposed and supported by experiment and density functional theory calculations.
Table 1. Catalytic dehydrocoupling reactions
Entry
E-H
Silane
T, h
Yield, %*
1
Ph2NH
Et3SiH
2
Ph2NH
ClMe2SiH
3
Ph2NH
Ph3SiH
20
>99
4
Ph2NH
PhMe2SiH
48
>99
5
Ph2NH
iPr3SiH
96
0
6
p-Me(C6H4)2NH
Et3SiH
30
>99
7
p-Me(C6H4)2NH
ClMe2SiH
16
>99
8
p-Me(C6H4)2NH
Ph3SiH
36
40
9
iPr2NH
Et3SiH
48
0
10
PhNH2
Et3SiH
48
0
11
PhSH
Et3SiH
99
12
p-Me(C6H4)SH
Et3SiH
99
13
p-Cl(C6H4)SH
Et3SiH
99
14
p-F(C6H4)SH
Et3SiH
99
15
C6F5SH
Et3SiH
168
>99
PhOH
Et3SiH
2
>99
o-(Me)2(C6H3)OH
Et3SiH
2
>99
p-OMe(C6H4)OH
Et3SiH
18
>99
p-Me(C6H4)OH
Et3SiH
3
>99
C6F5OH
Et3SiH
24
>99
p-C8H17(C6H4)CO2H
Et3SiH
1
>99
16
†
17† 18
†
19† 20 21
†
10
>99
1
>99
Conditions: To a solution of the catalyst (1.5 mol%) in C6D5Br or CD2Cl2 (1.0 mL) were added silane (1.1 Eq) and E-H (1.0 Eq) at 25 °C. *Yields measured by 1H-NMR. † 1.0 mol% of catalyst was used.
the OH group and the methoxy moiety was replaced by silyl fragments; however, one equivalent of silane leads selectively to the product of dehydrocoupling of the phenol (entry 18). This observation is similar to that reported for the hydrosilylation of methoxy-α-methylstyrenes (23). The carboxylic acid p-C8H17(C6H4)CO2H and Et3SiH gave the corresponding ester in less than 1 h (entry 21). The mechanism for these dehydrocoupling reactions is thought to be analogous to that previously proposed for 1 catalyzed olefin hydrosilylation (22). Thus, the initial step involves activation of the Si-H bond by the fluorophosphonium Lewis acid 1. Backside attack of the LUMO at Si center by the Lewis base (N, O, S) generates transient hypervalent silicon species that has hydridic (Si-H) and protic (E-H, E = N, O, S) hydrogens prompting loss of dihydrogen (Fig. 1). This mechanism is conceptually similar to related Lewis acid mediated hydrosilylations described by Piers and coworkers (24), Oestreich and coworker (25), and Gevorgyan, Yamamoto, and coworkers (26). Furthermore, the proposed mechanism is consistent with the experimental observation that basic amines and sterically less encumbered anilines do not dehydrocouple with Et 3 SiH, suggesting that 2 of 5 | www.pnas.org/cgi/doi/10.1073/pnas.1407484111
these bases interact stronger with the Lewis acid 1 than Et3SiH. At the same time, sterically demanding silanes slow or even stop the dehydrocoupling reaction (Table 1, entry 5), which is consistent with the proposed role of hypervalent silicon species. Furthermore, the use of silane in excess (10.0 Eq) with respect to the pTol2NH, accelerates the reaction, whereas using a large excess of amine dramatically increases the reaction time. This supports the view that equilibrium involving binding of silane or aniline to P(V) is competitive, and the dehydrocoupling reaction proceeds via silane activation. The quest for theoretical insight regarding the dehydrocoupling reaction of silane and amine with the Lewis acid 1 prompted preliminary gas-phase density functional theory (DFT) calculations at the WB97XD/def2TZV (27, 28) level of theory using the cation of 1, Me3SiH, and Ph2NH. Phosphonium cation interacts with both Me3SiH and Ph2NH resulting in the Me3SiH∙∙∙P and Ph2(H)N∙∙∙P distances of 2.3 and 3.5 Å, respectively. Both interactions are exothermic by 15.19 (ΔG = −1.8) and 23.35 (ΔG = −6.3) kcal·mol−1, respectively. This suggests that these species exist in an equilibrium (Keq = 5 × 10−4), slightly favoring the amine–phosphonium interaction, which was Pérez et al.
also observed experimentally. Coordination of amine to the phosphonium bound silane generating a transient five-coordinate silicon intermediate is also exothermic with ΔH = −37.1 (ΔG = −23.7) kcal·mol−1, which yields (C6F5)3P(F)H and [Ph2N(H)SiMe3]+ in a slightly endothermic (ΔH = 2.8 kcal·mol−1), but exergonic (ΔG = −9.3 kcal·mol−1) step. The subsequent reaction of these two species to yield the dehydrocoupling product is driven by the liberation of H2 even though the reaction is somewhat endothermic and endergonic (ΔH = 43.5 and ΔG = 34.1 kcal·mol−1). Overall, this catalytic cycle is computed to be slightly exothermic (ΔH = −6.0 kcal·mol−1) with only a small overall ΔG of −0.6 kcal·mol−1. This dehydrocoupling mechanism is reminiscent of that described for analogous B(C6F5)3 catalyzed reaction in the classic work from Piers and coworkers (29), which subsequently further illuminated and expanded by the groups of Oestreich (30), Brook (31), Rubinsztajn (32), and Paradies (33). In the context of dehydrocoupling, it is also important to note the pioneering work from the Tilley and Manners research groups, who exploited transition metal catalysts for this purpose (34, 35), whereas Hill et al. (36) described Zn, Mg, Ca, and Sr catalyzed preparations of amino silanes and Baba and coworker (37) have used InBr3 to dehydrocouple silanes and carboxylic acids (Table 1 and Fig. 1). The above dehydrocoupling reactions proceed with the liberation of H2. We queried the possibility of capturing this H2 by the addition of an olefin. Thus, 1,1-diphenylethylene was added to a mixture of p-Me(C6H4)2NH, Et3SiH, or Ph3SiH and 1.5 mol% of 1; heating this mixture to 100 °C gave after 5 h the dehydrocoupled product and 1,1-diphenylethane with 60% and >99% yield, respectively (Table 2, entries 1 and 2). Reactions occurred at room temperature; however, the reaction times and conversions improved at elevated temperatures (SI Appendix). Transfer hydrogenation is not limited to anilines, as substituted thiophenols, phenols, and carboxylic acids are competent proton sources as well, resulting in complete reduction of the olefin in less than 1 h (Table 2, entries 3–8). In a similar fashion, the olefin, 1-methyl-4-(prop1-en-2-yl)benzene was catalytically reduced using either ditolylamine or tolylthiophenol in combination with Et3SiH, affording yields up to 75%. The aniline reactions were done at 100 °C, whereas the reductions using thiol were performed at 25 °C (Table 2, entries 9–12). 2-Methylstyrene was reduced in up to 67% yields under similar conditions using thiophenols as the proton source at 25 °C (Table 2, entries 13 and 14), whereas (E)-α-methylstilbene was hydrogenated in high yields using thiophenol or carboxylic acids (Table 2, entries 15–17). Interestingly, the carboxylic acid C6F5CO2H with Et3SiH resulted in complete reduction of internal Pérez et al.
PNAS Early Edition | 3 of 5
CHEMISTRY
Fig. 1. Proposed mechanistic pathways for dehydrocoupling of silane and amine and transfer hydrogenation of olefins (experiment: R3Si = Et3Si, ClMe2Si, Ph3Si, PhMe2Si; Ar2NH = Ph2NH, p-Me(C6H4)2NH; calculations: R3Si = Me3Si; Ar = Ph). Gibbs free energies and enthalpies in parentheses for every step are provided in kilocalories per mole.
olefins (E)-α-methylstilbene (Table 2, entry 18) and 1-phenyl-2,2diphenylethylene in less than 2 h (Table 2, entry 19). The reduction of 3,3-dimethylbut-1-ene using p-Cl(C6H4)SH and p-Cl(C6H4)OH as the proton sources yielded 76% and 69% of 2,3-dimethylbutane, respectively (Table 2, entries 20 and 21 results from methyl migration). Methylenecyclohexane and 1-methylcyclohex-1-ene were also successfully reduced to methylcyclohexane in good yields (Table 2, entries 22–26). The olefinic unit of dibutyl 2-methylenesuccinate was also hydrogenated using arylthiols and silane at 25 °C, displaying compatibility with unsaturated esters (Table 2, entries 27–29). Finally, using 2-methylenesuccinic acid with Et3SiH in catalytic presence of 1, resulting in reduction of the internal alkene. Subsequent hydrolysis of the silyl-ester affords the corresponding acid in 89% isolate yield (Table 2, entry 30). It is noteworthy that, in all cases, neither hydrogenation nor dehydrocoupling is observed in the absence of the fluorophosphonium catalyst 1. These reactions provide access to both dehydrocoupling products as well as the alkanes derived from olefin hydrogenation. It is also important to note that these reactions are highly chemoselective, providing exclusively the products of dehydrocoupling and transfer hydrogenation with no hydrosilylation of olefin byproducts. In addition, these reactions exhibit impressive functional group tolerance, proceeding in the presence of aryl-halides, amines, thiols, ethers, and ester fragments. The mechanism for these transfer hydrogenations is thought to proceed by the intervention of the olefin in dehydrocoupling reaction pathway. Thus, rather than the elimination of H2, the transient [Ar2N (H)SiR3]+ protonates the olefin, generating carbocation that subsequently abstracts the hydride from (C6F5)3P(F) H (Fig. 1), yielding the alkane and dehydrocoupling product. This view of the mechanism is supported by the observation that these transfer hydrogenations are favored by the 1,1-disubstituted olefins. In the case of tBuCH=CH2 (Table 2, entries 19 and 20), protonation prompts methyl migration, also generating tertiary carbocation. To support the proposed mechanism, the catalysis using Et3SiD in combination with a proton source was performed, leading exclusively to deuteration of the more electrophilic position of the olefin (SI Appendix). This is consistent with a Markovnikov delivery of proton to a less substituted carbon, forming tertiary carbocation followed by the reaction of the latter with deuteride. DFT calculations at the WB97XD/def2TZV (27, 28) level of theory for hydrogenation of 1,1-diphenylethylene show that the protonation of the 1,1-diphenylethylene by [Ph2N(H)SiMe3]+ is endothermic [ΔH = 14.1 (ΔG = 13.7) kcal·mol−1], whereas the subsequent hydride abstraction from (C6F5)3P(F)H affording 1,1-diphenylethane and regenerating catalyst is almost thermal neutral (ΔH = 1.1 and ΔG = 0.0 kcal·mol−1). Overall, concurrent transfer hydrogenation with dehydrocoupling is energetically favorable to loss of H2 as it requires 28.3 kcal·mol−1 less energy, and lower by 20.5 kcal·mol−1 Gibbs free energy. This was also supported by the competition experiment in which the dehydrocoupling reaction in the presence of 1,1-diphenylethylene was observed to be significantly faster than in its absence (SI Appendix). Transfer hydrogenations of ketones, aldehydes, and imines can be achieved by the classic Meerwein–Ponndorf–Verley protocol, catalyzed by a variety of transition metal complexes and more recently by FLPs (38). In contrast, transfer hydrogenation of olefins has received much less attention. Homogeneous catalysts based on Fe (39) and Ru (40) were demonstrated in the 1970s, whereas the groups of Berke (41), Peters (42), and Manners (43) have more recently reported the transfer hydrogenation of olefins using Re-, Co-, and Rh-based catalysts with ammonia-boranes as the H2 source. Heterogeneous catalysts for transfer hydrogenation of olefins include supported Ni, Pd, or Ir and have exploited isopropanol, formic acid, ammonium formate, and glycerol as sources of H2 (44, 45). Catalyst-free reductions of olefins have also been reported, although this
Table 2. Transfer hydrogenation of olefins with concurrent dehydrocoupling catalysis
Entry
E-H
R3SiH
T, °C
T, h
Yield, %*
Ph2C = CH2 1
p-Me(C6H4)2NH
Et3SiH
100
5
60
2
p-Me(C6H4)2NH
Ph3SiH
100
6
>99
3
p-Me(C6H4)SH
Et3SiH
25
99
4
p-Cl(C6H4)SH
Et3SiH
25
99 (94)
5
p-F(C6H4)SH
Et3SiH
25
99 (96)
6
p-MeO(C6H4)OH
Et3SiH
25
99
8
C6F5CO2H
Et3SiH
25
1
>99 (94)
>99 (93)
p-MeC6H4(Me)C = CH2 9
p-Me(C6H4)2NH
Et3SiH
100
3
10†
p-Me(C6H4)2NH
Et3SiH
100
3
56 74
11
p-Me(C6H4)SH
Et3SiH
25
99
>99 (95)
BuO2CCH2(BuO2C)C = CH2 27
p-Me(C6H4)SH
Et3SiH
25
1
28
p-Cl(C6H4)SH
Et3SiH
25
1
89
29
p-F(C6H4)SH
Et3SiH
25
1
98 (96)
Et3SiH
25
20
96 (89)¶
HO2CCH2(HO2C)C = CH2 30
—
Conditions: To a solution of silane (1.0 Eq), R-H (1.0 Eq), and olefin (1.0 Eq) in C6D5Br or CD2Cl2 (1.0 mL) was added the catalyst (1.5 mol%). *Yields determined by 1H-NMR spectroscopy (isolated yield). † 2.0 Eq of olefin were used. ‡ To a solution of R-H (1.0 Eq) and olefin (1.0 Eq) in C6D5Br (1.0 mL) were added a mixture of silane (1.0 Eq) and the catalyst (1.5 mol%). § For complete conversion 1.5 Eq of reagents were used with respect to the olefin. { Isolated yield of the corresponding acid.
required the generation of the energetic diimide from hydrazine monohydrate and O2 (46). Although FLP catalysts have been shown to effect olefin hydrogenation, the present system, which exploits concurrent dehydrocoupling, is (to our knowledge) the first main-group system to mediate transfer hydrogenations of olefins. 4 of 5 | www.pnas.org/cgi/doi/10.1073/pnas.1407484111
Summary In conclusion, we have described a general and efficient procedure for dehydrocoupling of silanes with anilines, phenols, thiophenols, and benzoic acids catalyzed by EPCs. Moreover, we showed that these reactions can be used for in situ transfer hydrogenation of olefins. Thus, the present work further Pérez et al.
Methods General Procedure for Dehydrocoupling Reactions. To a solution of the catalyst [(C6F5)3PF][B(C6F5)4] (1.0–1.5 mol%) in C6D5Br or CD2Cl2 (0.34 M) was added the corresponding silane (1.0–1.2 Eq) and RH (R = Ar2N, ArS, ArO, ArCO2) (1.0 Eq) at 25 °C. The reaction was monitored by NMR analysis until the reaction was complete. Yield was determined by 1H-NMR spectroscopy. Freshly prepared catalyst was used and resulted in the optimal yields.
25 °C. The reaction was monitored by NMR or TLC until completion. Yield was determined by 1H-NMR spectroscopy. For isolated yields, the reaction was quenched with a diluted solution of NaHCO3 and the mixture was extracted with CH2Cl2. The organic solution was dried over MgSO4, filtered, and evaporated. The crude was diluted with hexane and filtered over silica gel; products were eluted with hexane and Et2O for dibutyl 2-methylenesuccinate. The quality of the catalyst is again essential for the successful completion of the reaction. Supplementary Data. Details of the syntheses, spectroscopic data are included in the SI Appendix.
General Procedure for Olefin Transfer-Hydrogenation. To a solution of silane (1.0 Eq), RH (R = Ar2N, ArS, ArO, ArCO2) (1.0 Eq) and olefin (1.0–1.2 Eq) was added the [(C 6F5) 3PF][B(C 6F5) 4] (1.5 mol%) in C6D5 Br or CD2Cl 2 (0.5 M) at
ACKNOWLEDGMENTS. We thank the Natural Sciences and Engineering Research Council (NSERC) of Canada for financial support. D.W.S. is grateful for the award of a Canada Research Chair. C.B.C. is grateful for the award of an NSERC Postgraduate Scholarship and a Walter C. Sumner Fellowship.
1. Sabatier P (1926) How I have been led to the direct hydrogenation method by metallic catalysts. Ind Eng Chem 18(10):1005–1008. 2. Hallman PS, Evans D, Osborn JA, Wilkinson G (1967) Selective catalytic homogeneous hydrogenation of terminal olefins using tris(triphenylphosphine)hydridochlororuthenium(II); hydrogen transfer in exchange and isomerisation reactions of olefins. Chem Commun 1967(7):305–306. 3. Boerner A, Holz J (2004) Homogeneous hydrogenations: Olefin hydrogenations. Transition Metals for Organic Synthesis, eds Beller M, Bolm C (Wiley, New York), 2nd Ed, Vol 2, pp 3–13. 4. Pettinari C, Marchetti F, Martini D (2004) Metal complexes as hydrogenation catalysts. Comprehensive Coordination Chemistry II, eds McCleverty JA, Meyer TJ (Elsevier, Oxford), Vol 9, pp 75–139. 5. Friedfeld MR, et al. (2013) Cobalt precursors for high-throughput discovery of base metal asymmetric alkene hydrogenation catalysts. Science 342(6162):1076–1080. 6. Trovitch RJ, Lobkovsky E, Bill E, Chirik PJ (2008) Functional group tolerance and substrate scope in bis(imino)pyridine iron catalyzed alkene hydrogenation. Organometallics 27(7):1470–1478. 7. Zuo W, Lough AJ, Li YF, Morris RH (2013) Amine(imine)diphosphine iron catalysts for asymmetric transfer hydrogenation of ketones and imines. Science 342(6162): 1080–1083. 8. Jagadeesh RV, et al. (2013) Nanoscale Fe2O3-based catalysts for selective hydrogenation of nitroarenes to anilines. Science 342(6162):1073–1076. 9. Bullock RM (2013) Chemistry. Abundant metals give precious hydrogenation performance. Science 342(6162):1054–1055. 10. Hantzsch A (1881) Condensationprodukte aus Aldehydammoniak und Ketoniartigen Verbindungen. Chem Ber 14:1637–1638. 11. Birch AJ (1944) A new reagent for primary and secondary aliphatic amines. J Chem Soc 1944:314–315. 12. Stephan DW, Erker G (2013) Frustrated lewis pair mediated hydrogenations. Top Curr Chem 332:85–110. 13. Gudat D (2010) Diazaphospholenes: N-Heterocyclic phosphines between molecules and Lewis pairs. Acc Chem Res 43(10):1307–1316. 14. Burford N, Ragogna PJ (2002) New synthetic opportunities using Lewis acidic phosphines. Dalton Trans (23):4307–4315. 15. Yoshifuji M (2009) Product class 3: Phosphenium salts. Sci Synth 42:63–69. 16. Guerret O, Bertrand G (1997) Trigonal planar phosphorus cations. Acc Chem Res 30(12):486–493. 17. Dunn NL, Ha M, Radosevich AT (2012) Main group redox catalysis: Reversible P(III)/P(V) redox cycling at a phosphorus platform. J Am Chem Soc 134(28):11330–11333. 18. Wittig G, Schöllkopf U (1954) Über Triphenyl-phosphin-methylene als Olefinbildende Reagenzien I. Chem Ber 87:1318–1330. 19. Terada M, Kouchi M (2006) Novel metal-free Lewis acid catalysis by phosphonium salts through hypervalent interaction. Tetrahedron 62(2-3):401–409. 20. Sereda O, Tabassum S, Wilhelm R (2010) Lewis acid organocatalysts. Top Curr Chem 291:349–393. 21. Hudnall TW, Kim YM, Bebbington MWP, Bourissou D, Gabbaï FP (2008) Fluoride ion chelation by a bidentate phosphonium/borane Lewis acid. J Am Chem Soc 130(33): 10890–10891. 22. Caputo CB, Hounjet LJ, Dobrovetsky R, Stephan DW (2013) Lewis acidity of organofluorophosphonium salts: Hydrodefluorination by a saturated acceptor. Science 341(6152):1374–1377. 23. Pérez M, Hounjet LJ, Caputo CB, Dobrovetsky R, Stephan DW (2013) Olefin isomerization and hydrosilylation catalysis by Lewis acidic organofluorophosphonium salts. J Am Chem Soc 135(49):18308–18310. 24. Parks DJ, Blackwell JM, Piers WE (2000) Studies on the mechanism of B(C6F5)3-catalyzed hydrosilation of carbonyl functions. J Org Chem 65(10):3090–3098. 25. Rendler S, Oestreich M (2008) Conclusive evidence for an SN2-Si mechanism in the B(C6F5)3-catalyzed hydrosilylation of carbonyl compounds: Implications for the related hydrogenation. Angew Chem Int Ed Engl 47(32):5997–6000.
26. Gevorgyan V, Rubin M, Benson S, Liu JX, Yamamoto Y (2000) A novel B(C6F5)3-catalyzed reduction of alcohols and cleavage of aryl and alkyl ethers with hydrosilanes. J Org Chem 65(19):6179–6186. 27. Grimme S (2006) Semiempirical GGA-type density functional constructed with a longrange dispersion correction. J Comput Chem 27(15):1787–1799. 28. Chai JD, Head-Gordon M (2008) Long-range corrected hybrid density functionals with damped atom-atom dispersion corrections. Phys Chem Chem Phys 10(44):6615–6620. 29. Blackwell JM, Foster KL, Beck VH, Piers WE (1999) B(C6F5)3-catalyzed silation of alcohols: A mild, general method for synthesis of silyl ethers. J Org Chem 64(13): 4887–4892. 30. Hermeke J, Mewald M, Oestreich M (2013) Experimental analysis of the catalytic cycle of the borane-promoted imine reduction with hydrosilanes: Spectroscopic detection of unexpected intermediates and a refined mechanism. J Am Chem Soc 135(46): 17537–17546. 31. Brook MA, Grande JB, Ganachaud F (2011) New synthetic strategies for structured silicones using B(C6F5)3. Adv Polym Sci 235:161–183. 32. Cella J, Rubinsztajn S (2008) Preparation of polyaryloxysilanes and polyaryloxysiloxanes by B(C6F5)3 catalyzed polyetherification of dihydrosilanes and bis-phenols. Macromolecules 41(19):6965–6971. 33. Greb L, Tamke S, Paradies J (2014) Catalytic metal-free Si-N cross-dehydrocoupling. Chem Commun (Camb) 50(18):2318–2320. 34. Sadow AD, Tilley TD (2005) Synthesis and characterization of scandium silyl complexes of the type Cp*2ScSiHRR’. sigma-Bond metathesis reactions and catalytic dehydrogenative silation of hydrocarbons. J Am Chem Soc 127(2):643–656. 35. Waterman R (2013) Mechanisms of metal-catalyzed dehydrocoupling reactions. Chem Soc Rev 42(13):5629–5641. 36. Hill MS, Liptrot DJ, MacDougall DJ, Mahon MF, Robinson TP (2013) Hetero-dehydrocoupling of silanes and amines by heavier alkaline earth catalysis. Chem Sci 4(11): 4212–4222. 37. Motokura K, Baba T (2012) An atom-efficient synthetic method: Carbosilylations of alkenes, alkynes, and cyclic acetals using Lewis and Bronsted acid catalysts. Green Chem 14(3):565–579. 38. Farrell JM, Heiden ZM, Stephan DW (2011) Metal-free transfer hydrogenation catalysis by B(C6F5)3. Organometallics 30(17):4497–4500. 39. Nishiguchi T, Fukuzumi K (1971) Homogeneous transfer-hydrogenation of olefins catalyzed by FeII, CoII, and NiII complexes: o- and p-dihydroxybenzene as hydrogen donors. Chem Commun 1971(3):139–140. 40. Ohkubo K, Aoji T, Hirata K, Yoshinaga K (1976) Enantioselective dehydrogenation of secondary carbinols by ruthenium(II) chiral phosphine complexes in the transfer hydrogenation of olefins. Inorg Nucl Chem Lett 12(11):837–842. 41. Jiang Y, Hess J, Fox T, Berke H (2010) Rhenium hydride/boron Lewis acid cocatalysis of alkene hydrogenations: Activities comparable to those of precious metal systems. J Am Chem Soc 132(51):18233–18247. 42. Lin TP, Peters JC (2013) Boryl-mediated reversible H2 activation at cobalt: Catalytic hydrogenation, dehydrogenation, and transfer hydrogenation. J Am Chem Soc 135(41):15310–15313. 43. Leitao EM, Jurca T, Manners I (2013) Catalysis in service of main group chemistry offers a versatile approach to p-block molecules and materials. Nat Chem 5(10): 817–829. 44. Johnstone RAW, Wilby AH, Entwistle ID (1985) Heterogeneous catalytic transfer hydrogenation and its relation to other methods for reduction of organic-compounds. Chem Rev 85(2):129–170. 45. Blaser HU, et al. (2003) Selective hydrogenation for fine chemicals: Recent trends and new developments. Adv Synth Catal 345(1-2):103–151. 46. Pieber B, Martinez ST, Cantillo D, Kappe CO (2013) In situ generation of diimide from hydrazine and oxygen: Continuous-flow transfer hydrogenation of olefins. Angew Chem Int Ed Engl 52(39):10241–10244.
Pérez et al.
PNAS Early Edition | 5 of 5
CHEMISTRY
demonstrates the high versatility of fluorophosphonium cations in catalysis providing a main-group catalyst for the transfer hydrogenation of olefins.
fluorophosphonium
A
wide range of commodity chemicals, petrochemicals, pharmaceuticals, materials, and foods depend on industrial-scale hydrogenation reactions. Catalysts for this process are based on both heterogeneous transition metals materials and homogeneous transition metal complexes. Indeed, it was the discovery of Sabatier (1) in the early 20th century that revealed the utility of amorphous Ni and other metals in hydrogenation processes. Following the dawn of organometallic chemistry in the 1960s, homogeneous catalysts were developed principally based on precious metals (2–4). Although subsequent modifications and optimizations have led to numerous highly selective and efficient catalyst technologies, concerns over cost, natural abundance, and toxicity have prompted efforts to uncover catalysts based on “earth-abundant” elements. The principal targets in these efforts have been the first-row transition metals. Indeed, the groups of Chirik (5, 6), Morris (7), Beller (8), and others (9) have developed remarkably active and selective catalysts based on Fe. Strategies to develop reduction methods using organic or main-group–based reagents have also been explored. For example, use of Hantzsch esters (10), Birch reduction of arenes (11), or the use of B or Al hydrides are effective, albeit stoichiometric reductions. More recently, the advent of “frustrated Lewis pairs” (FLPs) have led to the discovery that combinations of sterically encumbered donors and acceptors act as catalysts for the reductions of imines, enamines, silylenol ethers, olefins, and alkynes (12). To uncover hydrogenation strategies using earth-abundant elements, we have focused our efforts on Lewis acidic phosphorus species. Although Gudat (13), Burford and coworker (14), Yoshifuji (15), and Bertrand and coworker (16), among others, have reported phosphenium cations that demonstrate Lewis acidity, Radosevich and coworkers (17) have exploited the reaction of the unique planar P(III) species with ammoniaborane to give a P(V)H2 derivative that effects the subsequent reduction of diazobenzene. Although the acidity of P(V) has been exploited previously in ylide reagents (18), Diels–Alder reactions catalysis (19), and addition reactions to polar unsaturates (20), Gabbaï and coworkers (21) have more recently exploited the acceptor capabilities of phosphonium cations, in fluoride sensor strategies. In our own efforts, we have recently reported the synthesis of highly electrophilic phosphonium www.pnas.org/cgi/doi/10.1073/pnas.1407484111
cations (EPCs) (22). The species [(C6F5)3PF][B(C6F5)4] 1 proved to be an effective catalyst for the hydrodefluorination of fluoroalkanes in the presence of silane (22). We subsequently showed that 1 could effectively catalyze the hydrosilylation of olefins and alkynes (23). In this report, we first exploit the Lewis acidity of 1 to catalyze the dehydrocoupling of silane with a variety of species including amines, thiols, phenols, and carboxylic acids, liberating H2. We then show that these intermolecular and intramolecular dehydrocoupling reactions can be used for transfer hydrogenation of olefins, thus providing an unprecedented route to metal-free hydrogenation of olefins. Results and Discussion Initially, 1.5 mol% of 1, Ph2NH, and Et3SiH were allowed to react for 10 h. This afforded near complete conversion to Ph2NSiEt3 (Table 1, entry 1). Interestingly, when Et3SiH was replaced by ClMe2SiH, the reaction required only 1 h for complete conversion (Table 1, entry 2). In contrast, reactions with more sterically encumbered silanes such as Ph3SiH or PhMe2SiH resulted in an increased reaction time (Table 1, entries 3, 4), whereas the bulkier silane, iPr3SiH (Table 1, entry 5) precluded the reaction. para-Methyl substituted anilines showed a slower reaction time (Table 1, entries 6–8). iPr2NH and aniline (Table 1, entries 9, 10) led to catalyst degradation and thus no catalytic activity. Compound 1 was also very effective in the catalytic dehydrocoupling of thiophenols (Table 1, entries 11–14) and phenols (Table 1, entries 16–21) with Et3SiH. In general, these reactions were complete in 1 h with the exception of C6F5SH and C6F5OH, which required a longer time for completion (Table 1, entries 15 and 20, respectively). In this prior case (C6F5SH), heating at 100 °C accelerated the reaction, resulting in complete conversion in 3 h. In the case when para-methoxy phenol was reacted with the excess of Et3SiH, both dehydrocoupling of Significance For more than a century, hydrogenation has been limited to the use of transition metal-based catalysts. With the emerging focus on the chemistry of earth-abundant elements, the 21st century has seen a renaissance in main-group chemistry. In this work, an electrophilic phosphonium cation is shown to act as main-group catalyst effecting the dehydrocoupling of silane and amines, phenols, thiols, and carboxylic acids with the concurrent release of H2. In addition, performing the reactions in the presence of olefins, dehydrocoupling occurs with simultaneous hydrogenation of the olefin. This chemistry provides an unprecedented avenue to metal-free transfer hydrogenation catalysis of olefins. Author contributions: M.P., C.B.C., R.D., and D.W.S. designed research; M.P., C.B.C., and R.D. performed research; M.P., C.B.C., R.D., and D.W.S. analyzed data; and M.P., C.B.C., R.D., and D.W.S. wrote the paper. The authors declare no conflict of interest. This article is a PNAS Direct Submission. 1
To whom correspondence should be addressed. E-mail: [email protected].
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. 1073/pnas.1407484111/-/DCSupplemental.
PNAS Early Edition | 1 of 5
CHEMISTRY
A major advance in main-group chemistry in recent years has been the emergence of the reactivity of main-group species that mimics that of transition metal complexes. In this report, the Lewis acidic phosphonium salt [(C6F5)3PF][B(C6F5)4] 1 is shown to catalyze the dehydrocoupling of silanes with amines, thiols, phenols, and carboxylic acids to form the Si-E bond (E = N, S, O) with the liberation of H2 (21 examples). This catalysis, when performed in the presence of a series of olefins, yields the concurrent formation of the products of dehydrocoupling and transfer hydrogenation of the olefin (30 examples). This reactivity provides a strategy for metalfree catalysis of olefin hydrogenations. The mechanisms for both catalytic reactions are proposed and supported by experiment and density functional theory calculations.
Table 1. Catalytic dehydrocoupling reactions
Entry
E-H
Silane
T, h
Yield, %*
1
Ph2NH
Et3SiH
2
Ph2NH
ClMe2SiH
3
Ph2NH
Ph3SiH
20
>99
4
Ph2NH
PhMe2SiH
48
>99
5
Ph2NH
iPr3SiH
96
0
6
p-Me(C6H4)2NH
Et3SiH
30
>99
7
p-Me(C6H4)2NH
ClMe2SiH
16
>99
8
p-Me(C6H4)2NH
Ph3SiH
36
40
9
iPr2NH
Et3SiH
48
0
10
PhNH2
Et3SiH
48
0
11
PhSH
Et3SiH
99
12
p-Me(C6H4)SH
Et3SiH
99
13
p-Cl(C6H4)SH
Et3SiH
99
14
p-F(C6H4)SH
Et3SiH
99
15
C6F5SH
Et3SiH
168
>99
PhOH
Et3SiH
2
>99
o-(Me)2(C6H3)OH
Et3SiH
2
>99
p-OMe(C6H4)OH
Et3SiH
18
>99
p-Me(C6H4)OH
Et3SiH
3
>99
C6F5OH
Et3SiH
24
>99
p-C8H17(C6H4)CO2H
Et3SiH
1
>99
16
†
17† 18
†
19† 20 21
†
10
>99
1
>99
Conditions: To a solution of the catalyst (1.5 mol%) in C6D5Br or CD2Cl2 (1.0 mL) were added silane (1.1 Eq) and E-H (1.0 Eq) at 25 °C. *Yields measured by 1H-NMR. † 1.0 mol% of catalyst was used.
the OH group and the methoxy moiety was replaced by silyl fragments; however, one equivalent of silane leads selectively to the product of dehydrocoupling of the phenol (entry 18). This observation is similar to that reported for the hydrosilylation of methoxy-α-methylstyrenes (23). The carboxylic acid p-C8H17(C6H4)CO2H and Et3SiH gave the corresponding ester in less than 1 h (entry 21). The mechanism for these dehydrocoupling reactions is thought to be analogous to that previously proposed for 1 catalyzed olefin hydrosilylation (22). Thus, the initial step involves activation of the Si-H bond by the fluorophosphonium Lewis acid 1. Backside attack of the LUMO at Si center by the Lewis base (N, O, S) generates transient hypervalent silicon species that has hydridic (Si-H) and protic (E-H, E = N, O, S) hydrogens prompting loss of dihydrogen (Fig. 1). This mechanism is conceptually similar to related Lewis acid mediated hydrosilylations described by Piers and coworkers (24), Oestreich and coworker (25), and Gevorgyan, Yamamoto, and coworkers (26). Furthermore, the proposed mechanism is consistent with the experimental observation that basic amines and sterically less encumbered anilines do not dehydrocouple with Et 3 SiH, suggesting that 2 of 5 | www.pnas.org/cgi/doi/10.1073/pnas.1407484111
these bases interact stronger with the Lewis acid 1 than Et3SiH. At the same time, sterically demanding silanes slow or even stop the dehydrocoupling reaction (Table 1, entry 5), which is consistent with the proposed role of hypervalent silicon species. Furthermore, the use of silane in excess (10.0 Eq) with respect to the pTol2NH, accelerates the reaction, whereas using a large excess of amine dramatically increases the reaction time. This supports the view that equilibrium involving binding of silane or aniline to P(V) is competitive, and the dehydrocoupling reaction proceeds via silane activation. The quest for theoretical insight regarding the dehydrocoupling reaction of silane and amine with the Lewis acid 1 prompted preliminary gas-phase density functional theory (DFT) calculations at the WB97XD/def2TZV (27, 28) level of theory using the cation of 1, Me3SiH, and Ph2NH. Phosphonium cation interacts with both Me3SiH and Ph2NH resulting in the Me3SiH∙∙∙P and Ph2(H)N∙∙∙P distances of 2.3 and 3.5 Å, respectively. Both interactions are exothermic by 15.19 (ΔG = −1.8) and 23.35 (ΔG = −6.3) kcal·mol−1, respectively. This suggests that these species exist in an equilibrium (Keq = 5 × 10−4), slightly favoring the amine–phosphonium interaction, which was Pérez et al.
also observed experimentally. Coordination of amine to the phosphonium bound silane generating a transient five-coordinate silicon intermediate is also exothermic with ΔH = −37.1 (ΔG = −23.7) kcal·mol−1, which yields (C6F5)3P(F)H and [Ph2N(H)SiMe3]+ in a slightly endothermic (ΔH = 2.8 kcal·mol−1), but exergonic (ΔG = −9.3 kcal·mol−1) step. The subsequent reaction of these two species to yield the dehydrocoupling product is driven by the liberation of H2 even though the reaction is somewhat endothermic and endergonic (ΔH = 43.5 and ΔG = 34.1 kcal·mol−1). Overall, this catalytic cycle is computed to be slightly exothermic (ΔH = −6.0 kcal·mol−1) with only a small overall ΔG of −0.6 kcal·mol−1. This dehydrocoupling mechanism is reminiscent of that described for analogous B(C6F5)3 catalyzed reaction in the classic work from Piers and coworkers (29), which subsequently further illuminated and expanded by the groups of Oestreich (30), Brook (31), Rubinsztajn (32), and Paradies (33). In the context of dehydrocoupling, it is also important to note the pioneering work from the Tilley and Manners research groups, who exploited transition metal catalysts for this purpose (34, 35), whereas Hill et al. (36) described Zn, Mg, Ca, and Sr catalyzed preparations of amino silanes and Baba and coworker (37) have used InBr3 to dehydrocouple silanes and carboxylic acids (Table 1 and Fig. 1). The above dehydrocoupling reactions proceed with the liberation of H2. We queried the possibility of capturing this H2 by the addition of an olefin. Thus, 1,1-diphenylethylene was added to a mixture of p-Me(C6H4)2NH, Et3SiH, or Ph3SiH and 1.5 mol% of 1; heating this mixture to 100 °C gave after 5 h the dehydrocoupled product and 1,1-diphenylethane with 60% and >99% yield, respectively (Table 2, entries 1 and 2). Reactions occurred at room temperature; however, the reaction times and conversions improved at elevated temperatures (SI Appendix). Transfer hydrogenation is not limited to anilines, as substituted thiophenols, phenols, and carboxylic acids are competent proton sources as well, resulting in complete reduction of the olefin in less than 1 h (Table 2, entries 3–8). In a similar fashion, the olefin, 1-methyl-4-(prop1-en-2-yl)benzene was catalytically reduced using either ditolylamine or tolylthiophenol in combination with Et3SiH, affording yields up to 75%. The aniline reactions were done at 100 °C, whereas the reductions using thiol were performed at 25 °C (Table 2, entries 9–12). 2-Methylstyrene was reduced in up to 67% yields under similar conditions using thiophenols as the proton source at 25 °C (Table 2, entries 13 and 14), whereas (E)-α-methylstilbene was hydrogenated in high yields using thiophenol or carboxylic acids (Table 2, entries 15–17). Interestingly, the carboxylic acid C6F5CO2H with Et3SiH resulted in complete reduction of internal Pérez et al.
PNAS Early Edition | 3 of 5
CHEMISTRY
Fig. 1. Proposed mechanistic pathways for dehydrocoupling of silane and amine and transfer hydrogenation of olefins (experiment: R3Si = Et3Si, ClMe2Si, Ph3Si, PhMe2Si; Ar2NH = Ph2NH, p-Me(C6H4)2NH; calculations: R3Si = Me3Si; Ar = Ph). Gibbs free energies and enthalpies in parentheses for every step are provided in kilocalories per mole.
olefins (E)-α-methylstilbene (Table 2, entry 18) and 1-phenyl-2,2diphenylethylene in less than 2 h (Table 2, entry 19). The reduction of 3,3-dimethylbut-1-ene using p-Cl(C6H4)SH and p-Cl(C6H4)OH as the proton sources yielded 76% and 69% of 2,3-dimethylbutane, respectively (Table 2, entries 20 and 21 results from methyl migration). Methylenecyclohexane and 1-methylcyclohex-1-ene were also successfully reduced to methylcyclohexane in good yields (Table 2, entries 22–26). The olefinic unit of dibutyl 2-methylenesuccinate was also hydrogenated using arylthiols and silane at 25 °C, displaying compatibility with unsaturated esters (Table 2, entries 27–29). Finally, using 2-methylenesuccinic acid with Et3SiH in catalytic presence of 1, resulting in reduction of the internal alkene. Subsequent hydrolysis of the silyl-ester affords the corresponding acid in 89% isolate yield (Table 2, entry 30). It is noteworthy that, in all cases, neither hydrogenation nor dehydrocoupling is observed in the absence of the fluorophosphonium catalyst 1. These reactions provide access to both dehydrocoupling products as well as the alkanes derived from olefin hydrogenation. It is also important to note that these reactions are highly chemoselective, providing exclusively the products of dehydrocoupling and transfer hydrogenation with no hydrosilylation of olefin byproducts. In addition, these reactions exhibit impressive functional group tolerance, proceeding in the presence of aryl-halides, amines, thiols, ethers, and ester fragments. The mechanism for these transfer hydrogenations is thought to proceed by the intervention of the olefin in dehydrocoupling reaction pathway. Thus, rather than the elimination of H2, the transient [Ar2N (H)SiR3]+ protonates the olefin, generating carbocation that subsequently abstracts the hydride from (C6F5)3P(F) H (Fig. 1), yielding the alkane and dehydrocoupling product. This view of the mechanism is supported by the observation that these transfer hydrogenations are favored by the 1,1-disubstituted olefins. In the case of tBuCH=CH2 (Table 2, entries 19 and 20), protonation prompts methyl migration, also generating tertiary carbocation. To support the proposed mechanism, the catalysis using Et3SiD in combination with a proton source was performed, leading exclusively to deuteration of the more electrophilic position of the olefin (SI Appendix). This is consistent with a Markovnikov delivery of proton to a less substituted carbon, forming tertiary carbocation followed by the reaction of the latter with deuteride. DFT calculations at the WB97XD/def2TZV (27, 28) level of theory for hydrogenation of 1,1-diphenylethylene show that the protonation of the 1,1-diphenylethylene by [Ph2N(H)SiMe3]+ is endothermic [ΔH = 14.1 (ΔG = 13.7) kcal·mol−1], whereas the subsequent hydride abstraction from (C6F5)3P(F)H affording 1,1-diphenylethane and regenerating catalyst is almost thermal neutral (ΔH = 1.1 and ΔG = 0.0 kcal·mol−1). Overall, concurrent transfer hydrogenation with dehydrocoupling is energetically favorable to loss of H2 as it requires 28.3 kcal·mol−1 less energy, and lower by 20.5 kcal·mol−1 Gibbs free energy. This was also supported by the competition experiment in which the dehydrocoupling reaction in the presence of 1,1-diphenylethylene was observed to be significantly faster than in its absence (SI Appendix). Transfer hydrogenations of ketones, aldehydes, and imines can be achieved by the classic Meerwein–Ponndorf–Verley protocol, catalyzed by a variety of transition metal complexes and more recently by FLPs (38). In contrast, transfer hydrogenation of olefins has received much less attention. Homogeneous catalysts based on Fe (39) and Ru (40) were demonstrated in the 1970s, whereas the groups of Berke (41), Peters (42), and Manners (43) have more recently reported the transfer hydrogenation of olefins using Re-, Co-, and Rh-based catalysts with ammonia-boranes as the H2 source. Heterogeneous catalysts for transfer hydrogenation of olefins include supported Ni, Pd, or Ir and have exploited isopropanol, formic acid, ammonium formate, and glycerol as sources of H2 (44, 45). Catalyst-free reductions of olefins have also been reported, although this
Table 2. Transfer hydrogenation of olefins with concurrent dehydrocoupling catalysis
Entry
E-H
R3SiH
T, °C
T, h
Yield, %*
Ph2C = CH2 1
p-Me(C6H4)2NH
Et3SiH
100
5
60
2
p-Me(C6H4)2NH
Ph3SiH
100
6
>99
3
p-Me(C6H4)SH
Et3SiH
25
99
4
p-Cl(C6H4)SH
Et3SiH
25
99 (94)
5
p-F(C6H4)SH
Et3SiH
25
99 (96)
6
p-MeO(C6H4)OH
Et3SiH
25
99
8
C6F5CO2H
Et3SiH
25
1
>99 (94)
>99 (93)
p-MeC6H4(Me)C = CH2 9
p-Me(C6H4)2NH
Et3SiH
100
3
10†
p-Me(C6H4)2NH
Et3SiH
100
3
56 74
11
p-Me(C6H4)SH
Et3SiH
25
99
>99 (95)
BuO2CCH2(BuO2C)C = CH2 27
p-Me(C6H4)SH
Et3SiH
25
1
28
p-Cl(C6H4)SH
Et3SiH
25
1
89
29
p-F(C6H4)SH
Et3SiH
25
1
98 (96)
Et3SiH
25
20
96 (89)¶
HO2CCH2(HO2C)C = CH2 30
—
Conditions: To a solution of silane (1.0 Eq), R-H (1.0 Eq), and olefin (1.0 Eq) in C6D5Br or CD2Cl2 (1.0 mL) was added the catalyst (1.5 mol%). *Yields determined by 1H-NMR spectroscopy (isolated yield). † 2.0 Eq of olefin were used. ‡ To a solution of R-H (1.0 Eq) and olefin (1.0 Eq) in C6D5Br (1.0 mL) were added a mixture of silane (1.0 Eq) and the catalyst (1.5 mol%). § For complete conversion 1.5 Eq of reagents were used with respect to the olefin. { Isolated yield of the corresponding acid.
required the generation of the energetic diimide from hydrazine monohydrate and O2 (46). Although FLP catalysts have been shown to effect olefin hydrogenation, the present system, which exploits concurrent dehydrocoupling, is (to our knowledge) the first main-group system to mediate transfer hydrogenations of olefins. 4 of 5 | www.pnas.org/cgi/doi/10.1073/pnas.1407484111
Summary In conclusion, we have described a general and efficient procedure for dehydrocoupling of silanes with anilines, phenols, thiophenols, and benzoic acids catalyzed by EPCs. Moreover, we showed that these reactions can be used for in situ transfer hydrogenation of olefins. Thus, the present work further Pérez et al.
Methods General Procedure for Dehydrocoupling Reactions. To a solution of the catalyst [(C6F5)3PF][B(C6F5)4] (1.0–1.5 mol%) in C6D5Br or CD2Cl2 (0.34 M) was added the corresponding silane (1.0–1.2 Eq) and RH (R = Ar2N, ArS, ArO, ArCO2) (1.0 Eq) at 25 °C. The reaction was monitored by NMR analysis until the reaction was complete. Yield was determined by 1H-NMR spectroscopy. Freshly prepared catalyst was used and resulted in the optimal yields.
25 °C. The reaction was monitored by NMR or TLC until completion. Yield was determined by 1H-NMR spectroscopy. For isolated yields, the reaction was quenched with a diluted solution of NaHCO3 and the mixture was extracted with CH2Cl2. The organic solution was dried over MgSO4, filtered, and evaporated. The crude was diluted with hexane and filtered over silica gel; products were eluted with hexane and Et2O for dibutyl 2-methylenesuccinate. The quality of the catalyst is again essential for the successful completion of the reaction. Supplementary Data. Details of the syntheses, spectroscopic data are included in the SI Appendix.
General Procedure for Olefin Transfer-Hydrogenation. To a solution of silane (1.0 Eq), RH (R = Ar2N, ArS, ArO, ArCO2) (1.0 Eq) and olefin (1.0–1.2 Eq) was added the [(C 6F5) 3PF][B(C 6F5) 4] (1.5 mol%) in C6D5 Br or CD2Cl 2 (0.5 M) at
ACKNOWLEDGMENTS. We thank the Natural Sciences and Engineering Research Council (NSERC) of Canada for financial support. D.W.S. is grateful for the award of a Canada Research Chair. C.B.C. is grateful for the award of an NSERC Postgraduate Scholarship and a Walter C. Sumner Fellowship.
1. Sabatier P (1926) How I have been led to the direct hydrogenation method by metallic catalysts. Ind Eng Chem 18(10):1005–1008. 2. Hallman PS, Evans D, Osborn JA, Wilkinson G (1967) Selective catalytic homogeneous hydrogenation of terminal olefins using tris(triphenylphosphine)hydridochlororuthenium(II); hydrogen transfer in exchange and isomerisation reactions of olefins. Chem Commun 1967(7):305–306. 3. Boerner A, Holz J (2004) Homogeneous hydrogenations: Olefin hydrogenations. Transition Metals for Organic Synthesis, eds Beller M, Bolm C (Wiley, New York), 2nd Ed, Vol 2, pp 3–13. 4. Pettinari C, Marchetti F, Martini D (2004) Metal complexes as hydrogenation catalysts. Comprehensive Coordination Chemistry II, eds McCleverty JA, Meyer TJ (Elsevier, Oxford), Vol 9, pp 75–139. 5. Friedfeld MR, et al. (2013) Cobalt precursors for high-throughput discovery of base metal asymmetric alkene hydrogenation catalysts. Science 342(6162):1076–1080. 6. Trovitch RJ, Lobkovsky E, Bill E, Chirik PJ (2008) Functional group tolerance and substrate scope in bis(imino)pyridine iron catalyzed alkene hydrogenation. Organometallics 27(7):1470–1478. 7. Zuo W, Lough AJ, Li YF, Morris RH (2013) Amine(imine)diphosphine iron catalysts for asymmetric transfer hydrogenation of ketones and imines. Science 342(6162): 1080–1083. 8. Jagadeesh RV, et al. (2013) Nanoscale Fe2O3-based catalysts for selective hydrogenation of nitroarenes to anilines. Science 342(6162):1073–1076. 9. Bullock RM (2013) Chemistry. Abundant metals give precious hydrogenation performance. Science 342(6162):1054–1055. 10. Hantzsch A (1881) Condensationprodukte aus Aldehydammoniak und Ketoniartigen Verbindungen. Chem Ber 14:1637–1638. 11. Birch AJ (1944) A new reagent for primary and secondary aliphatic amines. J Chem Soc 1944:314–315. 12. Stephan DW, Erker G (2013) Frustrated lewis pair mediated hydrogenations. Top Curr Chem 332:85–110. 13. Gudat D (2010) Diazaphospholenes: N-Heterocyclic phosphines between molecules and Lewis pairs. Acc Chem Res 43(10):1307–1316. 14. Burford N, Ragogna PJ (2002) New synthetic opportunities using Lewis acidic phosphines. Dalton Trans (23):4307–4315. 15. Yoshifuji M (2009) Product class 3: Phosphenium salts. Sci Synth 42:63–69. 16. Guerret O, Bertrand G (1997) Trigonal planar phosphorus cations. Acc Chem Res 30(12):486–493. 17. Dunn NL, Ha M, Radosevich AT (2012) Main group redox catalysis: Reversible P(III)/P(V) redox cycling at a phosphorus platform. J Am Chem Soc 134(28):11330–11333. 18. Wittig G, Schöllkopf U (1954) Über Triphenyl-phosphin-methylene als Olefinbildende Reagenzien I. Chem Ber 87:1318–1330. 19. Terada M, Kouchi M (2006) Novel metal-free Lewis acid catalysis by phosphonium salts through hypervalent interaction. Tetrahedron 62(2-3):401–409. 20. Sereda O, Tabassum S, Wilhelm R (2010) Lewis acid organocatalysts. Top Curr Chem 291:349–393. 21. Hudnall TW, Kim YM, Bebbington MWP, Bourissou D, Gabbaï FP (2008) Fluoride ion chelation by a bidentate phosphonium/borane Lewis acid. J Am Chem Soc 130(33): 10890–10891. 22. Caputo CB, Hounjet LJ, Dobrovetsky R, Stephan DW (2013) Lewis acidity of organofluorophosphonium salts: Hydrodefluorination by a saturated acceptor. Science 341(6152):1374–1377. 23. Pérez M, Hounjet LJ, Caputo CB, Dobrovetsky R, Stephan DW (2013) Olefin isomerization and hydrosilylation catalysis by Lewis acidic organofluorophosphonium salts. J Am Chem Soc 135(49):18308–18310. 24. Parks DJ, Blackwell JM, Piers WE (2000) Studies on the mechanism of B(C6F5)3-catalyzed hydrosilation of carbonyl functions. J Org Chem 65(10):3090–3098. 25. Rendler S, Oestreich M (2008) Conclusive evidence for an SN2-Si mechanism in the B(C6F5)3-catalyzed hydrosilylation of carbonyl compounds: Implications for the related hydrogenation. Angew Chem Int Ed Engl 47(32):5997–6000.
26. Gevorgyan V, Rubin M, Benson S, Liu JX, Yamamoto Y (2000) A novel B(C6F5)3-catalyzed reduction of alcohols and cleavage of aryl and alkyl ethers with hydrosilanes. J Org Chem 65(19):6179–6186. 27. Grimme S (2006) Semiempirical GGA-type density functional constructed with a longrange dispersion correction. J Comput Chem 27(15):1787–1799. 28. Chai JD, Head-Gordon M (2008) Long-range corrected hybrid density functionals with damped atom-atom dispersion corrections. Phys Chem Chem Phys 10(44):6615–6620. 29. Blackwell JM, Foster KL, Beck VH, Piers WE (1999) B(C6F5)3-catalyzed silation of alcohols: A mild, general method for synthesis of silyl ethers. J Org Chem 64(13): 4887–4892. 30. Hermeke J, Mewald M, Oestreich M (2013) Experimental analysis of the catalytic cycle of the borane-promoted imine reduction with hydrosilanes: Spectroscopic detection of unexpected intermediates and a refined mechanism. J Am Chem Soc 135(46): 17537–17546. 31. Brook MA, Grande JB, Ganachaud F (2011) New synthetic strategies for structured silicones using B(C6F5)3. Adv Polym Sci 235:161–183. 32. Cella J, Rubinsztajn S (2008) Preparation of polyaryloxysilanes and polyaryloxysiloxanes by B(C6F5)3 catalyzed polyetherification of dihydrosilanes and bis-phenols. Macromolecules 41(19):6965–6971. 33. Greb L, Tamke S, Paradies J (2014) Catalytic metal-free Si-N cross-dehydrocoupling. Chem Commun (Camb) 50(18):2318–2320. 34. Sadow AD, Tilley TD (2005) Synthesis and characterization of scandium silyl complexes of the type Cp*2ScSiHRR’. sigma-Bond metathesis reactions and catalytic dehydrogenative silation of hydrocarbons. J Am Chem Soc 127(2):643–656. 35. Waterman R (2013) Mechanisms of metal-catalyzed dehydrocoupling reactions. Chem Soc Rev 42(13):5629–5641. 36. Hill MS, Liptrot DJ, MacDougall DJ, Mahon MF, Robinson TP (2013) Hetero-dehydrocoupling of silanes and amines by heavier alkaline earth catalysis. Chem Sci 4(11): 4212–4222. 37. Motokura K, Baba T (2012) An atom-efficient synthetic method: Carbosilylations of alkenes, alkynes, and cyclic acetals using Lewis and Bronsted acid catalysts. Green Chem 14(3):565–579. 38. Farrell JM, Heiden ZM, Stephan DW (2011) Metal-free transfer hydrogenation catalysis by B(C6F5)3. Organometallics 30(17):4497–4500. 39. Nishiguchi T, Fukuzumi K (1971) Homogeneous transfer-hydrogenation of olefins catalyzed by FeII, CoII, and NiII complexes: o- and p-dihydroxybenzene as hydrogen donors. Chem Commun 1971(3):139–140. 40. Ohkubo K, Aoji T, Hirata K, Yoshinaga K (1976) Enantioselective dehydrogenation of secondary carbinols by ruthenium(II) chiral phosphine complexes in the transfer hydrogenation of olefins. Inorg Nucl Chem Lett 12(11):837–842. 41. Jiang Y, Hess J, Fox T, Berke H (2010) Rhenium hydride/boron Lewis acid cocatalysis of alkene hydrogenations: Activities comparable to those of precious metal systems. J Am Chem Soc 132(51):18233–18247. 42. Lin TP, Peters JC (2013) Boryl-mediated reversible H2 activation at cobalt: Catalytic hydrogenation, dehydrogenation, and transfer hydrogenation. J Am Chem Soc 135(41):15310–15313. 43. Leitao EM, Jurca T, Manners I (2013) Catalysis in service of main group chemistry offers a versatile approach to p-block molecules and materials. Nat Chem 5(10): 817–829. 44. Johnstone RAW, Wilby AH, Entwistle ID (1985) Heterogeneous catalytic transfer hydrogenation and its relation to other methods for reduction of organic-compounds. Chem Rev 85(2):129–170. 45. Blaser HU, et al. (2003) Selective hydrogenation for fine chemicals: Recent trends and new developments. Adv Synth Catal 345(1-2):103–151. 46. Pieber B, Martinez ST, Cantillo D, Kappe CO (2013) In situ generation of diimide from hydrazine and oxygen: Continuous-flow transfer hydrogenation of olefins. Angew Chem Int Ed Engl 52(39):10241–10244.
Pérez et al.
PNAS Early Edition | 5 of 5
CHEMISTRY
demonstrates the high versatility of fluorophosphonium cations in catalysis providing a main-group catalyst for the transfer hydrogenation of olefins.
