Hydrocyanation of alkenes and alkynes
Hydrocyanation of alkenes and alkynes refers to the transition-metal-mediated or -catalyzed addition of hydrogen cyanide across a carbon-carbon π bond. This reaction may be used to synthesize nitriles from alkenes or alkynes in a Markovnikov or anti-Markovnikov fashion.
- 1 Introduction
- 2 Mechanism and Stereochemistry
- 3 Scope and Limitations
- 4 Synthetic Applications
- 5 Comparison to Other Methods
- 6 Experimental Conditions and Procedure
- 7 References
The addition of hydrogen cyanide across activated carbon-carbon π bonds, such as the C=C bond of α,β-unsaturated carbonyl compounds, is a well-known, synthetically useful transformation. Because this process requires a sufficiently electrophilic substrate, unactivated alkenes will not undergo addition under conditions useful for activated substrates. However, the addition of hydrogen cyanide across a π bond is a thermodynamically favorable process, and the high activation barriers associated with addition to unactivated alkenes and alkynes may be surmounted using transition-metal catalysis. Transition-metal catalyzed addition of cyanide across π bonds may occur in a Markovnikov or anti-Markovnikov fashion to provide fully saturated nitriles or vinyl nitriles (Eq. 1).(1)
The most common catalysts used to effect hydrocyanation are nickel(0) and palladium(0) complexes. The industrial development of nickel-catalyzed hydrocyanation, which produced several concepts useful to the field of organometallic chemistry, was motivated by the need to mass-produce adiponitrile (1,4-dicyanobutane) for nylon synthesis. Although many site-selective variants of hydrocyanation have been developed, fewer efficient asymmetric variants exist.
Mechanism and Stereochemistry
Nickel-catalyzed Hydrocyanation of Alkenes
Because the mechanism of hydrocyanation involves a number of ligand substitution processes, appropriately tuned ligand association and dissociation kinetics are essential for catalyst turnover. The most common ligands used for nickel-catalyzed hydrocyanation are triaryl phosphites, for which the kinetics of ligand association and dissociation on nickel are optimal.(2)
The mechanism of nickel-catalyzed hydrocyanation begins with ligand dissociation (Eq. 2). The tetracoordinate complex NiL4 may lose two ligand molecules and coordinate alkene to directly form catalytic intermediate I, which oxidatively adds hydrogen cyanide to give intermediate II. Alternatively, oxidative addition of HCN may occur before alkene coordination, if dissociation of the ligand is slow. After oxidative addition to form compound VII, ligand dissociation and alkene coordination yield intermediate II via the alternative pathway. A second ligand dissociation produces 16-electron complex III, which may coordinate another molecule of alkene upon migratory insertion into the nickel-hydrogen bond. Association of a ligand back on to the metal center encourages reductive elimination of the nitrile product, regenerating intermediate II. Reductive elimination is the turnover-limiting step of this reaction, and is too slow to result in turnover when electron-deficient alkenes (e.g., acrylonitrile or tetrafluoroethylene) are used.
When excess hydrogen cyanide is present, intermediate IV is subject to decomposition into catalytically inactive Ni(CN)2, ethane, ethylene, and ligand. Because of this side reaction, hydrogen cyanide is often added slowly to a solution of the pre-catalyst and substrate.
When a Lewis acid is present, it may coordinate to cyanide and produce the cationic nickel hydride complex IX from intermediate VIII. This cationic complex coordinates an alkene at a much faster rate than VIII and leads to greater selectivity for the anti-Markovnikov addition product in many cases, owing to a steric bias introduced by the Lewis acid. For terminal alkenes, a Lewis acid is necessary to promote association of the π bond to nickel, and anti-Markovnikov addition products typically predominate when Lewis acids are used.
Nickel-catalyzed Hydrocyanation of Butadiene
The hydrocyanation of butadiene is more rapid than hydrocyanation reactions of unconjugated alkenes. In addition, a significant preference for the branched isomer X over the linear 1,4-adduct XI (2:1) is observed and no 4-pentenenitrile forms (Eq. 3). The formation of a relatively stable π-allylnickel complex upon migratory insertion of butadiene into the nickel-hydrogen bond accounts for all of these observations. Related η3-benzyl complexes form in hydrocyanation reactions of vinylarenes.(3)
Addition of a Lewis acid to the reaction mixture after the first hydrocyanation results in isomerization of XI to 4-pentenenitrile. Under conditions of kinetic conrol, rapid anti-Markovnikov hydrocyanation of 4-pentenenitrile takes place to yield adiponitrile.
Hydrocyanation of Alkynes
The nickel-catalyzed hydrocyanation of alkynes is mechanistically similar to the alkene hydrocyanation mechanism described above. Hydrogen and cyanide add across the triple bond in a syn fashion. When terminal alkynes are employed as substrates, internal nitriles are the major products in the absence of (often important) steric factors. This preference is explained by the greater stability of bonds between nickel and more highly substituted carbons.
Copper(I) salts may be used in conjunction with ammonium chloride and HCN to hydrocyanate acetylene in a rather specialized application. A plausible mechanism of this reaction is shown below (Eq. 4).(4)
The hydrocyanation of alkenes by nickel(0) complexes proceeds stereospecifically syn (Eq. 5). Migratory insertion occurs in a syn fashion and reductive elimination occurs with retention of configuration. Evidence for syn addition in hydrocyanations of dienes and vinylarenes has also been observed.(5)
Studies of asymmetric hydrocyanation reactions have revealed an intriguing electronic effect on the enantioselectivity of the reaction. Rendering the nickel center electron deficient through the use of electron-poor ligands increases enantioselectivity. For ligand L1 (Eq. 6), the enantioselectivity is highest when the aryl group is 3,5-di(trifluoromethyl)phenyl. However for ligand L2, the enantioselectivity is highest when the Ar and Ar' groups are electronically dissimilar.(6)
Evidence accumulated to date suggests that removing electron density from the nickel center increases the rate of reductive elimination relative to β-hydride elimination, which re-forms the starting alkene. Furthermore, when a chiral ligand is used, the effect is greater for one diastereomeric pathway than the other, resulting in increased enantioselectivity. This electronic effect supplements the steric constraints imposed by the shape of the ligand.
Scope and Limitations
At elevated temperatures, unactivated terminal alkenes may be hydrocyanated by nickel(0)-bisphosphine complexes in the presence of a Lewis acid. Aluminum trichloride and a number of other Lewis acids provide the anti-Markovnikov addition selectively (Eq. 7). Rigid, chelating phosphine ligands with bite angles in excess of 100 degrees (such as Xantphos) are needed for reactions with isolated alkenes.(7)
The DuPont adiponitrile process nicely highlights the scope of the hydrocyanation of butadienes (Eq. 8). In the presence of a Lewis acid, isomerization to 4-petenenitrile may occur, and a kinetically controlled second hydrocyanation produces adiponitrile. Ideal conditions establish the second hydrocyanation as the most rapid step in this process.(8)
Ordinarily the branched, allylic nitrile is favored in hydrocyanations of butadienes (see the first hydrocyanation above, which favors the branched isomer by a ratio of 2:1). However, the use of bulky, rigid ligands leads to the linear 1,4-adduct without the requirement of a Lewis acid (Eq. 9).(9)
Hydrocyanation of terminal alkynes using nickel(0) complexes typically produce the internal nitrile as the major product (Eq. 10).(10)
Trimethylsilyl cyanide is less toxic than hydrogen cyanide, and may be used successfully in metal-catalyzed additions to alkynes. For reactions with terminal alkynes completely site selectivity is observed (the internal nitrile is the observed product). Nearly complete stereoselectivity is observed as well (Eq. 11).(11)
Asymmetric hydrocyanation reactions involving nickel(0) catalysts most often employ chiral, chelating bis-triarylphosphite ligands. Strained alkenes may be hydrocyanated to yield exo products with moderate enantioselectivity. The example in Eq. 12 highlights the use of acetone cyanohydrin as an alternative source of HCN.(12)
Vinylarenes may be hydrocyanated with moderate enantioselectivity by carbohydrate-derived chelating phosphinites, such as the ligand shown in Eq. 13. The use of electron-poor aryl groups enhances the enantioselectivity of this process.(13)
Although hydrocyanation reactions of butadienes occur readily, enantioselective hydrocyanation reactions are still inefficient. The example in Eq. 14, which makes use of a bulky BINOL-based phosphite ligand, results in only 33% ee.(14)
Hydrocyanation is useful in the production of the non-steroidal anti-inflammatory drugs naproxen and ibuprofen (Eq. 15). Because a general method for the asymmetric hydrocyanation of vinylarenes is lacking, most preparations to date that use hydrocyanation yield these drugs in racemic form.(15)
Comparison to Other Methods
A variety of methods exist to install the cyano group on organic molecules. Related to the metal-catalyzed method described in this article is the palladium-catalyzed cross-coupling of vinyl halides and cyanide (Eq. 16). Hydrogen cyanide is not neeeded for this method and it is completely stereospecific; however, the vinyl halide substrates used in these reactions are more complex than simple alkenes and alkynes.(16)
Dehydration of aldehyde oximes may also produce nitriles in a straightforward manner (Eq. 17). Again, however, extra synthetic steps may be required to prepare the oxime.(17)
Nucleophilic addition of cyanide to alkyl halides is an alternative method for the preparation of saturated nitriles. Phase-transfer catalysis has been used in this context.
Experimental Conditions and Procedure
Hydrogen cyanide is highly toxic, volatile, and may polymerize explosively in the presence of base. Thus, extreme care should be taken when using this reagent in the laboratory. Nickel(0) catalysts are air sensitive and should be prepared in a glovebox; often, the active catalyst is prepared from a readily available nickel(0) source such as Ni(cod)2 prior to hydrocyanation. Alternatively, nickel(0) complexes can be prepared by reduction of nickel(II) salts with activated zinc in the presence of a ligand. Because decomposition to nickel(II) cyanide is a problematic side reaction, hydrogen cyanide is often added slowly to catalyst/substrate solutions. In some cases (reactions of vinylarenes, for instance), slow addition of the substrate is also necessary to prevent oligomerization.
Tetrakis(tri-4-tolylphosphite)nickel(0) (1.50 g, 1.0 mmol) and tri-4-tolyl phosphite (0.30 mL, 1.0 mmol) were dissolved in toluene (30 mL). Zinc chloride (0.06 g, 0.5 mmol) was dissolved in propionitrile (0.5 mL) and then added to the catalyst mixture, which was subsequently heated to 88◦ under nitrogen. 4-Isobutylstyrene (3.45 g, 21.5 mmol) was added by syringe pump; 0.30 g was added initially and the remainder added at 1.33 mL/h. After the initial addition of the substrate, HCN/N2 was fed at 3 mL/min for 3 hours and then at 1 mL/min for 3 hours. Silica gel chromatography of the reaction mixture (hexane/EtOAc) afforded the title product (2.66 g, 14.2 mmol, 68% yield) as a colorless liquid; 1H NMR (360 MHz, CDCl3) δ 0.90 (d, J = 7 Hz, 6 H), 1.61 (d, J = 7 Hz, 3 H), 1.76–1.94 (m, 1 H), 2.47 (d, J = 7 Hz, 2 H), 3.86 (q, J = 7 Hz, 1 H), 7.14 (d, J = 8 Hz, 2 H), 7.25 (d, J = 8 Hz, 2 H). Anal. Calcd for C13H17N: C, 83.37; H, 9.15; N, 7.48. Found: C, 83.31; H, 8.95; H, 7.47.
- ↑ Rajanbabu, T. V. Org. React. 2011, 75, 1.
- ↑ Podlech, J. Science of Synthesis 2004, 19, 311.
- ↑ Drinkard, W. C.; Lindsey, R. V., Jr. U.S. Patent 3,496,215 (1970); Chem. Abstr. 1972, 77, 4982.
- ↑ Tolman, C. A.; McKinney, R. J.; Seidel, W. C.; Druliner, J. D.; Stevens, W. R. Adv. Catal. 1985, 33, 1.
- ↑ a b Tolman, C. A.; Seidel, W. C.; Druliner, J. D.; Domaille, P. J. Organometallics 1984, 3, 33.
- ↑ McKinney, R. J.; Roe, D. C. J. Am. Chem. Soc. 1986, 108, 5167.
- ↑ Jackson, W. R.; Lovel, C. G. J. Chem. Soc., Chem. Commun. 1982, 1231.
- ↑ Parshall, G. W.; Ittel, S. D. Homogeneous Catalysis, 2nd ed.; Wiley: New York, 1992; pp 201 – 202.
- ↑ Bäckvall, J.-E.; Andell, O. S. Organometallics 1986, 5, 2350.
- ↑ RajanBabu, T. V.; Casalnuovo, A. L. Pure Appl. Chem. 1994, 66, 1535.
- ↑ a b c Casalnuovo, A. L.; RajanBabu, T. V.; Ayers, T. A.; Warren, T. H. J. Am. Chem. Soc. 1994, 116, 9869.
- ↑ Goertz, W.; Keim, W.; Vogt, D.; Englert, U.; Boele, M. D. K.; van der Veen, L. A.; Kamer, P. C. J.; van Leeuwen, P. W. N. M. J. Chem. Soc., Dalton Trans. 1998, 2981.
- ↑ McKinney, R. J.; Nugent, W. A. Organometallics 1989, 8, 2871.
- ↑ Bini, L.; Muller, C.; Wilting, J.; von Chrzanowski, L.; Spek, A. L.; Vogt, D. J. Am. Chem. Soc. 2007, 129, 12622.
- ↑ Chatani, N.; Takeyasu, T.; Horiuchi, N.; Hanafusa, T. J. Org. Chem. 1988, 53, 3539.
- ↑ Baker, M. J.; Pringle, P. G. J. Chem. Soc., Chem. Commun. 1991, 1292.
- ↑ Wilting, W.; Janssen, M.; Mu ̈ller, C.; Lutz, M.; Spek, A. L.; Vogt, D. Adv. Synth. Catal. 2007, 349, 350.
- ↑ a b c Nugent, W. A.; McKinney, R. J. J. Org. Chem. 1985, 50, 5370.
- ↑ Yamamura, K.; Murahashi, S.-I. Tetrahedron Lett. 1977, 4429.
- ↑ March, J. Advanced Organic Chemistry: Reactions, Mechanisms, and Structure, 4th ed.; Wiley: New York, 1992; pp 1038–1042.
- ↑ Starks, C. M. J. Am. Chem. Soc. 1971, 93, 195.