Copper(I) Cyanide1

CuCN

[544-92-3]  · CCuN  · Copper(I) Cyanide  · (MW 89.57) (.NaCN)

[13715-19-0]  · C2CuN2Na  · Copper(I) Cyanide-Sodium Cyanide  · (MW 138.58) (.LiCl)

[59219-07-7]  · CClCuLiN  · Copper(I) Cyanide-Lithium Chloride  · (MW 131.96) (.2LiCl)

[121340-53-2]  · CCl2CuLi2N  · Copper(I) Cyanide-Lithium Chloride  · (MW 174.35) (.2LiBr)

[129126-28-9]  · CBr2CuLi2N  · Copper(I) Cyanide-Lithium Bromide  · (MW 263.25)

(useful precursor for organocopper(I) and organocuprate(I) reagents1)

Alternate Name: cuprous cyanide.

Physical Data: mp 474 °C; d 2.920 g cm-3.

Solubility: insol H2O and most organic solvents. The lithium halide complexes are sol THF.

Form Supplied in: off-white solid.

Handling, Storage, and Precautions: highly toxic; must be handled with care.

General Discussion.

The first investigation of organocopper chemistry was carried out by Kondyreva and Fomin,2 who in 1915 studied the effect of heavy metal salts on organomagnesium compounds. They found that organocopper compounds, prepared from CuCl, CuBr, CuI, CuSCN, and CuCN, decomposed to give equimolar amounts of alkane and alkene when the organic group was n-alkyl, and biaryl when it was aryl. These observations were confirmed by Whitesides et al. a half century later.3 The Russian authors also noted that the solutions prepared from Grignard reagents and copper salts no longer reacted with acetone to give t-butanol. While they did not find a synthetic application for their solutions, it should be noted that Gilman did not investigate the reaction chemistry of the solutions he prepared from 2 equiv of MeLi and copper(I) salts, either.4 Nevertheless, organocuprates are now commonly called Gilman reagents.

The pioneering work on synthetic applications of organocopper reagents was done almost exclusively with Copper(I) Iodide. Unfortunately, nearly two decades were to pass before it was found that Gilman reagents prepared from CuCN and 2 equiv of RLi were purer, more thermally stable,5 and often gave higher yields of the desired product than Gilman reagents prepared from CuI.1,7 Gilman reagents such as Bu2CuLi.LiI appear to be more prone to electron transfer and radical reactions than the corresponding reagents prepared from CuCN.6

Based upon their apparent higher reactivity, it was originally claimed that the reagents prepared from 2 equiv of RLi and CuCN were higher order cyanocuprates, R2Cu(CN)Li2;1a-c,7 however, NMR studies,9-11 EXAFS investigations,12 and theoretical calculations13 have recently converged on the conclusion that the CN is not bonded to Cu. Lipshutz has recently conceded that the CN is not s-bonded to Cu.8 Thus these reagents should be represented as R2CuLi.LiCN by analogy with R2CuLi.LiI and R2CuLi.LiBr for the cuprates prepared from CuI and Copper(I) Bromide, respectively. It has been demonstrated by 13C NMR and 6Li NMR that LiI is not free in the Gilman reagent prepared from CuI and 2PhLi in DMS,14 and that LiI and LiCN are not part of the cuprate bonding in THF.9-11 It has recently been shown how the aggregation states of organocopper reagents in solution may be assigned by using 13C NMR spectroscopy.11

In contrast, both NMR and EXAFS studies confirm that CN is bonded to Cu in the so-called lower order cyanocuprates, RCu(CN)Li,10,12 prepared from 1 equiv of RLi and CuCN. While they are relatively unreactive towards a-enones and alkyl halides,15 they have been used successfully with more reactive substrates such as allylic carboxylates16 and vinyloxiranes.17 The introduction of additives such as Boron Trifluoride18,19 should give these reagents broader applicability. Trost used lower order cyanocuprates when he introduced the concept of chirality transfer in acyclic systems,20 which has been applied brilliantly by Ibuka and Yamamoto.18

Recently, several novel lower order cyanocuprates have found application. For example, Piers has converted 2-alkynoates into either (Z)- or (E)-3-trimethylstannyl-2-alkenoates by treatment with Me3SnCu(CN)Li under the appropriate conditions (eq 1).21a He also compared lower order phenyl cyano- and thiocuprates in developing an efficient methylenecyclopentane annulation (eq 2).21b

In a most exciting development from the viewpoint of molecular complexity, Knochel has prepared highly functionalized cyanocuprates RCu(CN)ZnX, where X = I, Br, Cl, OMs, OTs, or OP(O)(OR)2, from the corresponding alkylzinc reagents.22 In a related development, Rieke has reported the direct formation of highly functionalized allylic organocopper reagents via the reduction of CuCN.2LiBr, CuCN.LiCl, or CuCN.2LiCl with Lithium Naphthalenide to give a highly reactive form of Cu metal.23 Linderman has prepared a-alkoxyorganocuprate reagents from the corresponding organotin compounds and reacted them with a,b-unsaturated carbonyl compounds (eq 3).24

Corey has used Bu4NCu(CN)2 as a precursor to a new class of cuprate reagents, which he symbolizes as RCu(CN)2(NBu4)Li,25 implying that they are higher order cuprates with two CN ligands bonded to Cu, although this remains to be proven. The 13C NMR spectroscopy of organocopper reagents has now been developed to the point where conjectures concerning the structures of new organocopper species should not be made without making sure that they are consistent with the NMR data.11 At the time of this writing there is still a backlog of structures that need to be checked. Theory is also being developed to the point where it can shed considerable light on whether structures are feasible or not.13

As discussed above, when the Gilman reagents prepared from CuI were introduced, side-by-side comparisons were not made with the corresponding cuprates prepared from other CuI salts. (Or, if comparisons were made, they were not reported.) In other words, a depth-first search was done, rather than a breadth-first one.26 Unfortunately, the same pattern was repeated when the cuprates prepared from CuCN were introduced: papers with tables of yields for typical substrates were published, but very few direct comparisons with other cuprates were made.1b,c,7a Thus the best advice we can give the synthetic chemist is to try several of the most successful versions of the Gilman reagent, viz, those prepared from CuI, CuBr.SMe2 (see Copper(I) Bromide), and CuCN.1a-c It is also useful to compare organocopper reagents prepared from lithium reagents with those prepared from the corresponding Grignard reagents (See Copper(I) Chloride).

In one of the few direct comparisons, Fleming has shown that the silylcupration of alkynes with (Me2PhSi)2CuLi.LiCN gives the opposite regiochemistry to (Me2PhSi)2CuLi.LiBr, prepared from CuBr.DMS.27 It is interesting to note that Fleming's work predates Lipshutz's, and that he uses the correct (lower order) formulation. Bertz has compared CuCN with CuI, CuBr, CuBr.DMS, CuCl, CuSCN and CuOTf as far as the formation of alkyl and aryl Gilman reagents is concerned,28a and also reacted a number of them, including those made from CuCN, with dithioesters, where they give carbophilic addition.28b For the preparation of stoichiometric cuprates from Grignard reagents, Copper(I) Trifluoromethanesulfonate (copper(I) triflate) was demonstrated to be superior to the other CuI salts, including CuCN.28b

While CuCN has been used in catalytic amounts with Grignard reagents (where the ate complex may be presumed to be the active species),16b,29,30 CuCN has not been shown to be superior to the other CuI salts, and even some CuII salts appear to be equally effective.1a For example, in the addition of Grignard reagents to nitriles, CuCN, CuI, CuBr, CuBr.DMS, and CuCl all gave the same yield of product to within ±3%. Thus the yields of ketimine product from t-BuMgCl and 4-methoxybenzonitrile were 87-93% (i.e. 90 ± 3%) after 2 h in refluxing THF in the presence of 2 mol % of the CuI salt.30 This is a very useful extension and generalization of Bertz's work on the activation of the carbon-nitrogen double bond by CuI.31

Lipshutz has described stoichiometric higher order, mixed lithio magnesio organocuprates prepared from CuCN, 2-lithiothiophene, and Grignard reagent (1:1:1), although no evidence that a higher order species is present in such solutions has been adduced.32a In fact, the authors admit that these reagents are not discrete.32a The corresponding organocuprates prepared from CuCN, 2-lithiothiophene, and a lithium reagent (or lithium 2-thienyl(cyano)cuprate(I) and a lithium reagent) have been assigned a higher order structure,1b but are probably lower order R(2-thienyl)CuLi.LiCN reagents.9-11 The 2-thienyl group was introduced as a nontransferred ligand for mixed cuprates by the Swedish group of Malmberg, Nilsson, and Ullenius.32b Whatever their structure may be, they have been applied to some interesting synthetic problems, such as the prostaglandin synthesis shown in eq 4.33

As mentioned above, Bu4NCu(CN)2 has been used as a cuprate precursor. The related alkali metal dicyanocuprates, especially NaCu(CN)2, have been used in typical cuprate coupling reactions with aryl and vinyl halides.34,35 In other typical organocopper reactions, CuCN itself reacts with acid chlorides to yield a-ketonitriles,36 and with aryl iodides to afford aryl cyanides.37

It is noteworthy that the first sodium organocuprate, Bu2CuNa.NaCN, was prepared from BuNa and CuCN,38 and Seyferth prepared the first acyl cuprates from R2CuLi.LiCN and CO.39 The latter development enables the synthesis of 1,4-diketones directly from a-enones via conjugate addition.

Whitesides et al. introduced the oxidative coupling of CuI ate complexes, which they prepared from CuI.40 Bertz and Gibson demonstrated that the oxidation of organocuprates from CuCN gave significantly different results,41 which Lipshutz et al. developed into a synthesis of unsymmetrical biaryls.42 Oxidative couplings of heterocuprates usually proceed in poor yields;1e,43 however, Snieckus et al. have developed useful synthetic methodology for the N-arylation of amines based upon the oxidation of aryl amidocuprates.44

In one of his many pioneering studies, House showed that the enolates formed upon the addition of Me2CuLi.LiBr to a-enones were basically lithium enolates.45 Nevertheless, Posner showed that the addition of 1 equiv of CuCN to a lithium enolate could dramatically decrease polyalkylation in cyclopentanone derivatives.46

Finally, CuCN has been used to catalyze the reactions of diazoacetic esters;1f however, Copper(I) Chloride is the CuI salt most frequently used for this purpose.

Related Reagents.

See Copper(I) Bromide, Copper(I) Chloride, Copper(I) Iodide (and their combination reagents); also Copper(I) Trifluoromethanesulfonate. For examples of cyanocuprates, see Lithium Butyl(cyano)cuprate, Lithium Cyano(methyl)cuprate, etc.


1. (a) Lipshutz, B. H.; Sengupta, S. OR 1992, 41, 135. (b) Lipshutz, B. H. S 1987, 325. (c) Lipshutz, B. H.; Wilhelm, R. S.; Kozlowski, J. A. T 1984, 40, 5005. (d) Chapdelaine, M. J.; Hulce, M. OR 1990, 38, 225. (e) Kauffmann, T. AG(E) 1974, 13, 291. (f) Dave, V.; Warnhoff, E. W. OR 1970, 18, 217.
2. Kondyreva, N. V.; Fomin, D. A. Zh. Russ. Fiz.-Khim. O-va, Chast Khim. 1915, 47, 190 (CA 1915, 9, 1473).
3. Whitesides, G. M.; Stedronsky, E. R.; Casey, C. P.; San Filippo, Jr., J. JACS 1970, 92, 1426.
4. Gilman, H.; Jones, R. G.; Woods, L. A. JOC 1952, 17, 1630.
5. Bertz, S. H.; Dabbagh, G. CC 1982, 1030.
6. Bertz, S. H.; Dabbagh, G.; Mujsce, A. M. JACS 1991, 113, 631.
7. (a) Lipshutz, B. H.; Wilhelm, R. S.; Floyd, D. M. JACS 1981, 103, 7672. (b) Lipshutz, B. H.; Sharma, S.; Ellsworth, E. L. JACS 1990, 112, 4032.
8. Lipshutz, B. H.; James, B. JOC 1994, 59, 7585.
9. Bertz, S. H. JACS 1990, 112, 4031.
10. Bertz, S. H. JACS 1991, 113, 5470.
11. Bertz, S. H.; Dabbagh, G.; He, X.; Power, P. P. JACS 1993, 115, 11 640.
12. Stemmler, T.; Penner-Hahn, J. E.; Knochel, P. JACS 1993, 115, 348.
13. Snyder, J. P.; Spangler, D. P.; Behling, J. R.; Rossiter, B. E. JOC 1994, 59, 2665.
14. Bertz, S. H.; Dabbagh, G. JACS 1988, 110, 3668.
15. Gorlier, J.-P.; Hamon, L.; Levisalles, J.; Wagnon, J. CC 1973, 88.
16. (a) Underiner, T. L.; Goering, H. L. JOC 1988, 53, 1140. (b) Underiner, T. L.; Goering, H. L. JOC 1990, 55, 2757. (c) Tseng, C. C.; Paisley, S. D.; Goering, H. L. JOC 1986, 51, 2884.
17. Marino, J. P.; Jaén, J. C. JACS 1982, 104, 3165.
18. (a) Ibuka, T.; Tanaka, M.; Nishii, S.; Yamamoto, Y. CC 1987, 1596. (b) Ibuka, T.; Tanaka, M.; Nishii, S.; Yamamoto, Y. JACS 1986, 108, 7420.
19. (a) Kang, S.-K.; Lee, D.-H.; Sim, H.-S.; Lim, J.-S. TL 1993, 34, 91. (b) Oppolzer, W.; Moretti, R.; Godel, T.; Meunier, A.; Löher, H. TL 1983, 24, 4971.
20. Trost, B. M.; Klun, T. P. JOC 1980, 45, 4256.
21. (a) Piers, E.; Wong, T.; Ellis, K. A. CJC 1992, 70, 2058. (b) Piers, E.; Karunaratne, V. CC 1983, 935.
22. (a) Knochel, P.; Chou, T.-S.; Jubert, C.; Rajagopal, D. JOC 1993, 58, 588. (b) Jubert, C.; Knochel, P. JOC 1992, 57, 5425. (c) Achyutha Rao, S.; Knochel, P. JOC 1991, 56, 4591. (d) Knochel, P.; Yeh, M. C. P.; Berk, S. C.; Talbert, J. JOC 1988, 53, 2390.
23. Stack, D. E.; Dawson, B. T.; Rieke, R. D. JACS 1992, 114, 5110.
24. Linderman, R. J.; Godfrey, A.; Horne, K. TL 1987, 28, 3911.
25. Corey, E. J.; Kyler, K.; Raju, N. TL 1984, 25, 5115.
26. Cormen, T. H.; Leiserson, C. E.; Rivest, R. L. Introduction to Algorithms; MIT Press: Cambridge, MA, 1990; pp 469-485.
27. (a) Fleming, I.; Newton, T. W.; Roessler, F. JCS(P1) 1981, 2527. (b) Fleming, I.; Roessler, F. CC 1980, 276.
28. (a) Bertz, S. H.; Gibson, C. P.; Dabbagh, G. TL 1987, 28, 4251. (b) Bertz, S. H.; Dabbagh, G.; Williams, L. M. JOC 1985, 50, 4414.
29. Trost, B. M.; Merlic, C. A. JACS 1988, 110, 5216.
30. Weiberth, F. J.; Hall, S. S. JOC 1987, 52, 3901.
31. (a) Bertz, S. H. JOC 1979, 44, 4967. (b) Bertz, S. H. TL 1980, 21, 3151.
32. (a) Lipshutz, B. H.; Parker, D. A.; Nguyen, S. L.; McCarthy, K. E.; Barton, J. C.; Whitney, S. E.; Kotsuki, H. T 1986, 42, 2873. (b) Malmberg, H.; Nilsson, M.; Ullenius, C. TL 1982, 23, 3823.
33. Okamoto, S.; Kobayashi, Y.; Kato, H.; Hori, K.; Takahashi, T.; Tsuji, H.; Sato, F. JOC 1988, 53, 5590.
34. House, H. O.; Fisher, W. F., Jr. JOC 1969, 34, 3626.
35. Procházka, M.; &SScaron;iroký, M. CCC 1983, 48, 1765.
36. Normant, J. F.; Piechucki, C. BSF(2) 1972, 2402.
37. Suzuki, H.; Hanafusa, T. S 1974, 53.
38. Bertz, S. H.; Gibson, C. P.; Dabbagh, G. OM 1988, 7, 227.
39. (a) Seyferth, D.; Hui, R. C. JACS 1985, 107, 4551. (b) Seyferth, D.; Hui, R. C. TL 1986, 27, 1473.
40. Whitesides, G. M.; San Filippo, Jr., J.; Casey, C. P.; Panek, E. J. JACS 1967, 89, 5302.
41. Bertz, S. H.; Gibson, C. P. JACS 1986, 108, 8286.
42. Lipshutz, B. H.; Siegmann, K.; Garcia, E.; Kayser, F. JACS 1993, 115, 9276.
43. &AAring;kermark, B.; Almemark, M.; Jutand, A. ACS 1982, B36, 451.
44. Iwao, M.; Reed, J. N.; Snieckus, V. JACS 1982, 104, 5531.
45. House, H. O.; Wilkins, J. M. JOC 1976, 41, 4031.
46. Posner, G. H.; Lentz, C. M. JACS 1979, 101, 934.

Steven H. Bertz & Edward H. Fairchild

LONZA, Annandale, NJ, USA



Copyright 1995-2000 by John Wiley & Sons, Ltd. All rights reserved.