A titanium(III)-catalyzed redox umpolung reaction for the reductive cross-coupling of enones with acrylonitriles

Article abstract of DOI:10.1002/chem.201100501

Titanium does the job: 1,6-Difunctionalized alkyl units, which are traditionally hard to access, have been successfully obtained through a reductive cross-coupling of activated alkenes catalyzed by Ti^III. A plausible reaction mechanism that pro

Full text of DOI:10.1002/chem.201100501

COMMUNICATION
DOI: 10.1002/chem.201100501
A TitaniumACTHNUTRGNE(UNG III)-Catalyzed Redox Umpolung Reaction for the Reductive
Cross-Coupling of Enones with ACHUTNGRNENGU crylonitriles
Jan Streuff*[a]
The installation of carbonyl or nitrile functionalities at the
powder as stoichiometric reductant in THF. First attempts
achieved only low turnover numbers and 3a was received in
a maximum yield of 35%.[10] However, it was quickly discov-
ered that addition of trimethylsilyl chloride successfully cir-
cumvents product inhibition by formation of the corre-
sponding TMS-enol ether 4, which is easily hydrolized to
the desired cyclohexanone product during workup.[11] After
careful experimentation, reaction conditions were identified
that give complete conversion and 87% yield in the pres-
ence of 10 mol% of the titanium catalyst in THF as optimal
reaction medium within only 4 hours (Scheme 1). The cata-
lyst amount was successfully reduced to 5 mol% and a yield
of 71% was received after a reaction time of 10 hours. How-
ever, an excess of inexpensive acrylonitrile is necessary to
assure suppression of reduction and homo-coupling prod-
ucts. Using other hydrochlorides such as collidine hydro-
chloride was less successful as well as the substitution of
zinc with manganese as terminal reductant. To demonstrate
the scope of this double-reductive alkylation reaction, a
series of substrates was submitted to the optimized condi-
tions on a 1 mmol scale. Enones with different ring sizes
such as cyclopentenone and cycloheptenone also worked
well (3b,c) and quaternary carbon centers were formed
smoothly when 3-methyl-2-cyclohexen-1-one and the 3,5-di-
methyl derivative were employed (3d,e).[12] With 2-phenyl-
cyclohexenone the corresponding trans-product 3 f was
formed in a clean reaction in 51% yield while the syn-prod-
À
1,2-, 1,4-, or even 1,6- positions of an alkyl chain by C C
bond formation is a challenging task for modern synthetic
methods because it requires the formal coupling of two simi-
larly polarized synthons. The concept of “umpolung”, which
found wide application in synthesis and methodology devel-
opment is the classical way to address this issue.[1] In the
past decades, electrochemical oxidative or reductive cou-
pling protocols have been applied in organic synthesis to
perform such connections, and important industrial applica-
tions like the Baizer–Monsanto adiponitrile process have
been established.[2,3] However, a related transition-metal-cat-
alyzed reductive cross-coupling reaction of activated alkenes
was not reported to date. In contrast, the transition-metal
catalyzed or mediated reductive conjugate alkylations or
alkyl-alkyl cross-coupling reactions known in the literature
usually require laborious synthesis of halogenated or meta-
lated coupling precursors, instead of employing a readily
available alkene.[4,5] To this end, a double reductive 1,4-alky-
lation concept was worked out that forges a 1,6-difunctional-
ized carbon skeleton starting from two different three-
carbon acceptor synthons (a3) in one redox economic reac-
tion [Eq. (1)].[6,7] It was envisioned that low valent titanium
catalysts, which have found versatile application in pinacol
couplings, epoxide opening and cyclization reactions, or con-
jugate reductions, could perform this redox umpolung reac-
tion.[8,9]
Screening experiments were started by reacting 2-cyclo-
hexen-1-one (1a) with an excess of acrylonitrile (2a) in the
presence of titanocene dichloride as catalyst and zinc
uct 3g was received from 4-tert-butyl cyclohexenone (3JHH
=
3 Hz). The polysubstituted and sterically congested terpe-
noids (S)-verbenone and (S)-carvone formed the desired
products in 44% and 80% yield, respectively (3h–j). In
À
both cases, the diastereoselectivity for the C C bond form-
ing step was high (>10:1 and 9:1, respectively) and the
remote olefin functionality of carvone remained unaffected.
The stereochemical outcome in case of 3i and 3j is in very
good agreement with previously reported radical mediated
1,4-alkyl additions to carvone and related molecules.[13] In
addition, 3j was partially epimerized into 3i by treatment
with KOH in methanol for 30 minutes and thus the overall
diastereomeric ratio was increased to 4.2:1.[14] An acyclic
enone and unsaturated oxazolidinoneamides also formed
the desired products in moderate yields (3k–m). By replac-
ing TMSCl with AlEt2Cl, the diastereomeric ratio of prod-
uct 3m was slightly improved to 1.6:1.[6a] In the following,
the flexibility of the reaction towards the nitrile acceptor
was also probed and it was found that crotononitrile was sig-
[a] Dr. J. Streuff
Institut fꢀr Organische Chemie und Biochemie
Albert-Ludwigs-Universitꢁt Freiburg
Albertstrasse 21, 79104 Freiburg i. Br. (Germany)
Fax : (+49)761-203-8715
E-mail: jan.streuff@ocbc.uni-freiburg.de
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201100501.
Chem. Eur. J. 2011, 17, 5507 – 5510
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5507

Scheme 2. Isolation of the silyl ether intermediate as single regioisomer.
Treatment with mCPBA yields a-hydroxyketone 5. CPBA=chloro-
AHCTUNGTREGpNNUN eroxybenzoic acid.
ing e-ketoamide 3q in 73% yield [Eq. (2)]. It was observed
that catalytic amounts of cinnamonitrile help to promote
this reaction.
When the catalyzed 1,4-alkylation was attempted on a
series of substrates that would form a second, vicinal quater-
nary carbon, the corresponding tertiary allylic alcohol was
isolated instead (Figure 1). A 3-tert-butyl substituent, for ex-
Scheme 1. Scope and limitations of the reductive coupling. [a] Reaction
conditions: (1.0 mmol), (5.0 mmol), [Cp2TiCl2] (10 mol%), zinc
1
2
powder (2.0 mmol), HClNEt3 (1.3 mmol), TMSCl (1.5 mmol), THF (c=
0.4m), T=358C. Then aq. HCl or TBAF. [b] Carried out with [Cp2TiCl2]
(5 mol%). [c] Reaction was run in the dark. [d] Reaction run at 258C.
[e] Combined yield. [f] TMSCl was substituted by AlEt2Cl (2.0 equiv),
which was precomplexed to the substrate at 08C. brsm=based on recov-
ered starting material, Cp=cyclopentadiene, d.r.=diasteromeric ratio,
TBAF=tetra-n-butylammonium fluoride, TMS=trimethylsilyl.
Figure 1. Increased sterical bulk shifts the reaction outcome towards 1,2-
addition. For the reaction conditions, see Scheme 1.
nificantly less reactive than acrylonitrile but still gave an
isolable amount of the product diastereomers (27%). How-
ever, methacrylonitrile showed similar reactivity to acryloni-
trile (3o,p) and again, a quaternary center was formed. In
principle, the TMS-enol ether intermediates can be isolated
from the reaction mixture. For example, enol ether 4 was
obtained in 87% yield as a single regioisomer (Scheme 2).
The corresponding TES- and TBS-enol ethers were synthe-
sized in the same manner with good regioselectivities.[14]
Further transformation of these silyl enol ethers was exem-
plified by Rubottom oxidation of 4.
ample, showed the same effect and led to formation of 6,
which is prone to water elimination. Interestingly, a free
ketone is easily tolerated by the reaction and products 7 and
8 were produced in 46% and 55% yield with moderate dia-
stereoselectivity from the common diketone precursors. No-
tably, the Wieland–Miescher diketone also led to a small
amount of the 1,4-addition product (14%) containing two
adjacent quaternary carbons. In case of progesterone, a 10:1
mixture of both C-3 epimers was isolated in 65% yield. The
relative configuration of the major diastereomer (9) was un-
ambiguously determined by X-ray crystallographic analy-
sis.[14,15]
The acrylonitriles can be substituted with acrylamides as
long as they have a sufficiently low-lying LUMO. A proof of
principle was carried out by employing N-methyl-N-tosyl-
ACHTUNGTRENNUNGacrylamide, which was smoothly transformed into the result-
5508
www.chemeurj.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 5507 – 5510

TitaniumACHTUNGTRENNUNG(III)-Catalyzed Umpolung Reactions
COMMUNICATION
acrylonitrile was added. After 4 h, workup was carried out by addition of
1n aqueous HCl (10 mL) and stirring for 30 min. Dichloromethane
(20 mL) was added and the organic layer was separated followed by ex-
traction of the aqueous phase with dichloromethane (3ꢃ10 mL). The
combined organic extracts were dried (Na2SO4), concentrated, and puri-
fied by flash chromatography (silica gel, hexanes/ethyl acetate=3:2, Rf =
0.45). The product (3a) was obtained as a colorless oil in 87% yield
(131.1 mg). 1H NMR (250 MHz, CDCl3): d=1.31 (ddd, J=3.1, 10.9,
14.0 Hz, 1H), 1.52–1.71 (m, 1H), 1.60 (dd, J=5.4, 7.2 Hz, 1H), 1.66 (dd,
J=4.3, 7.0 Hz, 1H), 1.74–2.06 (m, 4H), 2.13–2.41 (m, 3H), 2.32 ppm (t,
J=7.3 Hz, 2H); 13C NMR (62.5 MHz, CDCl3): d=14.42, 24.48, 30.12,
31.37, 37.49, 40.90, 46.82, 119.07, 209.84 ppm; IR (NaCl): n˜ =2939, 2868,
2245, 1711, 1451, 1425, 1348, 1318, 1285, 1230, 1170, 1104, 1060, 949, 868,
755 cmÀ1; MS (EI, 70 eV): m/z (%): 151.1 [M+] (76), 122.1 (18), 108.1
(100), 97.1 (41), 80.1 (8), 55.3 (40), 41.3 (14); HRMS (EI) calcd for
C9H13ON+: 151.09971, found: 151.09960.
Scheme 3. Proposed catalytic cycle. The turnover of the second catalyst
equivalent is not drawn for clarity reasons.
Acknowledgements
I wish to thank Prof. B. Breit for his generous support and the Fonds der
Chemischen Industrie for financial support in form of a Liebig Fellow-
ship. Dr. Manfred Keller is acknowledged for X-ray crystallographic
analysis.
A reasonable catalytic cycle could start with the genera-
tion of a titaniumACHTUNGTRENNUNG(III) catalyst by reduction of [Cp2TiCl2]
with zinc (Scheme 3). This species (for simplification the di-
meric form is not drawn), which has been studied in detail
before,[16] transfers a single electron to the enone substrate
and an allylic radical is generated. This electron-rich radical
adds to acrylonitrile at the b position of the former enone,
thus generating a less nucleophilic radical next to the cyano
group.[17] This intermediate is quickly reduced and protonat-
ed by a second TiIII species and the resulting TiIV enolate is
displaced by TMS chloride. Finally, the catalyst is regenerat-
ed in another reduction step.
Keywords: cross-coupling · enones · radical reactions ·
titanium · umpolung
[1] a) R. W. Hoffmann, Elemente der Syntheseplanung, Elsevier, Mꢀn-
chen, 2006; b) A. B. Smith, III, C. M. Adams, Acc. Chem. Res.
2004, 37, 365–377; c) M. Schmittel, Top. Curr. Chem. 1994, 169,
183–230; d) D. Seebach, Angew. Chem. 1979, 91, 259–278; Angew.
Chem. Int. Ed. Engl. 1979, 18, 239–258.
[2] a) K. D. Moeller, Synlett 2009, 1208–1218; b) Encyclopedia of Elec-
trochemistry, Vol. 8, Organic Electrochemistry (Ed.: H. J. Schꢁfer),
Wiley-VCh, Weinheim, 2004; c) Organic Electrochemistry (Eds.: H.
Lund, O. Hammerich), Marcel Dekker, New York, 2000; for a relat-
ed cobalt-catalyzed dimerization, see: d) C.-C. Wang, P.-S. Lin, C.-H.
Cheng, Tetrahedron Lett. 2004, 45, 6203–6206.
In conclusion, a titanium-catalyzed double-reductive alky-
lation of enones was developed that employs readily avail-
able alkene precursors and leads to 1,6-difunctionalized ke-
tonitriles by a redox umpolung process. The regioselectivity
can be shifted to yield tertiary 1,4-cyanoalcohols by in-
creased sterical demand. This reaction does not only allow
[3] a) K. Weissermel, H.-J. Arpe, Industrielle Organische Chemie, VCH,
Weinheim, 1988; b) M. M. Baizer, Tetrahedron 1984, 40, 935–969;
c) M. M. Baizer, Chem. Ind. 1979, 435–439.
3
3
À
for the catalytic formation of sp sp carbon–carbon bonds,
but also gives rise to regioselectively formed silyl enol
ethers, which can be intercepted and employed in follow-up
chemistry. We are currently exploring this chemistry further
toward an enantioselective protocol and other coupling part-
ners. A detailed mechanistic investigation will follow.
[4] For reviews, see: a) T. Jerphagnon, M. G. Pizzuti, A. J. Minnaard,
B. L. Feringa, Chem. Soc. Rev. 2009, 38, 1039–1075; b) A. Rudolph,
M. Lautens, Angew. Chem. 2009, 121, 2694–2708; Angew. Chem.
Int. Ed. 2009, 48, 2656–2670; c) A. Alexakis, J. E. Bꢁckvall, N.
Krause, O. Pꢄmies, M. Diꢅguez, Chem. Rev. 2008, 108, 2796–2823.
[5] Recent contributions: a) W. Li, N. Chen, J. Montgomery, Angew.
Chem. 2010, 122, 8894–8898; Angew. Chem. Int. Ed. 2010, 49, 8712–
8716; b) S. Ogoshi, T. Haba, M. Ohashi, J. Am. Chem. Soc. 2009,
131, 10350–10351; c) C.-Y. Ho, H. Ohmiya, T. F. Jamison, Angew.
Chem. 2008, 120, 1919–1921; Angew. Chem. Int. Ed. 2008, 47, 1893–
1895; d) H.-T. Chang, T. T. Jayanth, C.-C. Wang, C.-H. Cheng, J.
Am. Chem. Soc. 2007, 129, 12032–12041; e) T. Hayashi, K. Yamasa-
ki, Chem. Rev. 2003, 103, 2829–2844; f) C.-C. Wang, P.-S. Lin, C.-H.
Cheng, J. Am. Chem. Soc. 2002, 124, 9696–9697.
[6] For related stoichiometric umpolung reactions, see: a) M. P. DeMar-
tino, K. Chen, P. S. Baran, J. Am. Chem. Soc. 2008, 130, 11546–
11560; b) G. Pandey, M. K. Ghorai, S. Hajra, Tetrahedron Lett. 1998,
39, 8341–8344; c) M. Schmittel, A. Burghart, W. Malisch, J. Reising,
R. Sçllner, J. Org. Chem. 1998, 63, 396–400.
[7] For a review on redox economy, see: N. Z. Burns, P. S. Baran, R. W.
Hoffmann, Angew. Chem. 2009, 121, 2896–2910; Angew. Chem. Int.
Ed. 2009, 48, 2854–2867.
Experimental Section
Representative procedure (3a): A flame-dried 50 mL-Schlenk tube was
charged under argon atmosphere with a magnetic stir bar and triethyl-
ACHTUNGTRENNUNGamine hydrochloride (179 mg, 1.3 mmol, 1.3 equiv). Stirring was started
and the vessel was evacuated and backfilled with argon after a few mi-
nutes. Zinc powder (131 mg, 2.0 mmol, 2.0 equiv) was added, followed by
titanocene dichloride (25 mg, 0.1 mmol, 10 mol%) and absolute THF
(2.5 mL). The reaction vessel was placed in an oil bath with a tempera-
ture of 358C. Once the color change of the slurry from purple-brownish
to lime-green was complete, cyclohexenone (97 mL, 1.0 mmol, 1.0 equiv)
was rapidly added, followed by acrylonitrile (332 mL, 5.0 mmol, 5 equiv)
and TMS-Cl (190 mL, 1.5 mmol, 1.5 equiv). Upon addition of substrate
the mixture turned deep-green and changed to deep-purple when the
Chem. Eur. J. 2011, 17, 5507 – 5510
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5509

J. Streuff
[8] Selected reviews: a) A. Gansꢁuer, C.-A. Fan, F. Keller, P. Karbaum,
Chem. Eur. J. 2007, 13, 8084–8090; b) A. Gansꢁuer, J. Justicia, C.-A.
Fan, D. Worgull, F. Piestert, Top. Curr. Chem. 2007, 279, 25–52.
[9] a) A. Gansꢁuer, L. Shi, M. Otte, J. Am. Chem. Soc. 2010, 132,
11858–11859; b) A. D. Kosal, B. L. Ashfeld, Org. Lett. 2010, 12,
44–47; c) A. Gansꢁuer, D. Worgull, K. Knebel, I. Huth, G. Schna-
kenburg, Angew. Chem. 2009, 121, 9044–9047; Angew. Chem. Int.
Ed. 2009, 48, 8882–8885; d) J. Justicia, I. Sancho-Sanz, E. ꢆlvarez-
Manzaneda, J. E. Oltra, J. M. Cuerva, Adv. Synth. Catal. 2009, 351,
2295–2300; e) R. E. Estꢅvez, J. Justicia, B. Bazdi, N. Fuentes, M.
Paradas, D. Choquesillo-Lazarte, J. M. Garcꢇa-Ruiz, R. Robles, A.
Gansꢁuer, J. M. Cuerva, J. E. Oltra, Chem. Eur. J. 2009, 15, 2774–
2791; f) R. E. Estꢅvez, M. Paradas, A. Millꢈn, T. Jimꢅnez, R.
Robles, J. M. Cuerva, J. E. Oltra, J. Org. Chem. 2008, 73, 1616–
1619; g) A. Gansꢁuer, C.-A. Fan, F. Keller, J. Keil, J. Am. Chem.
Soc. 2007, 129, 3484–3485; h) R. E. Estꢅvez, J. L. Oller-Lꢉpez, R.
Robles, C. R. Melgarejo, A. Gansꢁuer, J. M. Cuerva, J. E. Oltra,
Org. Lett. 2006, 8, 5433–5436; i) A. Gansꢁuer, Chem. Commun.
1997, 457–458.
[11] TMSCl has shown a benifitial role in previous TiIII-catalyses (see ref-
erences 8b, 9c and 9i for example).
[12] Cyclopentenone is known to add to acrylonitriles in a [2+2] addi-
tion, therefore the reaction was carried out in the dark: D. I. Schus-
ter, G. Lem, N. A. Kaprinidis, Chem. Rev. 1993, 93, 3–22.
[13] A. Arase, Y. Masuda, A. Suzuki, Bull. Chem. Soc. Jpn. 1976, 49,
2275–2279.
[14] See the Supporting Information for details.
[15] CCDC-801986 contains the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
[16] R. J. Enemærke, J. Larsen, T. Skrydstrup, K. Daasbjerg, J. Am.
Chem. Soc. 2004, 126, 7853–7864.
[17] For an overview on the selectivity in C C bond forming radical re-
À
actions, see: a) B. Giese, Angew. Chem. 1985, 97, 555–567; Angew.
Chem. Int. Ed. Engl. 1985, 24, 553–565; b) B. Giese, Angew. Chem.
1983, 95, 771–782; Angew. Chem. Int. Ed. Engl. 1983, 22, 753–764.
[10] For an alternative route to products 3, see: M. C. P. Yeh, P. Knochel,
Tetrahedron Lett. 1988, 29, 2395–2396.
Received: February 14, 2011
Published online: April 12, 2011
5510
www.chemeurj.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 5507 – 5510