Enantioselective synthesis of cyclic, quaternary oxonitriles

Article abstract of DOI:10.1021/jo1011202

Quaternary oxonitriles are stereoselectively generated from the union of five-, six-, and seven-membered 2-chloroalkenecarbonitriles with chiral alcohols via a Claisen rearrangement. The strategy rests on a new conjugate addition-elimination of allylic alkoxides to 2-chlorocycloalkenecarbonitriles to afford substituted 2-alkoxyalkenenitriles. Subsequent thermolysis unmasks a cyclic oxonitrile while selectively forming a new quaternary center with enantiomeric ratios typically greater than 9:1. The overall alkylation strategy addresses the challenge of enantioselectively generating hindered, quaternary centers while simultaneously installing ketone, nitrile, and olefin functionalities.

Full text of DOI:10.1021/jo1011202

pubs.acs.org/joc
Enantioselective Synthesis of Cyclic, Quaternary Oxonitriles
†
Yakup Gunes-, M. Fatih Polat, Ertan Sahin, Fraser F. Fleming,* and
†
†
,‡
€
Ramazan Altundas*,†
†Department of Chemistry, College of Sciences, Ataturk University, 25240 Erzurum, Turkey, and
‡Department of Chemistry and Biochemistry, Duquesne University, Pittsburgh,
Pennsylvania 15282-1530, United States
ramazanaltundas@atauni.edu.tr
Received June 9, 2010
Quaternary oxonitriles are stereoselectively generated from the union of five-, six-, and seven-membered
2-chloroalkenecarbonitriles with chiral alcohols via a Claisen rearrangement. The strategy rests on a new
conjugate addition-elimination of allylic alkoxides to 2-chlorocycloalkenecarbonitriles to afford sub-
stituted 2-alkoxyalkenenitriles. Subsequent thermolysis unmasks a cyclic oxonitrile while selectively
forming a new quaternary center with enantiomeric ratios typically greater than 9:1. The overall alkyla-
tion strategy addresses the challenge of enantioselectively generating hindered, quaternary centers while
simultaneously installing ketone, nitrile, and olefin functionalities.
Introduction
transition-metal-mediated coupling,4 aldol condensation,5
and allylic alkylation,6 for example, the aim is not only to
selectively install quaternary centers but to simultaneously
install multiple, easily differentiated functional groups.
Mixed carbonyl-nitrile nucleophiles have featured in several
quaternization methods.7 The attraction lies in the nitrile’s
small steric demand,8 which eases the steric compression, and
the facile chemoselective descrimination between carbonyl and
nitrile functionalities.9 Chiral alkylations of cyclic ketonitriles
are particularly attractive because of their rich history as
synthetic intermediates.10 For this reason, oxonitriles have been
employed in alkylations with chiral auxiliaries11 and chiral
bases12 and as pronucleophiles in chiral conjugate additions.13
Arguably the most robust method for installing hindered,
quaternary centers is the Claisen rearrangement.14 The predict-
ability and efficiency of the Claisen rearrangement appeared
Installing hindered, quaternary centers remains one of the
most challenging bond constructions in synthesis.1 Forging
quaternary centers with high enantiomeric ratios has stimu-
lated diverse strategies, many of which harness intra-
molecular assistance to overcome the unfavorable steric
compression of the central carbon atom.2 Within several
representative methods, asymmetric conjugate additions,3
(1) (a) Jiang, B.; Dong, J. J.; Si, Y. G.; Zhao, X. L.; Huang, Z. G.; Xu, M.
Adv. Synth. Catal. 2008, 350, 1360. (b) Jautze, S.; Peters, R. Angew. Chem.,
Int. Ed. 2008, 47, 9284. (c) Kita, Y.; Fujioka, H. Pure Appl. Chem. 2007, 79,
701. (d) Christoffers, J.; Mann, A. Angew. Chem., Int. Ed. 2001, 40, 4591.
(e) Corey, E. J.; Guzman-Perez, A. Angew. Chem., Int. Ed. 1998, 37, 388. (f) Fuji,
K. Chem. Rev. 1993, 93, 2037.
(2) (a) Trost, B. M.; Jiang, C. Synthesis 2006, 369. (b) Douglas, C. J.;
Overman, L. E. Proc. Natl. Acad. Sci. U.S.A. 2004, 111, 5363.
(3) (a) Jautze, S.; Peters, R. Synthesis 2010, 365. (b) Alexakis, A.;
Vuagnoux-d’Augustin, M.; Martin, D.; Kehrli, S.; Palais, L.; Henon, H.;
Hawner, C. Chimia 2008, 62, 461.
(4) (a) Liao, X.; Weng, Z.; Hartwig, J. F. J. Am. Chem. Soc. 2008, 130,
195. (b) Doyle, A. G.; Jacobsen, E. N. J. Am. Chem. Soc. 2005, 127, 62.
(5) (a) Das, J. P.; Chechik, H.; Marek, I. Nat. Chem. 2009, 1, 128. (b) Luo,
S.; Xu, H.; Zhang, L.; Li, J.; Cheng, J.-P. Org. Lett. 2008, 10, 653.
(c) Guizzetti, S.; Benaglia, M.; Raimondi, L.; Celentano, G. Org. Lett.
2007, 9, 1247. (d) Wu, Y.; Zhang, Y.; Yu, M.; Zhao, G.; Wang, S. Org. Lett.
2006, 8, 4417.
(6) (a) Usui, I.; Schimidt, S.; Breit, B. Org. Lett. 2009, 11, 1453.
(b) Fukuda, Y.; Kondo, K.; Aoyama, T. Tetrahedron Lett. 2007, 48, 3389.
(c) Plietker, B. Angew. Chem., Int. Ed. 2006, 45, 1469. (d) Trost, B. M.;
Crawley, M. L. Chem. Rev. 2003, 103, 2921.
(7) Fleming, F. F.; Iyer, P. S. Synthesis 2006, 893.
(8) The A-value is only 0.2 kcal mol-1: Eliel, E. L.; Wilen, S. H.; Mander,
L. N. In Stereochemistry of Organic Compounds: Wiley: New York, 1994;
pp 696-697.
(9) North, M. In Comprehensive Organic Functional Group Transforma-
tions II; Elsevier Science: New York, 2005; Chapter 3.18, pp 621-655.
(10) (a) Varner, M. A.; Grossman, R. B. Tetrahedron 1999, 55, 13867.
(b) Gawley, R. E. Synthesis 1976, 777. (c) Jung, M. E. Tetrahedron 1976, 32, 3.
(11) Enders, D.; Zamponi, A.; Raabe, G.; Runsik, J. Synthesis 1993, 725.
(12) (a) Nojiri, A.; Kumagai, N.; Shibasaki, M. J. Am. Chem. Soc. 2009,
131, 3779. (b) Nojiri, A.; Kumagai, N.; Shibasaki, M. J. Am. Chem. Soc.
2008, 130, 5630.
7092 J. Org. Chem. 2010, 75, 7092–7098
Published on Web 10/11/2010
DOI: 10.1021/jo1011202
r
2010 American Chemical Society

€
Gunes- et al.
JOCArticle
ideally matched with the versatility of oxonitriles7 for installing
functionalized quaternary centers. The strategy rests on con-
densing five-, six-, and seven-membered cyclic oxonitriles 1 with
a chiral allylic alcohol 2to afford an 2-alkoxyalkenenitrile 3for a
subsequent rearrangement (Scheme 1). The oxonitrile-Claisen
strategy holds a 3-fold attraction: the predictable configuration
of the quaternary center; the commercial availability of many
chiral allylic alcohols as an inexpensive source of chirality; and
installation of orthogonal ketone, nitrile, and alkene function-
alities. Described here is the successful crafting of chiral qua-
ternary oxonitriles through this efficient, stereoselective, and
predictable strategy.
chloroalkenenitriles was achieved by heating the reaction in a
sealed tube at 110 °C, which effectively provides 2-chlorocyclo-
hexenecarbonitrile (5b) and 2-chlorocycloheptenecarbonitrile
(5c) in 82% and 76%, respectively.
Exploratory addition-eliminations to 5a employed allyl
alcohol and sodium hydride (Scheme 2). Although the
desired 2-alkoxyalkenenitrile 3a was obtained, the isolation
was hampered by hydrolysis of the sensitive vinyl ether
during silica gel chromatography.19 An expedient solution
was to simply perform the thermolysis on the crude 2-alkoxy-
alkenenitrile 3a to directly access the allyl ketonitrile 4a (40%
yield, Scheme 2).
SCHEME 1. Oxonitrile-Claisen Rearrangement Strategy
SCHEME 2. Claisen-Oxonitrile Alkylation Strategy
Results and Discussion
A major challenge in executing the oxonitrile Claisen strategy
resides in accessing substituted 2-alkoxyalkenenitriles. O-Allyla-
tion of oxonitriles is often frustrated by competitive C-allylation
and generally requires primary allylic electrophiles.15 For assem-
bling the more challenging secondary β-allyloxyalkenenitriles
a conceptually different strategy was developed based on an
addition-elimination of an alkoxide to 2-chlorocycloalkenecar-
bonitriles 5a-5c (Scheme 2).16 Facile access to 2-chlorocyclo-
pentenecarbonitrile (5a) was readily achieved in 73% yield by
halogenating 1a17 with Ph3P and CCl4.18 Attemps to use PCl5,
POCl3, triphosgene, and oxalyl chloride either afforded signifi-
cantly lower yields or were completely ineffective. Although the
chlorination of 5a is relatively facile, the corresponding six- and
seven-membered oxonitriles (1b and 1c, respectively) are
unreactive toward Ph3P and CCl4 under identical condi-
tions. Access to the requisite six- and seven-membered
Close scrutiny of the crude reaction mixture before and
after the Claisen rearrangement revealed the emergence of
several side products during formation of the 2-alkoxyalk-
enenitrile, whereas the rearrangement faithfully generated
oxonitrile 4a of a similar purity. Consequently, subsequent
optimization focused on improving the conjugate addition-
elimination. Attempts to improve the alkoxide nucleophilicity
through solvent20 and temperature21 changes met with minimal
reward. The most efficacious conditions were to add 15-crown-5
to a room-temperature THF solution of the sodium alkoxide,
which raised the overall yield of 4a to 65%.22
Use of 15-crown-5 promotes the facile assembly of a variety of
substituted 2-alkoxyalkenenitriles (Table 1, column 4).23 Acyclic
and cyclic primary and secondary allylic alcohols react equally
well (Table 1, entries 1-8) with disubstituted olefins tolerating
terminal or internal substituents (Table 1, compare entries
2, 3, and 8). Tertiary alcohols, not surprisingly, are unable to
afford the 2-alkoxyalkenenitriles.24 Remarkably, the addition-
elimination proceeds with (S)-cis-verbenol (2i) to afford
alkenenitrile 3i despite the sterically hindered nature of
(13) (a) Kawato, Y.; Takahashi, N.; Kumagai, N.; Shibasaki, M. Org.
Lett. 2010, 12, 1484. (b) Li, H.; Song, J.; Deng, L. Tetrahedron 2009, 65, 3139.
(c) Marini, F.; Sternativo, S.; Del Verme, F.; Testaferri, L.; Tiecco, M. Adv.
Synth. Catal. 2009, 351, 1801. (d) Liu, T.-Y.; Li, R.; Chai, Q.; Long, J.; Li,
B.-J.; Wu, Y.; Ding, L.-S.; Chen, Y.-C. Chem.—Eur. J. 2007, 13, 319.
(e) Wang, B.; Wu, F.; Wang, Y.; Liu, X.; Deng, L. J. Am. Chem. Soc.
2007, 129, 768. (f) Wang, X.; Kitamura, M.; Maruoka, K. J. Am. Chem. Soc.
2007, 129, 1038. (g) Bell, M.; Poulsen, T. B.; Jørgensen, K. A. J. Org. Chem.
2007, 72, 3053. (h) Wang, Y.; Liu, X.; Deng, L. J. Am. Chem. Soc. 2006, 128,
3928. (i) Taylor, M. S.; Zalatan, D. N.; Lerchner, A. M.; Jacobsen, E. N.
J. Am. Chem. Soc. 2005, 127, 1313. (j) Li, H.; Song, J.; Liu, X.; Deng, L.
J. Am. Chem. Soc. 2005, 127, 8948. (k) Takenaka, K.; Minakawa, M.;
Uozumi, Y. J. Am. Chem. Soc. 2005, 127, 12273. (l) Taylor, M. S.; Jacobsen,
E. N. J. Am. Chem. Soc. 2003, 125, 11204. (m) Sawamura, M.; Hamashima,
H.; Ito, Y. J. Am. Chem. Soc. 1992, 114, 8295. (n) Giardina, A.; Marcantoni,
E.; Mecozzi, T.; Petrini, M. Eur. J. Org. Chem. 2001, 713.
(14) Castro, A. M. M. Chem. Rev. 2004, 104, 2939.
(15) (a) Ziegler, F. E.; Nangia, A.; Schulte, G. J. Am. Chem. Soc. 1987,
109, 3987. (b) Coates, R. M.; Rogers, B. D.; Hobbs, S. J.; Curran, D. P.; Peck,
D. R. J. Am. Chem. Soc. 1987, 109, 1160. (c) Coates, R. M.; Hobbs, S. J.
J. Org. Chem. 1984, 49, 1401. (d) Coates, R. M.; Shah, S. K.; Mason, R. W.
J. Am. Chem. Soc. 1982, 104, 2198.
(16) Under basic conditions, primary alcohols engage in a conjugate
addition-elimination: (a) Redman, A. M.; Dumas, J.; Scott, W. J. Org.
Lett. 2000, 2, 2061. (b) Friedrich, K.; Bechtold, L. W.; Fritz, H. J. Pract.
Chem. 2000, 342, 819.
(17) Readily prepared from adiponitrile by a Thorpe-Ziegler cyclization:
Schaefer, J. P.; Bloomfield, J. J. Org. React. 1967, 15, 1.
(18) The combination of Ph3P, CCl4 (a) was far superior than the
previously described use of PCl5 or POCl3 for preparing 2-chlorocyclopen-
tenecarbonitrile (b). (a) Kemp, D. S.; Carter, J. S. J. Org. Chem. 1989, 54, 109.
(b) Cariou, M.; Lamont, M. C. R. Chim. (C) 1974, 278, 1457.
(19) Triethylamine was incorporated as a trace cosolvent during the
isolation of several related Claisen substrates hinting at a general instability
of these enol ethers.15b
(20) Use of DMF alone or THF/DMF mixtures did not improve the yield.
(21) Performing the reaction at reflux did not improve the overall yield.
(22) KHwithandwithout18-C-6yieldssimilarresults, butthelowercostand
easier manipulation of NaH focused subsequent optimization on the latter base.
(23) The2-alkoxyalkenenitriles were identified from 1H NMR spectra of the
crude reaction mixture and subjected to the Claisen rearrangement. Although
the crude 2-alkoxyalkenenitriles from conjugate addition-elimination could be
directly thermolyzed, in our experience the efficiency is improved through an
aqueous work up prior to thermolysis.
(24) Use of linalool as a prototypical tertiary alcohol did not provide any
enol ether even at elevated temperatures.
J. Org. Chem. Vol. 75, No. 21, 2010 7093

€
Gunes- et al.
JOCArticle
TABLE 1. Alkenenitrile Addition-Claisen Rearrangements
the secondary alcohol (Table 1, entry 9). Unfortunately
the steric compression that likely protects the acid sensi-
tive enol ether of 3i during chromatography also prevents
Claisen rearrangement.
Despite the Claisen rearrangement of 2-alkyoxyalkeneni-
triles being retarded relative to the parent alkene,15,25 the
thermal sigmatropic bond relocation effectively generates a
variety of quaternary oxonitriles (Table 1).15,26 The diastereo-
selectivity is excellent for the rearrangement of the 2-alkoxyalk-
enenitriles derived from acyclic secondary allylic alcohols, with
only the trans-alkene being observed (Table 1, entries 3-7 and
11). Thermolysis of 3h, bearing a cyclic cis-alkene as part of the
allyl ether, proceeds with the same efficiency but without any
diastereofacial selectivity (Table 1, entry 8). Extending the
strategy to six- and seven-membered alkenenitriles faithfully
installs quaternary centers in the corresponding six- and seven-
membered oxonitriles (Table 1, entries 10 and 11, respectively).
Diastereomers only arise in the rearrangement of 3h, which
most likely arises from the cyclic Z-alkene of the allylic ether
moiety. Typically Claisen rearrangements occur through a chair-
like transition structure,14 as observed in closely related cyclic
enol ethers,27 which strongly suggests that the 2-alkoxyalkeneni-
triles typically rearrange through chairlike TSs. However, the
chairlike transition structure emanating from 3h thrusts the
allylic proton H* directly into the face of the approaching
cyclohexene ring (3h0, Scheme 3).28 Bond migration from the
opposite face, 3h00, necessarily requires rearrangement through a
boat-like TS. Although steric compression between the two rings
is relieved in the exo arrangement 3h00, the boat TS causes the
nitrile to eclipse the cyclohexene methylene protons. The close
interplay between these two different sets of steric interactions in
3h0 and 3h00 is reflected in the absence of diastereoselectivity
(Table 1, entry 8) but suggests substitution patterns to improve
the selectivity (vide infra).
SCHEME 3. Conformations Leading to Diastereomeric
Oxonitriles
Excellent enantioselectivity is observed for addition-rear-
rangements withchiral, secondary, allylic alcohols(Table 2).29
(25) Burrows, C. J.; Carpenter, B. K. J. Am. Chem. Soc. 1981, 103, 6983.
(26) Orsini, F.; Pelizzoni, F.; Forte, M. Gazz. Chim. Ital. 1986, 116, 115.
(27) Mikami, K.; Takahashi, K.; Nakai, T.; Uchimaru, T. J. Am. Chem.
Soc. 1994, 116, 10948.
(28) For an excellent overview and computational analysis of chair and
boat TSs in this type of system, see: (a) Gul, S.; Schoenebeck, F.; Aviyente,
V.; Houk, K. N. J. Org. Chem. 2010, 75, 2115. (b) Khaledy, M. M.; Kalani,
M. Y. S.; Khuong, K. S.; Houk, K. N.; Aviyente, V.; Neier, R.; Soldermann,
N.; Velker, J. J. Org. Chem. 2003, 68, 572.
(29) An analogous rearrangement with chiral 3k was successful but the
enantiomeric ratio was unable to be determined using a variety of HPLC and
GC methods.
aIsolated as a single isomer unless otherwise indicated. bIsolated as a
1:1 mixture of diastereomers. cYield of alkenenitrile.
7094 J. Org. Chem. Vol. 75, No. 21, 2010

€
Gunes- et al.
JOCArticle
Of the two more favorable
30
TABLE 2. Enantioselective Oxonitrile Syntheses
conformations 300 and 30000
.
conformers 30 and 3000, the primary difference lies in 30 progres-
sing through a boat-like TS, whereas 3000 progresses through a
chairlike TS. For the acyclic allylic alcohols the corresponding
2-alkoxyalkenenitriles exhibit a high preference for rearranging
through chairlike transition structures corresponding to 3000
(Table 2, entries 1-4).27
Thermolysis of the nitrile derived from 2n, in which the
alkene bears a methoxy substituent, is less selective. Rear-
rangement via 3000 engenders a diaxial-type interaction,
between R2 and the nitrile, that is avoided in the boat TS 30.
Despite the steric compression, the enantiomeric ratio remains
relatively high at 90:10 (Table 2, entry 4).
Claisen rearrangement of the 2-alkoxyalkenenitriles de-
rived from the cyclic allylic alcohols 2h, 2o, and 2p can only
proceed through the two transition structures 300 and/or 30000
because of the Z-geometry of the allylic ether moiety. For the
parent nitrile (3h, Table 1, entry 8) the transition structures
must be of comparable energy because no selectivity is
observed. Installing a bromine substituent proximal to the
alcohol is anticipated to strongly destabilize conformer 300
(R2 = Br) because of the steric compression between the
bromine and nitrile groups and the alignment of their di-
poles. Rearrangement via 30000 places the dipoles in opposite
directions and minimizes the steric interaction because the
cyclopentene ring moves the ring methylene away from a true
diaxial interaction with the proximal bromine substituent.
Crystallization of 4o allowed unequivocal determination of
the absolute and relative configuration with the assignment
being consistent with rearrangement via a boat-like transi-
tion structure.31 Although carveol (2p) might be expected to
exhibit a similar preference for bond translocation through a
boat-like transition structure 30000, the isopropyl substituent
projects into the cyclopentene ring in 30000, which is avoided in
the chairlike TS 300.28 The consequence of the two competing
effects is a lack of selectivity (Table 2, entry 6).
aThe enantiomeric ratio was determined on the alcohol (i) obtained
by cleaving the silyl ether. bThe enantiomeric ratio was determined on
the ketone (ii) obtained by hydrolysis of the enol ether.
Allylic alcohols 2f, 2l, and 2m lacking olefinic substituents
faithfully translate the carbinol stereochemistry with a mini-
mum enantiomeric ratio of 93:7 (Table 2, entries 1-3). Sub-
stitution on the olefin slightly diminishes the ratio to 90:10 for
2n, whereas with the cyclic, disubstituted allylic alcohol 2o
only a single enantiomer is detected (Table 2, entries 4 and 5,
respectively). Introducing an additional stereocenter in the
cyclohexenol ring, for 2p, leads to a complete loss of stereo-
selectivity (Table 2, entry 6).
SCHEME 5. Divergent Addition-Eliminations
Effective translation of the allylic alcohol chirality to the
quaternary center of the oxonitrile requires excellent discrimi-
nation between chair and boat transition structures (Scheme 4).
SCHEME 4. Claisen-Oxonitrile Alkylation Strategy
Investigating the structural diversity of the allylic alcohol
identified two limitations of the addition-rearrangement
strategy (Scheme 5). Attempts to condense the hydroxyni-
trile 2q32 with chloroalkenenitrile 5a were not successful,
possibly because of the lability of these “cyanoaldolates”
Comparing the four possible transition structures reveals severe
allylic strain between the two alkyl substituents R1 and R3 in
(31) Crystallographic details are provided in the Supporting Information.
(32) Prepared from (S)-3-buten-1,2-diol as described in the Supporting
Information.
(30) Hoffmann, R. W. Chem. Rev. 1989, 89, 1841.
J. Org. Chem. Vol. 75, No. 21, 2010 7095

€
Gunes- et al.
JOCArticle
toward retroaldol fragmentation under basic conditions.33 An
analogous condensation with the azide 2r afforded the diene 6,
presumably via elimination of hydroazoic acid from 3r. The
elimination appears delicately poised because the silyl ether-
containing allylic alcohol 2l (Table 2, entry 2) participates in the
rearrangement without elimination.
Unexpectedly (R)-1-phenylprop-2-en-1-ol (2s) affords the
bis-enol ether 8. Mechanistically 8 likely arises through an
initial conjugate addition-elimination to chloroalkeneni-
trile 5a to afford 3s. Presumably excess sodium alkoxide,
whose basicity is enhanced through the addition of 15-
crown-5, deprotonates 3s to afford the delocalized anion
7. Reprotonation affords the conjugated olefin 8 with the
methyl eclipsing the smaller ether oxygen rather than the
larger phenyl group.34
2-Chlorocyclopent-1-enecarbonitrile (5a). The general proce-
dure was employed with 2.00 g (18.34 mmol) of 1a to afford
1.72 g (73%) of 5a: IR (neat) 2220, 1622, 1434 cm-1; 1H NMR
(200 MHz, CDCl3) δ 2.76-2.60 (m, 4H), 2.17-2.01 (m, 2H);
13C NMR (50 MHz, CDCl3) δ 152.0, 116.5, 112.7, 40.5, 35.2,
23.8.
2-Chlorocyclohex-1-enecarbonitrile (5b). The general proce-
dure was employed with 1.50 g (12.18 mmol) of 1b to afford 1.41
g (82%) of 5b: IR (neat) 2217, 1631, 1450 cm-1; 1H NMR (400
MHz, CDCl3) δ 2.50-2.45 (m, 2H), 2.37-2.33 (m, 2H),
1.81-1.75 (m, 2H), 1.73-1.66 (m, 2H); 13C NMR (100 MHz,
CDCl3) δ 147.5, 116.9, 109.9, 33.8, 28.4, 22.5, 20.8.
2-Chlorocyclohept-1-enecarbonitrile (5c). The general proce-
dure was employed with 2.10 g (15.30 mmol) of 1c to afford
1.81 g (76%) of 5c: IR (neat) 2216, 1619, 1447 cm-1; 1H NMR
(400 MHz, CDCl3) δ 2.73-2.70 (m, 2H), 2.45-2.42 (m,
2H), 1.77-1.74 (m, 2H), 1.67-1.62 (m, 4H); 13C NMR (100
MHz, CDCl3) δ 152.5, 118.3, 114.3, 39.7, 31.6, 30.8, 25.8, 24.7;
HRMS (ESI) [M þ Hþ] calcd for 156.0580 (C8H11ClN), found
156.0577.
General Conjugate Addition-Claisen Rearrangement Procedure.
At rt, THF solutions of the allylic alcohol and neat 15-crown-5
were added sequentially to a suspension of NaH in THF or hexa-
nes. After 3 h, a THF or hexanes solution of 5 was added, and after
40 h H2O was added. The phases were separated, and the aqueous
phase was then extracted with EtOAc. The combined organic
phase was washed with brine, dried over Na2SO4 or MgSO4,
filtered, and then concentrated in vacuo. The crude 2-alkoxy-
cycloalkenenitrile was dissolved in toluene and was then heated
in a sealed tube at 110 °C for 20 h. After removal of the solvent in
vacuo, the residue was purified by radial chromatography or
column chromatography (EtOAc/Hexanes) to afford the pure
oxonitrile.
1-Allyl-2-oxocyclopentanecarbonitrile (4a). The general pro-
cedure was employed in THF with 200.0 mg (1.57 mmol) of 5a to
give 152 mg (65%) of 4a:35 IR (neat) 2239, 1755 cm-1; 1H NMR
(400 MHz, CDCl3) δ 5.85-5.74 (m, 1H), 5.26 (d, J = 9.0 Hz,
1H), 5.22 (d, J = 16.8 Hz, 1H), 2.61 (dd, J = 13.9, 7.0 Hz, 1H),
2.54-1.96 (m, 7H); 13C NMR (100 MHz, CDCl3) δ 209.1, 130.9,
121.2, 119.0, 48.5, 38.1, 36.8, 33.7, 19.4; HRMS (ESI) [M þ Hþ]
calcd for 150.0919 (C9H12NO), found 150.0913.
1-(2-Methylallyl)-2-oxocyclopentanecarbonitrile (4b). The gen-
eral procedure was employed in THF with 500.0 mg (3.92 mmol)
of 5a to give 463 mg (75%) of 4b: IR (neat) 2233, 1753 cm-1; 1H
NMR (400 MHz, CDCl3) δ 4.98 (s, 1H), 4.85 (s, 1H), 2.60 (d, J =
14.4 Hz, 1H), 2.53-2.31 (m, 3H), 2.21 (d, J = 14.4 Hz, 1H),
2.13-1.98 (m, 3H), 1.84 (s, 3H); 13C NMR (100 MHz, CDCl3) δ
209.3, 139.5, 119.5, 116.9, 47.9, 41.20, 36.4, 33.9 23.4, 19.4;
HRMS (ESI) [M þ Hþ] calcd for 164.1075 (C10H14NO), found
164.1070.
Conclusion
Installing quaternary centers in a diverse series of cyclic five-,
six-, and seven-membered oxonitriles is readily accomplished
by translating the stereochemistry of readily available chiral
alcohols via a Claisen rearrangement. Key tothis strategy is the
union of secondary allylic alkoxides to 2-chlorocycloalkene-
carbonitriles through a conjugate addition-elimination se-
quence. Subsequent thermolysis of the 2-alkoxyalkenenitriles
selectively installs a new quaternary center with enantiomeric
ratios typically greater than 9:1.
The strategy readily generates highly congested quaternary
centers with substitution patterns that complement those avail-
able by asymmetric allylic alkylation methods. For example,
transition metal coupling is ideal for the parent allylation,
whereas the Claisen chirality transfer is excellent for the union
of two cyclic substrates that forge a hindered quaternary-
tertiary array. The sigmatropic rearrangement precludes for-
mation of olefin regioisomers and always installs a trans-alkene
unless constrained within a medium-sized ring.
Experimentally the reaction is a two-step operation in
which an allylic alkoxide is condensed with a cyclic 2-chloro-
alkenecarbonitrile to afford a 2-alkoxyalkenenitrile that is
directly thermolyzed. Diverse primary and secondary, cyclic
and acyclic, allylic alcohols readily engage in the addition-
rearrangement to give good yields of quaternary oxonitriles.
Collectively the range of substituted allylic alcohols provide
mechanistic insight and define the substitution patterns required
for highly stereoselective rearrangements. The overall alkylation
strategy addresses the challenge of enantioselectively installing
hindered quaternary centers in cyclic hydroxynitriles while
simultaneously installing orthogonal ketone, nitrile, and
olefin functional groups.
(E)-1-(But-2-enyl)-2-oxocyclopentanecarbonitrile (4c). The gen-
eral procedure was employed in THF with 200.0 mg (1.57
mmol) of 5a to give 140 mg (55%) of 4c: IR (neat) 2235, 1755
1
cm-1; H NMR (400 MHz, CDCl3) δ 5.64 (dq, J = 15.2, 4.0
Hz, 1H), 5.40 (dt, J = 15.2, 6.4 Hz, 1 H), 2.45 (dd, J = 14.0, 6.8
Hz, 1H), 2.46-2.27 (m, 3H), 2.25-2.19 (dd, J = 14.0, 7.6 Hz,
1H), 2.12-1.91 (m, 3H), 1.6 (d, J = 6.4 Hz, 3H); 13C NMR (100
MHz, CDCl3) δ 209.5, 132.2, 123.3, 119.3, 48.9, 37.0, 36.8,
33.5, 19.4, 18.2; HRMS (ESI) [M þ Hþ] calcd for 164.1075
(C10H14NO), found 164.1070.
Experimental Section
General Procedure for 2-Chlorocycloalkene-1-carbonitriles
(5). Solid PPh3 was added to a CH2Cl2 solution of the cyclic
oxonitrile and CCl4. The resulting reaction mixture was refluxed
for 30 h or heated in a sealed tube for 48 h, the solvent was
removed, and then the crude product was purified by silica gel
column chromatography (1:4 EtOAc/hexanes) to afford the
pure 2-chlorocycloalkene-1-carbonitriles.
(E)-2-Oxo-1-(pent-2-enyl)cyclopentanecarbonitrile (4d). The
general procedure was employed in THF with 200.0 mg (1.57
mmol) of 5a to give 171 mg (62%) of 4d: IR (neat) 2235, 1770
cm-1 1H NMR (200 MHz, CDCl3) δ 5.75-5.16 (m, 1H),
;
(33) (a) Carlier, P. R.; Lo, C. W.-S.; Lo, M. M.-C.; Wan, N. C.; Williams,
I. D. Org. Lett. 2000, 2, 2443. (b) Carlier, P. R.; Lam, W. W.-F.; Wan, N. C.;
Williams, I. D. Angew. Chem. 1998, 37, 2252.
5.46-5.31 (m, 1H), 2.60-1.94 (m, 10H), 0.99 (t, J = 7.5 Hz,
(34) For an excellent compilation of steric compression based on A values
see ref 8.
(35) Spectrally identical to material previously characterized: Levy, L. M.;
Gotor, V. J. Org. Chem. 2004, 69, 2601.
7096 J. Org. Chem. Vol. 75, No. 21, 2010

€
Gunes- et al.
JOCArticle
3H); 13C NMR (50 MHz, CDCl3) δ 211.1, 141.0, 122.9, 121.0,
50.7, 38.8, 38.6, 35.33, 27.6, 21.1, 15.5; HRMS (ESI) [M þ Hþ]
calcd for 178.1232 (C11H16NO), found 178.1226.
5.2 Hz, 1H), 2.61-2.53 (m, 2H), 2.35 (dd, J = 13.9, 7.9 Hz, 1H),
2.15 (m, 1H), 1.96-1.87 (m, 3H), 1.78-1.58 (m, 2H), 1.38-1.20
(m, 8H), 0.87 (t, J = 7.0 Hz, 3H); 13C NMR (100 MHz, CDCl3) δ
205.3, 137.7, 121.1, 120.4, 55.3, 41.2, 40.2, 35.3, 32.7 31.5, 29.1,
29.0, 26.0, 24.7, 22.7, 14.2; HRMS (ESI) [M þ Hþ] calcd for
248.2014 (C16H26NO), found 248.2010.
(E)-1-(Hept-2-enyl)-2-oxocyclopentanecarbonitrile (4e). The
general procedure was employed in THF with 100.0 mg (0.78
mmol) of 5a to give 107 mg (65%) of 4e: IR (neat) 2235, 1753
cm-1; 1H NMR (400 MHz, CDCl3) δ 5.71 (dtt, J = 15.4, 6.6, 1.5
Hz, 1H), 5.38 (dtt, J = 15.4, 6.6, 1.5 Hz, 1H), 2.54 (m, 10H),
1.40-1.24 (m, 4H), 0.87 (t, J = 7.1 Hz, 3H); 13C NMR (100
MHz, CDCl3) δ 209.5, 137.8, 122.0, 119.3, 48.9, 37.0, 36.8, 33.5,
32.4, 31.5, 22.4, 19.4, 14.1; HRMS (ESI) [M þ Hþ] calcd for
206.1545 (C13H20NO), found 206.1539.
(þ)-(E)-1-(4-Hydroxybut-2-enyl)-2-oxocyclopentanecarbonitrile
(4l). The general procedure was employed in THF with 72.0 mg
(0.56 mmol) of 5a to give 133 mg (55%) of 4l. The enantiomeric
ratio was determined on the alcohol derived by deprotecting 4l
(260 mg, 0.62 mmol) with Bu4NF, which afforded alcohol i (39.0
mg, 35%). 93:7 er was determined by HPLC (Daicel OB-H, 10%
isopropanol/hexanes, 0.9 mL/min 207 nm, tR (major) = 40.9 min,
(þ)-(E)-1-(Oct-2-enyl)-2-oxocyclopentanecarbonitrile (4f). The
general procedure was employed in THF with 200.0 mg (1.57
mmol) of 5a to give 192 mg (65%) of 4f: 95:5 er was determined by
HPLC (Daicel AS, 5% isopropanol/hexanes, 1.0 mL/min 295 nm,
tR (major) = 9.7 min, tR (minor) = 8.6 min). [R]25D = þ30.15 (c
tR (minor) = 32.9 min). [R]20 = þ16.77 (c 1.55, CHCl3); IR
D
1
(neat) 3411, 2238, 1751 cm-1; H NMR (200 MHz, CDCl3) δ
5.82-5.51 (m, 2H), 4.04 (dd, J = 4.9, 1.1 Hz, 2H), 2.66-1.88 (m,
8H); 13C NMR (50 MHz, CDCl3) δ210.9, 137.7, 125.2, 120.8, 64.4,
50.5, 38.4, 35.5, 21.0; HRMS (ESI) [M þ Hþ] calcd for 180.1025
(C10H14NO2), found 180.1019.
1.26, ethyl acetate); IR (neat) 2236, 1755 cm-1; H NMR (400
1
MHz, CDCl3) δ 5.63 (dt, J = 15.2, 7.3 Hz, 1H), 5.38 (dt, J = 15.2,
7.4 Hz, 1H), 2.55 (dd, J = 14.2, 7.2 Hz, 1H), 2.50-2.29 (m, 3H),
2.25 (dd, J = 14.2, 8.0 Hz, 1H), 2.16-1.93 (m, 5H), 1.40-1.20 (m,
6H), 0.88 (t, J = 7.0 Hz, 3H); 13C NMR (100 MHz, CDCl3) δ
209.4, 137.8, 122.0, 119.2, 49.0, 37.1, 37.0, 33.5, 32.7, 31.5, 29.0,
22.7, 19.4, 14.3; HRMS (ESI) [M þ Hþ] calcd for 220.1701
(C14H22NO), found 220.1696.
(E)-2-Oxo-1-(5-phenylpent-2-enyl)cyclopentanecarbonitrile (4g).
The general procedure was employed in THF with 200.0 mg (1.57
mmol) of 5a to give 119 mg (60%) of 4g: IR (neat) 2234, 1753
cm-1; 1H NMR (200 MHz, CDCl3) δ 7.33-7.16 (m, 5H), 5.65 (dt,
J = 15.3, 6.3 Hz, 1H), 5.45-5.31 (dt, J = 15.3, 7.3 Hz, 1H), 2.72 (t,
J = 7.5 Hz, 2H) 2.60-1.87 (m, 10H); 13C NMR (50 MHz, CDCl3)
δ 210.0, 142.4, 137.3, 129.5, 129.3, 126.9, 123.8, 120.0, 49.5, 37.7,
37.6, 36.4, 35.1, 34.1, 20.1; HRMS (ESI) [M þ Hþ] calcd for
254.1545 (C17H20NO), found 254.1539.
1-(Cyclohex-2-enyl)-2-oxocyclopentanecarbonitriles (4h). The
general procedure was employed in THF with 200.0 mg (1.57
mmol) of 5a to give 350 mg (60%) of 4h. Careful chromatogra-
phy gave two pure fractions of the first and second eluting
isomers (85 mg and 45 mg, respectively). First isomer: light
yellow oily liquid, Rf = 0.3 (1:9 EtOAc/hexanes). IR (neat)
2233, 1755 cm-1; 1H NMR (400 MHz, CDCl3) δ 5.91-5.88 (m,
1H), 5.13 (d, J = 10.4 Hz, 1H), 2.87-2.4 (m, 1H), 2.59-2.52 (m,
1H), 2.37-1.97 (m, 8H), 1.87-1.82 (m, 1H), 1.60-1.39 (m, 2H);
13C NMR (100 MHz, CDCl3) δ 209.9, 133.0, 124.3, 119.0, 52.9,
39.5, 38.1, 31.1, 26.0, 25.1, 21.6, 19.8. Second isomer: colorless
liquid, Rf = 0.23 (1:9 EtOAc/hexanes). IR (neat) 2234, 1750
cm-1; 1H NMR (400 MHz, CDCl3) δ 5.97-5.92 (m, 1H), 5.81
(d, J = 12.0 Hz, 1H), 2.70-2.65 (m, 1H), 2.55-2.49 (m, 1H),
2.37-2.25 (m, 2H), 2.18-1.98 (m, 5H), 1.89-1.72 (m, 2H),
1.58-1.47 (m, 1H), 1.37-1.23 (m, 1H); 13C NMR (100 MHz,
CDCl3) δ 209.4, 131.8, 125.2, 119.1, 52.4, 39.4, 37.7, 31.6, 24.9,
24.7, 22.8, 19.5; HRMS (ESI) [M þ Naþ] calcd for 212.1051
(C12H15NaNO), found 212.1046.
(-)-(E)-2-Oxo-1-(4-phenylbut-2-enyl)cyclopentanecarbonitrile
(4m). The general procedure was employed in THF with 167.0
mg (1.30 mmol) of 5a to give 151 mg (48%) of 4m: 95:5 er was
determined by HPLC (Daicel OB-H, 20% isopropanol/hexanes,
1.0 mL/min 210 nm, tR (major) = 17.8 min, tR (minor) = 22.4
min); [R]20 = -10.4 (c 0.58; EtOAc); IR (neat) 2236, 1751
D
cm-1 1H NMR (400 MHz, CDCl3) δ 7.33-7.29 (m, 2H),
;
7.23-7.17 (m, 3H), 5.81 (dtt, J = 15.0, 6.8, 1.1 Hz, 1H), 5.53
(dddt, J = 15.0, 8.1, 6.8, 1.4 Hz, 1H), 3.41 (d, J = 6.8 Hz, 2H),
2.62-2.26 (m, 5H), 2.14-1.93 (m, 3H); 13C NMR (100 MHz,
CDCl3) δ 209.4, 140.0, 136.0, 128.8, 128.7, 126.5, 123.8, 119.2,
48.8, 39.2, 37.0, 36.8, 33.7, 19.4; HRMS (ESI) [M þ Hþ] calcd for
240.1388 (C16H18NO), found 240.1383.
(þ)-2-Oxo-1-(2-oxo-4-phenylbutyl)cyclopentanecarbonitrile (4n).
The general procedure was employed in THF with 210.0 mg (1.64
mmol) of 5a to give 350 mg (72%) of 4n. The enantiomeric ratio was
determined on the corresponding ketone obtained by hydrolysis
with 2 M HCl which afforded quantitatively ii (301.0 mg). 90:10 er
was determined by HPLC (Daicel OC, 10% isopropanol/hexanes,
1.0 mL/min 206 nm, tR (major) = 72.6 min, tR (minor) = 66.2
min). [R]20D = þ2.8 (c 2.86; EtOAc); IR (neat) 2230, 1755, 1716
cm-1; 1H NMR (400 MHz, CDCl3) δ 7.30-7.13 (m, 5H), 3.09 (s,
2H), 2.87 (t, J = 7.5 Hz, 2H), 2.74 (t, J = 7.5 Hz, 2H), 2.65-2.40
(m, 3H), 2.17-2.04 (m, 3H); 13CNMR (100 MHz, CDCl3) δ208.8,
204.8, 140.3, 128.9, 128.5, 126.6, 118.6, 46.2, 45.1, 44.1, 36.7, 33.7,
29.8, 20.1; HRMS (ESI) [M þ Hþ] calcd for 256.1338 (C16H18-
NO2), found 256.1332.
(S)-(-)-1-((S)-2-Bromocyclohex-2-enyl)-2-oxocyclopentane-
carbonitrile (4o). The general procedure was employed in THF
with 69.0 mg (0.54 mmol) of 5a to give 79 mg (55%) of 4o:
[R]20 = -124 (c 1.00; CHCl3); IR (KBr) 2229, 1745 cm-1
;
D
1H NMR (400 MHz, CDCl3) δ 6.34-6.32 (m, 1H), 3.22-
3.16 (m, 1H), 2.63-2.52 (m, 2H), 2.32 (m, 1H), 2.20-2.02
(m, 5H), 1.93 (m, 1H), 1.80 (m, 1H), 1.56 (m, 1H), 1.37 (m,
1H); 13C NMR (100 MHz, CDCl3) δ 208.7, 136.3, 120.7,
119.6, 52.2, 45.3, 38.6, 31.5, 27.7, 27.2, 20.4,19.6; HRMS
(ESI) [M þ Hþ] calcd for 268.0337 (C12H15BrNO), found
268.0332.
1-(2-Methylallyl)-2-oxocyclohexanecarbonitrile (4j). The gen-
eral procedure was employed in hexanes with 100.0 mg (0.71
mmol) of 5b to give 88 mg (70%) of 4j: IR (neat) 2218, 1730
1
cm-1; H NMR (400 MHz, CDCl3) δ 4.98 (m, 1H), 4.85 (m,
1H), 2.84 (ddd, J = 13.7, 11.1, 5.7 Hz, 1H), 2.75 (d, J = 14.3
Hz, 1H), 2.49 (m, 1H), 2.38-2.32 (m, 2H), 2.13-1.98 (m, 2H),
1.93-1.66 (m, 6H); 13C NMR (100 MHz, CDCl3) δ 202.9,
139.3, 120.2, 116.6, 50.3, 40.9, 39.0, 38.1, 28.4, 23.5, 21.9;
HRMS (ESI) [M þ Hþ] calcd for 178.1232 (C11H16NO), found
178.1220.
(þ)-(E)-1-(Oct-2-enyl)-2-oxocycloheptanecarbonitrile (4k). The
general procedure was employed in hexanes with 185.0 mg (1.19
mmol) of 5c to give 191 mg (65%) of 4k: [R]20D = þ13 (c 1.00;
CHCl3); IR (neat) 2239, 1715 cm-1; 1H NMR (400 MHz, CDCl3)
δ 5.60 (dt, J = 14.9, 6.8 Hz, 1H), 5.37 (m, 1H), 2.69 (dt, J = 12.5,
(þ)-1-((5S)-2-Methyl-5-(prop-1-en-2-yl)cyclohex-2-enyl)-2-
oxocyclopentanecarbonitrile (4p). The general procedure was
employed in THF with 160.0 mg (1.26 mmol) of 5a to give 224
mg (73%) of 4p. The diastereomers were carefully separated to
obtain pure first and second eluting isomers for characteriza-
tion. First isomer: 1H NMR (400 MHz, CDCl3) δ 5.68 (m, 1H),
4.75 (m, 1H), 4.70 (m, 1H), 3.08 (m, 1H), 2.66-1.80 (m, 10H),
1.74 (s, 3H), 1.40 (s, 3H). Second isomer: IR (neat) 2229, 1755
1
cm-1; H NMR (400 MHz, CDCl3) δ 5.68 (m, 1H), 4.76 (s,
1H), 4.67 (s, 1H), 2.67 (m, 1H), 2.55 (m, 1H), 2.39-2.00 (m,
11H), 1.86 (s, 3H), 1.70 (s, 3H); 13C NMR (100 MHz, CDCl3) δ
J. Org. Chem. Vol. 75, No. 21, 2010 7097

€
Gunes- et al.
JOCArticle
210.1, 209.9, 148.8, 132.0, 130.7, 130.2, 130.1, 128.9, 120.9,
118.8, 109.5, 109.4, 54.2, 50.5, 44.2, 42.5, 41.5, 40.8, 38.8, 36.9,
34.0, 31.2, 31.0, 30.5, 29.8, 29.7, 24.2, 23.2, 21.1, 21.0, 19.8,
19.2; HRMS (ESI) [M þ Hþ] calcd for 244.1701 (C16H22NO),
found 244.1696.
and mass spectral analyses from Dr. A. C. Goren and G. Bilsel
(TUBITAK-UME, Turkey) are gratefully acknowledged.
Supporting Information Available: Experimental procedures,
analytical data for all new compounds, and a CIF file for 4o.
This material is available free of charge via the Internet at
http://pubs.acs.org.
Acknowledgment. Financial support for this research from
TUBITAK (106T100) and Ataturk University (BAP-2005-230)
7098 J. Org. Chem. Vol. 75, No. 21, 2010