Dauben-Michno oxidative transposition of allylic cyanohydrins - Enantiomeric switch of (-)-carvone to (+)-carvone

Article abstract of DOI:10.1139/v11-026

Allylic cyanohydrins were subjected to Dauben-Michno oxidation at low temperatures to provide β-cyanoenones in good to excellent yields. The potential of this oxidative transposition as a means of an enantiomeric switch of enones containing a latent plane of symmetry was tested by conversion of (-)-carvone to its enantiomer.

Full text of DOI:10.1139/v11-026

535
AWARD LECTURE / CONFÉRENCE D'HONNEUR
Dauben–Michno oxidative transposition of allylic
cyanohydrins — Enantiomeric switch of
(–)-carvone to (+)-carvone*
Jason R. Hudlicky, Lukas Werner, Vladislav Semak, Razvan Simionescu, and
Tomas Hudlicky
Abstract: Allylic cyanohydrins were subjected to Dauben–Michno oxidation at low temperatures to provide b-cyanoenones
in good to excellent yields. The potential of this oxidative transposition as a means of an enantiomeric switch of enones
containing a latent plane of symmetry was tested by conversion of (–)-carvone to its enantiomer.
Key words: Dauben–Michno reaction, oxidative 1,3-transposition of allylic alcohols, cyanoenones, enantiomeric switch of
carvone.
Résumé : On a soumis des cyanohydrines allyliques à une oxydation de Dauben–Michno, à basse température, pour obtenir
des b-cyanoénones avec des rendements allant de bons à excellents. Dans le but d’évaluer le potentiel de cette transposition
oxydante comme méthode de réaliser une transposition énantiomère dans les énones contenant un plan de symétrie latent,
on a transformer la (–)-carvone en son énantiomère.
Mots‐clés : réaction de Dauben–Michno, transposition oxydante 1,3 d’alcools allyliques, cyanoénones, transposition énantio-
mère de la carvone.
[Traduit par la Rédaction]
such oxidation was the conversion of allylic alcohol 1 to
enone 2 during our recently completed synthesis of oseltami-
vir (3) (Tamiflu), Fig. 1.5
To establish whether the reaction is general for allylic alco-
hols substituted with other electron-withdrawing groups we
investigated a series of cyanohydrins as means of synthesis
of b-cyanoenones. In this paper we report the results of this
investigation, along with a procedure for an enantiomeric
switch of (–)-carvone to (+)-carvone.
Introduction
The Dauben–Michno oxidation constitutes a reliable proto-
col for oxidative transposition of tertiary allylic alcohols sub-
stituted with electron-donating groups.1,2 The reaction
proceeds by a [3,3] sigmatropic rearrangement of the initially
formed chromate ester followed by further oxidation of the
rearranged chromate ester to the final enone. The process is
analogous to the well-known chromium(VI) oxidations of
olefins to enones that proceed via an ene reaction of the
chromium oxide followed by oxidation. The initially formed
chromate ester is either oxidized to an enone or undergoes
[3,3] sigmatropic rearrangement prior to the final oxidation.
Thus, mixtures of regioisomeric enones are frequently en-
countered in the oxidation of cyclic olefins.3,4 Because only
one product can arise from the rearrangement of a tertiary al-
lylic alcohol, the Dauben–Michno reaction does not suffer
from regiochemical issues. Although well-documented for al-
lylic alcohols flanked by electron-donating groups, this reac-
tion has not been reported for allylic alcohols substituted
with esters or nitriles. To our knowledge the first example of
Results and discussion
We began the investigation with a series of cyclic enones
(Table 1, entries 1–3) and their conversion to cyanohydrins
following a recently published procedure that uses proline
catalysis for the formation of the intermediate trimethylsilyl
ethers of cyanohydrins.6 Following the hydrolysis, the free
cyanohydrins were obtained in essentially quantitative yield
and used without purification in the Dauben–Michno oxida-
tion. The proline-catalyzed protocol is exceedingly efficient
and cleanly affords the cyanohydrins, thus obviating any
Received 13 November 2010. Accepted 14 January 2011. Published at www.nrcresearchpress.com/cjc on 21 April 2011.
J.R. Hudlicky, L. Werner, V. Semak, R. Simionescu, and T. Hudlicky. Department of Chemistry and Centre for Biotechnology, Brock
University, 500 Glenridge Avenue, St. Catharines, ON L2S 3A1, Canada.
Corresponding author: T. Hudlicky (e-mail: thudlicky@brocku.ca).
*Based on the 2010 Bader Award Lecture.
Can. J. Chem. 89: 535–543 (2011)
doi:10.1139/V11-026
Published by NRC Research Press

536
Can. J. Chem. Vol. 89, 2011
Fig. 1. Dauben–Michno oxidation of a-hydroxybutenoate.
Fig. 2. Proposed enantiomeric switch of (–)-carvone via Dauben–Michno oxidation.
separation issues prior to subsequent reactions. In the case of
cyclopentenone, the cyanohydrin was obtained in 60% yield
at the expense of the competing 1,4-addition of the cyanide
to the enone. Similarly, cyanohydrins were obtained from
methylvinyl ketone (Table 1, entry 4), 3-penten-2-one (Ta-
ble 1, entry 5), 4-phenylbutenone (Table 1, entry 6), aceto-
phenone (Table 1, entry 7), and (–)-carvone (Table 1, entry
8).
the oxidant were used. We did not detect the corresponding
acid even after careful repetition of the experiments; however,
over-oxidation of an aldehyde to an acid requires the forma-
tion of a hydrate, which cannot take place under the condi-
tions of the reaction. If the corresponding acid is desired, a
simple dilution of the reaction with water and addition of an
additional equivalent of chromium oxide should provide the
corresponding carboxylic acid in high yield.
With pent-3-en-2-one (16) the corresponding cyanohydrin
17 was isolated in 76% yield and rearranged cleanly to the
b-cyanoenone 18 in 84% yield (66% over two steps); only the
Z-isomer 18a was observed. In the case of 4-phenylbutenone
(19), the rearrangement of the corresponding cyanohydrin 20
produced only traces of the transposed enone 21, and the ma-
jor product isolated was benzaldehyde (22). The formation of
benzaldehyde may be rationalized by the oxidative cleavage
of the intermediate 1,2-chromate diester formed from the
styrene double bond in competition with the Dauben–Michno
reaction.
For acetophenone (23) and the oxidation of its cyanohy-
drin 24, only the formation of an intractable polymer was ob-
served at prolonged reaction times or at higher temperatures.
The cyanohydrin derived from acetophenone was still present
in the reaction mixture after 16 h at room temperature. We
expected the Dauben–Michno to proceed with disruption of
aromaticity to provide 25, in analogy with reported cases of
reactivity of this type;7 however, attempts at trapping inter-
mediates of type 25 with external dienophiles such as maleic
anhydride proved unsuccessful.
The Dauben–Michno procedure required the treatment of
CrO3 with acetic anhydride at 80 °C before cooling and addi-
tion of the cyanohydrin at or below 0 °C. The best results
were obtained by performing the oxidation below 0 °C to
avoid competing reactions at the site of the double bond that
may result from the generation of peroxyacetic acid during
the reaction. Such competition is absent when the oxidation
is performed at –40 °C or below. The yields of transposed
cyanoenones were in the range of 70%–80% except in the
case of cyclopentenone, where the conversion was excellent
and essentially quantitative (vide NMR); however, the volatil-
ity of the product contributed to low isolated yields (Table 1).
With acyclic enones the results were different; in the case
of methylvinyl ketone the corresponding unsaturated alde-
hyde 15 (a consequence of the oxidation of a primary chro-
mate ester formed by the sigmatropic rearrangement) was
isolated as a mixture of Z- and E-isomers in the ratio of 7:1.
(Note: On repetition of these experiments at higher tempera-
tures or on a larger scale only the Z-isomer was detected.)
The volatility of the product made the accurate determination
of yield difficult. It was somewhat surprising that little or no
over-oxidation took place whether 1 equiv or up to 3 equiv of
The oxidative 1,3-transposition could, in principle, be used
Published by NRC Research Press

Hudlicky et al.
537
Table 1. Dauben–Michno oxidation of cyanohydrins.
Entry Enone Cyanohydrin
Product(s) (%)
aIsolated yield over two steps.
bYield of 1,2-adduct, 1,4-adduct of cyanide was observed during the cyanohydrin synthesis.
cIsolated yield, product is volatile and losses occur on evaporation.
dProduct content as estimated by NMR.
eFormed only in reactions above 0 °C.
to provide enantiomers of homochiral enones that contain a
latent plane of symmetry, as shown in Fig. 2 with carvone.
To test this idea we applied the Dauben–Michno protocol to
carvone with the expectation that further manipulations of the
b-cyanoenone 28 would lead to the formation of ent-carvone,
as depicted in Fig. 2.
The Dauben–Michno oxidation of cyanohydrin 27 derived
from carvone provided the corresponding b-cyanoenone 28 in
75% yield. At 0–4 °C, considerable amount of side products,
such as the methyl ketone 29 and epoxide 30, were isolated.
The former compound is likely a result of the oxidative
cleavage of the 1,2-diol, formed by the addition of CrO3 to
the isopropylidine double bond, whereas the latter is prob-
ably formed by epoxidation mediated by peroxyacetic acid
generated by in situ oxidation of acetic acid. When the reac-
tion was performed at –40 to –50 °C, only oxidative rear-
rangement took place, and no side products arising from
over-oxidation were observed.
Published by NRC Research Press

538
Can. J. Chem. Vol. 89, 2011
Fig. 3. Synthesis of ent-dihydrocarvone from enone 28.
Fig. 4. Enantiomeric switch of (–)-carvone.
Fig. 5. Djerassi’s assignment of diastereomeric dihydrocyanocarvone 31.
The proof of principle for the enantiomeric switch was
demonstrated on a small scale by the conversion of cyano-
enone 28 to the fully saturated b-cyano ketone 32 as shown
in Fig. 3. Hydrogenation of 28 at 1 atm (1 atm =
101.325 kPa) produced ketone 32 (34% yield), whose struc-
ture was assigned by the analysis of coupling constants of Ha,
as shown in Fig. 3.
The low yield of the hydrogenation was attributed to the
formation of a by-product (2-methyl-3-cyano-5-isopropylphe-
nol) resulting from full aromatization under the conditions of
the hydrogenation. Treatment of 32 with t-BuOK in THF
provided ent-dihydrocarvone (33). Enone 28 was also re-
duced under Luche conditions to the allylic alcohol 34 in an-
ticipation of a possible rhodium-catalyzed isomerization8
(RhCl3, EtOH) to the saturated ketone 31; however, this iso-
merization proved unsuccessful. The stereochemistry assign-
ment of 34 was evident from the large coupling constants of
the axial Hb, as shown in Fig. 3.
The enantiomeric switch of carvone was accomplished in
two steps via a reduction–elimination sequence as shown in
Fig. 4. The cyanoenone 28 was reduced with sodium dithio-
nite11 to a mixture of the saturated cyanoketones 31. All four
diastereomers of 31 are known compounds, generated over
the years in investigations of various reactions of enones12–14
that utilized carvone as a convenient starting material. The
major product from the dithionite reduction turned out to be
the diastereomer 31b (Fig. 5), as evidenced by melting point
as well as by detailed NMR analysis.
In the early 1900s Lapworth15–17 studied the addition of
cyanide to ent-carvone and isolated two diastereomers of 31,
one from the reaction of potassium cyanide in ethanol, water,
and acetic acid conducted at room temperature, and the other
from a reaction performed in refluxing ethanol without acetic
acid. (These two compounds would later be identified as 31a
and 31b, respectively; Fig. 5) When Djerassi et al.18 repeated
Lapworth’s experiments in 1963 with (–)-carvone, they found
that the physical constants of the two diastereomers matched
Lapworth’s results. They then studied the equilibration of the
two isomers and isolated a third when 31a was treated with
p-TsOH in EtOAc at room temperature for 1 week. The third
isomer, a much lower melting compound, was later assigned
as structure 31c, Fig. 4.
For the direct conversion of cyanoenone 28 to ent-carvone
(26) several methods were investigated, among them hydrolysis
to the corresponding acid and thermal decarboxylation,9 and
Birch reduction,10 as shown in Fig. 2. Neither method proved
very efficient in our hands.
Published by NRC Research Press

Hudlicky et al.
539
1
1
Fig. 6. Assignment of stereochemistry of 31b by NMR methods. (A) Resolution enhanced H spectrum. (B) Resolution enhanced H spec-
trum selectively decoupled from H7 methyl.
Fig. 7. Attempted Dauben–Michno synthesis of (+)-carvone from (–)-carvol 35.
Attempted equilibration of 31b was unsuccessful save for a
trace amount of a new compound tentatively assigned as 31d.
Djerassi et al.18 assigned the stereochemistry of the four dia-
stereomers of 31 by use of detailed conformational analysis
and optical rotary dispersion. In addition, they proved by
simple yet ingenious experiments that the equilibration of
31a occurs at C-2 and not at C-3 by conversion of nitriles to
benzamides and equilibration of C-2 positions. The stereo-
chemistry of 31b was evident by matching the melting point
with the compound identified by Djerassi et al.18
The assignment of 31b by NMR methods proved to be
nontrivial because of overlapping signals of the key protons
and required the dispersion field of a 600 MHz spectrometer
with advanced methods to separate the overlapping signals.
The assignment of stereochemistry is shown in Fig. 6. In the
normal proton spectrum, the six protons from the ring are all
grouped at 1.9–2.6 ppm with one proton at 2.03 ppm, a mul-
tiplet at 2.37 ppm corresponding to three protons, and an-
other multiplet at 2.57 ppm also corresponding to three
protons. The proton at 2.03 ppm shows a d–d–d splitting pat-
tern with three equal coupling constants (J = 12.2 Hz). Given
the ring structure of cyclohexane there is only one possible
case for the assignment of this proton, H3a. It is the only
possible case with two large vicinal couplings (axial–axial)
and one large geminal coupling. This also establishes that
protons H4a and H2a are axial protons, implying that both
the CN group and the propylene are in equatorial positions
on the ring.
For the assignment of the H1 proton two-dimensional
NMR methods had to be employed. From the H–C HSQC
and COSY spectra, the locations of the other ring protons
were determined. H1, H5, (H2a or H4a) are grouped at
2.6 ppm and H5 and H3e (H4a or H2a) are grouped at
2.37 ppm. H1 is identified from the coupling to the H7
methyl group in the COSY spectrum as being located on
the left side of the multiplet, as shown in Fig. 6A. To in-
crease spectral resolution, Gaussian line broadening was
employed with a shift of 0.5 and a line-broadening factor
of –1 Hz. From the proton spectrum a coupling constant of
6.2 Hz, corresponding to the vicinal coupling of H1 to H7
protons, was determined. This coupling alone would gener-
ate a quartet splitting pattern for H1. However, a more com-
plicated pattern is observed for H1, therefore, the coupling
H1 to H2 cannot be extracted directly from the proton 1-D
NMR spectrum.
A homonuclear proton experiment was employed to meas-
ure the H1–H2 vicinal coupling constant. A proton spectrum
was acquired selectively decoupling the H7 methyl group
protons at 1.30 ppm from H1. As a result the H1 coupling
pattern was simplified and reduced to H1–H2 vicinal cou-
pling, as shown in Fig. 6B. By overlapping the normal pro-
ton spectrum with the selectively decoupled spectrum, the
H1–H2 coupling constant of 12.2 Hz was extracted. This cor-
responds to an axial position for H1, confirming that the C-1
methyl group occupies an equatorial position. At this point
the right side of the 2.57 ppm multiplet can be assigned as
Published by NRC Research Press

540
Can. J. Chem. Vol. 89, 2011
H2a showing the d–d–d pattern expected for H2a, as H4a
would generate a d–d–d–d pattern. As this spectral assign-
ment for 31b would, of course, not have been possible in the
1960s, one must admire the brilliant deductive logic em-
ployed by Djerassi et al.18 in accurately assigning all four di-
astereomers of 31.
In preliminary experiments the treatment of 31b with
NaOEt in ethanol provided (+)-carvone, identical in all re-
spects to (–)-carvone except for the sign of optical rotation,
thus accomplishing the intended enantiomeric switch. Under
these conditions, however, the yields of carvone were low
(∼35%) at the expense of condensation products. The condi-
tions that employ Fe(OH)2 (shown in Fig. 4) provided the
much improved yield of 79% for (+)-carvone.
Finally, we subjected carvol (35), obtained by Luche re-
duction of carvone according to literature protocol,19 to the
Dauben–Michno conditions at –50 °C to investigate the pos-
sibility of competition between chromate ester re-oxidation to
(–)-carvone versus the Dauben–Michno transposition–oxidation
to (+)-carvone (Fig. 7). The rate-determining step in chro-
mium(VI) oxidation of alcohols is the collapse of the chro-
mate ester20 and not its formation. We speculated that at low
temperature, the sigmatropic rearrangement of the initially
formed chromate ester required for the transposition may
compete with the oxidation even for secondary allylic alco-
hols.
If the transition state of the [3,3] rearrangement is not
symmetrical along the latent plane of symmetry in carvone,
then the Dauben–Michno protocol applied to carvol would
provide the enantiomer directly. The initial experiment, per-
formed at –55 °C, indeed provided a scalemic mixture of car-
vone with an optical rotation lower than that of the
(–)-enantiomer, indicating that the two processes may indeed
compete.
Experimental section
General procedure for the synthesis of cyanohydrins6
To a vial charged with the lithium salt of L-proline
(0.1 mmol) was added the enone (1.0 mmol) followed by
trimethylsilyl cyanide (1.0 mmol, 133 µL). After the suspen-
sion was stirred at room temperature for 12 h, an additional
portion of trimethylsilyl cyanide (1.0 mmol, 133 µL) was
added, and the reaction mixture was subsequently stirred for
another 12 h. Upon completion, THF (2 mL) and 1 mol/L aq
HCl (2 mL) were added, and the resulting solution was
stirred for 1 h. Following the hydrolysis, the reaction mixture
was washed with diethyl ether (30 mL) and water (5 mL),
and the aqueous layer was re-extracted with diethyl ether
(2 × 15 mL). The combined organic layers were dried over
magnesium sulfate with a spatula tip of sodium bicarbonate,
then filtered and concentrated under reduced pressure to af-
ford the essentially pure cyanohydrin, which was used with-
out further purification.
General procedure for the Dauben–Michno oxidations5
Acetic anhydride (5.3 mmol, 0.5 mL) was added to a vial
charged with chromium(VI) oxide (2.5 mmol). The mixture
was stirred at 80 °C until complete dissolution (5 min). After
cooling to room temperature, the solution was diluted with
dichloromethane (2 mL) and added dropwise to a solution of
cyanohydrin (1.0 mmol) in dichloromethane (3 mL) at 4 °C.
The resulting solution was stirred for 10 min before addition
of ethanol (3 mL). The solution was allowed to stir for
30 min at room temperature before concentration under re-
duced pressure and co-evaporated with toluene. The crude
mixture was passed through a column of silica gel using di-
chloromethane as a solvent to yield b-cyanoenone.
1-Hydroxycyclohex-2-enecarbonitrile (5)6,22
1H (300 MHz, CDCl3) d: 6.03–6.09 (m, 1H), 5.79 (d, 1H,
J = 9.9 Hz), 1.79–2.21 (m, 6H).
We performed three experiments at different temperatures
(–55, –78, and –100 °C) and noted some decrease in the
value of rotation for the (–)-enantiomer. However, verifica-
tion of true er by HPLC (Daicel Chiralcel OB-H; flow =
0.6 mL/min (hexane–i-PrOH, 95:5; retention time for
(R)-(–)-carvone = 9.56 min and for (R)-(–)-carvone =
10.41 min)21 provided at best an indication of an ∼1%–2%
increase in the (+)-enantiomer. Thus, no conclusion can be
drawn from these experiments. For now, the three-step high-
yielding protocol for the enantiomeric switch via Dauben–
Michno oxidation of cyanohydrins should suffice.
3-Oxocyclohex-1-enecarbonitrile (6)23,24
1H (300 MHz, CDCl3) d: 6.53 (s, 1H), 2.50–2.59 (m, 4H),
2.10–2.19 (m, 2H).
1-Hydroxycyclopent-2-enecarbonitrile (8)
1H (300 MHz, CDCl3) d: 6.17 (dt, 1H, J= 2.3, 5.4 Hz),
5.82 (dt, 1H, J = 2.0, 5.4 Hz), 2.14–2.59 (m, 5H). 13C
(75 MHz, CDCl3) d: 138.81, 130.74, 120.88, 67.91, 31.01,
25.52. A by-product of this reaction was the 1,4-adduct, 3-
oxocyclopentanecarbonitrile:25 1H (300 MHz, CDCl3) d:
3.15–3.25 (m, 1H), 2.42–2.67 (m, 4H), 2.24–2.36 (m, 2H).
Conclusions
We have developed a reliable procedure for the conver-
sion of enones to the corresponding b-cyanoenones with
the attendant 1,3-functional transposition of the carbonyl
group of the original enone. The possibility of using this
transposition to generate enantiomers of enones that possess
a latent plane of symmetry was tested and reduced to prac-
tice with the enantiomeric switch of (–)-carvone. Future at-
tention should be given to a more detailed study of the
Dauben–Michno oxidation of other allylic alcohols, includ-
ing secondary ones, and the dependence of the outcome of
such oxidation on conformational bias and stereochemical
features of the substrates.
3-Oxocyclopent-1-enecarbonitrile (9)26
1H (300 MHz, CDCl3) d: 6.74 (s, 1H), 2.58–2.91 (m, 2H),
2.53 (t, 2H, J = 4.8 Hz). 13C (75 MHz, CDCl3) d: 206.24,
143.45, 143.04, 114.81, 34.10, 30.20.
1-Hydroxycyclohept-2-enecarbonitrile (11)
1H (300 MHz, CDCl3) d: 5.99 (ddd, 1H, J = 6.3, 6.3,
11.4 Hz), 5.68 (ddd, 1H, J = 1.2, 1.2, 11.7 Hz), 3.65 (bs,
1H), 2.19–2.25 (m, 1H), 1.82–2.07 (m, 6H), 1.70–1.75 (m,
1H). 13C (75 MHz, CDCl3) d: 135.04, 133.16, 120.55,
71.39, 39.28, 27.50, 26.18, 25.21.
Published by NRC Research Press

Hudlicky et al.
541
3-Oxocyclohept-1-enecarbonitrile (12)27
and allowed to warm to room temperature. The mixture was
then concentrated and co-evaporated three times with toluene
(10 mL). Chromatography of the residue (silica, 20 mL; di-
chloromethane–hexane, 1:1) afforded enone 28 as a light yel-
low solid; yield: 260 mg (75%). RF = 0.6 (dichloromethane–
1H (300 MHz, CDCl3) d: 6.56 (s, 1H), 2.63–2.67 (m, 4H),
1.82–1.93 (m, 4H). 13C (300 MHz, CDCl3) d: 201.17,
142.81, 127.36, 118.95, 42.96, 32.14, 25.36, 21.14.
2-Hydroxy-2-methylbut-3-enenitrile (14)
20
hexane, 1:1). ½aꢀD = 82.8 (c 1, CHCl3); mp 25–26 °C (neat).
1H (300 MHz, CDCl3) d: 5.95 (dd, 1H, J = 10.2, 17.1 Hz),
5.67 (d, 1H, J = 16.8 Hz), 5.35 (d, 1H, J = 10.2 Hz), 1.67
(s, 3H). 13C (300 MHz, kCDCl3) d: 137.49, 120.45, 116.77,
68.76, 28.06.
IR (KBr, cm–1) n: 3747, 30.85, 2970, 2925, 2218, 1688,
1
1648, 1439, 1383, 1318, 1295, 1140, 1108, 1070, 900. H
NMR (CDCl3, 300 MHz) d: 4.81 (s, 1H), 4.72 (s, 1H),
2.73–2.53 (m, 3 H), 2.45 (ddq, 1H, J = 3 × 2.4, 12.3,
17.4 Hz), 2.37 (dd, 1H, J= 12.9, 16.2 Hz), 2.00 (dd, 3H,
J = 1.8, 2.4 Hz), 1.70 (s, 3H). 13C NMR (CDCl3, 75 MHz)
d: 196.66, 146.31, 144.74, 124.58, 116.95, 111.86, 42.60,
41.26, 33.07, 20.40, 14.86. MS (+FBA) m/z (%): 176 (100).
HRMS could not be obtained; the compound proved to be
too labile. Structure was confirmed by HRMS after reduction
of the keto group.
(Z)-2-Methyl-4-oxobut-2-enenitrile (15a) (major)
IR (film, cm–1) n: 2925, 2222, 1688, 1618, 1448, 1381,
1227, 1171, 1095, 1041, 1019, 861, 787, 665. 1H
(300 MHz, CDCl3) d: 10.02 (d, 1H, J = 7.8 Hz), 6.52 (dq,
1H, J = 1.5, 7.8 Hz), 2.25 (d, 3H, J = 1.5 Hz). 13C
(300 MHz, CDCl3) d: 189.62, 141.64, 129.26, 115.41,
21.68. MS (EI+) m/z % 94 (M+ – H): 83 (15), 68 (63), 52
(29), 44 (100).
Oxidation of 27 at elevated temperature
(E)-2-Methyl-4-oxobut-2-enenitrile (15b) (minor)
IR (film, cm–1) n: 2925, 2222, 1688, 1618, 1448, 1381,
1227, 1171, 1095, 1041, 1019, 861, 787, 665. 1H
(300 MHz, CDCl3) d: 10.07 (d, 1H, J = 7.8 Hz), 6.52 (dq,
1H, J = 1.5, 7.8 Hz), 2.38 (d, 3H, J = 1.5 Hz). 13C
(300 MHz, CDCl3) d: 189.62, 141.64, 129.26, 115.41,
21.68. MS (EI+) m/z % 94 (M+ – H): 83 (15), 68 (63), 52
(29), 44 (100).
To a solution of crude cyanohydrin 27 (235 mg,
1.33 mmol) in dichloromethane (5 mL) cooled to 0 °C was
added a solution of CrO3 (332 mg, 3.33 mmol) and Ac2O
(0.78 mL) in dichloromethane (2.5 mL). After 20 min at 0 °C
a new portion of CrO3 (332 mg, 3.33 mmol) and Ac2O
(0.78 mL) in dichloromethane (2.5 mL) was added. After an
additional 70 min, the reaction was quenched by the addition
of ethanol (2 mL) and allowed to warm to room temperature
over 45 min. The mixture was then concentrated. Chromatog-
raphy of the residue (silica, 25 mL; Et2O–hexane, 1:1)
afforded diketone 29 (yield: 17.6 mg), epoxide 30 (yield:
73.2 mg), and a mixed fraction (yield: 61 mg).
(E)-2-Hydroxy-2-methyl-3-pentenenitrile (17)
RF = 0.76 (hexanes–EtOAc, 5:1). 1H NMR (300 MHz,
CDCl3, ppm) d: 6.10 (dq, J = 15.3, 6.6 Hz, 1H), 5.55 (dd,
J = 15.4, 1.7 Hz, 1H), 3.08 (brs, 1H), 1.76 (dd, J = 6.6,
1.7 Hz, 3H), 1.64 (s, 3H); –OH signal confirmed by D2O ex-
change. 13C NMR (75 MHz, CDCl3, ppm) d: 130.81, 129.02,
120.80, 68.42, 28.27, 17.26; NMR data are in agreement
with those previously reported.28
5-Acetyl-2-methyl-3-oxocyclohex-1-enecarbonitrile (29)
20
½aꢀ = 25.1 (c 1, CHCl3). IR (film, cm–1) n: 3022, 2964,
2220,D1685, 1416, 1358, 1314, 1265, 1216, 1104, 752, 668,
1
629. H (600 MHz, CDCl3) d: 3.23 (dddd, 1H, J = 4.2, 7.2,
10.5, 10.5 Hz), 2.82–2.78 (m, 3H), 2.65 (dd, 1H, J = 11.4,
16.8 Hz), 2.25 (s, 3H), 2.10 (s, 3H). 13C (600 MHz, CDCl3)
d: 205.98, 194.49, 146.66, 122.89, 116.56, 46.69, 38.99,
29.71, 27.88, 14.97. MS (EI+) m/z % 177 (M+): 135 (45),
107 (6), 79 (7), 43 (100), HRMS calcd for C10H11NO2:
177.1998; found: 177.0790.
(Z)-2-Methyl-4-oxopent-2-enenitrile (18)
RF = 0.15 (hexanes–EtOAc, 5:1); mp 37–38 °C (white
crystals, pentane–Et2O). IR (KBr, cm–1) n: 2359, 1732,
1
1637, 1385. H NMR (600 MHz, CDCl3, ppm) d: 6.60 (d,
J = 1.5 Hz, 1H), 2.39 (s, 3H), 2.16 (d, J = 1.5 Hz, 3H). 13C
NMR (151 MHz, CDCl3, ppm) d: 194.25, 140.10, 119.64,
117.12, 30.06, 22.18. MS (+EI) m/z (%): 109 (22), 94 (100),
66 (15), 43 (40). HRMS (+EI) calcd for C6H7NO:
109.05276; found: 109.05246.
2-Methyl-5-(2-methyloxiran-2-yl)-3-oxocyclohex-1-
enecarbonitrile (30)
20
½aꢀD = 57.5 (c 1, CHCl3). IR (film, cm–1) n: 2927, 2218,
1
1-Hydroxy-2-methyl-5-(prop-1-en-2-yl)cyclohex-2-
enecarbonitrile (27)
1686, 1436, 1386, 1313, 1140, 1070, 894, 820, 682, 622. H
(300 MHz, CDCl3) d: 2.70 (dd, 1H, J = 4.1, 14.2 Hz), 2.66–
2.57 (m, 3H), 2.50–2.43 (m, 1H), 2.33–2.26 (m, 1H), 2.26–
2.14 (m, 1H), 2.04 (s, 3H), 1.32 (d, 3H, J = 8.4 Hz). 13C
(300 MHz, CDCl3) d: 195.81, 146.45, 124.28, 116.78,
57.39, 52.64, 40.13, 39.41, 29.85, 18.82, 14.87. MS (EI+)
m/z % 176 (M+ – CH3): 148 (51), 134 (100), 122 (53), 58
(81), 43 (96). HRMS calcd for C11H13NO2: 191.2264; found:
176.0712.
1H (300 MHz, CDCl3) d: 5.70 (dd, 1H, J = 1.2, 3.0 Hz),
4.79 (d, 2H, J = 12.3 Hz), 3.24 (bs, 1H), 2.49–1.90 (m, 5H),
1.86 (s, 3H), 1.74 (s, 3H). 13C (300 MHz, CDCl3) d: 146.85,
131.76, 128.40, 121.09, 110.18, 70.21, 41.27, 38.96, 30.67,
20.57, 17.01.
b-Cyanocarvone (28)
To a solution of crude cyanohydrin 27 (352 mg,
1.99 mmol) in dichloromethane (3 mL) cooled to –55 °C
was added a solution of CrO3 (499 mg, 4.99 mmol) and
Ac2O (2 mL) in dichloromethane (6 mL). After 30 min the
reaction was quenched by the addition of methanol (10 mL)
(1R,2R,5S)-2-Methyl-3-oxo-5-(prop-1-en-2-yl)
cyclohexanecarbonitrile (32)
A solution of enone 28 (0.1g, 0.57 mmol) with Pd/C
(20 mg, 10%) in ethanol (3 mL, 95%) was hydrogenated in
Published by NRC Research Press

542
Can. J. Chem. Vol. 89, 2011
an H2 atmosphere at 80 °C for 16 h. The reaction mixture
was then filtered through Celite and concentrated. Chroma-
tography (silica, 8 mL; dichloromethane–hexane, 1:1) of
crude product afforded 35 mg (34%) of fully saturated ketone
32 as a white solid. The remaining mass balance consisted of
fully aromatized material identified as 2-methyl-3-cyano-5-
isopropylphenol. RF = 0.5 (dichloromethane–hexane, 1:1).
layer was extracted with CH2Cl2 (3 × 30 mL). The combined
organic extracts were washed with 0.5 N HCl (5 mL), then
water (5 mL), and dried over magnesium sulfate. After filtra-
tion and evaporation of the solvent, the crude mixture was
purified via flash column chromatography (EtOAc in hex-
anes, 10%–25%) to yield 145 mg (82%) of the mixture con-
sisting of three of the four isomers of cyanoketone 31. This
mixture was suitable for use in the elimination to (+)-carvone.
20
½aꢀD = –49.26 (c 1, CHCl3); mp 77 °C (hexane). IR (KBr,
cm–1) n: 3412, 2979, 2959, 2938, 2916, 2874, 2241, 1713,
Identification of isomers 31a, 31b, and 31c
1477, 1469, 1448, 1429, 1381, 1369, 1346, 1313, 1245,
1
1231, 1218, 1147, 1076, 870, 628. H NMR (CDCl3, 300
To a round-bottomed flask containing water (60 mL), so-
dium bicarbonate (4.31 g, 51.30 mmol), and tetrabutylammo-
nium bromide (275 mg, 0.855 mmol) was added a solution
of enone 28 (500 mg, 2.85 mmol) in toluene (60 mL). The
reaction mixture was heated to reflux, and sodium dithionite
(6 × 875 mg, 4.28 mmol) was added in six portions over 3 h.
After an additional 1 h of stirring at reflux, the reaction mix-
ture was allowed to cool to room temperature. The organic
layer was separated from the aqueous layer and then dried
over magnesium sulfate. After filtration and evaporation of
solvent, the crude mixture was purified via flash column
chromatography (hexane – diethyl ether, 7:1) to provide
75 mg (15%) of saturated ketone 31b as a white solid. Fur-
ther separation provided the remaining isomers 31a and 31c.
MHz) d: 2.58–2.44 (m, 3 H), 2.27 (dddd, 1H, J = 2.4, 2.7,
2.7, 13.5 Hz), 2.14 (dd, 1H, J =13.2, 12.9 Hz), 1.81 (ddd,
1H, J = 12.6, 12.6, 12.6 Hz), 1.63 (dd, 1H, J = 12.6,
14.1 Hz), 1.61 (m, 1H), 1.25 (d, 3H, J = 6.0 Hz), 0.93 (d,
3H, J = 6.6 Hz), 0.92 (d, 3H, J = 5.4 Hz). 13C NMR
(CDCl3, 75 MHz) d: 207.54, 120.28, 46.35, 44.53, 43.71,
36.13, 32.78, 32.36, 19.28, 19.06, 12.71. MS (+EI) m/z (%):
179 (100). HRMS (+EI) calcd for C11H17N1O1: 179.13101;
found: 179.13136.
(3S,5S)-3-Hydroxy-2-methyl-5-(prop-1-en-2-yl)cyclohex-1-
enecarbonitrile (34)
To the cooled (4 °C) solution of enone 28 (50 mg,
29 mmol) in methanol (2 mL) was added CeCl3·7H2O
(106 mg, 29 mmol). After 5 min of stirring, NaBH4 (11 mg,
29 mmol) was added portionwise. After an additional 10 min
stirring, the reaction was quenched by the addition of a citric
acid solution (0.5 mL, 5% w/w) and concentrated. The resi-
due was dissolved in dichloromethane (25 mL) and washed
with citric acid (5 mL, 5% w/w). The aqueous layer was re-
extracted with additional dichloromethane (5 mL), the com-
bined organic layers were dried over MgSO4, and then con-
centrated to yield essentially pure product. Chromatography
(silica, 5 mL, dichloromethane) yielded 48 mg (95%) of the
allylic alcohol 34 as a colourless oil. RF = 0.2 (dichloro-
(1S,2S,5S)-2-Methyl-3-oxo-5-(prop-1-en-2-yl)
cyclohexanecarbonitrile (31a)
RF = 0.17 (hexanes–EtOAc, 5:1). 1H NMR (300 MHz,
CDCl3, ppm) d: 4.86 (s, 1H), 4.80 (s, 1H), 3.34 (dt, J = 6.0,
3.7 Hz, 1H), 2.86–2.72 (m, 1H), 2.66–2.47 (m, 2H), 2.36–
2.18 (m, 2H), 2.05–1.86 (m, 1H), 1.77 (s, 3H), 1.25 (d, J =
6.6 Hz, 3H).
1H NMR data are in agreement with the literature.14
(1R,2S,5S)-2-Methyl-3-oxo-5-(prop-1-en-2-yl)
cyclohexanecarbonitrile (31b)
20
methane). ½aꢀD = 40.42 (c 1, CHCl3). IR (KBr, cm–1) n:
White solid. RF = 0.42 (hexane – ethyl acetate, 5:1).
3446, 3084, 2926, 2859, 2215, 1646, 1441, 1378, 1288,
20
½aꢀD = –56 (c 1, CHCl3). mp 82–84 °C (first crystallization:
1
1261, 1071, 1056, 1025, 895. H NMR (CDCl3, 300 MHz)
hexanes – diethyl ether; second crystallization: mp 85–86 °C
(lit.18 mp 86.5). IR (CHCl3, cm–1) n: 2244, 1719, 1648,
d: 4.81 (s, 1H), 4.76 (s, 1H), 4.24 (m, 1H), 2.36–2.13 (m, 5
H), 2.11 (s, 3H), 1.75 (s, 3H), 1.55 (ddd, 1H, J = 10.2, 12.0,
12.0 Hz). 13C NMR (CDCl3, 75 MHz) d: 154.69, 146.75,
118.34, 110.52, 108.08, 70.25, 39.28, 36.62, 32.71, 20.43,
18.33. MS (+EI) m/z (%): 177 (3), 159 (100), 144 (33), 134
(59). HRMS (+EI) calcd for C11H15N1O1: 177.11536; found:
177.11525. HRMS (+EI) calcd for C11H13N1: 159.10480;
found: 159.10496.
1
1452, 901. H NMR (CDCl3, 600 MHz, ppm) d: 4.85 (s,
1H), 4.78 (s, 1H), 2.61–2.52 (m, 3 H), 2.40–2.36 (m, 3H),
2.01 (ddd, 1H, J = 10.9, 11.9, 12.4 Hz), 1.75 (s, 3H), 1.28
(d, 3H, J = 6.4 Hz). 13C NMR (CDCl3, 150 MHz, ppm) d:
206.66, 145.27, 119.98, 111.30, 46.28, 44.14, 36.00, 34.25,
20.23, 12.67. MS (+EI) m/z (%): 177 (12), 95 (55), 82 (54),
68 (100), 67 (74), 41 (41). HRMS (+EI) calcd for
C11H15NO: 177.11536; found: 177.11509.
Reduction of 28 to 31a, 31b, and 31c
To a round-bottomed flask containing water (5 mL) and
tetrabutylammonium bromide (100 mg, 0.3 mmol) was added
a solution of enone 28 (175 mg, 1.0 mmol) in toluene
(5 mL). The emulsion was degassed in an ultrasonic bath for
20 min. A condenser was attached, and the system was
flushed with nitrogen. Sodium bicarbonate (1.52 g,
18 mmol) was added, and the reaction mixture was heated to
reflux. Sodium dithionite (7 × 200 mg, 85%, 6.95 mmol)
was added in seven portions over 2 h. After an additional
30 min of stirring at reflux, the reaction mixture was allowed
to cool to room temperature. The reaction mixture was di-
luted with CH2Cl2 (50 mL) and water (10 mL). The aqueous
(1S,2R,5S)-2-Methyl-3-oxo-5-(prop-1-en-2-yl)
cyclohexanecarbonitrile (31c)
Viscous oil; assignment is tentative (31c has a mp of 13 °C).
IR (KBr, cm–1) n: 2359, 1650, 1385, 1088. 1H NMR
(600 MHz, CDCl3, ppm) d: 4.87 (s, 1H), 4.79 (s, 1H), 3.14
(td, J = 5.1, 2.5 Hz, 1H), 2.64 (t, J = 12.8 Hz, 1H), 2.40
(dt, J = 12.8, 2.3 Hz, 1H), 1.87–1.80 (m, 1H), 1.78 (s, 1H),
1.61–1.49 (m, 1H), 1.45 (d, J = 6.8 Hz, 1H). MS (+EI) m/z
(%): 177 (27), 95 (72), 82 (52), 69 (29), 68 (100), 67 (70),
41 (42).; HRMS (+EI) calcd for C11H15NO: 177.11536;
found: 177.11498.
Published by NRC Research Press

Hudlicky et al.
543
(5) Werner, L.; Machara, A.; Hudlicky, T. Adv. Synth. Catal. 2010
352 (1), 195. doi:10.1002/adsc.200900844.
(6) Shen, Z.-L.; Ji, S.-J. Synth. Commun. 2009 39 (5), 775. doi:10.
1080/00397910802431149.
(7) Wenkert, E.; Hudlicky, T. Synth. Commun. 1977 7 (8), 541.
doi:10.1080/00397917709409274.
(8) [Review:]van der Drift, R. C.; Bouwman, E.; Drent, E. J.
Organomet. Chem. 2002 650 (1–2), 1. doi:10.1016/S0022-
328X(02)01150-6.
ent-Carvone
The mixture of saturated ketones 31a, 31b, and 31c ob-
tained (445 mg, 2.5 mmol) was dissolved in MeOH
(15 mL). The solution was degassed in an ultrasonic bath for
5 min (to avoid oxidation of Fe(II) by air). A condenser was
attached, and FeSO4·7H2O (1.39 g, 5.0 mmol) was added.
KOH (2.25 g, 35 mmol, 10 + 4 equiv) was added to a vigo-
rously stirred suspension to generate Fe(OH)2. The reaction
flask was placed in a preheated oil bath, and the mixture
heated at reflux for 20 min. The resulting dark black mixture
was quickly cooled to room temperature and diluted with
EtOAc (100 mL) and water (10 mL). The organic layer was
washed with 0.2 N HCl (3 × 10 mL) and dried over magne-
sium sulfate. After filtration and evaporation of solvent, the
crude product was filtered through a short pad of silica gel
(eluted with 10% EtOAc in hexanes) to yield S-(+)-carvone
(296 mg, 79%).
(9) Arnold, R. T.; Elmer, O. C.; Dodson, R. M. J. Am. Chem. Soc.
1950 72 (10), 4359. doi:10.1021/ja01166a007.
(10) Marshall, J. A.; Bierenbaum, R. J. Org. Chem. 1977 42 (20),
3309. doi:10.1021/jo00440a029.
(11) For reduction of enones with sodium dithionite see: Louis-
Andre, O.; Gelbard, G. Tetrahedron Lett. 1985 26 (7), 831.
doi:10.1016/S0040-4039(00)61940-8.
(12) Cocker, W.; Grayson, D. H.; Shannon, P. V. R. J. Chem. Soc.
Perkin Trans. 1995 1 (9), 1153. doi:10.1039/p19950001153.
(13) Verstegen-Haaksma, A. A.; Swarts, H. J.; Jansen, B. J. M.; de
Groot, A. Tetrahedron 1994 50 (33), 10073. doi:10.1016/
S0040-4020(01)89623-X.
S-(+)-Carvone
19
1
½aꢀD = +54.2 (c 1, CHCl3). H NMR (300 MHz, CDCl3)
d: 6.82–6.65 (m, 1H), 4.80 (s, 1H), 4.75 (s, 1H), 2.78–2.51
(m, 2H), 2.51–2.37 (m, 1H), 2.37–2.19 (m, 2H), 1.85–1.76
(m, 3H), 1.75 (s, 3H). (Commercially available (+)-carvone
is 96% optically pure; (–)-carvone is 98% optically pure).
(14) Lagriffoul, P.-H.; Tadros, Z.; Taillades, J.; Commeeyras, A. J.
Chem. Soc. Perkin Trans. 1992, 2, 1279. doi:10.1039/
P29920001279.
(15) Lapworth, A. J. Chem. Soc. 1906 89, 945.
(16) Lapworth, A. J. Chem. Soc. 1906 89, 1819.
(17) Lapworth, A.; Steele, V. J. Chem. Soc. 1911, 99, 1877.
(18) Djerassi, C.; Schneider, R. A.; Vorbrueggen, H.; Allinger, N.
L. J. Org. Chem. 1963 28 (6), 1632. doi:10.1021/jo01041a049.
(19) Valeev, R. F.; Vostrikov, N. S.; Miftakhov, M. S. Russ. J. Org.
Chem. 2009 45 (6), 810. doi:10.1134/S1070428009060025.
(20) Lee, D. G.; Downey, W. L.; Maass, R. M. Can. J. Chem. 1968
46 (3), 441. doi:10.1139/v68-070.
(21) Hareau, G. P.-J.; Koiwa, M.; Hikichi, S.; Sato, F. J. Am. Chem.
Soc. 1999 121 (15), 3640. doi:10.1021/ja9843122.
(22) Wada, M.; Takahashi, T.; Domae, T.; Fukuma, T.; Miyoshi, N.;
Smith, K. Tetrahedron: Asymmetry 1997 8 (23), 3939. doi:10.
1016/S0957-4166(97)00570-3.
Acknowledgements
The authors are grateful to the following agencies for fi-
nancial support of this work: Natural Sciences and Engineer-
ing Research Council of Canada (NSERC) (Idea to
Innovation and Discovery Grants); Canada Research Chair
Program; Canada Foundation for Innovation (CFI); Research
Corporation; Noramco, Inc.; TDC Research, Inc.; TDC Re-
search Foundation; and Brock University, St. Catharines, On-
tario. Thanks are also due to the Ministry of Education
(Spain) for fellowships to V.S. (AP2006-03468ESTANCIA-
2010). The authors thank K. Ridolfo for preliminary experi-
ments performed as part of his fourth-year Honours Thesis,
Spring 2010.
(23) Cronyn, M. W.; Goodrich, J. E. J. Am. Chem. Soc. 1952 74
(13), 3331. doi:10.1021/ja01133a034.
(24) Zhao, Y.; Yeung, Y.-Y. Org. Lett. 2010, 12 (9), 2128. doi:10.
1021/ol100603q.
References
(25) Farcasiu, D.; Schleyer, P. V. R.; Ledlie, D. B. J. Org. Chem.
(1) Luzzio, F. A. Org. React. 1998, 53, 1.
(2) Dauben, W. G.; Michno, D. M. J. Org. Chem. 1977 42 (4),
682. doi:10.1021/jo00424a023.
(3) Hudlicky, M. Oxidations in Organic Chemistry; ACS Mono-
graph 186; American Chemical Society: Washington, DC,
1990; pp 86–87.
(4) Wender, P. A.; Eissenstat, M. A.; Filosa, M. P. J. Am. Chem.
Soc. 1979 101 (8), 2196. doi:10.1021/ja00502a042.
1973 38 (20), 3455. doi:10.1021/jo00960a002.
(26) Zimmerman, H. E.; Pasteris, R. J. J. Org. Chem. 1980 45 (24),
4864. doi:10.1021/jo01312a013.
(27) Mease, R. C.; Hirsch, J. A. J. Org. Chem. 1984 49 (16), 2925.
doi:10.1021/jo00190a017.
(28) Kirsten, C. N.; Herm, M.; Schrader, T. H. J. Org. Chem. 1997
62 (20), 6882. doi:10.1021/jo970854a.
Published by NRC Research Press