Computational modeling of a stereoselective epoxidation: Reaction of carene with peroxyformic acid

Article abstract of DOI:10.1021/jo702722g

(Chemical Equation Presented) Electronic structure theory was used to model the epoxidation of 3-carene by peroxyformic acid. Reactants, products, and transition states were optimized at the B3LYP/6-31G* level of theory, followed by B3LYP/6-311+G** and MP2/6-311+G** single point calculations. The reaction pathway yielding the trans-epoxide product was found to have a significantly lower reaction barrier (7.8 kcal/mol) than that leading to the cis-epoxide product (9.4 kcal/mol), in agreement with expectations. Magnetic shieldings of the two isomeric carene epoxides were also calculated, using the GIAO method, and compared to experimental ¹H and ¹³C NMR spectra. Although the calculated carbon spectra proved inconclusive, the proton shieldings calculated for the trans-epoxide correlated much more closely to the experimental values for the major epoxidation product than did the shieldings calculated for the cis-epoxide, serving to verify the identity of the major product.

Full text of DOI:10.1021/jo702722g

Computational Modeling of a Stereoselective Epoxidation: Reaction
of Carene with Peroxyformic Acid
Stephanie M. Koskowich, Winslow C. Johnson, Robert S. Paley, and Paul R. Rablen*
Department of Chemistry and Biochemistry, Swarthmore College, 500 College AVenue,
Swarthmore, PennsylVania 19081
prablen1@swarthmore.edu
ReceiVed December 21, 2007
Electronic structure theory was used to model the epoxidation of 3-carene by peroxyformic acid. Reactants,
products, and transition states were optimized at the B3LYP/6-31G* level of theory, followed by B3LYP/
6
-311+G** and MP2/6-311+G** single point calculations. The reaction pathway yielding the trans-
epoxide product was found to have a significantly lower reaction barrier (7.8 kcal/mol) than that leading
to the cis-epoxide product (9.4 kcal/mol), in agreement with expectations. Magnetic shieldings of the
two isomeric carene epoxides were also calculated, using the GIAO method, and compared to experimental
1
13
H and C NMR spectra. Although the calculated carbon spectra proved inconclusive, the proton shieldings
calculated for the trans-epoxide correlated much more closely to the experimental values for the major
epoxidation product than did the shieldings calculated for the cis-epoxide, serving to verify the identity
of the major product.
3
Introduction
debate, it now seems well-established that the reaction indeed
occurs via a concerted mechanism (as Bartlett suggested) but
Ab initio and density functional electronic structure theories
have previously been used to model the epoxidation of alkenes
by peracids. In the classical transition state, all the atoms of
the peracid and the two carbons of the alkene are contained in
a single plane that is perpendicular to the plane of the alkene
and parallel to the axis of the alkene. This geometry corresponds
to Bartlett’s “butterfly mechanism”, whimsically named for the
shape of the transition structure. For many years, experimental
studies supported this mechanism. More recently, however,
computational methods have been used to determine the
transition state structures, and although there has been some
that the transition state has a symmetrical spiro geometry rather
than a planar geometry.
4–14
In this spiro geometry, the peracid
(
4) (a) Bach, R. D.; Canepa, C.; Winter, J. E.; Blanchette, P. E. J. Org. Chem.
1997, 62, 5191–5197. (b) Houk, K. N.; Liu, J.; DeMello, N. C.; Condroski,
K. R. J. Am. Chem. Soc. 1997, 119, 10147–10152.
(
5) Bach, R. D.; Glukhotsev, M. N.; Gonzalez, C.; Marquez, M.; Estévez,
1
C. M.; Baboul, A. G.; Schlegel, H. B. J. Phys. Chem. A 1997, 101, 6092–6100.
(6) Bach, R. D.; Dmitrenko, O.; Adam, W.; Schambony, S. J. Am. Chem.
Soc. 2003, 125, 924–934.
2
(
7) Freccero, M.; Gandolfi, R.; Sarzi-Amadè, M.; Rastelli, A. J. Org. Chem.
2002, 67, 8519–8527.
(
8) Adam, W.; Bach, R. D.; Dmitrenko, O.; Saha-Möller, C. J. Org. Chem.
(
(
(
1) Bartlett, P. D. Rec. Chem. Prog. 1957, 18, 111.
2000, 65, 6715–6728.
2) Woods, K. W.; Beak, P. J. Am. Chem. Soc. 1991, 113, 6281–6283.
(9) Washington, I.; Houk, K. N. Angew. Chem., Int. Ed. 2001, 40, 4485–
4488.
3) (a) Yamabe, S.; Konduo, C.; Minato, T. J. Org. Chem. 1996, 61, 616–
6
20. (b) Deubel, D. V.; Frenking, G.; Gisdakis, P.; Herrmann, W. A.; Rösch,
N.; Sundermeyer, J. Acc. Chem. Res. 2004, 37, 645–652. (c) Okovytyy, S.; Gorb,
L.; Leszczynski, J. Tetrahedron Lett. 2002, 43, 4215–4219.
(10) (a) Freccero, M.; Gandolfi, R.; Sarzi-Amadè, M.; Rastelli, A. J. Org.
Chem. 2000, 65, 2030–2042. (b) Freccero, M.; Gandolfi, R.; Sarzi-Amadé, M.;
Rastelli, A. J. Org. Chem. 2000, 65, 8948–8959.
3
492 J. Org. Chem. 2008, 73, 3492–3496
10.1021/jo702722g CCC: $40.75  2008 American Chemical Society
Published on Web 03/26/2008

Computational Modeling of a StereoselectiVe Epoxidation
TABLE 1. Calculated Reaction Barriers for Epoxidation by
Peroxyformic Acid (kcal/mol)
carene carene
theoretical method
ethylene propene
cis
trans
FIGURE 1. Epoxidation of 3-carene (1) by peroxyformic acid to yield
trans-carene epoxide (2) and cis-carene epoxide (3).
B3LYP/6-31G*//B3LYP/6-31G*
B3LYP/6-311+G**//B3LYP/6-31G*
MP2/6-311+G**//B3LYP/6-31G*
14.7
15.6
19.3
18.8
12.3
12.7
15.4
16.0
10.9
10.7
9.4
9.2
8.9
7.8
a
QCISD(T)/6-31G*//QCISD/6-31G*
lies in a plane that is perpendicular to both the plane of the
alkene (the horizontal plane) and the axis of the alkene double
bond.
a
Values taken from ref 5.
Bach’s group has extensively studied the transition states for
5
the epoxidations of ethylene, propene, and 2-butene. Both this
group and also those of Freccero and of Sarzi-Amadè have
successfully extended the approach to more complex alkenes,
as well, including bicyclic structures such as norbornene in
6
,7
which issues of facial selectivity arise.
The frequently
diastereoselective and synthetically useful peroxyacid epoxida-
tions of allylic and homoallylic systems have also attracted
8
–11
considerable computational attention.
We wished to see if
the same methodology could successfully be applied to the case
of 3-carene (Figure 1). Carene, one of the monoterpenes that
1
5
constitutes turpentine, has often been epoxidized as the first
step in a synthetic pathway, either as a precursor for ligand
preparation or as the starting point for natural product total
1
6
synthesis. Carene epoxidation has also recently been shown
1
7
to have biological significance.
The laboratory epoxidation of carene is frequently carried
1
8,19
out using a peracid, most commonly MCPBA,
which
produces exclusively the trans-carene epoxide (also known as
FIGURE 2. Reaction coordinate energy diagram for the epoxidation
of carene showing both possible reaction pathways. Labeled energies
were calculated at MP2/6-311+G**//B3LYP/6-31G*.
1
8
the R-epoxide). Here, we have modeled the epoxidation
reaction using the much smaller peroxyformic acid as the
oxidant. It has been reported that, for the reaction involving
simple peracids, the major product makes up roughly 80% of
We were interested to see whether computational methods
could reproduce the experimentally observed stereoselectivity
of this reaction. Furthermore, we wished to determine whether
computation of the NMR spectroscopic properties of the two
isomeric products, followed by comparison to the experimental
spectra, could be used to verify the identity of the major product.
Both procedures are potentially useful for predicting and
evaluating the outcomes of stereoselective epoxidation reactions
of practical experimental interest.
2
0
the mixture. Using GC/MS and NMR, this major product has
been identified with a fair degree of certainty as the trans- rather
2
1
than the cis-epoxide. However, it is worth noting that under
different reaction conditions the ratio of products can vary
2
2
greatly.
(
11) Freccero, M.; Gandolfi, R.; Sarzi-Amadè, M. Tetrahedron 1999, 55,
1309–11330.
12) Singleton, D. A.; Merrigan, S. R.; Liu, J.; Houk, K. N. J. Am. Chem.
Soc. 1997, 119, 3385–3386.
13) Bach, R. D.; Glokhovtsev, M. N.; Gonzalez, C. J. Am. Chem. Soc. 1998,
20, 9902–9910.
1
(
(
Results and Discussion
1
(
(
(
14) Bach, R. D.; Dmitrenko, O. J. Phys. Chem. A 2003, 107, 4300–4306.
15) Johansson, A.; Lundborg, M. Toxicology 1997, 120, 99–104.
16) (a) Krishnaswamy, D.; Govande, V. V.; Gumaste, V. K.; Bhawal, B. M.;
As a starting point, energies were calculated for the reactants,
products, and transition states for epoxidation of ethylene and
propene, using the structures reported by Bach et al. These
5
Deshmukh, A. R. A. S. Tetrahedron 2002, 58, 2215–2225. (b) Sasai, H.; Arai,
T.; Emori, E.; Shibasaki, M. J. Org. Chem. 1995, 60, 465–467. (c) Barluenga,
J.; Vázquez-Villa, H.; Ballesteros, A.; González, J. M. Org. Lett. 2002, 4, 2817–
authors showed that reaction barriers calculated at QCISD(T)/
6
-31G*//B3LYP/6-31G* closely approximated those calculated
2
819.
17) Duisken, M.; Benz, D.; Peiffer, T. H.; Blömeke, B.; Hollender, J. Curr.
Drug Metab. 2005, 6, 593–601.
18) Kauffman, G. S.; Harris, G. D.; Dorow, R. L.; Stone, B. R. P.; Parsons,
1
3
(
at QCISD(T)/6-31G*//QCISD/6-31G*. The calculations re-
ported here indicate that even MP2/6-311+G**//B3LYP/6-31G*
reaction barriers serve as a good approximation of those
(
R. L.; Pesti, J. A.; Magnus, N. A.; Fortunak, J. M.; Confalone, P. N.; Nugent,
5
W. A. Org. Lett. 2000, 220, 3119–3121.
calculated at QCISD(T)/6-31G*//QCISD/6-31G* for the eth-
(
19) (a) Joshi, S. N.; Malhotra, S. V. Tetrahedron: Asymmetry 2003, 14,
ylene and propene epoxidations (Table 1). Accordingly, the more
economical MP2/6-311+G**//B3LYP/6-31G* procedure was
used to compute barriers in the larger carene system, for which
QCISD(T) calculations would have been impractical.
The activation barriers for epoxidation of carene to yield the
cis and trans products are listed in Table 1, and a reaction energy
diagram is shown in Figure 2. As expected, the barrier for
epoxidation leading to the trans product (7.8 kcal/mol) is
substantially lower than that for epoxidation leading to the cis
product (9.4 kcal/mol). Since the reaction is under kinetic
1
1
763–1766. (b) Paquette, L. A.; Ross, R. J.; Shi, Y. J. Org. Chem. 1990, 55,
589–1598.
(
20) 84.6%, epoxidized with peracetic acid. (a) Gollnick, K.; Schroeter, S.;
Ohloff, G.; Schade, G.; Schenck, G. O. Ann. 1965, 687, 14–25. (b) 78%,
epoxidized with meta-chloroperbenzoic acid) as reported in ref 2.
(21) (a) Martins, R. R. L.; Neves, M. G. P. M. S.; Silvestre, A. J. D.; Simões,
M. M. Q.; Silva, A. M. S.; Tomé, A. C.; Cavaleiro, J. A. S.; Tagliatesta, P.;
Crestini, C. J. Mol. Catal. A 2001, 172, 33–42. (b) Brown, H. C.; Suzuki, A.
J. Am. Chem. Soc. 1967, 89, 1933–1941.
(
22) (a) Rebelo, S. L. H.; Pereira, M. M.; Simões, M. M. Q.; Neves,
M. G. P. M. S.; Cavaleiro, J. A. S. J. Catal. 2005, 234, 76–87. (b) Barluenga,
J.; Marco-Arias, M.; González-Bobes, F.; Ballesteros, A.; González, J. M.
Chem.—Eur. J. 2004, 10, 1677–1682.
J. Org. Chem. Vol. 73, No. 9, 2008 3493

Koskowich et al.
FIGURE 5. Atom numbering scheme for carene epoxide. For hydrogen
atoms, exo designates on the same face of the cyclohexane ring as the
cyclopropane moiety, and endo as on the opposite face from the
cyclopropane moiety.
FIGURE 3. Transition state structure for the trans-epoxidation of
carene determined at B3LYP/6-31G*.
FIGURE 6. Calculated 13C NMR chemical shifts for trans- and cis-
carene epoxides compared to the experimental values obtained from
the carene epoxidation major product. Isotropic shielding constants were
calculated at B3LYP/6-311++G**//B3LYP/6-31G*. For the trans-
carene epoxide best fit line: slope ) 0.980, intercept ) 4.90, correlation
FIGURE 4. Transition state structure for the cis-epoxidation of carene
determined at B3LYP/6-31G*.
2
Experimental and Calculated 13C NMR Resonances of
coefficient (R ) ) 0.990. For the cis-carene epoxide best fit line: slope
TABLE 2.
2
the Carene Epoxides (ppm)
) 0.954, intercept ) 6.48, correlation coefficient (R ) ) 0.985.
atom exptl
calcd/cis
calcd/trans
1
TABLE 3. Experimental and Calculated H NMR Resonances of
the Carene Epoxides (ppm)
C1
C2
C3
C4
C5
C6
C7
C8
C9
C10
15.9
23.2
56.0
58.2
19.2
13.8
16.0
23.1
14.6
27.7
23.5
28.2
59.8
62.6
25.0
22.3
23.5
26.6
16.3
32.4
5.6
22.4
27.7
59.8
62.8
24.3
20.0
22.0
25.0
17.1
30.4
4.6
atom/group
exptl
calcd/cis
calcd/trans
H1
H2exo
H2endo
H4
H5exo
H5endo
H6
C8 methyl
C9 methyl
C10 methyl
rms deviation
0.54
2.15
1.49
2.84
2.31
1.64
0.46
1.27
0.73
1.01
0.48
2.05
1.98
2.69
2.17
2.14
0.43
1.19
1.14
0.93
0.27
0.42
2.04
1.42
2.58
2.27
1.60
0.31
1.14
0.75
1.02
0.12
rms deviation
control, the relative energies of the transition structures are
expected to determine the stereochemical outcome. The actual
transition structures are shown in Figures 3 and 4. Peroxyformic
acid would appear to be an adequate model oxidant, despite
being much smaller than those typically used experimentally,
since the hydrogen of the formyl group, where a bulkier group
would in practice be located, points away from the rest of the
structure.
epoxide. Comparisons between the calculated and experimental
carbon chemical shifts were inconclusive; calculations for both
cis- and trans-carene epoxide correlated equally well with the
experimental spectra (Table 2, Figure 6; atomic numbering
scheme in Figure 5). However, the H NMR results (Table 3)
were more helpful. The isotropic shielding values calculated
1
Calculated isotropic magnetic shieldings were used to further
corroborate the previously reported experimental conclusion that
the major reaction product is trans- rather than cis-carene
for trans-carene epoxide correlated far more closely with the
2
experimental shifts (R ) 0.990) than did the isotropic values
2
calculated for cis-carene epoxide (R ) 0.885) (Figure 7). This
3
494 J. Org. Chem. Vol. 73, No. 9, 2008

Computational Modeling of a StereoselectiVe Epoxidation
are possible, but for which the direction and degree of preference
are not known in advance. Furthermore, calculation of the proton
magnetic shieldings for the isomeric epoxides, followed by
comparison to the experimental proton NMR spectra of the
major (and perhaps minor) products, can be used to verify the
experimental outcome.
Computational Methods
24
The Gaussian 03 package was used to carry out all calculations.
2
5
Standard Pople-type basis sets were employed. All geometries
were optimized at the B3LYP/6-31G* level. It has previously been
shown that, for alkene epoxidations, B3LYP optimizations are the
best of the more affordable methods; MP2 optimizations, for
5,12,13
example, yield incorrect structures.
Carene epoxidation transi-
tion state structures were determined through a multistep process.
First, the two carbon-oxygen distances were fixed at the values
for the 2-methyl-2-butene transition state, and all other parameters
were allowed to optimize. Second, starting from these partially
optimized geometries, unconstrained optimizations (opt ) calcfc,ts)
were carried out, leading to the fully optimized transition state
structures. All transition structures were verified as transition states
by the calculation of vibrational frequencies at the B3LYP/6-31G*
level of theory; that is, one imaginary frequency was found.
Single point energies were calculated at B3LYP/6-311+G** and
MP2/6-311+G** using the B3LYP/6-31G* optimized geometries.
Zero point energies calculated at B3LYP/6-31G* and corrected by
1
FIGURE 7. Calculated H NMR chemical shifts for trans- and cis-
carene epoxides compared to the experimental values obtained from
the carene epoxidation major product. Isotropic shielding constants were
calculated at B3LYP/6-311++G**//B3LYP/6-31G*. For the trans-
carene epoxide best fit line: slope ) 0.961, intercept ) –0.03, correlation
coefficient (R ) ) 0.990. For the cis-carene epoxide best fit line: slope
0.923, intercept ) 0.19, correlation coefficient (R ) ) 0.885.
26
a scale factor of 0.9804 were included, as well.
Isotropic shielding values were computed using the GIAO
2
27
method in conjunction with the B3LYP/6-311++G**//B3LYP/
2
)
6
-31G* level of theory. Calculated isotropic shielding values for
each of the three hydrogen atoms in a methyl group were averaged
to produce a single observable value for the methyl group as a
whole. The calculated isotropic shielding values were subtracted
from the corresponding value for TMS to yield calculated chemical
shift values, relative to TMS.
correspondence provides strong additional evidence that the
major product of carene epoxidation is indeed the trans-epoxide.
13
In addition, the observation that calculated C NMR shifts were
1
not useful for structure assignment in this case, but that the H
NMR shifts were, provides a perhaps interesting counter-
example to some of the successes recently reported with regard
Experimental Section
1
3
to the use of calculated C NMR shifts for structure determi-
To obtain experimental 13C and 1H NMR spectra, 117 µL of
2
3
nation.
(+)-3-carene was reacted with 437 mg of magnesium bis(monop-
eroxyphthalate) in 1 mL of methanol. The product was isolated by
diluting the mixture with 10 mL of ether and 2 mL of water. The
aqueous layer was removed and the organic layer washed three
Conclusions
The two reaction pathways for the epoxidation of 3-carene
by peroxyformic acid, leading to isomeric cis- and trans-epoxide
products, were explored via electronic structure calculations.
In agreement with expectations, the transition state leading to
the trans-epoxide was substantially lower in energy than that
leading to the cis-epoxide. Furthermore, although calculated
carbon NMR spectra did not prove useful, proton magnetic
shieldings computed for the trans-epoxide correlated much more
closely with the experimentally derived spectrum of the major
isomer than did the proton shieldings calculated for the cis
isomer. The present findings further illustrate the potential
usefulness of computational study of epoxidation reactions of
structurally and stereochemically complex alkenes using the
times and dried over MgSO
through a MgSO filter to remove the drying agent. The drying
agent was washed with 5 mL of ether, which was added to the
4
. The organic solution was then run
4
(24) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.;
Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci,
B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada,
M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.;
Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian,
H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski,
J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg,
J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.;
Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.;
Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.;
Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill,
P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.
Gaussian 03, revision C.02; Gaussian, Inc.: Wallingford, CT, 2004.
4–8,13,14
methodology pioneered by Bach and co-workers
and by
7
,10,11
4,9
the groups of Freccero and Sarzi-Amadè
and of Houk.
Perhaps most usefully, this methodology can likely be used to
predict with a fair degree of confidence the outcomes of
epoxidations in which different stereoisomeric epoxide products
(
25) Hehre, W. J.; Radam, L.; Schleyer, P. v. R. ; Pople, J. A. Ab Initio
Molecular Orbital Theory; Wiley: New York, 1986.
26) Foresman, J. B.; Frisch Æ. Exploring Chemistry with Electronic Structure
Methods; Gaussian: Pittsburgh, PA, 1996.
27) (a) London, F. J. Phys. Radium 1937, 8, 397. (b) McWeeny, R. Phys.
(
(
(
23) Two recent examples: (a) Rychnovksy, S. Org. Lett. 2006, 8, 2895–
898. (b) Porco, J. A., Jr.; Su, S.; Lei, X.; Bardhan, S.; Rychnovsky, S. D. Angew.
Chem., Int. Ed. 2006, 45, 5790–5792.
ReV. 1962, 126, 1028. (c) Ditchfield, R. Mol. Phys. 1974, 27, 789. (d) Dodds,
J. L.; McWeeny, R.; Sadlej, A. J. Mol. Phys. 1980, 41, 1419. (e) Wolinski, K.;
Hilton, J. F.; Pula, P. J. Am. Chem. Soc. 1990, 112, 8251.
2
J. Org. Chem. Vol. 73, No. 9, 2008 3495

Koskowich et al.
this work to recent uses of calculated 13C NMR spectra for
natural product structure determination (ref 23).
solution. Finally, the ether was removed by rotary evaporation,
1
3
1
leaving a colorless liquid as the product. C, H NMR, COSY,
and HETCOR spectra were collected on a 400 MHz spectrometer.
They were assigned using three-bond and four-bond couplings.
Supporting Information Available: Cartesian coordinates,
Acknowledgment. P.R.R. thanks The Henry Dreyfus Teacher-
Scholar Awards Program administered by The Camille and
Henry Dreyfus Foundation, and also Swarthmore College for
support of this research. The authors also thank an anonymous
reviewer who carefully pointed out some important references
that had been missed in our original literature search, and also
another anonymous reviewer who pointed out the relation of
imaginary frequencies (where applicable), and energies of
1
all stationary points; experimental NMR spectra ( H NMR,
1
3
C NMR, HETCOR, COSY) of carene epoxidation product.
This material is available free of charge via the Internet at
http://pubs.acs.org.
JO702722G
3
496 J. Org. Chem. Vol. 73, No. 9, 2008