Hydrophobic effects are dominant over secondary orbital interactions for a simple Diets-Alder reaction in salt solutions

Article abstract of DOI:10.1021/ol060741q

The stereoselectivity ratios for a Diels-Alder reaction between cyclopentadiene with methyl trans-crotonate carried out in salt solutions demonstrate the dominance of hydrophobic effects over secondary orbital interactions.

Full text of DOI:10.1021/ol060741q

ORGANIC
LETTERS
2006
Vol. 8, No. 10
2199-2202
Hydrophobic Effects Are Dominant over
Secondary Orbital Interactions for a
Simple Diels−Alder Reaction in Salt
Solutions
Diganta Sarma and Anil Kumar*
Physical Chemistry DiVision, National Chemical Laboratory, Pune 411008, India
a.kumar@ncl.res.in
Received March 27, 2006
ABSTRACT
The stereoselectivity ratios for a Diels−Alder reaction between cyclopentadiene with methyl trans-crotonate carried out in salt solutions
demonstrate the dominance of hydrophobic effects over secondary orbital interactions.
The pioneering contribution made by Rideout and Breslow1
on the special role of water on Diels-Alder reactions has
stimulated numerous investigations on promoting otherwise
sluggish organic reactions and understanding the forces
responsible for remarkable rate enhancement.2 It was also
demonstrated by Breslow and co-workers3 that some salts
promote Diels-Alder reactions, while others inhibit them.
The salts that promote Diels-Alder reactions are prohydro-
phobic and those which inhibit these reactions are antihy-
drophobic in nature.3g,h The former class of salts enhances
hydrophobic effects, while the latter decreases them. The
possible origin of forces includes hydrophobic packing,
solvent pressure, hydrogen bonding, hydrophobic hydration,
and salting-out (S-O) and salting-in (S-I) effects, etc.2a,4 As
part of our studies to delineate and quantify the origin of
these forces, we have investigated the reactions of cyclo-
pentadiene 1 with methyl acrylate 2a and with methyl
methacrylate 2b (Scheme 1) in aqueous salt solutions.5,6
Scheme 1. Reactions of Cyclopentadiene 1 with Methyl
Acrylate 2a, Methyl Methacrylate 2b, and Methyl
trans-Crotonate 2c
(1) Rideout, D. C.; Breslow, R. J. Am. Chem. Soc. 1980, 102, 7816.
(2) For example, see: (a) Breslow, R. Acc. Chem. Res. 1991, 24, 159.
(b) Fringuelli, F.; Tatichi, A. Dienes in Diels-Alder Reactions; Wiley: New
York, 1990. (c) Pindur, U.; Lutz, G.; Otto, C. Chem. ReV. 1993, 93, 741.
(d) Organic Synthesis in Aqueous Media; Grieco, P. A., Ed.; Blackie:
Glasgow, 1998. (e) Breslow, R. Water as SolVent for Chemical Reactions;
Anastas, P. T., Willimason, T. C., Eds.; Oxford University Press: Oxford,
1998. (e) Lindstrom, U. M. Chem. ReV. 2002, 102, 2751. (f) Breslow, R.
In Structure and ReactiVity in Aqueous Solutions; Cramer, C. J., Truhlar,
D. G., Eds.; ACS Symposium Series 568; American Chemical Society:
Washington, D.C., 1994 and references therein.
(3) (a) Breslow, R.; Maitra, U.; Rideout, D. C. Tetrahedron Lett. 1983,
24, 1901. (b) Breslow, R.; Maitra, U.; Rideout, D. C. Tetrahedron Lett.
1984, 25, 1239. (c) Breslow, R.; Guo, T. J. Am. Chem. Soc. 1988, 110,
5613. (d) Breslow, R.; Rizzo, C. J. J. Org. Chem. 1991, 113, 4340. (e)
Rizzo, C. J. J. Org. Chem. 1992, 57, 6382. (f) Breslow, R. In Structure
and ReactiVity in Aqueous Solutions; Cramer, C. J., Truhlar, D. G., Eds.;
ACS Symposium Series 568; American Chemical Society: Washington,
D.C., 1994; and references therein. (g) Breslow, R.; Connors, R. V. J. Am.
Chem. Soc. 1995, 117, 6601. (h) Breslow, R.; Guo, T. J. Am. Chem. Soc.
1995, 117, 9923.
Original contributions from Breslow and his group on the
spectacular role of water and its salt solutions1,3 and
subsequent investigations from our group on the salt effect
(4) Kumar, A. Chem. ReV. 2001, 101, 1.
(5) Pawar, S. S.; Phalgune, U.; Kumar, A. J. Org. Chem. 1999, 64, 7055.
(6) Deshpande, S. S.; Kumar, A. Tetrahedron 2002, 58, 8759.
10.1021/ol060741q CCC: $33.50
© 2006 American Chemical Society
Published on Web 04/18/2006

on the kinetics of Diels-Alder reactions have established
that salts such as LiCl, NaCl, NaBr, KCl, MgCl2, CaCl2,
and Na2SO4 enhance the rates and endo/exo ratios of Diels-
Alder reactions in water, whereas guanidinium chloride
(GnCl) and LiClO4 lower them.5-7 Accordingly, LiCl, NaCl,
CaCl2, MgCl2, etc. are the S-O agents, while GnCl, LiClO4,
urea, etc. are the S-I agents. These effects are discussed in
detail elsewhere.2a,4 The salt effects on the reactions of 1
with 2a and with 2b follow the expected results within the
framework of Breslow’s work on the salting effect and its
relation with kinetic data of Diels-Alder reactions.2a,3 The
salt effect on the endo/exo ratios for the reaction of 1 with
2b was less pronounced than that of 1 with 2a.
In this paper, we demonstrate a simple Diels-Alder
reaction between 1 with methyl trans-crotonate 2c (Scheme
1) for which the salt effects on the endo/exo ratios are
opposite to what Diels-Alder reactions in general exhibit.
It will be shown that “apparent contrasting” salt effects are
actually normal effects with respect to the correct definitions
of endo and exo stereoisomers.
In organic solvents, the reaction of 1 with 2a is endo-
selective, while that with 2b is exo-selective. The reaction
of 1 with 2c in organic solvents gives the endo/exo ratio as
nearly 1:1.8,9 Some comments on the definitions of endo and
exo stereoisomers are in order here. In the case of 2c, the
products are defined with respect to the stereochemistry of
carboxylate as is the convention. Here, the conventional endo/
exo refers to endocarboxy/exocarboxy. Thus, for this reaction, the
endocarboxy/exocarboxy ratio is equivalent to exomt/ endomt. The
subscripts carboxy and mt stand for carboxylate and methyl
group, respectively. It should be noted here that for the
reaction involving 2c, endo and exo stereoisomers should
be referred to with explicit use of carboxylate or methyl
groups, as the case may be. A simple statement of endo and
exo does not convey appropriate meanings of stereoisomers
formed as a result of the reaction of 1 with 2c.
The endocarboxy/exocarboxy ratios (the same ratio for exomt/
endomt is implied) shown in Figure 1 were obtained for the
reaction of 1 with 2c carried out in aqueous solutions of LiCl,
NaCl, KCl, MgCl2, CaCl2, GnCl, LiClO4, and urea up to 5
M (except for KCl up to 4 M due to its restricted solubility
in water). 1H NMR and GC analyses were extensively used
to determine the endocarboxy/exocarboxy ratios of the adduct.10
The precision of the endocarboxy/exocarboxy ratios as determined
from the average of triplicate reactions was observed as
(0.03. The maximum deviation at some salt concentration
did not exceed (0.05. An examination of Figure 1 shows
that the salts such as LiCl, NaCl, KCl, MgCl2, and CaCl2
that are known to increase the endocarboxy/exocarboxy ratios of
simple reactions of 1 with 2a and with 2b showed opposite
effects on the reaction of 1 with 2c.
Figure 1. Plots of (endo/exo) vs salt concentration for the reaction
of 1 with 2c in KCl (0), LiCl (b), NaCl (4), MgCl2 (3), CaCl2
(1), urea (O), GnCl (2), and LiClO4 (×).
ratios from that in water (endocarboxy/exocarboxy )1.28) alone.
This reaction in n-heptane, a nonpolar solvent, gave the
endocarboxy/exocarboxy ratio as 0.97. This decrease is monotonic
in nature but shows the signs of tapering off at higher salt
concentration possibly due to the solubility of these salts in
the solutions. The endocarboxy/exocarboxy ratios are consistently
lower in these salts solutions when compared to that in water.
MgCl2 is the most effective salt in decreasing the endocarboxy
/
exocarboxy ratios of this reaction, while KCl is the least
effective.
When the reactions were performed in urea, GnCl, and
LiClO4, higher endocarboxy/exocarboxy ratios of the products were
obtained, as compared to that in water. As noted previously,
these results are opposite to the trends obtained for the
reactions of 1 with 2a and 1 with 2b or other Diels-Alder
reactions in general. LiClO4 showed the least effect on the
increase of endocarboxy/exocarboxy ratios. This observation is
again in contrast to the observations made in the cases of
reactions of 1 with 2a and with 2b. GnCl, which was the
least effective salt for the reactions of 1 with 2a and 1 with
2b, was noted to be the most effective of the three salts
studied. Urea is the most effective agent for this reaction.
Intrigued by the above observations in water solutions,
we decided to perform the reaction in other two self-
associated solvents, namely ethylene glycol and formamide.
The reaction in ethylene glycol gave an endocarboxy/exocarboxy
ratio of 2.14, which was reduced to 1.18 in 5 M LiCl-
ethylene glycol solution (see Table 1). A gradual reduction
in the endocarboxy/exocarboxy ratios is seen from pure ethylene
glycol to 1, 3, and 5 M LiCl solutions. On the other hand,
this reaction gave higher endocarboxy/exocarboxy ratio in 5 M
LiClO4-ethylene glycol as compared to that in ethylene
glycol. The endocarboxy/exocarboxy ratio of 2.09 obtained in pure
formamide was lowered to 1.53 in 5 M LiCl solution of
formamide. Further, when the reaction was carried out in 5
As can be seen in the plots shown in Figure 1, LiCl, NaCl,
KCl, MgCl2, and CaCl2 decrease the endocarboxy/exocarboxy
(7) Kumar, A.; Pawar, S. S. Tetrahedron 2002, 58, 1745.
(8) Berson, J. A.; Hamlet, Z.; Mueller, W. A. J. Am. Chem. Soc. 1962,
84, 297.
(9) Kobuke, Y.; Fueno, T.; Furukawa, J. J. Am. Chem. Soc. 1970, 92,
6548.
(10) Evans, D. A.; Chapman, K. T.; Bisaha, J. J. Am. Chem. Soc. 1988,
110, 1238.
2200
Org. Lett., Vol. 8, No. 10, 2006

endocarboxy/exocarboxy ratio due to the S-O, while LiClO4
decreased them owing to the S-I effects. But this is not the
case for the reaction of 1 with 2c. It is therefore clear that
the reaction of 1 with 2c cannot be interpreted in terms of
salting phenomena.
Table 1. Endocarboxy/Exocarboxy Ratiosa,b for the Reaction of 1
with 2c in Ethylene Glycol (EG), Formamide (FM), and Their
Salt Solutions
endocarboxy
/
endocarboxy
/
These interesting salt effects can be explained within the
framework of hydrophobic packing as proposed by Breslow
in the past.2a Thus, accordingly, the lower endocarboxy/exocarboxy
values for this reaction in aqueous LiCl, NaCl, KCl, MgCl2,
and CaCl2 as compared to in water shown above suggest
that exomt is less than endomt in water. Since the methyl group
is more hydrophobic than the carboxylate group, the hydro-
phobic packing of the methyl group in the transition state
would lead to higher endomt, which corresponds to higher
exocarboxy. These prohydrophobic salts enhance the hydro-
phobic packing of diene and dienophile as compared to water
alone. This is exactly what we have noted in this investigation
of the reaction of 1 with 2c in the above salt solutions. On
the other hand, the antihydrophobic salts such as GnCl,
LiClO4, and urea decrease the hydrophobic packing resulting
in lower endomt corresponding to lower endocarboxy. This
suggests that hydrophobic effects dominate during the
stabilization of the geometry of transition state rather than
secondary orbital interactions13 as advocated in the past. For
the past three decades, secondary orbital interactions have
been employed to explain the stereoselectivites of Diels-
Alder reactions. However, the explanation of stereoselec-
tivities of Diels-Alder reactions on this basis been ques-
tioned.14 It is established from the theoretical calculations
that the atoms presumed to be involved in secondary orbital
interactions are situated relatively far (ca. 2.8 Å) in the
corresponding transition-state structures. 14,15 This finding is
against the existence of these interactions as the calculated
geometries for transition-state structures of Diels-Alder
reactions can lead to the estimation of the presence of other
attractive effects.16 Garcia et al. have deliberated on this issue
to conclude that the hypothesis of secondary orbital interac-
tions is not necessary to explain the stereoselectivity ratios.14
Accordingly, a combination of solvent effects, steric interac-
tions, hydrogen bonds, electrostatic interactions, etc. can be
used in their place.
solvent system
exocarboxy
solvent system
exocarboxy
EG
2.14
1.08
1.33
1.18
2.16
3 M LiClO4-EG
5 M LiClO4-EG
FM
5 M LiCl-FM
5 M LiClO4-FM
2.57
2.83
2.09
1.53
2.75
1 M LiCl-EG
3 M LiCl-EG
5 M LiCl-EG
1 M LiClO4-EG
a Also implies the same ratios for exomt/endomt
data with a standard deviation of 0.03.
.
b An average of triplicate
M LiClO4-formamide, the endocarboxy/exocarboxy ratio in-
creased to 2.75.
As for conformity, LiCl, NaCl, KCl, MgCl2, and CaCl2
should have increased the endocarboxy/exocarboxy ratios of this
reaction as they did for the reactions of 1 with 2a and with
2b. These results indicate that these salts show a different
behavior and no longer act as the S-O agents for the reaction
of 1 with 2c. Similarly, LiClO4, GnCl, and urea have
increased the endocarboxy/exocarboxy ratios and therefore no
longer act as the S-I agents. This observation is noted in all
three solvents.
The S-O and S-I phenomena are directly related to the
solubility of reactants in a salt solution when compared to
that in the pure solvent.11,12 We attempted to confirm our
observations on endocarboxy/exocarboxy ratios by first measuring
solubility of 2c in water, aqueous LiCl, and LiClO4 (Table
2). The solubility of 2c in aqueous 5 M LiCl was substantially
Table 2. Solubility of 2c in Different Solvent Systems
solubility of
2c (mM)
solubility of
2c (mM)
solvent system
solvent system
water
aq 5 M LiCl
aq 5 M LiClO4
36.70
5.56
39.10
EG
18.21
11.70
19.31
1 M LiCl-EG
1 M LiClO4-EG
A similar change in the endocarboxy /exocarboxy selectivity
has been observed for the reaction of 1 with 2b (Scheme 1);
however, this reaction is exo-selective in water and its salt
solutions.6 In the absence of methyl group substitution, as
in the reaction of 1 with methyl acrylate 2a (Scheme 1),
hydrophobic interactions become less important and SOI
lower in aqueous 5 M LiCl than that in water alone,
suggesting the role of LiCl as a S-O agent. On the other
hand, the solubility of 2c in aqueous 5 M LiClO4 was higher
than that in water, indicating LiClO4 to be a S-I agent. The
solubility of 2c in 1 M LiCl-ethylene glycol is less than
that noted in ethylene glycol alone, while 2c is more soluble
in 1 M LiClO4-ethylene glycol than in ethylene glycol itself.
The infinite miscibility of 2c in formamide and its salt
solutions did not allow us to measure its solubility in these
media. The solubility data of 2c in water and ethylene glycol
media demonstrate that LiCl should have increased the
(13) (a) Hoffman, R.; Woodward, R. B. J. Am. Chem. Soc. 1965, 87,
4388. (b) Hoffman, R.; Woodward, R. B. J. Am. Chem. Soc. 1965, 87,
4389. (c) Saur, J.; Sustmann, R. Angew. Chem., Int. Ed. Engl. 1980, 19,
779.
(14) Garcia, J. I.; Mayoral, J. A.; Salvetella, L. Acc. Chem. Res. 2000,
33, 658 and references therein.
(15) (a) Jorgensen, W. L.; Lim, D.; Blake, J. F. J. Am. Chem. Soc. 1993,
115, 2936. (b) Ruiz-Lopez, M. F.; Assfeld, X.; Garcia, J. I.; Mayoral, J.
A.; Salvetella, L. J. Am. Chem. Soc. 1993, 115, 8780. (c) Garcia, J. I.;
Mayoral, J. A.; Salvetella, L. Tetrahedron, 1997, 53, 6057. (d) Calvo-
Losada, S.; Suarez, D. J. Am. Chem. Soc. 2000, 122, 390. (e) Beno, B. R.;
Houk, K. N.; Singleton, D. A. J. Am. Chem. Soc. 1996, 118, 9984.
(16) Singleton, D. A. J. Am. Chem. Soc. 1992, 114, 6563.
(11) (a) Shoor, S. K.; Gubbins, K. E. J. Phys. Chem. 1969, 73, 498. (b)
Collins, K. D.; Washabaugh, M. W. Q. ReV. Biophys. 1985, 18, 323. (c)
McDevitt, W. F.; Long, F. A. J. Am. Chem. Soc. 1952, 74, 1773.
(12) Grover, P. K. Ryall, P. K. Chem. ReV. 2005, 105, 1.
Org. Lett., Vol. 8, No. 10, 2006
2201

directs the reaction to be endocarboxy selective leading to a
higher endocarboxy/exocarboxy ratio in water (Table 1).5
It is important to state here that we have carried out Diels-
Alder reactions of 1 with other dienophiles, but such types
of salt effects both in aqueous and nonaqueous salt solutions
were not observed. In summary, it is possible to invoke the
role of hydrophobic packing over that of secondary orbital
interactions to explain the stereoselective ratio of a particular
Diels-Alder reaction in aqueous salt solutions. The “apparent
contrasting” salt effects could be interpreted in terms of
hydrophobic effects.
Acknowledgment. We thank an anonymous reviewer for
making important suggestions with regard to the interpreta-
tion of the results discussed here. D.S. thanks the Council
for Scientific and Industrial Research, New Delhi, for
awarding him a research fellowship.
Supporting Information Available: Experimental de-
1
tails, stereochemical assignments, and H NMR spectra of
concerned compounds. This material is available free of
charge via the Internet at http://pubs.acs.org.
OL060741Q
2202
Org. Lett., Vol. 8, No. 10, 2006