Experimental evidence to support viscosity dependence of rates of Diels-Alder reactions in solvent media

Article abstract of DOI:10.1021/jo034221+

This note is aimed at ascertaining whether rates of Diels-Alder reactions depend on the viscosity of solvent media in which the reactions are performed. On the basis of the data collected from the literature and in this laboratory, it is seen in general that the rates increase in the solvents with their viscosities ranging up to ～1.2 cP. In solvents possessing viscosities above 1.2 cP, a drop in the reaction rates is observed in all cases. The effect of temperature on the above phenomena is also examined.

Full text of DOI:10.1021/jo034221+

The role of viscosity on the kinetics of Diels-Alder
reactions was investigated by Firestone and Vitale and
others.6 These authors showed that the intramolecular
Diels-Alder cyclization of N-propargyl-9-anthramide
increases with viscosity of several glymes.6 It is important
to note that they carried out the reaction in tetraglyme,
the viscosity of which is two times higher than that of
glyme itself. These authors noted that the slope of
relative rates versus relative viscosities for the above
intramolecular Diels-Alder reaction was much greater
than the one noted during the kinetic study of Claissen
rearrangement in these solvents. Diels-Alder reactions
share some common features with the Claissen rear-
rangement like low solvent dependence, unrelated to
polarity. The dimerization of cyclopentadiene was also
examined with respect to the viscosities of different
solvents, and a linear increase in the dimerization rate
was seen.7,8 From these examples, it is clear that in the
lower end of the viscosity scale, the increase in the rates
is sharp with viscosity. The reaction rates level off at ∼1.3
cP before dropping with increasing viscosity above ∼1.3
cP. The 1,3 dipolar cycloaddition of diphenyldiazomethane
with ethyl phenylpropiolate was found to be enhanced
with the rise in viscosity of different solvents to about 1
cP. Above 1 cP, a decline in the reaction rates was obser-
ved.7 According to these authors, since pressure increased
viscosities, a part of the pressure-induced rate accelera-
tions of these reactions was due to the enhanced viscosity.
The above examples demonstrate that rates of Diels-
Alder reactions and viscosity are correlated with each
other. However, very recently, questions have been raised
whether viscosity contributes to the rate acceleration in
Diels-Alder reactions.9,10 Firestone and Swiss, on the
other hand, have attempted to establish that the rates
of dimerization of cyclopentadiene in several solvents are
dependent on the viscosities of the solvents.11 Though the
dependence of rates of Diels-Alder reactions on the
solvent viscosity has been supported and criticized,9,10 we
strongly feel that this issue, when supported by experi-
mental data, should receive attention of organic chemists
through a more appropriate vehicle of communication.
In view of the current work from this laboratory in
identifying the origin of forces responsible for rate
enhancement of Diels-Alder reactions in different sol-
vents and their salt solutions,12 we set out to assess the
Exp er im en ta l Evid en ce to Su p p or t
Viscosity Dep en d en ce of Ra tes of
Diels-Ald er Rea ction s in Solven t Med ia
Anil Kumar* and Suvarna S. Deshpande
Physical Chemistry Division,
National Chemical Laboratory, Pune 411008, India
akumar@ems.ncl.res.in
Received February 20, 2003
Abstr a ct: This note is aimed at ascertaining whether rates
of Diels-Alder reactions depend on the viscosity of solvent
media in which the reactions are performed. On the basis
of the data collected from the literature and in this labora-
tory, it is seen in general that the rates increase in the
solvents with their viscosities ranging up to ∼1.2 cP. In
solvents possessing viscosities above 1.2 cP, a drop in the
reaction rates is observed in all cases. The effect of temper-
ature on the above phenomena is also examined.
In a recent paper, we addressed the issue of using 5 M
LiClO4-diethyl ether (LPDE) solution in achieving maxi-
mum rate enhancement of several Diels-Alder reac-
tions.1 We demonstrated for the first time a substantial
fall in rates of these reactions when carried out in the
LPDE solutions of concentrations above 5 M. Our results
offered experimental evidence for using 5 M LPDE in the
synthetic work for maximizing rates and yields of Diels-
Alder reactions.2 Originally, the rate enhancement of
Diels-Alder reactions in 5 M LPDE was explained by
Dailey and Forman in terms of the Lewis acid catalysis.3
We attributed the subsequent rate drops in LPDE
solutions of above 5 M to the decreased Lewis acid
catalytic activity of Li+ due to the increased formation
of dietherate and monoetherate complexes.4 As a result
of the complexation, fewer effective “naked” Li+ cations
become available to act as a catalyst, thereby decreasing
the rates to a substantial degree. Second, it was noted
that the rate decrease above 5 M LPDE was also related
to very high viscosity of LPDE solution. No dramatic
increase in the solution viscosity of LPDE was seen up
to a concentration of 5 M. However, the viscosity of nearly
saturated LPDE (≈6.06 M) increased by about 800%, as
compared to that of pure DE.5 Therefore, a highly viscous
environment was considered to play an inhibitive role due
to diffusional problems causing the reaction rates to fall.
(6) (a) Firestone, R. A.; Vitale, M. A. J . Org. Chem. 1981, 46, 2160.
(b) Sternbach, D. D.; Rossana, D. M. Tetrahedron Lett. 1982, 23, 303.
(c) Dumas, T.; Hoekstra, M.; Pentaleri, M.; Liotta, D. Tetrahedron Lett.
1988, 29, 3745. (d) Dolbier, W. R., J r.; Seabury, J . J . Am. Chem. Soc.
1987, 109, 4393. (e) Valgimigli, L.; Ingold, K. U.; Lusztyk, J . J . Org.
Chem. 1996, 61, 1, 7947.
(7) Swiss, R. A.; Firestone, R. F. J . Phys. Chem. A 1999, 103, 5369.
(8) Coster, G.; Pfeil, E. Chem. Ber. 1968, 101, 4248.
(1) (a) Kumar, A.; Pawar, S. S. J . Org. Chem. 2001, 66, 7646. (b)
For other possible origins of forces responsible for accelerating Diels-
Alder reactions with original citations, see: Kumar, A. Chem. Rev.
2001, 101, 1.
(2) (a) Grieco, P. A.; Nunes, J . J .; Gaul, M. D. J . Am. Chem. Soc.
1990, 112, 4595. (b) For a recent review on LPDE, see: Sankararaman,
S.; Nesakumar, J . E. Eur. J . Org. Chem. 2000, 2003. (c) Waldmann,
H. In Organic Synthesis Highlights III; Mulzer, J ., Waldmann, H., Eds.;
Wiley-VCH: Weinheim, Germany, 1998, 205.
(9) le Noble, W. J .; Asano, W. J . Phys. Chem. A 2001, 105, 3428.
(10) Weber, C. F.; van Eldik, R. J . Phys. Chem. A 2002, 106, 6904.
(11) Firestone, R. A.; Swiss, K. A. J . Phys. Chem. A 2002, 106, 6909.
(12) (a) Kumar, A.; Pawar, S. S. Tetrahedron 2002, 58, 1745. (b)
Kumar, A.; Pawar, S. S. J . Phys. Org. Chem. 2002, 15, 131. (c) Kumar,
A.; Pawar, S. S. Tetrahedron Lett. 2001, 42, 8681. (d) Kumar, A.;
Phalgune, U.; Pawar, S. S. J . Phys. Org. Chem. 2001, 14, 577. (e)
Kumar, A.; Phalgune, U.; Pawar, S. S. J . Phys. Org. Chem. 2000, 13,
555. (f) Pawar, S. S.; Phalgune, U.; Kumar, A. J . Org. Chem. 1999,
64, 7055. (g) Kumar, A. Pure Appl. Chem. 1998, 70, 615. (h) Kumar,
A. J . Phys. Org. Chem. 1996, 9, 287. (i) Kumar, A. J . Org. Chem. 1994,
59, 4612. (j) Kumar, A. J . Org. Chem. 1994, 59, 230.
(3) Forman, M. A.; Dailey, W. P. J . Am. Chem. Soc. 1991, 113, 2761.
(4) Pocker, Y.; Buchholz, R. F. J . Am. Chem. Soc. 1970, 92, 2075
and subsequent papers by Pocker.
(5) Willard, H. H.; Smith, G. F. J . Am. Chem. Soc. 1923, 45, 286.
10.1021/jo034221+ CCC: $25.00 © 2003 American Chemical Society
Published on Web 06/04/2003
J . Org. Chem. 2003, 68, 5411-5414
5411

TABLE 1. Diels-Ald er Rea ction s In vestiga ted in
Con n ection w ith th e Viscosities of Solven ts
reactions
t (°C)
100
ref
(1) intramolecular Diels-Alder reaction
of N-propargyl-9-anthramide
(2) dimerization of cyclopentadiene
(3) diphenyldiazomethane +
ethyl phenyl propionate
6
25, 40 7, 8
30
7
(4) cyclopentadiene + methyl acrylate
solvent (I)
3-170
22
13
14
16
15
15
13
13
14
17
18
solvent (II)
solvent (III)
solvent (IV)
(5) cyclopentadiene + methyl vinyl ketone
(6) cyclopentadiene + methyl methacrylate
(7) cyclopentadiene + methyl trans-crotonate
(8) cyclopentadiene + acrylonitrile
(9) cyclopentadiene + (-)-menthyl acrylate
(10) cyclopentadiene + methyl (E)-R-
cyanocinnamate
25
30
30
3-139
3-66
22
F IGURE 1. Plots of log(endo/exo) for the reactions of (1)
cyclopentadiene with methyl acrylate (4) at 30 °C, ref 13, (1)
at 22 °C, ref 14, and (O) at 30 °C, ref 15, and (2) cyclopenta-
diene with acrylonitrile at (×) 22 °C ref 14.
30
25
(11) 2,3-dimethyl butadiene + 1,4-
80
80
19
20
naphthoquinone
(12) N-(2′-4′-dichloro-6′-oxo-2′,4′-cyclohexadien-1-
ylidine)-4-nitrobenzamide +
2,3-dimethyl butadiene
or + ethyl vinyl ether
(13) 2,3-dimethyl butadiene + 5-substituted
(hydroxy, methoxy, acetyl)-1,4-naphthoquinone
80
21
existing data as a step to resolve the issue of the role of
viscosity on rates of Diels-Alder reactions. The main
objective of this work, therefore, is to examine the rate-
viscosity data to ascertain whether the viscosity of the
solvent plays any effective role in governing the kinetics
of Diels-Alder reactions. This issue, if resolved, will be
helpful in improving the reaction conditions, including
at high pressures, where viscosity of the solvent is likely
to influence the reaction kinetics.
F IGURE 2. Log(endo/exo) - η plots from our experiments
under identical conditions: cyclopentadiene + methyl acrylate
reaction (4); (X) and cyclopentadiene + acrylonitrile (1)
reactions.
exo values for the reaction of cyclopentadiene with methyl
acrylate decreased monotonically with an increase in
viscosity.
Before analyzing the published data, it should be
mentioned here that we have chosen the viscosity of a
solvent at the temperature at which the rates of a given
reaction have been measured. Therefore, both the y- and
x-axes belong to the same temperature. The experimental
data examined in this study are summarized in Table 1.
We first examine the simple reaction of cyclopentadi-
ene with methyl acrylate in different solvents.13-16 The
authors observed the endo/exo values to be dependent on
the solvent polarity. The endo/exo values, when plotted
against viscosity of solvents (Figure 1), show an increase
up to ∼1 cP before declining sharply. Interestingly, the
endo/exo plot for this reaction carried out in the aqueous
mixtures of many organic solvents also shows similar
behavior. If the endo/exo data collected at high temper-
atures are plotted against the corresponding viscosities
of solvents, the maximum in the endo/exo values shifts
to ∼0.8 cP.
Similarly, the reaction of cyclopentadiene with acry-
lonitrile showed a noticeable decrease in endo/exo values
with respect to viscosities of the solvents at 22 °C.14 In
view of the puzzling situation created by the opposite
trends, we decided to repeat the reactions of cyclopen-
tadiene with methyl acrylate and with acrylonitrile under
the reaction conditions specified by the original authors.
These results are shown in Figure 2 in the form of log-
(endo/exo) plotted against the solvent viscosities. The
earlier experiments on the reaction of cyclopentadiene
with methyl acrylate by Berson et al.13 at 30 °C were
found to be reproducible within experimental accuracy.
However, we observed a complete reversal of trend when
we carried out the reaction under the conditions reported
by Nagakawa et al. at 22 °C.14 The log(endo/exo) values
for the reaction of cyclopentadiene with methyl acrylate
first increased up to ∼1cP viscosity and then gradually
declined up to the viscosity range of 1 to 2 cP. The log-
(endo/exo) - η plot for the reaction of cyclopentadiene
with methyl acrylate from Berson et al.13 is steeper than
the one obtained from Nagakawa et al.14 as shown in
Figure 2.
We, however, observed a serious contrast to this
observation. Nakagawa et al. measured endo/exo for this
reaction and for the reaction of cyclopentadiene with
acrylonitrile at 22 °C.14 These data showed that endo/
(13) Berson, J . A.; Hamlet, Z.; Mueller, W. A. J . Am. Chem. Soc.
1962, 84, 297.
Another noteworthy observation was with regard to the
reaction of cyclopentadiene with acrylonitrile at 22 °C.
We obtained a completely different curve for this reaction.
Like other reactions as shown above, the log(endo/exo)
(14) Nakagawa, K.; Ishii, Y. Ogawa, M. Tetrahedron 1976, 32, 1427.
(15) Cativiela, C.; Garcia, J . I.; Mayoral, J . A.; Salvatella, L. J . Chem.
Soc., Perkin Trans. 2 1994, 847.
(16) Cativiela, C.; Garcia, J . I.; Mayoral, J . A.; Avenoza, A.; Pereg-
rina, J . M.; Roy, M. A. J . Phys. Org. Chem. 1991, 4, 48.
5412 J . Org. Chem., Vol. 68, No. 13, 2003

F IGURE 5. Plots of log k for the reactions of 2,3-dimethylb-
utadiene with 1,4-naphthoquinone, 80 °C.
F IGURE 3. Plots of log k vs η for the reactions of cyclopen-
tadiene with methyl acrylate (0), with (-) menthyl acrylate
(9), and with methyl (E)-R-cyanocinnamate (3) in different
solvents.
F IGURE 6. Dependence of log k on the solvent viscosity: 2,3-
dimethyl butadiene + 5-hydroxy-1,4-naphthoquinone (b), +
5-methoxy 1,4-naphthoquinone (4), + 5-acetyl 1,4-naphtho-
quinone (X); N-(2′-4′-dichloro-6′-oxo-2′-4′-cyclohexadien-1- yli-
dine)-4-nitrobenzamide + ethyl vinyl ether (2), and + 2,3-
dimethyl butadiene (O).
increase up to ∼0.75 cP and then show a fall in the
viscosity range of 0.75-1.25 cP. The log k - η plots for
both these reactions are quite steep. However, the rates
of the reaction of cyclopentadiene with (-)-menthyl
acrylate in several solvents with the viscosity range from
0.25 to 1.5 cP show a slow increase before reaching nearly
a maximum at ∼1 cP. There is a regular drop in the rates
in the viscosity range between 1 and 1.5 cP.
F IGURE 4. Plots of log(endo/exo) - η for the reaction of
cyclopentadiene (O) methyl trans-crotonate and with (b)
methyl methacrylate at 30 °C.13
values increased with η with the maximum being around
η ≈ 0.5-0.6 cP. Again a clear decrease in log(endo/exo)
values was witnessed from 1 to 2 cP.
The reaction of cyclopentadiene with methyl meth-
acrylate offers higher proportions of exo-products (anti-
Alder rule) as compared to that with methyl acrylate
(Alder rule), which yields higher proportions of endo-
products.13 Similarly, the reaction of cyclopentadiene with
methyl trans-crotonate gives slightly higher proportions
of endo-products in polar solvents and higher proportions
of exo-products (borderline Alder rule) in nonpolar sol-
vents.13 The log(endo/exo) - η plots for these two reactions
at 30 °C are shown in Figure 4, where we see a very small
positive rise in the endo/exo ratio in the viscosity range
of 0.37-1.1 cP. As the reactions were not carried out in
the solvents having viscosity above 1.2 cP, it was not
possible to see any decline in the rates. The log(endo/
exo) values for both these reactions shown in Figure 4
do not include the data in methanol due to the points
falling beyond the trend shown herein. The log(endo/exo)
values for the reactions of cyclopentadiene with methyl
methacrylate and methyl trans-crotonate are reported as
-0.056 and 0.272, respectively.13
The endo/exo values for reaction of cyclopentadiene
with methyl vinyl ketone also showed first an increase
with the viscosity up to 1 cP followed by a sharp decline
between η values of 1 and 2 cP.
To get a clear picture, we plotted log k (k ) rate
constant) as a function of solvent viscosity (Figure 3) for
the reactions of cyclopentadiene with methyl acrylate,16
with (-)-menthyl acrylate17 and with methyl (E)-R-
cyanocinnamate.18 The rates for the reaction of cyclopen-
tadiene with methyl acrylate16 in methanol, acetone,
dioxane, and their aqueous mixtures, ethanol, formamide,
etc., increased with viscosity up to ∼0.75 cP before declin-
ing in the viscosity range of 0.75-1.25 cP. The reaction
of cyclopentadiene with methyl (E)-R-cyanocinnamate has
been reported in several aqueous mixtures of acetone and
dioxane. If log k for this reaction is plotted against the
viscosities of these aqueous mixtures (Figure 3), the rates
(17) Cativiela, C.; Garcia, J . I.; Mayoral, J . A.; Royo, A. J .; Salvatella,
L.; Assfeld, X.; Ruiz-Lopez, M. F. J . Phys. Org. Chem. 1992, 5, 230.
(18) Cativiela, C.; Mayoral, J . A.; Avenoza, A.; Peregrina, J . M.; Roy,
M. A. J . Phys. Org. Chem. 1990, 3, 414.
We then examined the viscosity dependence of the rate
constants for the reaction of 2,3-dimethyl butadiene with
J . Org. Chem, Vol. 68, No. 13, 2003 5413

in the η range of ∼0.3 to ∼1 cP before showing a sharp
decline in these solvents with viscosities varying between
∼1.5 and ∼2 cP. The aim of conducting temperature-
dependent studies was to examine whether changes in
temperature in the reaction brought any changes in the
plot shown in the foregoing studies. It is clear from the
plot that as the temperature increases and the solvent
viscosity decreases, a drop in the log(endo/exo) value is
witnessed. With the rise in temperature, the solvent
viscosity decreases (in this temperature range) and the
maximum log(endo/exo) value shifts to the left: log(endo/
exo) ≈ 0.9 at η ) 1.3 cP at 2 °C; log(endo/exo) ≈ 0.83 at
η ) 1.2 cP at 25 °C; and log(endo/exo) ≈ 0.82 at η ) 0.91
cP at 33 °C.
F IGURE 7. Plots of log(endo/exo) against the solvent viscosity
for the reaction of cyclopentadiene with methyl acrylate, 2 °C
(9), 25 °C (2), 33 °C (×).
From the above examples, it is clear that rates of
different Diels-Alder reactions vary with the solvent
viscosities. Both the rate increase and decrease are seen
in the viscosity range observed above. Though this
observation is not yet understood on a molecular level,
the contents described in this work clearly dispel the
confusion over the relationship between rates of Diels-
Alder reactions and solvent viscosities. The strength of
the observations made in the above study lies in the fact
that both the reaction rates or endo/exo ratios and the
viscosities of solvents in which the reactions were carried
out were measured at the same temperature, thereby
eliminating any discrepancy in the correlation. In the
above text, it has not been our main objective to put
forward any theory in support of data but to demonstrate
by plotting the rate-viscosity data in a consistent man-
ner whether a correlation between rates of Diels-Alder
reactions and the solvent viscosity exists at all. This work
shows that there is such a correlation. The rate enhance-
ment in the low viscosity range cannot be accounted for
in terms of current kinetic theory, as the bond-forming
reactions are independent of viscosity in this collision-
controlled regime. Firestone and co-workers have inter-
preted this behavior in terms of the vibrational activation
theory.6,7,11 Accordingly, high vibrational and low trans-
lational energies (favored with increasing viscosity)
promote the bond making. The translational mode of a
molecule is retarded with increasing viscosity, resulting
in a shifting of the translational to the vibrational mode.
At very high viscosities, the situation pertains to encoun-
ter-controlled regime. In this regime, the relative freedom
of movement of the reactants in the microenvironment
of the encounter pair becomes limited. In such a highly
viscous environment, reactants cannot see each other,
and the reaction is consequently slowed.
1,4-naphthoquinone in different solvents at 80 °C.19 These
data shown in Figure 5 indicate that the rates show a
steep increase in the solvents possessing viscosity in the
range of 0.4-0.9 cP. A decrease in rates can be witnessed
in the viscosity range of 1.4-2 cP.
Further, the rates of the reactions of N-(2′-4′-dichloro-
6′-oxo-2′,4′-cyclohexadien-1-ylidine)-4-nitrobenzamide with
2,3-dimethylbutadiene or with ethyl vinyl ether20 show
prominent maxima at ∼1.2 cP before declining sharply
in the viscosity range up to 2 cP (Figure 6). The log k -
η plots for the reactions of 2,3-dimethylbutadiene with
5-substituted (hydroxy, methoxy, acetyl)-1,4-naphtho-
quinone in a variety of solvents21 are shown in Figure 6,
where we note a modest increase in rates with the solvent
viscosity reaching a maximum at ∼1.2 cP followed by a
leveling of the curves and then a mild decrease in the
rate of the reactions.
From the above, it is clear that both the endo/exo ratios
and the rates of Diels-Alder reactions show dependence
on viscosities of solvents. Another important observation
is that although the viscosity increases by about 3 times,
the rates can vary by as much as 100 times. This is
contradictory to the general observation that for a
diffusion-controlled bimolecular reaction, the ratio of rate
change should be comparable with that of the viscosity.
No explanation is available to describe this observation,
which is entirely based on the examination of the
experimental rate data (Figures 3 and 6), and no theo-
retical assumptions are involved in this observation.
Though Berson et al.13 have carried out temperature-
dependent studies of the reactions of cyclopentadiene
with different dienophiles at three or four temperatures,
their data cannot be deemed the best data set in the
present context, as temperatures selected for each solvent
are not common. In view of this, we carried out the
reaction of cyclopentadiene with methyl acrylate in five
solvents (acetone, methanol, acetic acid, 2-methoxyetha-
nol, and 1-propanol) of graded viscosities at three differ-
ent temperatures, i.e., 2, 25, and 33 °C (Figure 7). It is
interesting to note that log(endo/exo) increases sharply
Ack n ow led gm en t. This investigation, supported by
a generous grant-in-aid (No. SP/S1/G-19/99) from DST,
New Delhi, India, is a result of thoughtful comments
made by an anonymous reviewer, who examined our
work cited as ref 1a.
Su p p or tin g In for m a tion Ava ila ble: Names of solvents
used for reactions mentioned in Table 1, experimental proce-
dure and values of log(endo/exo) (plotted in Figure 7) for the
reaction of cyclopentadiene with methyl acrylate carried out
in different solvents at three temperatures. This material is
available free of charge via the Internet at http://pubs.acs.org.
(19) Coda, A. C.; Desimoni, G.; Ferrai, E.; Righetti, P. P.; Tacconi,
G. Tetrahedron 1984, 40, 1611.
(20) Desimoni, G.; Faita, G.; Righetti, P. P. Tetrahedron 1991, 30,
5857.
(21) Desimoni, G.; Faita, G.; Righetti, P. P.; Tornaletti, N.; Visigalli,
M J . Chem. Soc., Perkin Trans. 2 1989, 437.
J O034221+
5414 J . Org. Chem., Vol. 68, No. 13, 2003