Effect of pressure on sterically hindered reactions with late transition states

Article abstract of DOI:10.1002/poc.718

The effect of high pressure is examined in two types of sterically congested reactions presenting late transition states: isopolar Diels-Alder cycloadditions of isoprene and quinones of various steric environment and conjugate additions of tert-butylamine and acrylic compounds. In the Diels-Alder cycloadditions, pressure has no enhanced kinetic sensitivity in sterically demanding reactions compared with unhindered ones. This is due to the structural closeness of the transition state and product. At variance, in the conjugate addition of amines, the effect of pressure is magnified with increasing steric congestion at the reaction centres. This is ascribed to enhanced electrostriction with removal of steric hindrance to ionization. Copyright

Full text of DOI:10.1002/poc.718

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2004; 17: 221–225
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.718
Effect of pressure on sterically hindered reactions with
late transition states
Badra Gacem and G e´ rard Jenner*
Laboratoire de Pi e´ zochimie Organique (UMR 7123), Facult e´ de Chimie, Universit e´ Louis Pasteur, 67008 Strasbourg, France
ABSTRACT: The effect of high pressure is examined in two types of sterically congested reactions presenting late
transition states: isopolar Diels–Alder cycloadditions of isoprene and quinones of various steric environment and
conjugate additions of tert-butylamine and acrylic compounds. In the Diels–Alder cycloadditions, pressure has no
enhanced kinetic sensitivity in sterically demanding reactions compared with unhindered ones. This is due to the
structural closeness of the transition state and product. At variance, in the conjugate addition of amines, the effect of
pressure is magnified with increasing steric congestion at the reaction centres. This is ascribed to enhanced
electrostriction with removal of steric hindrance to ionization. Copyright # 2004 John Wiley & Sons, Ltd.
KEYWORDS: steric hindrance; pressure; transition state; Diels–Alder reaction; Michael reactions
INTRODUCTION AND THEORY
observed that the reactant bulkiness may severely influ-
ence the reactivity under pressure in the sense that greater
steric hindrance induced an enhanced pressure kinetic
response, confirming the results of the earlier and sparse
Steric hindrance is an old and wide concept encompass-
ing the effect of bulkiness of reactants and the steric
accessibility of reaction centres on reagent approach. The
effect of steric hindrance on reaction rates is well estab-
lished, although quantitative aspects may still need refin-
ing. The major contributions to the quantification of these
12
related literature. However, in contrast, sterically hin-
dered Baeyer–Villiger oxidations of ketones did not
13
conform to this kinetic scheme.
The aim of the this work was to clarify the relationship
between pressure and steric hindrance in specific reac-
tions with late transition states.
1
effects are due to early workers, and more recent
2
contributions refined the concept. Among the various
routes explored for the analysis of steric perturbations, it
was proposed to consider activation parameters as in-
dicators of steric congestion (see Refs 10 and 18–20
Steric effects can be described by the Hammond
14
postulate. This popular concept states that for a single
reaction step, the geometry of the transition state resem-
bles the side to which it is closer in energy. This means
that, as an example, the transition state of an endothermic
one-step reaction will be more product-like than that for
an exothermic similar reaction.
Space filling by substituents of gradual higher bulki-
ness slows the reaction and induces higher energy re-
quirements. In other words, a steric perturbation in
reactions of the same family is associated with a change
in the position of the transition state. Its location is best
defined by the nuclear positions on the reaction coordi-
nate. From this point of view, the only, although indir-
ectly, measurable activation quantity is the activation
volume ÁV*, which can be determined from a kinetic
study as a function of pressure:
3
quoted in our previous review paper ). In this respect, the
reaction volume and the activation volume are certainly
the most convincing quantities as they are related to the
size of the substituents, the spacing of the reactants and
the modification of the virtual volume of the activated
complex.
At the time when pressure was considered an interest-
ing physical parameter in organic synthesis, it was soon
recognized that, in some condensations, sterically con-
gested reactions were more accelerated by pressure than
would be expected from comparison with similar reac-
4
tions free of steric hindrance. This enhanced sensitivity
to pressure was quantified later via the activation vo-
3,5
lume.
We started some time ago a programme directed
towards high-pressure synthesis of hindered functiona-
ꢀ
ÁV ¼ ꢁRTð@ln k=@PÞ
ð1Þ
T
6
lized compounds using Knoevenagel, Passerini,
7
8
Strecker and Stetter reactions and cyanoalkylations
9
10
ÁV* results basically from two volume changes: mole-
11
and miscellaneous reactions. In all these processes we
ꢀ
cular reorganization (ÁV ) and interactions of reactants
ꢀ
s
and activated complex with the medium (ÁV ):
"
*
Correspondence to: G. Jenner, Laboratoire de Pi e´ zochimie Organique
(
UMR 7123), Facult e´ de Chimie, Universit e´ Louis Pasteur, 67008
Strasbourg, France.
E-mail: jenner@chimie.u-strasbg.fr
ꢀ
ꢀ
ꢀ
"
ÁV ¼ ÁV þ ÁV
ð2Þ
s
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 221–225

2
22
B. GACEM AND G. JENNER
ꢀ
ꢀ
"
ÁV is the structural term and ÁV the electrostriction
s
term (rigorously considered, ÁV* may also contain
viscosity contributions, which are certainly negligible
in the pressure range used for the present kinetic deter-
minations).
Taking into account this version of the Hammond
postulate, this should mean that when the transition state
of similar reactions is structurally very close to the final
state, an increase in steric constraints should not induce a
magnified kinetic pressure effect compared with the
normal effect expected for this type of reaction. However,
this is true in so far as the volume shrinkage due to the
application of pressure is the only consequence of bond
reorganization: the geometric volume change resulting
from bond breaking and forming. For polar reactions with
late transition states, one may wonder whether the addi-
Scheme 1
Table 1. Solvent effect in the Diels–Alder reactions of
a
isoprene with quinones 1–3
6
0 k (dm mol s )
3
ꢁ1 ꢁ1
1
2
b
Solvent
ꢃ
1
2
3
ꢀ
tional activation volume term ÁV can be affected by
steric effects.
"
Ethyl acetate
Tetrahydrofuran
Chloroform
83
83
86
6.53
4.58
—
—
14.58
—
—
0.87
0.52
2.23
0.16
—
0.36
—
As candidate reactions, we chose the Diels–Alder reac-
tion of dienes with quinones and the conjugate addition of
amines to acrylic compounds, both being characterized by
late transition states. In our context, the lateness of a
transition state is meant as the closeness of transition state
and product in terms of volume, e.g. the nuclei of the
activated complex are already near their final positions.
In [4 þ 2] cycloadditions with normal electron demand
Dichloromethane 104
161
Ethanol
0.81
a
T ¼ (303.3 K for reactions involving 1 and 2 and 312.8 K for those using
3.
b
Cohesion energy density.
According to Table 1, the solvent effect is small,
showing an amplification ratio of 2–5 from low-polarity
solvents to ethanol. Visibly, the reactions are quasi-
isopolar despite the relative polarity of quinones. This
implies weakly polarized transition states justifying the
(
reaction of electron-donating dienes with electron-with-
drawing dienophiles), the lateness of the transition state
has been amply demonstrated in several ways, convin-
15
cingly by pressure kinetics. In these reactions, the
values of ꢀ ¼ ÁV*/ÁV lie between 0.8 and 1.0
ꢀ
ꢀ
assumption ÁV* ¼ ÁV , although ÁV might take a
s
"
R
small value. As expected, the fastest reactions involve
the quinones having at least one unsubstituted double
bond (1 and 2) whereas 3 and 4 react more slowly by a
factor of 10–15.
(
ÁV ¼ reaction volume based on partial molar vo-
R
lumes). If the reactants have a low polarity, the transition
ꢀ
and the ground states are nearly isopolar (ÁV ꢂ 0).
"
The Michael-like conjugate addition of amines to
acrylic derivatives leading to ꢁ-amino compounds was
shown to be strongly accelerated by pressure, pointing to
The pressure coefficient of reaction rates (Table 2)
yields the activation volume (Table 3). All ÁV* values
were homogenized for T ¼ 298 K with the aid of El’ya-
16
a late and tight transition state. In these reactions,
ꢀ
18
nov’s relationship. The results underscore the remark-
ꢀ
ꢀ
depending on the medium.
¼ ÁV /ÁV ꢂ 1 and ÁV can take significant values
s
R
"
able uniformity of ÁV* values, which are all in the range
3
ꢁ1
ꢁ
32 to ꢁ38 cm mol . These are typical of normal
Accordingly, we determined ÁV* values from the
pressure effect on rate constants in the Diels–Alder reac-
tion of isoprene and quinones with variable steric envir-
onment and the conjugate addition of tert-butylamine and
acrylic compounds of increasing steric complexity.
electron demand Diels–Alder reactions. The reaction
volume ÁV based on partial molar volumes of reactants
R
and adduct also takes similar values so that ꢀ is very close
to unity.
Three conclusions can be drawn from these results:
1
. Interestingly, the ÁV* values do not vary within
uncertainty limits according to the medium and,
accordingly, are free of any electrostatic term.
RESULTS
Diels–Alder reaction of isoprene and quinones
The [4 þ 2] cycloaddition of 1,3-dienes and quinones was
2. The fact that ꢀ is very close to unity demonstrates the
extreme lateness of the transition state for these
cycloadditions, pointing to a concerted mechanism,
although the structures of the transition states are
certainly not symmetrical with respect to the lengths
of the two newly forming ꢂ-bonds.
1
7
studied a long time ago. For our purpose, we chose four
diversely substituted 1,4-quinones, 1–4 (Scheme 1). We
investigated the kinetic solvent effect, the pressure effect
and the temperature effect on the second-order rate
constant in their reactions with isoprene.
3. Finally, ꢀ is independent of the steric requirements in
the transition state. As suggested by a referee, the
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 221–225

EFFECT OF PRESSURE ON STERICALLY HINDERED REACTIONS
223
a
Table 2. Pressure effect on the Diels–Alder reactions of isoprene with quinones in different solvents
6
0 k (dm mol s )
3
ꢁ1 ꢁ1
1
1
(ethanol)
2
(ethanol)
2
(chloroform)
3
(ethanol)
3
(chloroform)
4
(ethanol)
4
(chloroform)
Pressure (MPa)
.1
0
14.6
—
18.9
21.4
—
—
—
31.3
34.8
35.2
—
22.3
—
29.7
—
—
35.9
—
40.2
44.0
—
8.7
—
—
—
11.8
—
14.6
18.2
21.7
25.7
—
1.57
2.05
—
—
—
—
—
3.29
—
0.69
0.96
—
—
—
—
—
1.72
—
1.05
—
—
—
1.44
—
—
1.82
—
2.42
2.98
0.56
—
—
—
0.78
—
—
1.16
—
1.38
2.05
1
1
5
8
2
4
0
5
0
0
5
2
2
3
3
5
6
7
5.50
—
2.02
—
1
00
—
a
T ¼ 303.3 K for reactions involving 1 and 2, 312.8 K for those using 3 and 314.2 K with 4 as the dienophile.
a
Table 3. Activation and reaction volumes of the Diels–Alder reactions of isoprene with quinones in different solvents
1
(ethanol)
2
2
3
3
(ethanol) (chloroform) (ethanol) (chloroform) (ethanol) (chloroform)
4
4
Parameter
ꢀ
3
ꢁ1
)
ÁV
(ꢃ2.0 cm mol
ꢁ37.2
nd
—
ꢁ35.2
ꢁ37.4
ꢁ37.2
ꢁ36.5
ꢁ38.1
ꢁ35.4
ꢁ35.6
nd
—
ꢁ32.3
nd
—
ꢁ35.3
ꢁ36.5
2
98
3
ꢁ1
)
ÁV (at 298 K) (ꢃ1.5 cm mol
R
b
ꢀ
(ꢃ0.10)
nd, Not determined.
0.94
1.02
1.08
0.97
a
b
ꢀ
¼ ÁV*/ÁV
R
.
effect of steric hindrance on the fictive volume of the
transition state caused by the higher degree of sub-
stitution on the concerted C atoms is compensated by
the effect of longer partial bonds. Quantum chemical
calculations could probably provide conclusive infor-
mation on the steric effects reported here.
results endorse our initial hypothesis based on the
Hammond postulate, that in sterically hindered isopolar
reactions with late transition states the pressure effect is
the normal effect expected for unhindered reactions, with
similar activation volumes.
These conclusions are corroborated by the other activa-
Conjugate additions of tert-butylamine with
tion quantities determined from a thermodynamic study.
Inspection of the E, ÁS*, ÁG* values in Table 4 shows
little change in the four quinone reactions. In particular,
there is no enthalpic penalty in the most sterically hin-
dered reactions. The ÁS* values are highly negative,
especially for the cycloadditions involving 3 and 4. They
are in accord with highly ordered transition-state com-
plexes, confirming the concertedness of the mechanism.
From this study, it is concluded that hindered isopolar
and concerted Diels–Alder reactions are no more en-
dothermic than those of their unhindered analogues.
Steric hindrance is manifested only in reduced rate
constants for the dimethylbenzoquinone reactions. The
acrylic esters and nitriles
The conjugate addition of amines to ethylenic compounds
activated by electron-attracting groups consists in a nu-
cleophilic attack taking place at the ꢁ-position of the
olefinic double bond resembling the Michael reaction.
The reaction mechanism has been investigated in a pre-
16
vious study. The rate-determining step is nucleophilic
attack on the activated double bond of the acrylic reactant
with complete development of zwitterionic-like species
1
6,19
which undergo rapid proton transfer (Scheme 2).
The reaction is extremely sensitive to the steric acces-
sibility of the reaction centres. In particular, ꢄ- and
Table 4. Activation parameters of the Diels–Alder reactions of isoprene with quinones in different solvents
1
(ethanol)
2
(ethanol)
2
(chloroform)
3
(ethanol)
3
(chloroform)
4
(ethanol) (chloroform)
4
Parameter
ꢁ
1
E (ꢃ2.5 kJ mol
Ln A (ꢃ1.2)
)
67.1
15.3
66.2
15.2
68.1
15.4
69.5
13.5
70.9
12.9
64.4 68.1
10.3 12.3
ꢁ159 ꢁ143
109 108
ꢁ
1
ꢁ1
)
ÁS* (ꢃ9 J mol
K
ꢁ118
ꢁ118
ꢁ117
ꢁ133
ꢁ137
ꢀ
ꢁ1
)
ÁG
(ꢃ10 kJ mol
100
101.5
100.5
106.5
109
2
98
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 221–225

2
24
B. GACEM AND G. JENNER
The pressure effect was examined in the addition
reactions of tert-butylamine to unsaturated esters and
nitriles. The kinetic behaviour of three unsaturated methyl
esters is portrayed in Fig. 1. Clearly, the kinetic sensitivity
to pressure is higher for the reactions involving methyl
methacrylate and methyl crotonate. From the k values, the
overall activation volumes can be deduced (Table 6).
In contrast with the above quinone Diels–Alder reac-
tions, comparison of ÁV* and ÁV values shows that the
R
overall activation volume is much lower than the reaction
volume; it follows, therefore, that Eqn (2) is applicable.
This indicates that electrostriction is the dominant deter-
mining factor in high-pressure kinetics. Electrostriction
interactions depend from the evidence on steric con-
straints imposed by R and R as ÁV* becomes more
Scheme 2
Table 5. Kinetics of the addition of tert-butylamine to
acrylic compounds at ambient pressure
a
1
2
negative (lower ÁV* values) with restricted access to the
ꢄ and ꢁ reaction centres in the acrylic compound. Such a
result can be interpreted as a pressure removal of steric
k0 : k
X
T (K)
X
Solvent
Crotonic Methacrylic
230 815
880 1495
10
hindrance to ionization. In other words, in the conjugate
additions considered in this work, the pressure effect on
steric hindrance is not manifested, or only slightly, in
3
3
08.9
18.0
COOCH3 Acetonitrile
CN Ethanol
a
k
acrylic compound (R
0
¼ rate constant of the corresponding reaction involving the unhindered
¼ R ¼ H).
ꢀ
ÁV owing to the lateness of transition states, but is an
s
integral part of ÁV .
1
2
ꢀ
"
For the methacrylic ester, the methyl group destabilizes
the anionic part of the dipolar transition state even if its
steric effect is small. For the crotonate, the methyl
substituent has a strong steric effect and a weak electronic
effect. The stronger pressure kinetic enhancement for the
methacrylate Michel reaction is due to enhanced pressure
stimulation of the formation of ionic species. In this case,
pressure compensates for the destabilizing effect of the
methyl group on the anionic part of the dipole in the
transition state.
The high negative ÁV* values listed in Table 6 are also
in line with the ÁH* and ÁS* values determined for two
of these reactions.
At variance, it should be emphasized that the steric
environment in the amine is much less crucial for the
pressure kinetics with ÁV* values between ꢁ48 and
Figure 1. Pressure effect on relative rate constants in the
addition of tert-butylamine to methyl acrylates (CH CN,
3
3
08.9 K)
3
ꢁ1
ꢁ
53cm mol (Table 7). This is probably due to a high
ꢁ
-methyl groups decrease the reactivity owing to a
degree of C—N bond formation in the transition state and,
simply, is in harmony with a late transition state.
The ÁV* values listed in Table 6 can differ from those
determined for the traditional base-catalysed Michael
reaction. In a previous study, we examined the addition
combination of electronic and, mainly, steric effects.
Table 5 reports some results. Interestingly, independent
of the nature of X, crotonic derivatives react faster than
methacrylic compounds.
Table 6. Activation parameters of tert-butylamine conjugate additions (homogeneized for T ¼ 298 K)
a
b
R
c
d
X
R1
R2
Medium
ÁV*
ÁV
ÁH*
ÁS*
COOCH3
COOCH3
COOCH3
CN
H
CH3
H
H
H
H
H
CH3
H
CH3
CH CN
3
CH CN
ꢁ39
ꢁ25.5
ꢁ25.4
ꢁ26.5
ꢁ29.1
ꢁ28.6
29.3
35.1
nd
nd
nd
ꢁ176
ꢁ217
nd
ꢁ43
ꢁ50
ꢁ35
ꢁ57
3
CH CN
3
CH OH
nd
nd
3
C H OH
CN
2
5
a
b
c
d
3
ꢁ1
.
ꢃ3–5 cm mol
3
ꢁ1
.
ꢃ1.5 cm mol
ꢁ1
ꢃ2.5 kJ mol
.
ꢁ1 ꢁ 1
ꢃ10 J mol
K
.
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 221–225

EFFECT OF PRESSURE ON STERICALLY HINDERED REACTIONS
Table 7. Activation volumes in the addition of amines to
225
CONCLUSION
methyl methacrylate (308.9 K, solvent acetonitrile)
0
3
ꢁ1
)
This study shows that for sterically hindered reactions
featuring late transition states the dependence of the
pressure effect on the magnitude of steric congestion
varies with the reaction type. It is small or non-existent as
long as the volume of activation is a one-component
structural expression. If steric hindrance to solvation or
ionization is likely to intervene, the pressure kinetic
acceleration is magnified with increasing steric conges-
tion at the reaction centres. The pressure effect on
sterically hindered reactions can, therefore, be clearly
described by the Hammond postulate, expressed as fol-
lows: the more endothermic activated complex should
involve more contraction along the reaction coordinate
provided that there is some space for the transition state
to progress toward the final state and, if occurring,
stronger electrostriction owing to the corresponding
higher polarization of the reactive species.
R
R
ꢁÁV* (ꢃ3–5 cm mol
Pr
H
H
H
H
H
Pr
48
50
49
53
53
51
iPr
Bu
iBu
t-Bu
Pr
of nitromethane to two differently crowded vinyl ketones
catalysed by tetrabutylammonium fluoride. The forma-
tion of the carbanion in this reaction does not require
20
ꢀ
pressure assistance (ÁV ꢂ 0). Since ÁV* ꢂ ÁV , steric
"
effects do not induce a stronger pressure acceleration and
R
should not be reflected in ÁV* values. This is in fact the
ꢀ
case since ÁV*/ÁV is close to unity for the Michael
reactions involving either the unhindered methyl vinyl
R
2
0
ketone or the highly congested mesityl oxide. In the
latter reaction, steric hindrance is evidenced only in the
considerable slowing of the rate.
In addition, the clear predominance of electrostrictive
and steric effects in the conjugate addition reactions is of
high value for synthetic purposes. Activation volumes of
3
ꢁ1
ꢁ
50, ꢁ60, ꢁ70 cm mol (values at 298 K) relate to rate
amplification ratios of about 60, 130 and 300, respec-
tively, at 300 MPa.
EXPERIMENTAL
Kinetic determinations. Isoprene, acrylic esters and ni-
triles were distilled before use. Other reagents were used
as received. Kinetic measurements were performed as
follows. Weighed reagents and internal standard (1,2,3-
trimethoxybenzene or bibenzyl, depending on the reac-
tion) were introduced in flexible PTFE tubes. After
completing the residual volume with the solvent, the
REFERENCES
1
. Charton M. Top. Curr. Chem. 1983; 114: 57–91; Gallo R. Prog.
Phys. Org. Chem. 1983; 14: 115–163; Hansch C, Leo A, Taft RW.
Chem. Rev. 1991; 91: 165–195.
2. Isizawa J, Sakakibara K, Hiroto M. Bull. Chem. Soc. Jpn. 1996;
9: 1003–1015.
. Jenner G. J. Chem. Soc., Faraday Trans. 1 1985; 81: 2437–2460.
. Gonikberg MG, Zhulin VM, El’yanov BS. The Physics and
Chemistry of High Pressure. Society of Chemical Industry:
London, 1963; 212–218.
. Kiselev VD, Konovalov AI, Asano T, Kashaeva EA, Iskhakova
GG, Shibab MS, Medvedeva MD. J. Phys. Org. Chem. 2001; 14:
636–643.
. Jenner G. Tetrahedron Lett. 2001; 42: 243–245.
. Jenner G. Tetrahedron Lett. 2002; 43: 1235–1238.
. Jenner G, Ben Salem R, Kim JC, Matsumoto K. Tetrahedron Lett.
2
. Jenner G. High Press. Res. in press.
6
tube was closed and introduced into the thermostated
ꢄ
3
4
(
ꢃ0.1 C) high-pressure vessel. After reaction, the sol-
vents were removed in vacuo. The residue was then
1
analysed by H NMR (300 MHz). The kinetic data were
5
reproducible to better than 5%. Activation volumes were
determined in two ways: (i) from the initial slope of the
smoothed log k values against pressure and (ii) from
6
7
8
2
derivatization of the polynomial (log k ¼ a þ bP þ cP ).
003; 44: 447–449.
9
1
1
0. Jenner G. Tetrahedron 2002; 58: 4311–4317.
1. Jenner G. High Press. Res. 2002; 22: 511–514.
12. Jenner G. High Press. Res. 1992; 11: 21–32.
Determination of reaction volumes. Precision density
measurements permit the apparent molar volume, ÈV,
to be determined according to the known equation
13. Jenner G. Tetrahedron Lett. 2001; 42: 8969–8971.
1
1
4. le Noble WJ, Asano T. J. Am. Chem. Soc. 1975; 97: 1778–
781.
5. Jenner G. In Organic High Pressure Chemistry, le Noble WJ (ed).
1
È ¼ M =d ꢁ ð1000=CÞðd ꢁ d Þ=d
ð3Þ
V
w
0
0
0
Elsevier: Amsterdam, 1988; 143–203.
16. Jenner G. New J. Chem. 1995; 19: 243–245.
where M is the molecular weight of the solute and C its
w
1
1
1
7. Onishenko AS. Diene Synthesis. Israel Program for Scientific
Translations: Jerusalem, 1964; 85–273.
8. El’yanov BS, Vasylvitskaya EM. Rev. Phys. Chem. Jpn. 1980; 50:
169–184.
9. Shenhav H, Rapoport Z, Patai S. J. Chem. Soc. B 1970; 469–
molar concentration and d and d are the densities of the
0
solution and the solvent, respectively. Extrapolation of
È to zero concentration gives the value of the partial
V
molar volume. Density measurements were performed at
ꢄ
4
2
76; Rappoport Z, Peled P. J. Am. Chem. Soc. 1979; 101:
682–2693.
20. Jenner G. New J. Chem. 1999; 23: 525–529.
25.0 C in the corresponding solvent with a digital
densimeter (Parr DMA 602).
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 221–225