Kinetic effects in water and ethylene glycol. Application to high pressure organic synthesis

Article abstract of DOI:10.1039/b000241k

The kinetic effect of various Diels-Alder and Michael reactions is studied in water and ethylene glycol vs. organic solvents. The rate enhancement is considerable in water, much less in ethylene glycol. It is proposed that strong solvophobic interactions operate in water whereas the kinetic results in glycol are best explained by hydrogen bonding and polarity effects. From a synthetic point of view, use of the properties of water (hydrophobic interactions) or ethylene glycol (ionogenic medium) associated with the kinetic effect of high pressure may constitute an interesting multiactivation method to increase chemical reactivity. Examples of triactivation (high pressure catalytic Diels-Alder reactions in ethylene glycol) are given.

Full text of DOI:10.1039/b000241k

View Article Online / Journal Homepage / Table of Contents for this issue
Kinetic e†ects in water and ethylene glycol. Application to high
pressure organic synthesis
Ge

rard Jennera and Ridha Ben Salemb
zochimie Organique (CNRS UMR 7509), Institut de Chimie, Universite L ouis

a L aboratoire de Pie

Pasteur, 1 rue Blaise Pascal, 67008 Strasbourg, France. Fax: ] 33 3.88.41.68.15;
e-mail: jenner=chimie.u-strasbg.fr
` 
b L aboratoire de Synthese et Physicochimie Organique, Faculte des Sciences de Sfax, Route de
Soukra, 3038 Sfax, T unisia. Fax ] 216 42.74.437; e-mail: ridha.bensalem=fss.rnu.tn
Received (in Montpellier, France) 6th January 2000, Accepted 16th February 2000
The kinetic e†ect of various DielsÈAlder and Michael reactions is studied in water and ethylene glycol vs. organic
solvents. The rate enhancement is considerable in water, much less in ethylene glycol. It is proposed that strong
solvophobic interactions operate in water whereas the kinetic results in glycol are best explained by hydrogen
bonding and polarity e†ects. From a synthetic point of view, use of the properties of water (hydrophobic
interactions) or ethylene glycol (ionogenic medium) associated with the kinetic e†ect of high pressure may
constitute an interesting multiactivation method to increase chemical reactivity. Examples of triactivation (high
pressure catalytic DielsÈAlder reactions in ethylene glycol) are given.
In our quest to optimize the chemical yield of organic reac-
tions, we turned to multiactivation modes involving high pres-
sure as the basic parameter.1 In earlier papers we described
the combined e†ect of lanthanide catalysis and high pressure.2
We report herein the e†ect of pressure on organic reactions
carried out in aqueous solution and in water-like media with
the aim of utilizing their solvophobic properties as driving
forces for synthetic purposes (physicochemical activation).
Despite its common use water is a peculiar liquid under
ambient and around ambient conditions (0È90 ¡C, 0È200
MPa). The liquid state is characterized by extended hydrogen
bonding. This physical singularity has been exploited in
chemical reactions by Breslow3 and since water has become a
popular medium to enhance, sometimes considerably, the rate
constant of many organic reactions.4 The kinetic e†ect of
water is ascribed to several causes (often associated or inter-
related), which basically are: (i) enforced hydrophobic inter-
actions,5 (ii) hydrogen bonding to carbonyl and nitrile bonds
when available6,7 and (iii) electrostatic interactions.8,9
Hydrophobic interactions result from the associative e†ect
forcing the organic molecules together to minimize the waterÈ
hydrocarbon interfacial area. The reduction of the solvent
accessible surface area as part of the activation process con-
tributes to the reduction of the Gibbs energy of activation.7
The relative insolubility of the substrates conditions the mag-
nitude of the hydrophobic interactions. Any factor, such as
addition of additives or application of pressure, that may
inÑuence these interactions leads to an alteration in the rate
through salting-in or salting-out e†ects. As the aqueous reac-
tion proceeds via the tiny amount of dissolved reactants, it
would mean that the chemical yield is strongly dependent on
their solubility. In addition, the hydration sphere must also
accommodate the product as long as it forms during the reac-
tion. This leads to important limitations to the use of water as
a medium in synthetic organic chemistry. It is therefore desir-
able to look for other media that might also be capable of
favoring molecular aggregation while allowing better misci-
bility of organic liquids or solubilization of solid reactants.
Such media are sometimes called ““water-likeÏÏ and include
some diols and formamide. Ethylene glycol and formamide
are both polar solvents and strongly self-associate by hydro-
gen bonding.10 They possess sufficient cohesive force to
promote association of solvophobic molecules, however to a
lesser extent than water does.
On the other hand, we investigated in a former study the
kinetic e†ect of pressure on Michael and DielsÈAlder reac-
tions in aqueous solution.11 It was observed that the rate con-
stant of aqueous reactions was less sensitive to pressure as
compared to the rate constant determined in hydrocarbon sol-
vents. The result, though complex, was ascribed to a detrimen-
tal e†ect of pressure on hydrophobic interactions. However,
the values of the activation volume were found negative
enough in such a way that it could be interesting to consider a
possible combination of pressure activation and hydrophobic
e†ects to stimulate the reactivity of sluggish molecules.
The purpose of this paper is to report (i) the comparative
kinetic e†ect of water and ethylene glycol in selected reactions
and (ii) the combined synthetic e†ect of pressure and solvo-
phobic interactions exerted by these media in some reactions
that do not proceed at normal pressure in hydrocarbon sol-
vents.
Results and discussion
Kinetic behavior in water and diols
The rate of a bimolecular reaction depends on the speciÐc rate
constant and the concentration of reactants, that is their solu-
bility if the reaction is carried out in water. As an example,
Fig. 1 portrays the variation of the yield vs. concentration of
reactants in an aqueous Michael-like reaction (conjugate addi-
tion of tert-butylamine to methacrylonitrile).
For high values of w (corresponding to diluted aqueous
solutions of methacrylonitrile), the medium is homogeneous
or pseudo-homogeneous and the yields of the Michael adduct
are low. Best results are obtained at the saturation limit. The
yield at constant reaction time obviously depends on the
amount of hydrophobic surface that becomes accessible in the
transition state. As shown in Fig. 1, increasing the concentra-
tion of reactants beyond the saturation limits (low w values)
DOI: 10.1039/b000241k
New J. Chem., 2000, 24, 203È207
203
This journal is ( The Royal Society of Chemistry and the Centre National de la Recherche ScientiÐque 2000

View Article Online
conforms to the behavior shown in Fig. 1; the yields in
aqueous solution decrease with increasing concentration of
reactants beyond saturation. Whereas under such conditions
as shown in Table 1 there is no or little reaction in acetone
and methanol, the adducts are formed in 20È30% yield in
diols and formamide. These results are apparently related to
solvophobic interactions. It is difficult to invoke a mere elec-
trostatic or polarity origin since the reaction proceeds slowly
in methanol compared to the yields obtained in diols, even
though these alcohols have similar values for the cohesive
energy density (208 for methanol and 213 for ethylene glycol).
Hydrogen bonding between the two carbonyl bonds of tolu-
quinone and water or diols is probably a major cause of the
enhanced reactivity.12
To get a better idea of the e†ect of water and ethylene
glycol, we followed the kinetics of several DielsÈAlder reac-
tions, with normal electronic demand, with the diene bearing
successively a keto, ester and cyano group, with neutral
Fig. 1 Addition of tert-butylamine to methacrylonitrile in water (0.1
MPa, 50 ¡C, 24 h, V
: V
\ 1.5, total volume: 3.5 mL). Yield of
amine nitrile
b-aminonitrile (based on methacrylonitrile) as a function of w (volume
of water to volume of methacrylonitrile).
(
dimerization of isoprene) and inverse (HCCP ] styrene) elec-
tronic demand (Table 2, Scheme 2).
lowers the overall yield. Such limitation prompted us to inves-
tigate organic reactions in diols and formamide. The medium
e†ect on yields (Table 1) was studied in the DielsÈAlder reac-
tion of isoprene and toluquinone shown in Scheme 1 (the two
other possible regioisomers were not formed). From the
results of Table 1, it is clear that water, formamide, ethylene
glycol and other diols are perfectly suitable for this cyclo-
addition, as surmised. Glycerol is an exception; the lower
yield is due to a very poor solubilization of the quinone. This
is also the case of water; however, hydrophobic aggregation of
reactants is much higher in water than in glycerol. The result
The data listed in Table 2 indicate that the rate enhance-
ment is very variable in ethylene glycol as well as in aqueous
solutions. In glycol the highest values for the kinetic ratio
k /k are found in DielsÈAlder reactions of unsaturated car-
bonyl compounds (entries 1, 2, 5). In cycloadditions involving
styrene or acrylic compounds (methyl acrylate, acrylonitrile)
the rate enhancement is two or three times lower (entries 3, 4,
S
ref
7). However, the enhancement of the rates is only one order of
magnitude at best when comparing the rate constants in the
chloroalkane and the diol respectively. This is hardly compat-
ible with solvophobic interactions as the actual cause of rate
acceleration. It is better explained by simple polarity e†ects
and hydrogen bonding between the hydroxyl groups of ethyl-
ene glycol and the carbonyl group of the ketone or quinone in
their cycloaddition with furan or isoprene. In a previous paper
we examined the solvent dependence of the rate constant in
the dimerization of isoprene (entry 6).13 Under the same con-
Scheme 1
Table 1 Medium e†ect in the [4 ] 2] cycloaddition of toluquinone
(
TQ) to isoprenea
Medium
TQ/M
Isoprene/M
Yieldb/%
Acetone
Methanol
Ethylene glycol
0.5
0.5
0.5
0.5
0.5
0.5
0.1
0.2
1.0
1.0
1.0
1.0
1.0
1.0
0.2
0.4
0
1
24
29
5
21
37
28
1
,3-Propanediol
Glycerol
Formamide
Water
Water
a P (0.1 MPa), T (22.0 ¡C), reaction time (2 h). b Based on initial TQ.
Scheme 2
Table 2 Kinetics of DielsÈAlder reactions in water and ethylene glycola
k/dm3 mol~1 s~1
Ratio k /k
refc
S
Entry
Reactionb
T/¡C
CH Cl
Glycol
Water
Glycol
Water
2
2
1
2
3
4
5
6
7
Furan ] MVK
30.1
41.2
62.0
65.5
30.1
83.0
50.4
0.8 ] 10~7
1.2 ] 10~7
7.6 ] 10~7 d
2.3 ] 10~7
3.4 ] 10~5
1.3 ] 10~7 d
2.1 ] 10~6
1.2 ] 10~6
1.32 ] 10~6
3.8 ] 10~6
1.2 ] 10~6
4.5 ] 10~4
9.2 ] 10~7
1.7 ] 10~5
5.22 ] 10~5
6.91 ] 10~5
7.72 ] 10~4
1.7 ] 10~5
1.68 ] 10~2
1.5 ] 10~4
2.5 ] 10~3
15
11
5
5
13
7
650
575
1015
75
495
1150
1200
Isoprene ] MVK
Isoprene ] MA
Isoprene ] AN
Isoprene ] TQ
Isoprene (dimerization)e
HCCP ] styrene
8
a At ambient pressure. Concentration of reactants in water was 10~3 M. b MVK (methyl vinyl ketone), MA (methyl acrylate), AN (acrylonitrile),
TQ (toluquinone) HCCP (hexachlorocyclopentadiene). c k , k : rate constant in glycol or water and CH Cl (reference), respectively.
S
ref
2 2
d Chloroform was the solvent. e Pressure (20 MPa).
204
New J. Chem., 2000, 24, 203È207

View Article Online
ditions the value of this rate constant in ethylene glycol per-
fectly correlates with the k values determined in other polar
and nonpolar solvents according to the theory of regular solu-
tions: 9.2 ] 10~7 dm3 mol~1 s~1 in glycol (E \ 56.3,
T
d2 \ 213) and 10.8 ] 10~7 dm3 mol~1 s~1 in formamide
(
E \ 56.6, d2 \ 369).
T
The kinetic ratios in water vs. chloroalkanes of the DielsÈ
Alder reactions involving isoprene are much higher, in agree-
ment with the results reported for the corresponding
cyclopentadiene [4 ] 2] cycloadditions.14 In this case hydro-
phobic interactions are obviously the major reason for the
kinetic alterations. The highest values are observed for the
least hydrophilic molecules (HCCP, styrene, methyl acrylate)
Scheme 3
pressure is used as a synthetic parameter in aqueous reactions.
With the given reactant concentrations, the results obtained-
for bimolecular reactions such as DielsÈAlder cycloadditions
and Michael reactions are shown in Table 4 (Scheme 3).
The following observations are in order:
(
entries 3, 7). When ketones are involved electric polarization
with enhanced solvation of a more polar transition state than
the initial state must also be considered.8
To distinguish hydrophobic e†ects from other e†ects it is
common practice to use additives that increase or decrease
hydrocarbon solubility.15 It is well known that lithium halides
have a salting-out e†ect in aqueous solution whereas lithium
perchlorate or urea increase this solubility. However, in ethyl-
(
i) From a general point of view, Table 4 clearly shows that
water promotes DielsÈAlder and Michael reactions in agree-
ment with the literature results. Notable exceptions are dis-
closed in entries 8 and 9 where the addition of amines to
acrylic esters operates in acetonitrile but not in water. The
reason was discussed in a former paper and unambigously
ascribed to the prevalence of the reverse reaction.16
ene glycol as solvent, it was shown that LiClO produced a
4
salting-out e†ect.16 Our results are listed in Table 3. With the
exception of the rate decrease observed in the furan reaction
in aqueous LiCl solution, the relative rate constants deter-
mined in water are indicative of hydrophobic interactions.
(
ii) The modest DielsÈAlder reactivity of furan was not
solved until the late 1970s when Dauben and Krabbenhoft
reported an efficient high pressure route permitting use of less
reactive dienophiles.17 Furan does not add to MVK at
ambient pressure in hydrocarbon solvents (entry 1). Inter-
estingly, water promotes the cycloaddition even at atmo-
spheric pressure. Increasing pressure to 300 MPa leads to an
excellent yield. For comparison, the same yield is obtained in
dichloromethane only at pressures in excess of 1000 MPa.18
(iii) As expected pressure has a positive e†ect in bringing
some aqueous reactions to completion at 300 MPa whereas
the reactivity is zero in organic solvents. In entries 8 and 9, the
absence of reactivity persists even at 300 MPa, meaning that
The k values are higher in LiCl solutions and lower in LiClO
solutions. However, the kinetic trends are rather erratic if eth-
ylene glycol is used as the solvent.
4
High pressure synthesis in water
Considering the rate enhancements reported in Table 2, we
were prompted to investigate the efficiency of multiactivation,
considered as the combination of pressure and solvophobic
activation. However, from the synthetic point of view, it is
evident that high reactant concentrations are desired. This
means heterogeneous conditions, which also prevail when
Table 3 Salt e†ect upon the relative rate constants of DielsÈAlder reactions in water and in ethylene glycola
Water
Ethylene glycol
Reactionb
No additive
3 M LiCl
3 M LiClO4
No additive
3 M LiClO4
Furan ] MVK
Isoprene ] MVK
Isoprene ] MA
HCCP ] styrene
1
1
1
1
0.88
1.10
1.40
1.10
0.93
0.87
0.71
0.75
È
1
1
È
1.18
0.90
1.09
1
a See conditions of Table 2. b MVK (methyl vinyl ketone), MA (methyl acrylate), HCCP (hexachlorocyclopentadiene).
Table 4 High pressure synthesis in aqueous solutiona
Yield/%
Entry
Reactionb
Cc/M
T /¡C
t/h
0.1 MPa
300 MPad
1
2
3
4
5
6
7
8
9
Furan ] MVK
Isoprene ] PBQ
Isoprene ] TQ
Isoprene ] DMBQ
1.20
0.92
0.98
0.88
1.25
1.25
1.20
1.10
0.95
30
20
20
20
30
30
50
50
30
16
5
5
24
24
24
24
24
24
27
21
15
12
6
50
19
0
87(0)
82(0)
65(0)
47(0)
45(0)
100(0)
95(2)
0(23)
0(11)
CCN ] ButNH
2
CCN ] Pri(Me)NH
MCN ] Pr NH
2
MMA ] Pri NH
2
MMA ] ButNH
2
0
a No reaction in organic solvents (chloroform for entry 1, acetone for entries 2È4, acetonitrile for entries 5È9) at the same T and P of (0.1 MPa).
b MVK (methyl vinyl ketone), PBQ (p-benzoquinone), MMA (methyl methacrylate) DMBQ (2,6-dimethylbenzoquinone), CCN (crotononitrile),
MCN (methacrylonitrile), TQ (toluquinone). c Total concentration of reactants (reactants in equimolar amounts), in water (3.5 mL). d In paren-
theses yield obtained under the same conditions in the organic solvents listed in footnote (a).
New J. Chem., 2000, 24, 203È207
205

View Article Online
pressure is unable to shift the equilibrium toward the amino-
ester in water.
As a provisional conclusion, operation in water under pres-
sure can be a powerful means to drive organic synthesis. It
can be viewed as a combination of physical activation
sure is a fairly sensitive parameter. The yields obtained in the
diol at 300 MPa are comparable to those obtained in aque-
oussolution. However, in harmony with the results of Table 4,
ethylene glycol is a dissociating medium promoting gener-
ation of zwitterions in the same way as water does. The
Michael reaction between amines and acrylic esters is sluggish
in ethylene glycol. The yield is not enhanced by pressure, sug-
gesting again predominant reverse reactions. We checked this
reversal in entry 1. The b-aminoester synthesized in aceto-
nitrile at 300 MPa from methyl methacrylate and diisopropy-
lamine was dissolved in ethylene glycol and exposed to a 300
MPa pressure. After 24 h at 50 ¡C, 65% reversal was observed.
There was no retro-DielsÈAlder reactions in entries 7È14.
Entry 8 deserves particular interest since furans are very reluc-
tant to enter into [4 ] 2] cycloadditions unless highly reactive
dienophiles are involved. Thus, methyl vinyl ketone does not
add to furan under ambient pressure in ether, chlorinated sol-
vents, acetonitrile or methanol. An 18% yield is observed in
ethylene glycol. Formamide produces a comparable yield
(Table 6). Interestingly, the endo : exo ratio (61 : 39) is not
altered in water, ethylene glycol and formamide, whereas it
varies with the polarity of the medium.8 This would give
support to the existence of enforced hydrophobic interactions.
The yields in Table 6 are quite fair at 300 MPa, making the
method synthetically useful. In comparison, a yield of 90%
was reached only above 1000 MPa in ether or dichloro-
methane.18 Similar results relative to pressure acceleration in
ethylene glycol are obtained in entries 9 and 10 in Table 5.
However, the additional methyl group in toluquinone imposes
severe steric hindrance leading to low yields at 300 MPa. The
hetero-DielsÈAlder reaction (entry 11) is signiÐcantly pro-
moted by pressure in ethylene glycol. Lastly, DielsÈAlder reac-
tions with inverse electronic demand involving HCCP (entries
(
pressure) and physicochemical activation (hydrophobic stabil-
ization of the transition state). This is probably true for DielsÈ
Alder reactions. However, there may be another reason for the
speeding up of the Michael-like reactions reported in Table 4.
This type of reaction involves zwitterions whose generation is
strongly promoted by highly polar media.19 Another
comment is concerned with the general pressure e†ect in
aqueous solution. As evoked in the introduction, the activa-
tion volume relative to homogeneous aqueous reactions
(
*V E ) was invariably found to be higher than the activation
water
volume relative to the corresponding reactions carried out in
less polar organic solvents (*V E).11 This means a lower pres-
s
sure dependence of the rate constant in water and reveals the
limits of the pressure multiactivation process. An outstanding
example is the DielsÈAlder reaction of hexachloro-
cyclopentadiene and styrene described in a previous paper.11
High pressure synthesis in ethylene glycol
Ethylene glycol, which possesses an extensive hydrogen
bonding network, may also be used at high pressure.
However, in this case, at variance with aqueous reactions, the
solution is usually homogeneous. The results are listed in
Table 5 (Scheme 4).
In the conjugate addition of amines to acrylic nitriles pres
1
2È14) give low yields of adducts at 300 MPa in ethylene
glycol.
Some of these results (Table 5) contrast with those obtained
in aqueous solutions (Table 4) while others show similarities:
i) For normal electronic demand DielsÈAlder reactions
(
carried out in ethylene glycol (entries 7È11 in Table 5), the
yield ratio is modiÐed in nearly comparable proportion as in
water when pressure is varied from ambient to 300 MPa.
However, the inverse electronic demand reactions (entries
12È14) are less sensitive to pressure.
(ii) The conjugate addition of amines to acrylic compounds
proceeds in the same way in both media.
A possible explanation may be o†ered with the involvement
of polarity and hydrogen bond donating e†ects. Polarity
e†ects are strongly manifested in Michael reactions, either in
water or diol solution. Enforced hydrophobic e†ects would
accelerate the rate of normal electronic demand DielsÈAlder
reactions in aqueous solution and to a much lesser degree in
ethylene glycol in accordance with the comments relative to
Table 2. Hydrogen bonding and polarity e†ects would also
a†ect these reactions (for a more detailed insight see our pre-
vious paper11).
Scheme 4
Table 5 High pressure addition reactions in ethylene glycola
Yield/%
Entry
Reactionb
T/¡C
0.1 MPa
300 MPa
1
2
3
4
5
6
7
8
9
1
1
1
1
1
MMA ] Pri NH
2
50
30
30
30
50
30
30
30
20
20
30
80
80
80
3
22
5
26
8
17
24
18
14
0
0
0
n.d.
20
2
26
21
67
51
100
85
86
59
6
28
2
11
29
MMA ] BuiNH
2
The rather modest results obtained in the DielsÈAlder reac-
CCN ] ButNH
2
CCN ] Pri(Me)NH
MCN ] ButNH
2
MCN ] Pr NH
2
Table 6 Solvent e†ect in the cycloaddition of methyl vinyl ketone
(MVK) and furana
Isoprene ] TQ
Furan ] MVK
DMFu ] PBQ
Yield/%
% endo product
0
1
2
3
4
DMFu ] TQ
MVK ] EVE
Medium
d2
0.1 MPa 300 MPa 0.1 MPa 300 MPa
HCCP ] 3,3-dimethyl-1-butene
HCCP ] 2-pentene
HCCP ] 4-methyl-1-pentene
Dichloromethane 104
0
0
17
18
86
97
87
È
È
60
62
60
74
52
61
62
61
Methanol
208
a Concentration of reactants (0.5È0.6 mmol each), volume of glycol (3 mL).
b MMA (methyl methacrylate), CCN (crotononitrile), MCN (methacrylonitrile),
TQ (toluquinone), MVK (methyl vinyl ketone), DMFu (2,5-dimethylfuran),
PBQ (p-benzoquinone), EVE (ethyl vinyl ether), HCCP (hexa-
chlorocyclopentadiene).
Ethylene glycol
Formamide
Water
213 18
369 23
547 23
a T (30 ¡C), t (16 h).
206
New J. Chem., 2000, 24, 203È207

View Article Online
Table 7 Lanthanide catalysis in water-like media at high pressurea
volume of the tube is adjusted with water, ethylene glycol or
the organic solvent as required. The tube is shaken for about
one minute and then introduced in the pressure vessel thermo-
regulated at the given temperature. In the case of pressure
runs, pressure is gradually generated by an oil driven intensi-
Ðer up to 300 MPa. Reaction pressure is then released and the
solution transferred into a separating funnel. The organic
layer is collected by two successive extractions with diethyl
ether and dried. After usual removal of the solvent the crude
Reactionb
T/¡C
30
Solvent
Catalyst
Yield/%
MVK ] EVE
Chloroform
Glycol
Glycol
Chloroform
Glycol
Glycol
None
None
Yb(OTf)
None
None
Yb(OTf)
None
None
12
28
100
3
40
100
0
3
3
Crotonaldehyde ] EVE
DMFu ] TQ
60
20
Dichloromethane
Glycol
6
residue is directly analyzed by 1H NMR (200 MHz, CDCl )
Glycol
Yb(OTf)3
26
3
and the yield determined from relative intensities of character-
a P (300 MPa), t (24 h). The catalytic experiments were carried out with ytter-
bium triÑate (2.5% molar). b MVK (methyl vinyl ketone), EVE (ethyl vinyl
ether), DMFu (2,5-dimethylfuran), TQ (toluquinone).
istic protons vs. methoxy groups of the internal standard.
Kinetic measurements are carried out as thoroughly described
previously.11
Acknowledgements
This work was carried out as part of the cooperation program
CNRS/DGRST (Tunisia).
tions listed in Table 5 prompted us to associate chemical acti-
vation via Lewis acid catalysis. Organic lanthanide
compounds as mild Lewis acid catalysts are particularly suit-
able in hetero cycloadditions involving substrates possessing
carbonyl groups.2 Table 7 reports the notable improvements
obtained when conducting the experiments in ethylene glycol
under 300 MPa in the presence of ytterbium triÑate. However,
the generality of such a triactivation process is limited.20
References
1
G. Jenner, in High Pressure Chemistry, Biochemistry and Material
Sciences, ed. R. Winter, Kluwer Academic Publishers, Dordrecht,
1999, pp. 291È311.
2
(a) G. Jenner, High Press. Res., 1995, 13, 321; (b) G. Jenner,
T etrahedron L ett., 1995, 36, 233; (c) G. Jenner, T etrahedron L ett.,
1
996, 37, 3691.
R. Breslow, Acc. Chem. Res., 1991, 24, 159.
A. Lubineau, J. Auge and Y. Queneau, Synthesis, 1994, 741.
Conclusions
3
4
5
The results reported in this paper point to the rather complex
e†ect of water in organic reactions. From a synthetic point of
view, it is related to the amount of hydrophobic interactionsÈ
in relation with reactant solubilitiesÈand to the magnitude of
the activation volume.9 Combination of pressure and hydro-
phobic activation may be an interesting method to increase
the reactivity of reluctant molecules.
Substitution of ethylene glycol for water leads in some cases
to interesting results. For isopolar reactions such as DielsÈ
Alder reactions, use of ethylene glycol as a medium is beneÐ-
cial only if carbonyl groups are involved, presumably via
hydrogen bonding and polarity e†ects rather than through
pure solvophobic interactions. In fact, our results show that
such interactions do not exist or at the best, very moderate.
The results shown in this paper should give full adherence
to high pressure as a synthetic parameter, possibly combined
with additional activation methods. The multiactivation pro-
cedure can efficiently complement other synthetic strategies as
clearly demonstrated in the cycloaddition of furan.

W. Blokzijl, M. J. Blandamer and J. B. Engberts, J. Am. Chem.
Soc., 1991, 113, 4241.
J. F. Blake, D. Lim and W. L. Jorgensen, J. Org. Chem., 1994, 59,
6
7
8
03.
S. Otto, W. Blokzijl and J. B. Engberts, J. Org. Chem., 1994, 59,
5372.
8
9
G. Jenner, T etrahedron L ett., 1994, 35, 1189.
G. Jenner and R. Ben Salem, Rev. High Press. Sci. T echnol., 1998,
7, 1265.
1
0
1
(a) J. Thomas and D. F. Evans, J. Phys. Chem., 1970, 74, 3812; (b)
A. Ray, J. Am. Chem. Soc., 1969, 91, 6511.
G. Jenner, J. Phys. Org. Chem., 1999, 12, 619.
1
12 T. Dunams, W. Hoekstra, M. Pentaleri and D. Liotta, T etra-
hedron L ett., 1988, 29, 3745.
1
1
3
4
G. Jenner and R. Ben Salem, T etrahedron, 1997, 53, 4637.
N. K. Sangwan and H. J. Schneider, J. Chem. Soc., Perkin T rans.
2
, 1989, 1223.
C. J. Rizzo, J. Org. Chem., 1992, 57, 6382.
1
1
5
6
G. Jenner, T etrahedron, 1996, 52, 13557.
17 W. Dauben and H. O. Krabbenhoft, J. Am. Chem. Soc., 1976, 98,
1992.
1
8
J. Rimmelin, G. Jenner and P. Rimmelin, Bull. Soc. Chim. Fr.,
1
978, 461.
1
2
9
0
G. Jenner, New J. Chem., 1995, 19, 173.
G. Jenner and R. Ben Salem, Rev. High Press. Sci. T echnol., in
the press.
Experimental
Ambient and high pressure runs are carried out as follows. A
3.5 mL Ñexible PTFE tube is Ðlled up with 1,2,3-tri-
methoxybenzene (standard) and substrates as required. The
Paper b000241k
New J. Chem., 2000, 24, 203È207
207