Gas-phase Diels-Alder cycloaddition reaction in the presence of methanol and water vapor

Article abstract of DOI:10.1002/poc.577

Hydrogen bonding effects of protic solvents, apart from bulk properties, on the reaction rate of the cycloaddition of cylopentadiene and vinyl acetate in the presence of water and methanol in the gas phase were investigated. The results showed that methan

Full text of DOI:10.1002/poc.577

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2003; 16: 79–83
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.577
Gas-phase Diels±Alder cycloaddition reaction in the pre-
sence of methanol and water vapor
M. R. Gholami* and B. A. Talebi
Department of Chemistry, Sharif University of Technology, Tehran, Iran
Received 29 October 2001; revised 23 August 2002; accepted 28 August 2002
ABSTRACT: Hydrogen bonding effects of protic solvents, apart from bulk properties, on the reaction rate of the
epoc
cycloaddition of cylopentadiene and vinyl acetate in the presence of water and methanol in the gas phase were
investigated. The results showed that methanol increases the reaction rate in the gas phase more than the water. This is
attributed to the stronger hydrogen bonding effect of methanol in this phase. Ab initio and semi-empirical calculations
show that methanol stabilizes the transition state of the reaction more than water. This arises from two different
origins, distribution of charge and geometry of the hydrogen bond. Copyright  2002 John Wiley & Sons, Ltd.
Additional material for this paper is available from the epoc website at http://www.wiley.com/epoc
KEYWORDS: cycloaddition reaction; hydrogen bonding; solvent effect; theoretical study; gas phase kinetics
INTRODUCTION
No experimental work has been reported on hydrogen
bonding effects of protic solvents on DA reactions
separately from their bulk properties. For this purpose,
we have carried out DA reactions of cyclopentadiene and
vinyl acetate in the gas phase as a model reaction in the
presence of water and methanol vapor (Scheme 1).
Theoretical calculations were also carried out to achieve
a better understanding of this effect, especially to
compare the hydrogen bonding abilities of methanol
and water in the stabilization of the TS of this DA
reaction.
Interest in the role of solvents in organic reactions has
increased over the last few years. A considerable
influence of solvents on chemical processes (reaction
rate, selectivity, etc.) has been well established in the last
decade.1–5 Studies have revealed that a number of
organic reactions proceed more rapidly in aqueous
solutions than in organic solvents.6–11 The use of water
as a solvent has obvious environmental and economic
advantages over other solvents.
Diels–Alder (DA) cycloaddition reactions have be-
come a benchmark for theoretical and experimental
studies of solvent effects.2–5,12 In previous studies, it has
been reported that the acceleration of DA reactions in
water is mainly a result of a combination of hydrophobic
interactions and hydrogen bonding.13 For experimental
investigation of the influence of hydrogen bonding on DA
reactions, Otto et al. compared the DA reaction of
cyclopentadiene (CPD) and methyl vinyl ketone with the
reaction of the corresponding sulfone in some protic and
aprotic solvents.14 Blake et al. calculated extent of
hydrogen bonding to the dienophile and also its effect on
the stability of the transition state (TS) in the DA reaction
using ab initio calculations. As in previous calculations,
they noted that hydrogen bonding is particularly sensitive
to small charge variations, and consequently the largest
solvent effects will be found in hydrogen bond donor
solvents.2
RESULTS AND DISCUSSION
The experimental results of the kinetic study of the
reaction are given in Table 1. The reaction rate was
obtained in the presence of methanol, water, a mixture of
methanol and water vapor and without any solvent vapor.
For the three vapor phases, the reaction rates were higher
than that in the absence of any solvent vapor.
Since the reaction was carried out in the presence of a
saturated vapor of solvents, it may be mistakenly
concluded that this increase could arise from the pressure
effect of the solvent vapor. We selected the reaction
*Correspondence to: M. R. Gholami, Department of Chemistry, Sharif
University of Technology, Tehran, Iran.
E-mail: gholami@sharif.edu
Contract/grant sponsor: Sharif University of Technology.
Scheme 1
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 79–83

80
M. R. GHOLAMI AND B. A. TALEBI
Table 1. Pseudo-®rst-order rate constants of the reaction in
the gas phase at 296.1 K
Conditions
Kobs (sÀ1
)
Without any solvent vapor
2.70 Â 10À4
4.60 Â 10À4
7.92 Â 10À4
3.47 Â 10À4
In presence of water vapor
In presence of methanol vapor
In presence of both water and methanol vapor
conditions to be pseudo-first order, and first-order
reactions are independent of pressure.15 Furthermore,
when we compare the vapor pressure of water
(20.9 mmHg at 296.1 K) with that of a vapor mixture of
water and methanol (19.1 mmHg at 296.1 K) (for a 2:1
volume ratio of liquid phase of water and methanol,
obtained using the Ramsay–Yang method),16 we see they
are nearly the same. Hence we expect that the reaction
rate will not change on going from water vapor to a
mixture vapor of water and methanol unless there is a
difference in chemical origin. Consequently, pressure
cannot explain the variation of the reaction rate.
A common feature of methanol and water is that both
are protic solvents and could have hydrogen bond
interactions with the hydrogen bond sites on other
molecules. The dienophile of this reaction (vinyl acetate)
has an oxygen atom bonded to the vinyl group. Hence this
oxygen could have a short-length interaction with a protic
solvent molecule through hydrogen bond formation.
Therefore, the existence of hydrogen bond interactions
can increase the reaction rate in the presence of water,
methanol or their vapor mixture.
The difference between the reaction rates in the vapor
phase of methanol and water is due to the difference in
the hydrogen bond donor ability of the solvents in the gas
phase. If the strength of hydrogen bonding is a key factor
in increasing of the reaction rate, from the kinetic results
obtained (Table 1) it seems that methanol is a stronger
hydrogen bond donor than water in the gas phase. This
may be expected from the greater acidity of alcohols than
water in the gas phase.
Figure 1. Structural speci®cation of the reaction in the
absence of solvent molecules according to AM1 and HF
calculations
were considered for the CPD and VAC. Full geometry
optimization was performed at the semi-empirical and
HF/3–21G levels, and the other single-point calculations
were carried out on the HF/3–21G optimized structures.
Some structural information is given for the reaction in
the gas phase without any solvent vapor and for reaction
in the presence of water and methanol molecules in Figs
1–3.
It is well known that the energy of the hydrogen bond
depends on the YÁÁÁH distance and the X—HÁÁÁY angle
(where X is a hydrogen bond donor and Y is a hydrogen
bond accepting atom). Based on the YÁÁÁH distance, all
hydrogen bonds can be divided into strong, medium and
weak. An alternative approach to an analysis of the
hydrogen bond energy is provided by the electron density
distribution.19 It has been demonstrated that the value of
To investigate the effect of hydrogen bonding on the
activation energy of this reaction theoretically, we carried
out ab initio and AM1 calculations using Gaussian 98.17
For this purpose we considered complexes of the TS and
dienophile (vinyl acetate) with a solvent molecule near
the hydrogen bond site. Since the ether-type COC oxygen
of vinyl acetate was closer to the reaction center than the
carbonyl oxygen, it was considered as a hydrogen bond
acceptor site.
The synchronous transit-guided quasi-Newton (STQN)
method implemented by Schlegel et al.18 was used to
locate the TS. This method was performed with the QST2
and QST3 options. The location of the TS was done using
QST3, which requires three molecule specifications: the
reactants, the product(s) and an initial structure for the
transition state. In all calculations, identical conformers
Figure 2. Structural speci®cation of the reaction in the
presence of water molecules according to AM1 and HF
calculations
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 79–83

GAS-PHASE DIELS–ALDER CYCLOADDITION WITH PROTIC SOLVENTS
81
the electron density in the bond critical point correlates
with the bond energy.20 Therefore, a comparison of the
H-bond strength may also be carried out based on the
charges of atoms contributing to the hydrogen bond.
According to AM1 calculations, the activation energy
of the reaction is lower in the presence of methanol
(Table 2), but the water molecule geometrically is
somewhat better partitioned than methanol within the
hydrogen bonds of the TS (Figs 2 and 3). Mulliken
charges can help us to compare atomic charges of the
hydrogen bonding atoms in complexes of the dienophile
and the TS with solvent molecules. The calculated atomic
charges are presented in Tables 3 and 4. As atomic
charges calculated using AM1 show, going from vinyl
acetate to the TS of the reaction in the presence of water
is accompanied by more charge separation on atoms
contributing to the hydrogen bonding. In contrast, the
methanol molecule better distributes charges on the TS to
the rest of the complex. Therefore, according to the AM1
calculation, better distribution of charges is a reason why
the methanol molecule stabilizes the TS of the reaction
and decreases the activation energy more than the water
molecule (Table 2).
Figure 3. Structural speci®cation of the reaction in the
presence of methanol molecules according to AM1 and HF
calculations
Table 2. Calculated energy of the reactant and TS of the reaction with and without the solvent molecule
Energy (hartree)
Conditions
AM1
HF/3–21G HF/6–31G HF/6–31G* HF/6–31‡‡G* B3LYP/6–31‡‡G*
Pure gas phase
Reactant
TS
À0.04733 À494.6986 À497.2557 À497.4668
À497.4814
À497.4072
0.0742
À573.4918
À573.4241
0.0677
À500.5870
À500.5450
0.0420
À577.0079
À576.9713
0.0366
0.00155 À494.6449 À497.1845 À497.3949
TS–reactant
Reactant
TS
0.04888
0.0537
0.0712
0.0719
In presence of water
À0.14176 À570.2994 À573.2415 À573.4718
À0.09829 À570.2492 À573.1770 À573.4073
TS–reactant
0.04347
0.0502
0.0645
0.0645
In presence of methanol Reactant
TS
À0.14173 À609.1111 À612.2485 À612.5024
À612.5215
À612.4564
0.0651
À616.3148
À616.2816
0.0332
À0.11987 À609.0654 À612.1867 À612.4407
TS–reactant
0.02186
0.0457
0.0618
0.0617
Table 3. Calculated Mulliken charges of the dienophile and TS of the reaction in the presence of water molecules (Fig. 2)
Mulliken charge
AM1
HF/3–21G
HF/6–31G
HF/6–31G*
HF/6–31‡‡G*
B3LYP/6–31‡‡G*
Dienophile
O-5
H-6
O-7
À0.32459
0.23199
À0.46192
À0.77013
0.42925
À0.77215
À0.83818
0.48984
À0.86186
À0.70966
0.50772
À0.92584
À0.59289
0.66691
À1.11621
À0.44477
0.59655
À1.05007
Transition state
O-5
À0.30837
À0.75423
À0.82023
À0.69902
À0.47724
À0.32328
H-6
0.24294
0.42413
0.48594
0.50534
0.63554
0.60073
O-7
À0.47036
À0.77692
À0.86622
À0.92982
À1.11756
À1.05954
D (charge)a
O-5
0.01622
0.01094
À0.00844
0.01591
À0.00512
À0.00476
0.01795
À0.00391
À0.00436
0.01064
À0.00237
À0.00398
0.11565
À0.03137
À0.00135
0.12148
0.00419
À0.00946
H-6
O-7
a
D(charge) = charge on TS À charge on dienophile.
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 79–83

82
M. R. GHOLAMI AND B. A. TALEBI
Table 4. Calculated Mulliken charges of the dienophile and TS of the reaction in the presence of methanol molecules (Fig. 3)
Mulliken charge
AM1
HF/3–21G
HF/6–31G
HF/6–31G*
HF/6–31‡‡G*
B3LYP/6–31‡‡G*
Dienophile
O-5
À0.30793
0.24857
À0.39165
À0.75465
0.40091
À0.70171
À0.80399
0.45480
À0.78321
À0.68583
0.48155
À0.78331
À0.54079
0.59692
À0.78398
À0.40576
0.55989
À0.68037
H-6
O-7
Transition state
O-5
À0.30758
À0.73126
À0.78594
À0.67877
À0.43693
À0.27245
H-6
0.25152
0.42107
0.47834
0.50363
0.58609
0.61520
O-7
À0.39584
À0.71498
À0.79819
À0.79679
À0.77448
À0.68756
D(charge)a
O-5
0.00035
0.00295
À0.00419
0.02339
0.02016
À0.01327
0.01805
0.02354
À0.01498
0.00705
0.02208
À0.01349
0.10386
À0.00883
0.00950
0.13331
0.05531
À0.00719
H-6
O-7
a
D(charge) = charge on TS À charge on dienophile.
In spite of the AM1 results presented above, HF and
DFT calculations show that distribution of charges on the
TS occurs better in the presence of water molecules
(Tables 3 and 4). For example, the hydrogen atom is more
positive in the presence of methanol moleculs. On the
other hand, hydrogen bond formation in the presence of
methanol has a better geometric condition. Bond lengths
and angles derived from these calculations support this
argument. Methanol has a shorter hydrogen bond length
in TS than in the dienophile, whereas water has a greater
hydrogen bond length (Figs 2 and 3). From the change in
hydrogen bond length of water with TS relative to
dienophile, it would be expected that water would
increase the activation energy of the reaction. However,
since water decreases the negative charge on O-5 and
positive charge on H-6 (key atoms in the hydrogen bond),
the net result is that water also decreases the activation
energy of the reaction. The net contribution of the charge
distribution and the geometry can be summarized in the
energetics of the reaction (Table 2). Therefore, according
to HF calculations, geometric specification is the reason
why methanol has a stronger hydrogen bond interaction
with the TS relative to reactants than water.
dimer immediately before use. Methanol and vinyl
acetate (VAC) was distilled before use and water was
redistilled in a quartz distillation unit.
Kinetic measurements. The reaction was studied
pseudo-first order with respect to CPD. The concentration
of VAC was selected to be at least 10 times that of CPD.
The quantitative analysis of CPD was done by gas
chromatography with a Porapack-Q column and flame
ionization detection. Since the concentration of VAC was
nearly constant (for the pseudo-first-order conditions of
the kinetic study), it was also considered as an internal
standard in chromatographic analysis.
The reaction was carried out in a two-necked flask, one
neck being used to connect the flask to a vacuum system
with a controller valve and the other having a septum
inlet for injection of reactants and solvents into the
vessel. For each measurement, the system was evacuated
to near 1 mmHg, then the system was saturated with the
required solvents and finally the reactants were injected
separately into the vessel. The second inlet was also used
for sampling of the reaction mixture with a gas syringe at
different times to measure the concentration of CPD.
In summary, hydrogen bond donor solvents such as
methanol and water could affect the rate of DA reactions
through short-length interactions. According to experi-
mental and theoretical results obtained in this study, in
the gas phase methanol increases the rate of this reaction
more than water. Also, it can be concluded that, since the
hydrogen bond donor ability of water is higher than that
of methanol in the liquid phase,21 hydrogen bond
cooperativity could exist in liquid-phase water which
induces dramatic effects on the solute properties through
hydrogen bond interactions.
Acknowledgements
Gratitude is expressed to Sharif University of Technol-
ogy for support of this research. Thanks are due to Dr A.
L. Cooksy (San Diego State University) for his valuable
advices.
REFERENCES
1. Reichardt C. Solvents and Solvent Effects in Organic Chemistry
(2nd edn). Verlag Chemie: Wainheim, 1988.
2. Blake JF, Lim D, Jorgensen WL. J. Org. Chem. 1994; 59: 803.
3. Cativiela C et al. J. Phys. Org. Chem. 1992; 5: 230.
4. Ruiz-Lopez MF et al. J. Am. Chem. Soc. 1993; 115: 8780.
EXPERIMENTAL
Materials. Cyclopentadiene (CPD) was prepared from its
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 79–83

GAS-PHASE DIELS–ALDER CYCLOADDITION WITH PROTIC SOLVENTS
83
5. Kumar A, Phalgune U, Pawar SS. J. Phys. Org. Chem. 2000; 13:
555.
6. Gholami MR, Habibi YA. J. Chem. Res. (S) 1999; 226.
7. Gajeweski JJ. Acc. Chem. Res. 1997; 30: 219.
8. Wijnen JW, Engberts JBFN. J. Org. Chem. 1997; 62: 2039.
9. Wijnen JW, Steiner RA, Engberts JBFN. Tetrahedron Lett. 1995;
30: 5389.
10. Wijnen JW, Zavarise S, Engberts JBFN. J. Org. Chem. 1996; 61:
2001.
11. Blokzjil W, Blandamer MJ, Engberts JBFN. J. Am. Chem. Soc.
1991; 113: 4241.
12. Gholami MR, Habibi YA. Int. J. Chem. Kinet. 2000; 32: 431.
13. Blokzijl W, Engberts JBFN. J. Am. Chem. Soc. 1992; 114: 5440.
14. Otto S, Blokzijl W, Engberts JBFN. J. Org. Chem. 1994; 59: 5372.
15. Holbrook KA, Pilling MJ, Robertson SH. Unimolecular Reactions
(2nd edn). Wiley: Chichester, 1996.
16. Salzberg HW, Morrow JI, Cohen SR. Laboratory Course in
Physical Chemistry. Academic Press: New York, 1968.
17. 17Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA,
Cheeseman JR, Zakrzewski VG, Montgomery JA, Stratmann RE,
Burant JC, Dapprich S, Millam JM, Daniels AD, Kudin KN, Strain
MC, Farkas O, Tomasi J, Barone V, Cossi M, Cammi R, Mennucci
B, Pomelli C, Adamo C, Clifford S, Ochterski J, Petersson GA,
Ayala PY, Cui Q, Morokuma K, Malick DK, Rabuck AD,
Raghavachari K, Foresman JB, Cioslowski J, Ortiz JV, Stefanov
BB, Liu G, Liashenko A, Piskorz P, Komaromi I, Gomperts R,
Martin RL, Fox DJ, Keith T, Al-Laham MA, Peng CY,
Nanayakkara A, Gonzalez C, Challacombe M, Gill PMW, Johnson
BG, Chen W, Wong MW, Andres JL, Head-Gordon M, Replogle
ES, Pople JA. Gaussian 98 (Revision A.9). Gaussian: Pittsburgh,
PA, 1998.
18. Schlegel HB, Peng C, Ayala PY, Frisch MJ. J. Comput. Chem.
1996; 17: 49.
19. Topol IA, Burt SK, Rasin AA. Chem. Phys. Lett. 1995; 47: 112.
20. Badar RWF. Atoms in Molecules: a Quantum Theory. Oxford
University Press: Oxford, 1990.
21. Gholami MR, Habibi YA. J. Phys. Org. Chem. 2000; 13: 468.
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 79–83