Elimination kinetics of DL-mandelic acid in the gas phase

Article abstract of DOI:10.1002/(SICI)1099-1395(199702)10:2<121::AID-POC870>3.0.CO;2-8

DL-Mandelic add was pyrolyzed in a static reaction vessel over the temperature range 300-1-340-0 °C and pressure range of the substrate 15-2-52-1 Torr. The reaction, in a seasoned vessel and in the presence of the free radical inhibitor cyclohexene, is homogeneous, unimolecular and obeys a first-order rate law. The reaction yielded benzaldehyde, CO, H₂O and small amounts of benzyl alcohol and CO₂. The rate coefficients followed the Arrhenius equation: log k₁ (s ^-1)=(12·54±0·12)-(171·3±1· 4) kJ mol^-1 (2·303RT)^-1. The present result may imply a unimolecular elimination involving a semi-polar five-membered cyclic transition-state mechanism. Steric factor appears not to be important in rate enhancement.

Full text of DOI:10.1002/(SICI)1099-1395(199702)10:2<121::AID-POC870>3.0.CO;2-8

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 10, 121–124 (1997)
ELIMINATION KINETICS OF DL-MANDELIC ACID IN THE GAS PHASE
GABRIEL CHUCHANI* and IGNACIO MARTIN
Centro de Química, Instituto Venezolano de Investigaciones Científicas (IVIC), Apartado 21827, Caracas 1020-A, Venezuela
DL-Mandelic acid was pyrolyzed in a static reaction vessel over the temperature range 300·1–340·0 °C and pressure
range of the substrate 15·2–52·1 Torr. The reaction, in a seasoned vessel and in the presence of the free radical inhibitor
cyclohexene, is homogeneous, unimolecular and obeys a first-order rate law. The reaction yielded benzaldehyde, CO,
H2O and small amounts of benzyl alcohol and CO2. The rate coefficients followed the Arrhenius equation: log k1
(sϪ1)=(12·54±0·12)Ϫ(171·3±1·4) kJ molϪ1 (2·303RT)Ϫ1. The present result may imply a unimolecular elimination
involving a semi-polar five-membered cyclic transition-state mechanism. Steric factor appears not to be important in
rate enhancement. © 1997 by John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 121–124 (1997) No. of Figures: 0 No. of Tables: 9 No. of References: 8
Keywords: DL-mandelic acid; gas-phase elimination kinetics
Received 15 May 1996; revised 23 September 1996; accepted 4 November 1996
INTRODUCTION
the benzylic part of a molecule, such as mandelic acid, may
give rise to an effective assistance of the acidic hydrogen of
the COOH group for a faster dehydration process. Conse-
quently, this work was aimed at studying the homogeneous
pyrolytic elimination kinetics of DL-mandelic acid in the gas
phase.
In association with the results on the gas-phase elimination
of 2-halocarboxylic acids,1, 2 where the acidic H of the
COOH group assists the elimination of the halogen atom in
a semi-polar five-membered cyclic transition state, several
2-hydroxycarboxylic acid pyrolyses were examined. In this
respect, a correlation of rate increase in the homogeneous,
unimolecular gas-phase dehydration process from primary
to tertiary 2-hydroxycarboxylic acids was obtained.3 More-
over, later work showed that an increase in the bulkiness of
the alkyl group at the 2-position of the 2-hydroxycarboxylic
acids gave a small but a significant increase in rate
coefficients.4 In this respect, the electronic transmission of
the alkyl groups, without underestimating steric accelera-
tion as a possible important factor, was believed to enhance
the pyrolysis rate of these substrates. The mechanism of
these reactions was suggested to proceed via a semi-polar
five-membered cyclic transition state as shown in equation
(1).
It is interesting to consider several earlier studies where
high molecular weight aliphatic aldehydes were prepared by
distillation following 2-hydroxy acid decomposition,5, 6
while degradation of disubstituted glycolic acids in the
presence of lead tetraacetate led to the formation of the
corresponding ketone.7
To obtain additional information about the extent of the
nature of the transition state of these gas-phase eliminations,
with the premise that the C—OH bond polarization is rate
determining, the high stabilization of the C—OH bond by
Table 1. Stoichiometry based on the Pf/P0 ratioa
Temperature (°C)
P0 (Torr)
Pf (Torr)
Pf/P0
310·0
320·1
330·1
340·0
26
71
2·73
2·67
2·80
2·70
27
73
32·5
25·3
91
68·3
* Correspondence to: G. Chuchani.
a P0 =initial pressure; Pf =final pressure
© 1997 by John Wiley & Sons, Ltd.
CCC 0894–3230/97/020121–04 $17.50

122
GAS-PHASE KINETICS OF MANDELIC ACID
Table 2. Stoichiometry based on the pressure vs chromatographic
analysis at 320·1 °C
Table 6. Invariability of the rate coefficients with initial pressure at
320·1 °C
Time (min)
Reaction (%) (pressure)
Benzaldehyde (%) (chromatography)
1·5
3
6
9
P0 (Torr)
15·2
28·23
26·4
28·27
40·1
28·04
43·0
28·47
52·1
28·07
19·9 38·8 59·1 65·7
20·8 37·5 57·8 62·5
104k1 (sϪ1
)
[equation (2), path 1] demands Pf =3P0, where Pf and P0 are
the final and initial pressure of the substrate, respectively.
The average experimental Pf /P0, values measured after 10
half-lives and at four different temperatures was 2·72 (Table
1). The resulting departure from the required theoretical
stoichiometry was due to the parallel decarboxylation
process to give a small amount of benzyl alcohol in ca 11%
yield [equation (2), path 2]. Otherwise, the yields of
pyrolysis products, within the range of rate determination
and up to 60% decomposition, are only benzaldehyde, CO
and H2O.
RESULTS AND DISCUSSION
The molecular elimination of DL-mandelic acid (␣-hydroxy-
phenylacetic acid) in a seasoned static reaction vessel was
studied in the temperature range 300·1–340·0 °C and
pressure range 15·2–52·1 Torr. The reaction, after a very
long period of time, can be described by equation (2).
The verification of the stoichiometry of equation (2), path
1, was possible by comparing the percentage decomposition
of the substrate mandelic acid as predicted from pressure
measurements with the quantitative analysis of the product
benzaldehyde (Table 2). The formation of a small amount of
benzyl alcohol [equation (2), path 2] was monitored by the
quantitative chromatographic analysis of this product after
50% decomposition of the substrate (Table 3). The homoge-
neity of this reaction was examined by using a packed
reaction vessel with a six times greater surface-to-volume
ratio than the unpacked reaction vessel. The rate coefficients
were unaffected in packed and unpacked vessels seasoned
with the products of decomposition of allyl bromide.
However, after several runs of the substrate in the packed
and unpacked clean Pyrex vessels, no differences in the rate
coefficients were obtained (Table 4). Apparently, the surface
of the reaction was deactivated by pyrolyzing the substrate
under study until reproducible k values were obtained.
The absence of a free radical process in this elimination
process was examined by carrying out several runs in the
presence of different proportions of cyclohexene as inhib-
The theoretical stochiometry based on the main reaction
Table 3. Formation of benzyl alcohol at 320·1 °C
Time (min)
5·2 10 40 51 83
Reaction (%) (pressure)
55·8 68·5 74·8 78·5 90·1
Benzyl alcohol (%) (chromatography) 0·6 1·2 4·5 5·7 9·2
Table 4. Homogeneity of the reaction
Temperature
(°C)
S/V (cmϪ1
)
104k1 (sϪ1
)
Aa
Bb
300
320
300
1
6
1
6
1
6
8·77
8·89
28·18
28·61
88·56
88·98
8·66
8·55
28·39
28·58
88·98
88·28
Table 7. Temperature dependence of rate coefficients
Temperature (°C)
104k1 (sϪ1
300·1
8·66
310·0
15·58
320·1
28·39
330·1
50·34
340·0
88·98
)
a Clean Pyrex vessel and after several runs with the substrate.
b Seasoned vessel with allyl bromide.
Log k1 (sϪ1)=(12·54±0·12)Ϫ(171·3±1·4) kJ molϪ1 (2·303RT)Ϫ1
(r=0·999979)
Table 5. Effect of the inhibitor cyclohexene on rates at 320·1 °C
P0 (substrate)
(Torr)
P0 (Inhibitor)
104k1
a
(Torr)
Pi/P0
(sϪ1
)
Table 8. Variation of rate coefficient with temperature for the
benzyl alcohol formation
33·3
41·2
43·0
52·1
15·2
–
–
28·55
28·37
28·47
28·07
28·23
145·3
101·5
159·0
65·5
1·2
2·3
3·1
4·3
Temperature (°C)
104k1 (sϪ1
310·0
0·92
320·1
1·92
330·1
3·74
340·0
7·19
)
Log k1 (sϪ1)=(13·15±0·40)Ϫ(202·7±4·6) kJ molϪ1 (2·303RT)Ϫ1
(r=0·999907)
a Rates determined up to 55% reaction.
© 1997 by John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 10, 121–124 (1997)

G. CHUCHANI AND I. MARTIN
Table 9. Comparative rates at 340·0 °C
123
Substrate
104k1 (sϪ1
)
Rel rate Ea (kJ molϪ1
)
Log[A (sϪ1)]
Ref.
·
CH3CHOHCOOH
C6H5CHOHCOOH
4·57
88·98
1·0
19·5
182·8±1·3
171·3±1·4
12·24±0·22
12·54±0·12
3
This work
itor (Table 5). This substrate was always pyrolyzed in
seasoned vessels and in the presence of at least twice the
amount of the free radical suppressor cyclohexene. No
induction period was observed and the rates were reproduci-
ble with a standard deviation not greater than 5% at the
working temperature.
the pyrolysis of benzoylformic acid in the gas phase.8
The rate coefficients of the hydroxy acid substrate,
calculated from k1 =(2·303/t)log[2P0/(3P0 ϪPt)], were found
to be invariable with initial pressure (Table 6). Plotting
log(3P0 ϪPt) against time t gave a good straight line up to
50–60% reaction. The variations of the rate coefficients with
temperature are described in Table 7. The results given in
Table 7 lead by using the least-squares procedure and 90%
confidence limits, to the Arrhenius equation shown. The k
values for the formation of benzyl alcohol were determined
[equation (2), path 2] from k1 =(2·303/t)log [(2P0 ϪPt)/P0],
while the variation of the rate coefficient with temperature is
as given in Table 8 (with a confidence limit of 0.9).
The results of the present work suggest that the
elimination processes of 2-hydroxy acids in the gas phase
are polar in nature. Moreover, steric acceleration may not be
an important factor for a faster elimination rate with
increase in the bulkiness of the alkyl group at the 2-position
of 2-hydroxycarboxylic acids.4
EXPERIMENTAL
The consideration that the decomposition of 2-hydroxy-
carboxylic acids may proceed through a mechanism where
the C—OH bond polarization, in the sense of C␦+ · · · OH␦Ϫ
is rate determining,3, 4 appears to find support from the
results shown in Table 9. In this respect, because of the
greater stabilization of the benzylic C—OH bond polariza-
tion of mandelic acid in the transition state, the direct
participation of the acidic H of the C=O of the COOH
group leads to a faster dehydration rate compared with lactic
acid. Apparently, the mechanism of this reaction, as in
2-hydroxycarboxylic acid decomposition,3, 4 appears to
proceed through a semi-polar five-membered cyclic transi-
tion state as shown in equation (3).
The purity of better than 99·0% of the substrate DL-mandelic
acid (Aldrich) was checked by GLC (10% SP 1200–1%
H3PO4, Chromosorb W AW DMCS, 80–100 mesh) and
titration with a solution of 0·05
M
NaOH. The products
benzaldehyde (Aldrich) and benzyl alcohol (Aldrich) were
quantitatively analyzed with the above column. The analy-
ses of CO and CO2 were performed with a Carbosieve B
(60–80 mesh) column. The identities of the substrate and
products were further verified by NMR and mass spec-
trometry.
The hydroxy acid was pyrolyzed in a static reaction
vessel, seasoned with allyl bromide, and in the presence of
at least twice the amount with respect to pressure of the
inhibitor cyclohexene (Table 4). The rate coefficients were
determined manometrically up to 50–60% reaction. The
temperature was controlled within ±0·2 °C by a Shinko
DIC-PS resistance thermometer temperature controller and
measured with a calibrated platinum/platinum–13% rho-
dium thermocouple. No temperature gradient was found in
the reaction vessel when measured with a thermometer
introduced around the heated aluminum block. The sub-
strate mandelic acid, dissolved in dioxane (Merck, 99·5%
pure), was injected directly into the reaction vessel with a
syringe through a silicone-rubber septum.
REFERENCES
In the case of the small amounts of benzyl alcohol
obtained beyond the range of rate determination for 50%
decomposition [equation (2), path 2], the mechanism for the
process is different and may be rationalized according to
equation (4). This type of mechanism is nearly analogous to
1. G. Chuchani and R. M. Dominguez, Int. J. Chem. Kinet. 21, 367
(1989).
2. G. Chuchani, R. M. Dominguez and A. Rotinov, Int. J. Chem.
Kinet. 23, 779 (1991).
© 1997 by John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 10, 121–124 (1997)

124
GAS-PHASE KINETICS OF MANDELIC ACID
3. G. Chuchani, I. Martin, A. Rotinov and R. M. Domínguez, J.
Phys. Org. Chem. 6, 54 (1993).
6. R. R. Davies and H. H. Hodgson, J. Soc. Chem. Ind. (London)
62, 128 (1943).
4. G. Chuchani, I. Martín and A. Rotinov, Int. J. Chem. Kinet. 27,
849 (1995).
7. E. Rohrmann, R. G. Jones and H. A. Shonle, J. Am. Chem. Soc.
66, 1856 (1944).
5. E. E. Blaise, Bull. Soc. Chim. Fr. 31, 483 (1904).
8. R. Taylor, Int. J. Chem. Kinet. 23, 247 (1991).
© 1997 by John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 10, 121–124 (1997)