Solvent Effect in the Thermal Decomposition Reaction of trans-3,3-Dimethyl-5,6-tetramethylene-1,2,4-trioxacyclohexane

Article abstract of DOI:10.1021/jo980658j

The kinetic data of the thermal decomposition reaction of trans-3,3-dimethyl-5,6-tetramethylene-1,2,4-trioxacyclohexane has been measured in different solvents (benzene, toluene, 2-propanol, 2-methoxyethanol, and p-dioxane) at 0.02 mol kg^-1 initial concentration and in the temperature range of 135.0-165.0°C. The enthalpy and entropy of activation of the unimolecular reaction of this trioxane in several organic solvents have been correlated through "isokinetic relationships" to validate the existence of a genuine solvent effect on that reaction.

Full text of DOI:10.1021/jo980658j

J . Org. Chem. 1999, 64, 8457-8460
8457
Articles
Solven t Effect in th e Th er m a l Decom p osition Rea ction of
tr a n s-3,3-Dim eth yl-5,6-tetr a m eth ylen e-1,2,4-tr ioxa cycloh exa n e
Gladys N. Eyler,*,† Adriana I. Can˜izo,† Carmen M. Mateo,† Elida E. Alvarez,† and
La´zaro F. R. Cafferata‡
Area de Quı´mica, Departamento de Ingenierı´a Quı´mica, Facultad de Ingenier´ıa, Universidad Nacional
del Centro de la Provincia de Buenos Aires, Avenida del Valle 5737, (7400) Olavarr´ıa,
Provincia de Buenos Aires, Argentina, and Laboratorio LADECOR, (UNLP), La Plata, Argentina
Received April 8, 1998
The kinetic data of the thermal decomposition reaction of trans-3,3-dimethyl-5,6-tetramethylene-
1,2,4-trioxacyclohexane has been measured in different solvents (benzene, toluene, 2-propanol,
2-methoxyethanol, and p-dioxane) at 0.02 mol kg-1 initial concentration and in the temperature
range of 135.0-165.0 °C. The enthalpy and entropy of activation of the unimolecular reaction of
this trioxane in several organic solvents have been correlated through “isokinetic relationships” to
validate the existence of a genuine solvent effect on that reaction.
Ta ble 1. P seu d o-F ir st-Or d er Ra te Con sta n t Va lu es for
th e Th er m a l Decom p osition Rea ction of
tr a n s-3,3-Dim eth yl-5,6-tetr a m eth ylen e-
1,2,4-tr ioxa cycloh exa n e (0.02 m ol k g-1
In itia l Con cen tr a tion )
In tr od u ction
Cyclic peroxides of the substituted 1,2,4-trioxanes type
are structurally related to the six-membered ring 1,2,4,5-
tetroxanes and 1,2,4,5-trioxazines. Many derivatives of
a 1,2,4-trioxane are now available for study thanks to
new synthetic methods1 in an effort to obtain these cyclic
peroxides in good yields and to know in deep their
reactions. It is interesting to mention that the antima-
larial activity of the plant extract qinghaosu is associated
with the presence of the 1,2,4-trioxane ring in molecules
of compounds (Artemisinin) found in its composition.1
The 1,2,4,5-tetroxanes conformational aspects and
kinetic data of the thermolysis reaction in solution2-5
have been extensively investigated, and comparative
analysis3 of the reactivities of cyclic peroxides of the
substituted 1,2,4,5-trioxazines families of compounds was
also performed considering their reaction products and
available kinetic data in several media.
reaction solvent
2-propanol
T (°C)
kexp × 105 (s-1
)
ra
145.0
160.0
175.0
135.0
140.0
150.0
165.0
145.0
154.5
162.0
175.5
135.0
140.0
147.0
156.0
165.4
171.2
135.0
143.0
143.0
150.0
150.0
165.5
2.12
5.53
27.6
1.96
2.22
6.63
14.9
0.51
2.24
2.54
12.5
0.11
0.204
0.40
1.02
2.39
4.15
1.98
4.11
3.19
7.07
7.20
15.8b
0.986
2-methoxyethanol
benzene
0.987
0.986
0.999
toluene
On the other hand, it has been reported a further effect
of solvent polarity on the thermal decomposition rate of
acyclic peroxides6,7 and cyclic diperoxides4,8 (1,2,4,5-
tetroxanes). In particular, that effect was observed on the
p-dioxane
0.988
* To whom correspondence should be addressed. Tel: 54-2284-
451055/6. Fax: 54-2284-450628. E-mail: neyler@fio.unicen.edu.ar.
† Universidad Nacional del Centro de la Provincia de Buenos Aires.
‡ Laboratorio LADECOR.
(1) J efford, C. W.; Rossier, J . C.; Boukouvalas, J . J . Chem. Soc.,
Chem. Commun. 1987, 713; 1987, 1593 and references therein.
(2) Cafferata, L. F. R.; Eyler, G. N.; Svartman, E. L.; Can´izo, A. I.;
Borkowski, E. J . J . Org. Chem. 1990, 55, 1058.
(3) Cafferata, L. F. R. In Trends in Organic Chemistry; Menon, L.,
Ed.; Council of Scientific Research: Trivandrum, India, 1993; Vol. 4,
pp 773-791.
(4) Cafferata, L. F. R.; Eyler, G. N.; Svartman, E. L.; Can˜izo, A. I.;
Alvarez, E. E. J . Org. Chem. 1991, 56, 411.
a
Correlation coeficient according to the Arrhenius equation (eqs
b
1-5). kexpvalue obtained up to 43% 1 decomposition, because at
higher conversions the first-order law is not obeyed.
activation parameter values. All the cases previously
cited constitute a common “reaction series”9 where the
rate-determining step is the O-O bond rupture in each
molecule.
Here, available kinetic data on the thermal decomposi-
tion reaction of trans-3,3-dimethyl-5,6-tetramethylene-
(5) Cafferata, L. F. R.; Furlong, J . J . Adv. Oxygenated Processes
1995, 4, 83-105.
(6) Cafferata, L. F. R.; Mir´ıfico, M. V. An. Asoc. Quı´m. 1982, 70, 29.
(7) Cafferata, L. F. R.; Quinta´ns, M. T. An. Asoc. Quı´m. Argent. 1987,
75, 461-474.
(9) Bunnett, J . F. Investigation of rates and mechanism of reaction.
In Technique of Chemistry; Weissberger, A., Ed.; Wiley: New York,
1974; Part I, Vol. VI, Chapter VIII.
(8) Eyler, G. N.; Can˜izo, A. I.; Alvarez, E. E.; Ullman, R.; Cafferata,
L. F. R. To be published.
10.1021/jo980658j CCC: $18.00 © 1999 American Chemical Society
Published on Web 10/21/1999

8458 J . Org. Chem., Vol. 64, No. 23, 1999
Eyler et al.
Ta ble 2. Activa tion P a r a m eter s for th e Un im olecu la r Th er m a l Decom p osition Rea ction a t 160 ˚C of
tr a n s-3,3-Dim eth yl-5,6-tetr a m eth ylen e-1,2,4-tr ioxa cycloh exa n e (1) in Solu tion
∆t
(°C)
solvent
∆Hq (kcal mol-1
)
∆Sq (cal mol-1 K-1
)
∆Gq (kcal mol-1
)
ref
2-propanol
2-methoxyethanol
benzene
toluene
p-dioxane
n-octane
30.0
30.0
30.0
36.2
30.5
27.9
30.7 ( 2.0
24.6 ( 1.8
37.6 ( 1.1
34.9 ( 1.0
24.0 ( 1.1
39.2 ( 0.2
-5.4 ( 4.6
-18.3 ( 4.3
8.6 ( 2.6
33.0 ( 2.0
32.6 ( 1.8
33.8 ( 1.1
34.4 ( 1.0
32.6 ( 1.1
33.9 ( 0.2
this work
this work
this work
this work
this work
10
1.1 ( 1.0
-19.8 ( 2.6
12.2 ( 0.5
1,2,4-trioxacyclohexane (1) in n-octane10 solution were
compared with those obtained in toluene, 2-propanol,
benzene, p-dioxane, and 2-methoxyethanol to learn about
the solvent effect on the thermolysis of this trioxane.
cleavage in n-octane solution (hydrogen atom abstraction,
R-cleavage, and â-cleavage). They observed that upon
thermolysis the trioxane 1 undergoes unimolecular cleav-
age to form acetone and adipaldehyde (=95%) and
reported that the diradical formed from homolysis of the
O-O bond must rearrange rapidly (most probably) by an
R-cleavage reaction.
Resu lts a n d Discu ssion
Rate measurements on the thermal decomposition
reaction of trans-3,3-dimethyl-5,6-tetramethylene-1,2,4-
trioxacyclohexane (1) in all the solvents at 0.02 mol kg-1
initial concentration in the temperature range of 135.0-
165.0 °C (Table 1) show a first-order kinetic law behavior
up to ca. 60% trioxane conversions.
The temperature effect on the experimental rate
constant values (kexp) in the solvents investigated can be
represented by the following Arrhenius equations (1-5)
where the errors shown are standard deviation from a
least-mean-squares data treatment11 and the activation
The reaction products analysis for 1 thermolysis in
benzene solution in the initial concentrations range
where induced decomposition is absent indicates adipal-
dehyde as the main product, acetone in low yield, and
traces of trans-1,2-cyclohexanediol. In toluene solvent,12
the benzaldehyde derived from solvent oxidation is the
main product, while acetone, adipaldehyde, and trans-
1,2-cyclohexanediol are minor products. In this case,
bibenzyl and benzyl alcohol were found in low yields. In
p-dioxane, the molar yield of acetone is ca. 30%. In
2-methoxyethanol, the yields of acetone, trans-1,2-cyclo-
hexanediol, and methane were qualitatively determined.
In 2-propanol, the main products derived from the 1
thermal decomposition are adipaldehyde and methane.
In this case, the molar yield of acetone is higher than 1
mol per mol of 1 decomposed, probably, because of the
oxidation of the solvent, as it was found in the thermal
decomposition reaction of acetone cyclic diperoxide in
2-propanol.13
energy is expressed in kcal mol-1
ln kexp2-propanol (s-1) ) (27.1 ( 4.6) -
(31 570 ( 2 004)/RT (1)
ln kexp2-methoxyethanol (s-1) ) (20.6 ( 4.3) -
(25 510 ( 1813)/RT (2)
ln kexpbenzene (s-1) ) (34.2 ( 2.6) -
(38 415 (1135)/RT (3)
ln kexptoluene (s-1) ) (35.0 ( 1.0) -
(35753 ( 1012)/RT (4)
ln kexpp-dioxane (s-1) ) (19.9 ( 2.6) -
(24900 ( 1107)/RT (5)
.
The activation enthalpy and entropy values corre-
sponding to the unimolecular thermal decomposition
reaction (eq 6) can be compared with already reported
data for analogous homolysis of 1 in n-octane10 solution
(Table 2). Those values indicate a marked solvent effect
on the unimolecular decomposition reaction of 1 in going
from n-octane to p-dioxane as solvent.
Benzoyl peroxide,6 di-tert-butyl peroxide,7 and cyclic
diperoxides derived from acetone4 and cyclohexanone8
show solvent effect on the rate of decomposition reaction
in solution. As an example, we report the rate constant
values of acetone cyclic diperoxide in differents solvents
(Table 3), where the least activation enthalphy and
entropy were found in protic solvent (2-propanol and
acetic acid) because of the hydrogen bonding formed
between them and the O-O bond of the cyclic diperoxide.
The linearity of those Arrhenius equations (Table 1)
over relatively large temperature ranges suggests that
the calculated activation parameters values for the 1
thermal decomposition reaction (Table 2) belong to a
single process, which could be its unimolecular thermal
cleavage of the O-O bond as the initial bond breaking
step (eq 6).
Nevertheless, 1 shows the least activation enthalpy
and entropy in p-dioxane, but this solvent and the O-O
bond cannot form hydrogen bonding, so it is possible to
assume that the actual interaction process is more
complex.
A linear relationship between the activation enthalpies
and entropies (∆Hq) ∆H° + â∆Sq, r ) 0.996) of the
unimolecular thermolysis reactions of the trioxane 1 can
Schuster and Bryant10 have postulated several reac-
tions paths to open to the diradical formed from this bond
(10) Schuster, G. B.; Bryant, L. A. J . Org. Chem. 1978, 43, 521-
522.
(12) J efford, C. W.; Cafferata, L. F. R.; Mateo, C. M. Unpublished
results.
(11) Huyberechts, S.; Halleux, A.; Kruys, P. Bull. Soc. Chim. Belg.
1955, 64, 203.
(13) Can˜izo, A. I.; Cafferata, L. F. R. An. Asoc. Quı´m. Argent. 1992,
80, 345-358.

Thermal Decomposition of Cyclic Peroxides
J . Org. Chem., Vol. 64, No. 23, 1999 8459
F igu r e 2. Dependence of the residual sum of squares Su on the supposed isokinetic temperature â for the 1 unimolecular
thermolysis in different solvents.
Ta ble 3. F ir st-Or d er Ra te Con sta n t Va lu es for th e
Un im olecu la r Th er m olysis of Aceton e Cyclic Dip er oxid e
solvent
benzene
toluene
acetonitrile
n-octane
acetophenone
2-propanol
acetic acid
T (°C)
kexp × 105 (s-1
)
ref
150.5
147.7
150.0
145.0
150.0
148.0
150.0
0.283
0.344
0.600
0.076
1.14
2
2
2
2
2
7.90
9.96
13, 14
14
be found according to Leffler’s treatment15 (Figure 1). It
should be noted that the “isokinetic temperature” (â =
211 °C) is far from the experimental temperatures where
the kinetic measurements were performed (135.0-171.2
°C), and the ranges of the ∆Hq and ∆Sq values (∆∆Hq)
15.2 kcal mol-1 and ∆∆Sq) 32.0 eu) are large compared
to the probable errors of those parameters.
F igu r e 1. “Isokinetic relationship” according to Leffler for the
thermal decomposition reaction of trans-3,3-dimethyl-5,6-
tetramethylene-1,2,4-trioxacyclohexane: 1, tolueno; 2, n-oc-
tane; 3, benzene; 4, 2-propanol; 5, p-dioxane; 6, 2-methoxy-
ethanol.
Later, Exner16,17 suggested that the single point of
intersection of lines in the Arrhenius plane could be used
for a sound statistical test since ln k and T are statisti-
cally independent; this is the basis of the isokinetic
relationship. The expected correlation would be not valid
if that intersection point is not found. To estimate the
isokinetic temperature (â) value, the residual sums of
squares (Su) are given and â is then defined by the
minimum of the Su when varying the abscissa of the
intersection point along the 1/T axis17,18 (Figure 2).
Accordingly, with this mathematical treatment the cor-
responding isokinetic temperature value for 1 thermoly-
sis is ca. 228 °C, which is in reasonable agreement with
the â value obtained according to Leffler treatment.
Without any doubt, an important solvent effect is opera-
tive on the rate of thermal decomposition of 1.
(14) Moryganov, B. N.; Kalinin, A. I.; Mikhotova, K. N. J . Gen. Chem.
(USSR) 1962, 32, 3414.
(15) Leffler, J . E. J . Org. Chem. 1955, 20, 1202.
(16) Exner, O. Nature 1970, 227, 366.
(17) Exner, O. Collect. Czechoslov. Chem. Commum. 1972, 37, 1425.

8460 J . Org. Chem., Vol. 64, No. 23, 1999
Eyler et al.
quantitative GC analysis (internal standard method, n-octane
in p-dioxane, and biphenyl in the other solvents) using a
methyl silicone gum rubber packed column at 70 °C for 5 min,
followed by a temperature increase of 10°/min up to 150 °C
for 15 min, injector temperature at 150 and 250 °C FID
temperature, with nitrogen as the carrier gas. Under these
conditions, the retention time of 1 was ca. 23 min. The
corresponding first-order rate constant values were obtained
by least-means-squares treatment of the GC data plotting the
values of ln [concentration of 1] vs time. The reaction products
were identified by GC and GC-MS analyses. Acetone was
identified by HPLC when 2-propanol was the reaction solvent.
The activation parameters were calculated according to the
Eyring equation, and the errors were worked out by the
Arrhenius equation method using a least-means-square data
treatment.11
Exp er im en ta l Section
Ma ter ia ls. trans-3,3-Dimethyl-5,6-tetramethylene-1,2,4-tri-
oxacyclohexane was prepared according to the method de-
scribed elsewhere.19 The solvents employed in the reactions
were purified by standard techniques,20 except the 2-propanol,
which was distilled from ethylenediaminetetraacetic acid
disodium salt dihydrate to remove traces of metallic ions,21
and their purity, suitable for kinetic studies, was checked by
GC analysis.
Kin etic Meth od s. Pyrex glass tubes (6 cm long × 4 mm
o.d.), half filled with the appropriate 1 solution were thor-
oughly degassed under vaccum at -196 °C and sealed with a
flame torch. To perform the runs, the ampules were immersed
in a thermostatic silicone oil bath ((0.1 °C) and withdrawn
after selected times, stopping the reaction by cooling at 0 °C.
The trioxane 1 remaining in the solution was determined by
Ack n ow led gm en t. This research project was finan-
cially supported by the Facultad de Ingenier´ıa, the
Secretar´ıa de Ciencia y Te´cnica de la Universidad
Nacional del Centro, and the Comisio´n de Investiga-
ciones Cient´ıficas de la Provincia de Buenos Aires (CIC).
(18) Linert, W.; J ameson, R. F. Chem. Soc. Rev. 1989, 18, 477.
(19) Payne, G. B.; Smith, C. W. J . Org. Chem. 1957, 22, 1682.
(20) Riddick, J . A.; Bunger, W. B. In Organic Solvent; Weissberger,
A., Ed.; Wiley-Interscience: New York, 1970.
(21) Wilson, T.; Landis, M. E.; Baumstark, A. L.; Bartlett, P. J . Am.
Chem. Soc. 1973, 95, 4765.
J O980658J