Thermal decomposition of formaldehyde diperoxide in aqueous solution

Article abstract of DOI:10.1134/S107036320910017X

Thermal decomposition of formaldehyde diperoxide (1,2,4,5-tetraoxane) in aqueous solution with an initial concentration of 6.22 × 10^-3 M was studied in the temperatures range from 403 to 439 K. The reaction was found to follow first-order kinet

Full text of DOI:10.1134/S107036320910017X

ISSN 1070-3632, Russian Journal of General Chemistry, 2009, Vol. 79, No. 10, pp. 2187–2190. © Pleiades Publishing, Ltd., 2009.
Published in Russian in Zhurnal Obshchei Khimii, 2009, Vol. 79, No. 10, pp. 1692–1696.
Thermal Decomposition of Formaldehyde Diperoxide
1
in Aqueous Solution
a
a
a
a
S. A. I. Cazut , E. H. Ramírez Maisuls , M. R. Delfino , J. M. Romero ,
a
b
N. L. Jorge , and E. A. Castro
a
Área Fisicoquímica, Facultad de Ciencias Exactas y Naturales y Agrimensura, UNNE, Campus Universitario,
Av. Libertad 5400, Corrientes, 3400 Argentina
phone/fax: +54(3783)457996; e-mail: lidianj@exa.unne.edu.ar
b
INIFTA, División Química Teórica, Departamento de Química, Facultad de Ciencias Exactas,
Diagonal 113 y 64, Suc. 4, C.C. 16, La Plata, Buenos Aires, 1900 Argentina
phone: +54(2214)257430; fax: +54(2214)254642; e-mail: castro@quimica.unlp.edu.ar; eacast@gmail.com
Received July 7, 2008
Abstract—Thermal decomposition of formaldehyde diperoxide (1,2,4,5-tetraoxane) in aqueous solution with
–
3
an initial concentration of 6.22×10 M was studied in the temperatures range from 403 to 439 K. The reaction
was found to follow first-order kinetic law, and formaldehyde was the major decomposition product. The
≠
–1
≠
–1 –1
activation parameters of the initial step of the reaction (ΔH = 15.25±0.5 kcal mol , ΔS = –47.78±0.4 cal mol K ,
–
1
E
a
= 16.09±0.5 kcal mol ) support a mechanism involving homolytic rupture of one peroxide bond in the
1
,2,4,5-tetraoxane molecule with participation of the solvent and formation of a diradical intermediate.
DOI: 10.1134/S107036320910017X
Thermal decomposition of 1,2,4,5-tetroxane-
substituted complexes has received considerable
attention in recent years [1–5]. The chemistry of
organic peroxides, which includes synthesis, charac-
terization, and transformations of hydrogen peroxide
derivatives, has a long history and rather strong
traditions [6, 7]. Cyclic di- and tetraoxanes derived
from aliphatic aldehydes and ketones, which were
prepared in our laboratory, inspired numerous studies
related to their application as explosives [8, 9]. In
addition, studies on thermal decomposition of acetone
diperoxide (ACDP) and formaldehyde diperoxide
(FDP) have been initiated.
The reactivity of peroxides is generally attributed to
lability of the O–O bond which readily undergoes
homolytic dissociation constituting the rate-deter-
mining stage of the overall process (Scheme 1).
Scheme 1.
R2
R2
/
C–O
R COR2 + O2
1
2
·
O
a
R1
O
R1
O ·
O
R1
O
R1
O
O
O
/
C–C
·
·
1
+ R2 + R C(O)OOC(O)R2
1
R
R2
R2
b
The reaction finally gives the corresponding ketone
or aldehyde and oxygen via C–O bond cleavage in the
primary diradical intermediate (path a). However, in
some solvents C–C bond cleavage is observed (path b).
When the reaction is carried out in hydrocarbons such
as benzene or toluene, the major products are carbonyl
compound and oxygen, while hydrocarbons are formed
as minor ones [1, 2, 6, 7, 10–12]. These findings
suggest recombination of intermediate radicals generated
by homolytic cleavage of C–C bonds in the initial
diradical. Thus, stepwise reaction mechanism has been
postulated for thermolysis of tetraoxanes in different
solvents, though a concerted process could not be ruled
out (Scheme 2). This type of reaction has been
1
The text was submitted by the authors in English.
2
187

2
188
CAZUT et al.
Table 1. First– order constant values for DPF in aqueous
solution
Table 2. Activation parameters for the thermolysis of DPF
in aqueous solution
a
3
6
–1
a
≠
≠
≠,
Temperature, K
[FDP] × 10 , M
k
exp × 10 , s
r
E
a
,
ΔH ,
ΔS ,
ΔG
kcal mol
Solvent
–1
–1
–1
–1
–1
kcal mol
kcal mol
cal mol
K
4
4
03.0
13.0
6.22
6.22
1.62
2.56
2.77
4.27
0.9970
0.9970
0.9970
0.9980
0.9990
Toluene
THF
22.80±0.9 21.90±0.9 –16.30±1.2
28.71±0.9
30.11±0.7
2
0.0
2
1.10±0.7 20.20±0.7 –23.70±0.6
4
4
23.0
39.0
6.22
6.22
Methanol
1
1
4.80±0.8 14.00±0.8 –39.40±0.8
6.09±0.5 15.25±0.5 –47.78±0.4
32.02±0.8
30.47±0.5
b
8.35
a
Water
Correlation coefficient from least mean square data treatment.
Up to at least 80% DPF conversion.
b
a
The standard deviations were calculated as described in [13].
proposed for thermal decomposition of other cyclic per-
oxides [2, 10] but not for thermolysis of tetraoxanes.
The thermolysis of FDP in aqueous solution at an
−3
initial concentration of 6.2 × 10 M in the temperature
range grom 403 to 439 K follows first-order kinetic
law until ~60% conversion (Table 1, Fig. 1). The
activation parameters for the thermolysis of FDP in
aqueous solution were similar to those obtained in
other solvents (Table 2). To exclude contribution of a
radical-induced decomposition as a competing me-
chanism, the kinetics of thermal decomposition of FDP
in water were studied at 140°C at a higher initial
substrate concentration (0.2 M, Table 1). The results
showed that the rate constants are independent of the
Scheme 2.
R1 R2
O
O
O
O
1
2
2
R COR + O2
R2 R1
In this work we examined the kinetics of thermal
decomposition of 1,2,4,5-tetraoxane (formaldehyde
diperoxide, FDP) in aqueous solution to assess the
influence of solvent polarity on the initial reaction
stage. The data were compared with those obtained in
previous studies.
initial concentration. The reaction carried out at 166°C
−3
(
initial FDP concentration 6.2 × 10 M) followed
first-order kinetics up to at least 80% conversion
Fig. 1). These data suggest that there is no
(
contribution from any secondary process to decom-
position of FDP at higher conversion.
ln [DFP]
–
–
–
–
–
4.80
5.20
5.60
6.00
6.40
The nature of the products (formaldehyde) indicates
the lack of other decomposition processes and supports
the assertion that the rate constants actually correspond
to unimolecular homolysis. However, radical-initiated
decomposition process cannot be ruled out at higher
temperatures and higher initial substrate concen-
trations. The temperature effect on the k values can be
represented by Eq. (1).
–1
–1
ln k (s ) = (6.76 ± 0.4) – (8098.24 ± 0.5) T .
(1)
1
The corresponding plot is linear (r = 0.999) over a
relatively wide temperature range (ΔT = 36 K);
therefore, the calculated activation parameters for FDP
decomposition in aqueous solution belong to a single
process which could be unimolecular homolysis. The
results of AM1 semiempirical calculations with full
geometry optimization [11, 12] also supported the
stepwise mechanism of formaldehyde diperoxide
decomposition through intermediate diradical species.
The calculated energy barrier for the concerted uni-
2
3
4
0
2000
4000
6000
8000 10000
Time, s
Fig. 1. Semilog kinetic plots for the decomposition of
formaldehyde diperoxide (c
solution at (1) 403, (2) 413, (3) 423, and (4) 439 K.
–
3
0
= 6.22×10 M) in aqueous
RUSSIAN JOURNAL OF GENERAL CHEMISTRY Vol. 79 No. 10 2009

THERMAL DECOMPOSITION OF FORMALDEHYDE DIPEROXIDE
2189
Scheme 3.
H
H
H
·
O
·
O
O
O ·
O ·
H
O
H
O
H
O
2 H CO + O2
2
O
H
H
H
O
O
O
H
H
H
Solvent cage
molecular decomposition turn out to be higher by
of the C–O bond with formation of formaldehyde and
oxygen molecules. Presumably, generation of the
diradical occurs in a solvent cage, and solvent mole-
cules facilitate the subsequent C–O bond cleavage.
–
1
~
20 kcal mol higher than the barrier to unimolecular
homolytic O–O bond cleavage (stepwise diradical-
–
1
initiated decomposition, ~21 kcal mol ).
Thus, the thermolysis of diperoxides in aqueous
solution can be represented by Scheme 3, where the
rate of tetraoxane ring opening inside a solvent cage is
relatively fast. The subsequent cleavage of C–O bonds
in the intermediate diradical yields two formaldehyde
and oxygen molecules.
Thus the thermal decomposition of formaldehyde
diperoxide in water solution follows first-order kinetic
law up to at least 60% substrate conversion. The
activation parameters correspond to unimolecular
process. The mechanism of FDP thermolysis in
aqueous solution conforms to that proposed previously
for tetraoxanes and includes initial homolytic
dissociation of the peroxide bond with formation of
intermediate diradical and subsequent C–O bond
cleavage to produce two formaldehyde molecules and
one oxygen molecule as final products. Solvents exert
a significant effect on the entropy of activation of the
examined reaction. The large and negative entropies of
A considerable solvent effect on the thermal
decomposition of FDP should be noted. Polar solvents
could facilitate peroxide bond rupture, thus lowering
–
1
the activation parameters by ~20 kcal mol . Weakly
polar solvents should not affect the transition state, so
that the activation energy may be expected to be
similar to that obtained in the gas phase [14]. In fact,
the thermal decomposition of FDP in toluene and
tetrahydrofuran showed analogous activation param-
≠
–1
–1
activation (ΔS = –49.8 cal mol
K
in water)
suggests formation of an adduct by solvent molecules
and FDP.
≠
eters (Table 2). The Gibbs energies of activation (ΔG )
found for the thermal decomposition of FDP in water
and other solvents turned out to be nearly similar
EXPERIMENTAL
Formaldehyde diperoxide was prepared by standard
methods [11, 12], and its purity was checked by GLC.
(
Fig. 2, Table 2), indicating qualitatively analogous
interactions between the solute and solvent molecules
in the initial thermolysis step. The enthalpy of
activation decreases in going from weakly polar
solvents to strongly polar ones. On the other hand, the
entropy of activation increases in absolute value, so
that the enthalpy–entropy compensation effect is
observed. The activation parameters are in agreement
with the stepwise mechanism involving homolytic
cleavage of one peroxide bond in FDP with formation
of diradical intermediate, as was observed for other
analogous reactions [15–17].
–17.60
–
18.00
–18.40
–
–
18.80
19.20
Quantitative analysis of the thermolysis products
showed formation of 2 mol of formaldehyde per mole
of FDP, regardless of the temperature. Insofar as
concerted mechanism of decomposition may be ruled
out, the formation of the thermolysis products can be
interpreted in terms of initial homolytic cleavage of the
O–O bond to give diradical A (Scheme 3) which can
rebuild FDP molecule or undergo subsequent cleavage
–19.60
.002250
0
0.002350
0.002450
–1
1/T, K
Fig. 2. Eyring plot for thermal decomposition of form-
aldehyde diperoxide in aqueous solution.
RUSSIAN JOURNAL OF GENERAL CHEMISTRY Vol. 79 No. 10 2009

2
190
CAZUT et al.
n-Octane of analytical grade (Fluka) was used as
internal standard in quantitative determination of FDP
and other products.
J. Org. Chem., 1984, vol. 49, p. 2107.
3. Jorge, N.L., Doctoral Thesis, 1997, UNLP, Argentina.
4
5
6
. Jorge, N.L., Gómez Vara, M.E., Castro, E.A., Auti-
no, J.C., and Cafferata, L.F.R., J. Mol. Struct.
Pyrex glass ampules (12 cm long and 4 mm i.d.)
were filled with 0.2 ml of FDP solutions, thoroughly
degassed under reduced pressure at 77 K, and sealed
using a flame torch. The ampules were immersed into
a silicone oil bath maintain at a required temperature
with an accuracy of ± 0.1K and withdrawn after 5–
(Theochem), 1999, vol. 459, p. 29.
. Jorge, N.L., Peruchena, N.M., Castro, E.A., and
Cafferata, L.F.R., J. Mol. Struct. (Theochem), 1998,
vol. 433, p. 311.
. McCullough, K.J., Morgan, A.R., Nonhebel, D.C., and
Pauson, P.L., J. Chem. Res., Synop., 1980, no. 2, p. 36.
1
0 min at each temperature, and the reaction was
stopped by cooling to 273 K. The unreacted FDP and
reaction products were determined by GLC analysis on
Hewlett–Packard 5890 Series II gas chromatograph
7. McCullough, K.J., Morgan, A.R., Nonhebel, D.C.,
Pauson, P.L., and White, C.J., J. Chem. Res., Synop.,
1
980, no. 2, p. 34.
(
3
flame ionization detector; HP-5 capillary column,
0 m × 0.25 mm, stationary phase 5% phenylmethyl-
silicone, carrier gas nitrogen; injector temperature
8. Dubnikova, F., Kosloff, R., Almog, J., Zeiri, Y., Boese, R.,
Itzhaky, H., and Keinan, E., J. Am. Chem. Soc., 2005,
vol. 127, p. 1146.
3
4
78 K, oven temperature programming from 313 to
23 K at a rate of 30 deg min ). The products were
9. Van Duin, A.C., Zeiri, Y., Dubnikova, F., Kosloff, R.,
and Goddard, W.A., J. Am. Chem. Soc., 2005, vol. 127,
p. 11053.
–
1
identified by comparing their retention times with
those of authentic samples or by comparing the
corresponding mass spectra (in this case helium was
used as carrier gas, and a Hewlett–Packard 5972A
mass-selective detector was connected to the
chromatograph).
10. Cañizo, A.I. and Cafferata, L.F.R., Anal. Asoc. Quim.
Argentina, 1992, vol. 80, no. 4, p. 345.
1
1
1
1. Jorge, N.L., Leiva, L.C.A., Romero, J.M., and Gómez
Vara, M.E., Rev. Int. Inform. Tecnol., 2002, vol. 13,
no. 2, p. 23.
2. Jorge, N.L., Leiva, L.C.A., Romero, J.M., Gómez-Vara,
M.E., Hernández-Laguna, A., and Cafferata, L.F.R.,
http://www1.unne.edu.ar/cyt/2002/cyt.htm.e037.
The first-order rate constants were calculated from
the slope of the plots of ln [DPF] values versus
reaction time. In all cases, the reaction progress was
monitored for at least one half-conversion period. The
corresponding energies of activation were determined
from the Arrhenius equation, followed by the least-
squares processing [13, 18]. The other activation
3. Huyberechts, S., Halleux, A., and Kruys, P., Bull. Soc.
Chim. Belg., 1955, vol. 64, p. 203.
14. Cafferata, L.F.R. and Lombardo, J.D., Int. J. Chem.
Kinet., 1994, vol. 26, p. 503.
≠
≠
≠
parameters (ΔH , ΔS , ΔG ) were calculated by the
Eyring equation.
1
5. Cafferata, L.F.R., Eyler, G.N., Svartman, E.L., Cañizo, A.I.,
and Alvarez, E.E., J. Org. Chem., 1991, vol. 56, p. 411.
1
1
6. Leiva, L.C., Castellanos, M.G., Jorge, N.L., Caffera-
ta, L.F.R., and Gómez Vara, M.E., Rev. Soc. Quim.
México, 1998, vol. 42, p. 223.
REFERENCES
1
. Cafferata, L.F.R. and Furlong, J.J., Advances in
Oxygenated Processes, Baumstark, A.L., Ed., Green-
wich, CT: JAI, 1995, vol. 4, p. 81.
7. Leiva, L.C., Cafferata, L.F.R., and Gómez Vara, M.E.,
Anal. Asoc. Quim. Argentina, 2000, vol. 88, no. 1/2, p. 9.
2
. Cafferata, L.F.R., Eyler, G.N., and Mirífico, M.V.,
18. Schaleger, L.L. and Long, F.A., Adv. Phys. Org. Chem.,
RUSSIAN JOURNAL OF GENERAL CHEMISTRY Vol. 79 No. 10 2009