Polymer Degradation and Stability 39 (1993) 299-303

Kinetics of polymer degradation involving the

splitting off of small molecules: Part 9
thermal dehydrobromination of polyvinyl
bromide
Peter Simon
Department of Physical Chemistry, Faculty of Chemical Technology, Slovak Technical University, Radlinsk~ho 9,
812 37 Bratislava, Czechoslovakia

(Received 13 February 1992; accepted 1 March 1992)

High conversion kinetic runs of the thermal dehydrobromination of polyvinyl

bromide have been treated using the theory developed previously. The results
suggest that the dehydrobromination occurs most probably via the mechanism
of gradual zip growth where the initiation step of the zip reaction is not
autocatalytic. Good agreement between experimental and theoretical kinetic
curves has been achieved for this mechanism. The activation energy and
pre-exponential factor for random elimination of HBr are Ea = 153 kJ mo1-1
and log(A/s -1) = 13-7. The ratio of the rate constants of zip propagation and
random elimination decreases with increasing temperature.

INTRODUCTION corresponds to quantitative dehydrobrom-

ination. 6 During dehydrobromination, long pol-
Although polyvinyl bromide (PVBr) was one of yene sequences are formed which are longer than
the first vinyl polymers to be discovered, 1 it has in PVC dehydrochlorination. 9-11 This indicates
not been the subject of much study. Attention that the dehydrobromination occurs v i a the zip
has turned toward PVBr and vinyl bromide reaction. ~1The high-conversion kinetic curves are
copolymers only in the last two decades because S-shaped 12 which is typical of autocatalytic
of their flame-retardant properties. 2-8 Several reactions.
papers have been devoted to the degradation of In Part 3 of this series, the kinetic curves for
vinyl chloride/vinyl bromide copolymers for a the thermal dehydrochlorination of PVC in an
better understanding of the degradation of inert atmosphere were considered. 15 In Parts 4
PVC 9-n but we can only find one paper dealing and 5, autocatalytic mechanisms of polymer
with the thermal degradation of PVBr up to degradation which occur via elimination of
high conversions over a wide range of tempera- low-molecular weight compounds have been
tures. 12 analyzed 16,17 and the theory has been applied to
PVBr is substantially less stable than PVC but the study of the thermal degradation of PVC in
in most respects it shows analogous degradation HCI and oxidative atmospheres 18"19 and to the
behaviour. 6 Decomposition of PVBr begins at degradation of polyvinyl esters TM (Parts 6-8). The
about 100C 12,13 and up to 270C, de- aim of this present paper is to use the theory for
hydrobromination is the dominating reaction. 14 the treatment of experimental data on the
Ostromysslenskii was the first to report that large thermal dehydrobromination of PVBr in an inert
quantities of HBr are liberated when PVBr is atmosphere published by Mazon-Arechederra et
heated to 220C. 1 The weight loss of PVBr al. 12

299
300 Peter ~imon

S I M U L A T I O N OF T H E KINETIC C U R V E S Part 8 and the minimized parameters are A, Ea,

AA~, AEy, Po and xm.
For the simulation of kinetic curves the same Two minima have been found for the model of
procedure is used as in Parts 3, 6, 7 and 8, i.e., gradual zip growth without the autocatalytic
the whole set of kinetic runs measured at various initiation step. The values of parameters
temperatures is treated simultaneously. Models corresponding to the minima and the standard
of both immediate and gradual zip growth 16'17 are deviations per mesh point, or, are summarized in
tested to fit the experimental data. In the case of Table 1. As the values of 0 show, the first
the immediate zip growth, the kinetic equations minimum is not so deep as the other. Despite
are eqns (7) and (8) in Part 4 and the parameters this, the values of the parameters seem to be
for the minimization of the sum of squares more reasonable for the first minimum where the
between experimental and calculated values of activation energy of random dehydrobromina-
conversion are A, E~, AA, AEa, AA m AE m Po tion, Ea, is lower than the corresponding
and xm. For gradual zip growth, the kinetics are activation energy for PVC dehydrochlor-
described by eqns (13) and (14) in Part 5 and the ination. 15'18'19 This should be the case since the
minimized parameters are A, Ea, AAr, AE~, C--CI bond energy is higher than the energy of
AA#, AEa, Po and x,,,. The meanings of the C - - B r bond. 6 The difference between the
individual parameters are given in Part 6. activation energies of propagation and initiation
steps, AEr, is - 5 7 kJ mo1-1, which is approxim-
ately equal to the conjugation energy of two
double bonds. 21 This decrease is in accordance
RESULTS AND DISCUSSION with the idea that the origin of the allylic
activation consists in the conjugation of the
The possibility of the occurrence of the transition state of dehydrohalogenation with the
mechanisms of immediate and gradual zip adjoining C----C bond. 22 For these reasons, the
growths 16'17 has been examined here. In both kinetic parameters corresponding to the first
cases, a minimum has been found for the minimum in Table 1 appear to be more
combination of the parameters AA and AEa reasonable and acceptable even though the other
leading to the value m ~ l or, for the minimum is a little deeper. The deviations
combination of Ay and E~ leading to 7--->0. between the calculated and experimental values
Then, both eqn (7) in Part 4 as well as eqn (13) of conversion for the first minimum are shown in
in Part 5 degenerate into a simple autocatalytic Fig. 1. The agreement between theory and
kinetic equation where the main feature of the experiment is very good. The deviations do not
degradation of PVBr, i.e. the zip reaction, is not exhibit any regularity, thus pointing to the fact
involved. that the theory adequately describes the ex-
For the model of immediate zip growth, the perimental data.
minimum corresponding to the zip reaction has The activation energy of random de-
been found for the combination of parameters hydrobromination of PVBr is less, by
leading to very low values of B and to values of 17kJmo1-1, than the corresponding activation
m of the order 105 where the value of Po is about energy for PVC dehydrochlorination. ~5 More-
10-5-10 -4 . The growth of such long zips seems over, the pre-exponential factor for de-
hardly possible to be immediate. For the model hydrobromination of PVBr is higher. It follows
of gradual zip growth, the parameters AAa and from eqn (1) in Part 3 that the rate constant of
AEa have always led to the value fl---)0, which random elimination, B, is higher for PVBr than
means that the initiation step of the zip reaction for PVC at any temperature. For example, at
is not autocatalytic. Equations (13) and (14) in 190C, the ratio of rate constants of random
Part 5 are thus reduced to eqns (1) and (2) in elimination for PVBr and PVC is about 800.
Table 1. Kinetic parameters for the degradation of PVBr
Iog(A/s-1) Ea/kJ mol ' Iog(A~/s-~) EJkJ tool-1 po x,. o
13.7 153 10-4 96 0-014 4 0-999 0.024 1
15-9 206 4-9 14 6 10 - 7 0-959 0-016 4
 Thermal dehydrobrominationof polyvinyl bromide 301

O'OB

A
A & AA Z~
0"04 AA
A
A
A
x X
u +4- o ,x X
: O
++ XAX x &&A D
X th
4-+
+ N []
0"04 -
-t-
+

0"0B - []

I I I I
Fig. 1. Deviations between the calculated and experimental values of conversion for the dehydrobromination of P V B r 12
Temperatures: (O) 130C; (O) 140C; (+) 145C; (x) 150C; (A) 160C; (A) 170C; (D) 190C.

Since 7 is the ratio of the rate constants of zip Table 1 shows that E~ is less than Ea.
propagation and random initiation, ~7 the values Consequently, according to eqn (2) in Part 6, 7
of A~ and Er can be calculated as decreases with increasing temperature. For the
temperature region of PVBr dehydrobromina-
A ~ = A . AA~ (1)
tion, 7 has values 1 200-11 000. In Part 5 we have
Ey = E, + AEr (2) examined how the conversion at which the
maximum rate is reached, ~, depends on 7. For
f l = 0 and Po=O it has been found that the
greater the value of 7, the greater is ~. As
illustrated in Fig. 2, an opposite trend is
8
encountered in the dehydrobromination of
PVBr. The value of 7 decreases with increasing
temperature, but the conversion increases. This
behaviour is obviously the consequence of
structural defects in the polymer chain causing
premature zip termination, zd. In eqn (1) in Part
8, the fraction of growing zips is given by the
term ( 1 - zJz), where zd/z is expressed by eqn
(21) in Part 5. As shown by Fig. 3, the fraction of
growing zips exhibits a maximum, the position of
which is almost identical with the position of the
maximum rate. This shows that not only the
! u value of 7, but also the probability of the
0 ,2 .4 .6 .8 !. occurrence of structural defects greatly affects the
x
position of the maximum rate. The positions of
Fig. 2. Dependences of the rate of dehydrobromination on
conversion (calculated values). Temperatures: (1) 150C; (2) the maximum rate found experimentally 12 coin-
160C; (3) 170C. Arrows indicate the conversion at which cide perfectly with the calculated values.
the maximum rate is reached, . Comparison of eqns (9) and (10) in Part 5
302 Peter ~imon

!
.4 40

.3 30

20

10

0 .2 .4 .,,, 1. 0 .2 .4 .6 1.
X X

Fig. 3. Fraction of growing zips as a function of conversion. Fig. 4. Ratio of the rates of zip propagation and random
Temperatures and arrows as in Fig. 2. elimination as a function of conversion. Temperatures and
arrows as in Fig. 2.

leads to the result that the ratio of the rate of termination. In the model of immediate zip
dehydrobromination due to zip propagation and growth, zip termination is considered. On the
the rate of random elimination of HBr is given by contrary, in the model of gradual zip growth it is
assumed that zip termination does not take place
p = (1 - za/z)yp (3)
or is delayed. 17 Our results thus indicate that
at first sight, p increases with the probability p. the rate of zip propagation in PVBr deo
This is undoubtedly the case for an ideal polymer hydrobromination is incomparably higher than
chain when zd = 0. As Fig. 4 demonstrates, for a the rate of zip termination. This conclusion is in
real chain the dependences of p on x also exhibit agreement with Braun et ai.11 who have observed
maxima, the position of which depends on that the ratio of rate constants of zip propagation
temperature. However, considerable differences and termination is higher for vinyl chloride/vinyl
exist between the positions of the maximum rate bromide copolymers than for PVC.
and the maximum of p. Obviously, at the
conversion corresponding to the maximum
dehydrobromination rate, interconnection of zips REFERENCES
predominates over the formation of new zips thus
leading to the decrease in the rate of 1. Ostromysslenskii, I., J. Russ. Phys. Chem. Soc., 414
dehydrobromination. The ratio p attains its (1912) 204, 240.
maximum at much higher conversions not only 2. Neumeyer, J. P., Wadsworth, J. I., Knoepfler, N. B. &
when the interconnection of zips is more intense, Mack, C. H., Thermochim. Acta, 16 (1976) 133.
3. Hobart, S. R., Mack, C. H., Donaldson, D. J. &
but also when premature zip termination due to Neumeyer, J. P., Thermochim. Acta, 16 (1976) 149.
structural irregularities plays an important role. 4. Boylston, E. K. & Muller, L. L., J. Appl. Polym. Sci.,
Then the number of HBr molecules produced by 19 (1975) 1079.
5. Diab, M. A., Acta Polymerica, 37 (1986) 390.
propagation per random elimination decreases 6. McNeill, I. C., Straiton, T. & Anderson, P., J. Polym.
rapidly. Due to higher values of y, the values of Sci. : Polym. Chem. Ed., 18 (1980) 2085.
p are higher for lower temperatures. 7. Diab. M. A., J. Polym. Sci. : Polym. Chem. Ed., 21
(1983) 3249.
Dehydrobromination of PVBr is described by 8. Zhubanov, B. A., Nazarova, S. A., Karzhaubaeva, R.
the model of gradual zip growth whereas G. & Gibov, K. M., Vysokomol. Soedin, BI$ (1976) 50.
dehydrochlorination of PVC in an inert atmos- 9. Braun, D. & Thallmaier, M., Makromol. Chem., 99
(1966) 59.
phere obeys the model of immediate zip 10. Braun, D. & Thallmaier, M., J. Polym. Sci. : Part C, 16
growth. 15 The main difference between the two (1967) 2351.
models resides in the consideration of zip 11. Braun, D., B6hringer, B., Iv~in, B., M6ri, E., Kelen, T.
 Thermal dehydrobromination of polyvinyl bromide 303

& Tiid6s, F., Angew. Makromol. Chem., 143 (1986) 16. Simon, P., Poly. Deg. and Stab., 35 (1992) 45.
165. 17. Simon, P., Poly. Deg. and Stab., 35 (1992) 157.
12. Maz6n-Arechederra, J. M., Delgado Quintero, M. P. & 18. Simon, P. & Valko, L., Poly. Deg. and Stab., 35 (1992)
Barrales-Rienda, J. M., J. Polym. Sci. : Polym. Chem. 249.
Ed., 2,0 (1982) 25. 19. Simon, P., Poly. Deg. and Stab., 36 (1992) 85.
13. O'Shea, M. L., Low, M. J. D. & Monterra, C., Mater. 20. Simon, P. & Ryb~r, M., Poly. Deg. and Stab., 38
Chem. Phys., 2,3 (1989) 499. (1992) 255.
14. Montaudo, G., Puglisi, C., Scamporrino, E. & Vitalini, 21. Valko, L. & Simon, P,, Chem. Phys., 99 (1985) 447.
D., J. Polym. Sci. : Polym. Chem. Ed., 24 (1986) 301. 22. Simon, P., (~ernay, P. & Valko, L., Europ. Polym. J.,
15. Simon, P., Gatial, A. & Valko, L., Poly. Deg. and 25 (1989) 245.
Stab., 29 (1990) 263.