JOURNAL OF APPLIED POLYMER SCIENCE VOL. 17, PP.

2311-2327 (1973)

A Kinetic Study of the Emulsion Polymerization

of Vinyl Acetate

N. F R I I S and L. NYHAGEN, Danish Atomic Energy Commission Research

Establishment Ris#, DK-4000 Roskilde, Denmark

Synopsis
The emulsion polymerization of vinyl acetate was studied a t 50°C. It was found that
the rate of polymerization was proportional to the 0.5 power of the initiator concentra-
tion and the 0.25 power of the number of particles. The number of particles was propor-
tional to the power 0.5 f 0.05 of the emulsifier concentration, but independent of the in-
itiator concentration. The limiting viscosity number of the polymers produced waa
independent of the initiator concentration and number of polymer particles. It is sug-
gested that the mechanism of vinyl acetate emulsion polymerization is similar to that of
vinyl chloride. The linearity of the conversion-versus-time curve is explained as being
due partly to a decrease in the desorption rate of radicals from the polymer particles and
partly t o a decrease in the termination rate constant.

INTRODUCTION
I n the literature i t is generally concluded that the emulsion polymerization
of vinyl acetate does not fall within the classic theories proposed by Smith
and Ewart.' It is believed that this is due partly t o the high transfer
constant to monomer in vinyl acetate polymerization and partly to the
relatively high solubility of vinyl acetate in water. In vinyl acetate
emulsion polymerization, it is often observed that the rate of polymerization
remains constant until 85% conversion, although the separate monomer
phase has vanished already at or before 30%. This anomaly has caused
much speculation concerning the mechanism of vinyl acetate emulsion
polymerization, and during the past decade several investigations have been
performed in order t o explain this behavior.*-'
Some investigators4t6have proposed a mechanism in which most of the
polymerization takes place in the water phase. The constant rate behavior
is thus explained by assuming the monomer-polymer particles to act like
reservoirs, keeping the monomer concentration in the water phase constant.
However, the water phase hypothesis is not compatible with the high rates
of polymerization usually found in emulsion systems. Thus, Gershberg6
had to assumc a termination rate constant in the water phasr several orders
of magnitude lower than that reported for vinyl acetate in bulk polymeriza-
tion. Furthermore, in a study of the polymerization of vinyl acetate in
aqueous media, Napper and Partss have observed a marked increase in
2311
@ 1973 by John Wiley & Sons, Inc.
2312 FRIIS AND NYHAGEN

polymerization rate as soon as the initially formed polymer precipitated.

This suggests that the major locus of polymerization is the monomer-
swollen polymer particles and not the water phase.
Recently, three different models have been proposed for vinyl acetate
emulsion p o l y m c r i ~ a t i o n . ~ ~
Common
'~~ t o these models is that they all
involve a mechanism allowing radicals t o escape the polymer particles.
This mechanism was originally introduced by Smith and Ewart' and is
necessary to explain the low concentration of radicals per particle usually
observed in vinyl acetate emulsion polymerization. However, the implica-
tions of the three models are vastly different, and the experimental results
on which the models are based are contradictory. We therefore decided t o
reexamine the emulsion polymerization of vinyl acetate.

EXPERIMENTAL
Materials
Vinyl acetate was distilled 24 hr prior to use on a 2-ft column filled with
glass helices. The initiator used in the emulsion experiments was an
analytical-grade potassium persulfate obtained from Merck, and in the bulk
experiments an analytical-grade a,a'-azoisobutyronitrile obtained from
Fluka AG. The emulsifier was a purified-grade sodium lauryl sulfate
(Quolac ON WD), obtained from the American Alcolac Corporation.
Polymerization Procedure
The polymerization was carried out in a 2-liter Pyrex vessel provided
with stirrer, thermometer, and Nz inlet. The emulsion was purged with
nitrogen 30 min proir t o the addition of initiator. The nitrogen was
obtained from a standard cylinder and deprived of any oxygen by passing
through a 5% solution of pyrogallol in 2N NaOH. The temperature was
controlled within *0.2"C.
I n all experiments the emulsion was composed of 550 ml vinyl acetate,
1150 ml redistilled water, and varying amounts of emulsifier and initiator.
The degree of conversion was determined from samples withdrawn a t
regular intervals. The emulsion was broken by freezing in liquid nitrogen.
The precipitated polymer was washed thoroughly with distilled water and
dried t o constant weight a t 50°C. Elemental analysis proved that the
dried polymer contained less than O.Ol~oinitiator and emulsifier.
The number of polymer particles was determined by light scattering and
electron microscopy. High-quality micrographs were obtained by using
the hardening technique developed by Vanzo.

RESULTS
Disappearance of the Separate Monomer Phase
I n order t o establish the point of which the separate monomer phase
disappears, latex samples were withdrawn a t various conversions early in
 VINYL ACETATE POLYMERIZATION 2313

-E
.
-
0
0 020-
*
L

-2
0
0.15 -
.->
Q,
.g 0.x) -
c
0
-
0.05-

0.05 0.10 015 0.20 0.25

X

Fig. 1. Amount of vinyl acetate present as a separate monomer phase as function of

conversion.

the polymerization. The samples were centrifuged for 15 min a t 2000 rpm.
The amount of vinyl acetate separated in this way was measured by a grad-
uated scale and taken to equal the amount of vinyl acetate present as
monomer droplets in the emulsion. Figure 1 shows a plot of this quantity,
in ml vinyl acetate/ml emulsion, as a function of conversion. From this
curve it appears that the separate monomer phase vanishes at 20% conver-
sion.
Further evidence of the validity of the results obtained by this technique
is the fact that extrapolation of the straight line gives a value of 0.32 ml
vinyl acetate/ml emulsion, which is the initial composition of the emulsion.
It should be mentioned that results from similar investigations reported
in the literature are somewhat scattered. Thus, French'O has found that
the separate monomer phase disappeared a t 13.5y0 conversion, while
Vanzo2 on the basis of vapor pressure measurements has found a value of
32%. Recently, Nomura et al.' have reported a value of 0.23.
The disappearance of the separate monomer phase should be reflected
in the conversion-versus-time curve. The data obtained in the present
work show that the curves become linear between 15Qj,and 20% conversion,
and a value of 20% is therefore not unreasonable.
Shape of the Polymerization Curve
Figure 2 shows typical conversion-versus-time plots obtained a t different
initiation rates. It appears that the rate of polymerization is constant in
the interval 20-90% conversion and that the shape of the curves is inde-
pendent of the initiation rate.
2314 FRIIS AND NYHAGEN

1.0

0.8

0.6

0.1

0.2

LO 80 120 160 200 210

t,min
Fig. 2. Conversion-vs.-time curves a t different initiation rates: ( 0 )4 X (0)2 X
( 8 ) 10-3 moles K&O8/1. HzO, 9.5 g sodium lauryl sulfate/l. H20.

1.o

0.8

0.6

0.1

0.2

LO 80 120 160 200

t, min
Fig. 3. Conversion-vs.-time curves a t different emulsifier concentrations: ( 8 )24.0: ( 0 )
9.5; (0)2.4 g sodium lauryl sulfate/l. H20. 2 X moles KaSzOs/l. HzO.

I n a study of the effect of emulsifier Concentration on the polymerization

rate, we observed that the emulsifier concentration exerts a certain effect on
the shape of the polymerization curve. At high emulsifier concentrations,
the rate of polymerization begins to decrease from 80% conversion. At
low concentrations, i.c., when the particles are relatively large, we observed
a slight acceleration in polymerization rate, beginning a t 70’% conversion.
This effect, which we reproduced several times, is shown in Figure 3.
 VINYL ACETATE POLYMERIZATION 2315

1,min
Fig. 4. Bulk polymerization of vinyl acetate at 50°C. [AIBN] = 4 X moles/l.
HpO: ( 0 )experimental; (-) theoretical.

In emulsion polymerization a single polymer particle can be regarded as a

locus of bulk polymerization with intermittent initiation. A decrease in
the termination rate which is observed in bulk polymerization should there-
fore also occur in a single polymer particle. For the sake of comparison
vinyl acetate was polymerized in bulk at 50°C with a,cu'-azoisobutyro-
nitrile as intiator, and Figure 4 shows a conversion-versus-time plot ob-
tained from such an experiment. Together with the experimental curve is
shown the theoretical curve obtained by integration of the rate expression
given in eq. (l),assuming that the initiation rate is constant:
-d[Ml/dt = kp[Ml ( ~ r j [ I l / k i p ) l ' * (1)
where k , = 3500 l./mole-sec," k,, = lo8 l./mole-sec,ll and k r j = 1.7 X
lO+/sec. l 2 It appears that there is an appreciable autoacceleration,
beginning at approximately 10% conversion. On the assumption that k ,
remains constant, k,, can be obtained as a function of conversion from the
experimental curve in Figure 4. Equation (2) gives the relation between
k,, and convcrsion x thus obtained, and Figure 5 shows the corresponding
plot of k,, versus conversion:
kl, = 2 exp(A + A ~ +x A2x2+ A3x3) l./mole-sec (2)
2316 FRIIS AND NYHAGEN

105
02 0.4 0.6 0.8 1 I
X

Fig. 5. Termination rate constant vs. conversion in bulk polymerization of vinyl acetate
a t 50°C.

where A = 17.6620, A , = -0.4407, A , = -6.7530, and AS = -0.3495.

The application of eq. (2) to emulsion polymerization will be discussed in a
later section.
Effect of Emulsifier Concentration
The effect of emulsifier concentration on the polymerization rate was
investigated a t three different initiation rates. Figure 6 shows log-log plots
of R, versus emulsifier concentration. R, is calculated from the slope of
the linear portion of the conversion-versus-time plot. It appears that the
effect of emulsifier concentration is independent of the initiation rate.
From the slope of the straight lines, the emulsifier dependence exponent is
calculated t o be 0.12. Although this exponent is much smaller than the 0.6
power predicted in the classical theory of styrene emulsion polymerization,
there is, nevertheless, good agreement in the literature that the emulsifier
concentration does not affect the rate of polymerization t o the same extent
in vinyl acetate as in styrene emulsion polymerization.
The effect of the emulsifier concentration on the number of polymer
particles was investigated at a persulfate concentration of moles/l.
HzO. From the log-log plot shown in Figure 7, it appears that the number
of polymer particles increases with increasing emulsifier concentration; and
from the slope of the straight line the emulsifier dependence exponent is
calculated t o be 0.5 f 0.05, a value not far from the 0.6 power predicted
by the theory of Smith and Ewart.' The value of 0.5 is also in good agree-
ment with the data recently reported by Nomura et al.? From the rela-
tionship between emulsifier concentration and number of polymer particles
and polymerization rate, respectively, i t can be deduced that the poly-
 VINYL ACETATE POLYMERIZATION 2317

0.5 1 2 5 10 20 50
g l l H20 sodium lauryl sulphate
Fig. 6. Effect of emulsifier concentration on polymerization rate at different rates of
initiation: ( 0 )4 X (0)2 X (8) moles KBzOJ. HzO.

c
s! 10

1 I , , , I
2 3 L 5 10 15 20 25
g l l H 2 0 sodium lauryl sulphate
Fig. 7. Effect of emulsifier concentration on number of polymer particles.

merization rate is proportional t o the 0.25 power of the number of particles.

This exponent is in good agreement with the value 0.20 reported by Pat-
siga.
The small effect of the number of particles is not unique for vinyl acetate
emulsion polymerization. Thus, Ugelstad13 has reported that the order
of reaction with respect t o the number of particles was 0.05 t o 0.15 in vinyl
chloride emulsion polymerization.
Figure 8 shows a plot of limiting viscosity number [TI against conversion
at two different emulsifier concentrations. It appears t h a t there is no
significant change in [ T ] as the emulsifier concentration and the number of
particles are changed. This was also concluded by O’Neill14in a study of
the radiation-induced emulsion polymerization of vinyl acetate. GPC
2318 FRIIS AND NYHAGEN

a 9.5g/1 H20 sodium lauryl sulphate

a
28 - 8 x 1017 particles/l 0

o 2.4g/I H20 sodium lauryl sulphate

26
- 4 x1017particles/[
a

2.4 - 0

.
-
D
01 2 2 -
.
0

fi 20 -
7
a0

01 0.2 03 0.4 0.5 0.6 0.7 0.8 0.9

X

Fig. 8. Limiting viscosity number of poly(viny1 acetate) as function of conversion:

moles KS2O8/1. HzO.

measurements are now in progress in order to investigate if the MWD of

the polymers is also unaffected by the number of particles. This will be the
subject of a future publication.
Effect of Initiator Concentration
The influence of initiator concentration was investigated a t two different
emulsifier concentrations. Figure 9 shows a log-log plot of polymerization
rate versus initiator concentration. From the slope of the lines it is found
that the initiator dependence exponent is 0.5. This is in good agreement
with the findings of Dunn and Taylor and Gershberg6 but contradictory to
the results reported by Stannett5and Vanzo,2 where the rate was found t o

, I I
I

[K2S20J (molesll H20) x lo3

Fig. 9. Effect of initiator concentration on polymerization rate at two different emulsifier

concentrations: (0)9.5; ( 0 )2.4 g sodium lauryl sulfate /l. HzO.
 VINYL ACETATE POLYMERIZATION 2319

moles/\ H,O K,S,

06

04

t ,rnin
Fig. 10. Effect of addition on extra amount of initiator.

0 0.2 0.1 0.6 0.8 1.0

X
Fig. 11. Number of particles vs. converion at three different initiator concentrations: ( 0 )
2 X (8) (0)5 X moles KzS2Os/l. HzO. 9.5 g sodium lauryl sulfate/l.
HzO.

be of first order with respect t o initiator concentration. For vinyl chloride,

Ugelstad13 has reported a n order of 0.5 with respect t o initiator concentra-
tion.
The effect of the addition of an extra amount of initiator a t different
conversions was also investigated. Figure 10 shows t h a t the polymeriza-
tion rate increases when more initiator is added. From the slopes of the
curves it is found that the ratio between the rates before and after the
addition is approximately proportional t o the square root of the ratio
between the respective initiator concentrations. This effect is also ob-
served in vinyl chloride emulsion polymerization, l6 but not in styrene
polymerization.
2320 FRIIS AND NYHAGEN

Figure 11shows the number of polymer particles as a function of conver-

sion a t three diffcrent initiator concentrations. Although the points are
somcwhat scattered, i t is reasonable t o conclude that the initiator concen-
tration does not affect the number of particles. Also a t this point the
emulsion polymerization of vinyl acetate resembles that of vinyl chloride16
but is diffcrent from styrene, where the number of particles depends on the
initiator concentration t o the 0.4 power. Figure 11 also shows that the

[K2S20$= 5 ~ 1 0 moledl
-~ H20
2.8 . [K2S20& ld3 molesll H 2 0 .0

o [K2 s208]= 4 x moledl H 2 0 C

0
2.6
0
0
2.4
0

0 2.2
\ 0
d
0
'13
0
2.0 0 0
U

I I I I 1 I I 1 I I
0.1 Q2 03 0.4 0.5 06 0.7 0.8 0.9
X
Fig. 12. Limiting viscosity number of poly(viny1 acetate) as function of conversion: 9.5 g
sodium lauryl sulfate/l. HzO.

number of polymer particlcs remains constant between 10% and 100~o. A

similar result has been reported by Napper and Parts.g
I n Figure 12 is plotted the limiting viscosity number [ T ] as a function of
conversion at three different initiator Concentrations, and i t appears that
[77] is independent of the initiator concentration. The independence of [ q ]
on initiator concentration and number of polymer particles suggests that
the molecular weight is controlled primarily by transfer t o monomer
and polymer. GPC measurements will confirm whether this is the case.
 VINYL ACETATE POLYMERIZATION 2321

Effect of Electrolytes
Stannett et al.5 have investigated the effect of adding electrolytes.
They have found that both phosphate buffer and potassium sulfate in-
creased the rate of polymerization. Furthermore, they have found that in
the presence of electrolytes the dependence of rate on initiator was de-
decreased. An attempt to reproduce these effects failed. In our experi-
ments, the presence of electrolytes in concentrations of 0.05 moles/l. H,O
did not affect the rate of polymerization, nor the dependence on initiator
concentration.

Discussion
From the data presented here, the following points can be established
concerning vinyl acetate emulsion polymerization :
1. The average concentration of radicals per particle increases with
increasing initiator concentration.
2. The rate of polymerization is approximately proportional to the
square root of the initiator concentration.
3. The rate of polymerization is proportional to the 0.25 power of the
number of particles.
4. The number of polymer particles remains constant in the interval
10% to 100~o conversion.
5. The number of polymer particles is independent of the initiator
concentration.
6. The number of polymer particles is proportional to the 0.5 power of
the emulsifier concentration.
7. The limiting viscosity number is independent of the initiator con-
centration, emulsifier concentration, and number of polymer particles.
This picture does not resemble that of styrene emulsion polymerization,
but rather that of vinyl chloride emulsion polymerization. Ugelstad13
has recently presented a model for vinyl chloride emulsion polymerization,
and it is suggested that the same model could be applied to the vinyl acetate
system.
Presentation of the Model
For vinyl chloride, Ugelstad18has deduced the rate expression

where k, = propagation rate constant, [M,] = monomer concentration

in the polymer particles, NA = Avogadro's number, k , = decomposition
rate constant of initiator, f = initiator efficiency factor, [I] = initiator
concentration, lip = total volume of the monomer-swollen polymer
particles, k,, = termination rate constant, N , = total number of polymer
particles, and kd = desorption rate constant.
2322 FRIIS AND NYHAGEN

0.0 I I I I I
1

I 1 I I I I
0.2 0.4 0.6 0.8 ID
X

Fig. 13. Plot of log ( D , / D p o )vs. conversion. Computed from experimental data with
aid of eqs. (3) and ( 5 ) .

To fit the rate expression given in eq. (3) t o experimental data, Ugelstad
defined the desorption rate constant k d as

~ transfer constant to monomer, D , is the self-diffusion

where l ~ is, the
constant of a monomer radical in a polymer particlc, and K is a numerical
constant. Equation (4) expresses that only radicals formed by transfer to
monomer can escape the particles.
I n this paper, k , will be defined in a different manner. Suppose a
monomer radical is formed by transfer to a monomer molecule. This
radical can either escape the particle, or it can add a monomer molecule to
form a dimeric radical. The probability for the radical to escape is given as

where the diffusion term is obtained from Einstein's diffusion equation"

and the radius of the particlc T is choscn as the mean displacement necessary
for the radical to escape the particle. If desorptiori of dimeric and larger
radicals is neglected owing t o a rapid decrease in D,, the desorption rate
constant is given as thc product of @ and the frequency a t which monomer
radicals are being formed :
kd = @ kfm [RIP]IN*- (6)
A certain fraction of the radicals entering the particles from the water phase
will escape the particles before adding a monomer molecule. This is
accounted for in the absorption rate constant, which, however, drops out in
the final rate expression when termination in the water phase is neglected.
 VINYL ACETATE POLYMERIZATION 2323

During the polymerization, both D, and k,, decrease. The decrease in

k,, is given as function of conversion in eq. (2). It is therefore possible t o
calculate the decrease in D, necessary for eq. (3) to fit the experimental
results. Figure 13 shows the relative decrease in D, thus obtained. It
appears that the diffusion coefficient of monomer radicals should decrease by
a factor 100 in the interval 0% t o 90% conversion. This is not unreason-
able. I n an investigation of the diffusion of water through swollen polymer
membranes, Peterlin et a1.18 have found a similar decrease in the diffusion
coefficient in going from a highly swollen t o a dry membrane. Peterlin et
al. l 8 deduced the following expression for the self-duffision coefficient of low
molecular weight compounds in polymers :

D, = DPoexp(-Px,(l - a)/(1 + ax,)) (7)

where DPois the self-diffusing coefficient of the diffusing compound in its
own medium, P = V*/V,,, a = V,,/V,,, and x , = (1 - H ) / H . V* is a
critical free volume fraction necessary for diffusion t o take place, V,, and
V,, are free volume fractions of monomer and polymer, respectively, and H
is the volume fraction of the low molecular weight compound. Since

where x is conversion, d , and d , are densities of monomer and polymer,

respectively, the following relationship between the self-diffusing coefficient
and conversion is obtained:

-Pxdm(l - a )
D, Dp0exp
=
+
(1 - ~ ) d , axd,

Substituting eq. (8) into eq. (5) and eq. (5) into eq. (3), the values of a and P
necessary t o describe the obtained data are found t o be a = 0.3 and /3 =
3.2. I n this computation, the following values of the constants were used:
k, = 3500 l./mole-sec, 2k$ = 10-6/sec, k,, = 0.75 l./mole-sec, d , = 1150
g/l., d , = 900 g/l., and DPo = dm2/sec. [M,] was calculated from
eqs. (9) and (10) :

where i,is the conversion a t which the separate monomer phase disappears;
lc,, was calculated from eq. (2).
The and P values cannot be calculated theoretically with great accu-
racy. However, comparison with data from the literature18 shows that the
2324 FRIIS AND NYHAGEN

values computed here are not unreasonable. One should expect the value
of a to lie within 0.15 and 0.40.
With the values of a and fi obtained by fitting eq. (3) to a single experi-
mental curve, we were able to simulate all our experimental data (more than
40 experiments) in the range of conversion from 0% to 100%. Figure 14

10

08

06
X

04

02 4 . 5 ~1017

40 80 120 160 200 240 280 320

1, min
Fig. 14. Comparison between calculated and experimental conversion-vs.-time plots.
The solid line represents values calculated from eq. (3).

I I I I

0.2 0.L 0.6 0.8

X

Fig. 15. Contribution of the term N A ~ V , / to

~ ~polymerization
, rate as function of con-
version: (A) 5 X 10'6 particles./l.; (B) 10l8particles/l.

shows a comparison between experimental and calculated conversion-

versus-time plots. It is seen that the acceleration observed a t high conver-
sions in experiments with large polymer particles is reflected by the mathe-
matical model. Furthermore, the model correctly reflects the effect of
number of particles.
 VINYL ACETATE POLYMERIZATION 2325

Although the termination constant decreases rapidly during the poly-

merization, the term N A 2 V p / k lonly
p plays a secondary role for the rate of
polymerization. This is illustrated in Figure 15, where the contribution of
this term is calculated in per cent of the total polymerization rate as function
of conversion at diff erent concentrations of polymer particles.
Thus, it can be concluded that the constant rate behavior in vinyl acetate
emulsion polymerization is mainly due to a decrease in the desorption rate
constant.

Comparison with Literature

Stannett et al.5 have performed a very comprehensive study of the emul-
sion polymerization of vinyl acetate. Although the recipe used by these
investigators is very similar to that used in the present work, a comparison
of the results shows large deviations. Therefore, an attempt to apply their
model to our data failed. The model deduced by Stannett et al. predicts
that the polymerization rate decreases approximately with the fourth
root of monomer concentration from the point of disappearance of the
separate monomer phase. This is not in agreement with the constant rate
behavior observed in this work. Also, the dependence of polymerization
rate on initiator and emulsifier concentration is different.
However, excellent agreement is obtained by comparing the data
reported by Nomura et al.? to our data. In a study of the effect of polymer
particles in emulsion polymerization, these investigators have deduced a
general rate expression for emulsion polymerization. l9 Application of this
expression to vinyl acetate7leads to an expression similar to that deduced by
Ugelstad for vinyl chloride polymerization. l 3

CONCLUSIONS
I n conclusion, we may say that the emulsion polymerization of vinyl
acetate is similar to that of vinyl chloride. In both systems, transfer
to monomer seems t o play an important role in comparison with styrene
emulsion polymerization.
However, one important problem still remains to be solved in the
emulsion polymerization of vinyl acetate and vinyl chloride, namely,
the mechanism by which the particles are formed. I n styrene emulsion
polymerization, the number of particles is proportional t o the 0.4 power of
initiator concentration. In the vinyl acetate system, there is no effect of
initiator concentration on the final number of particles.
In accordance with the classic theory for styrene emulsion polymeriza-
tion, the polymer particles are generated solely by the primary initiator
radicals. In the vinyl acetate system the ability of radicals to escape the
particles must give rise to formation of particles by such radicals in addition
to the particles being generated by the primary initiator radicals. This
difference might be contributory to the different effect of initiator concen-
tration on particle formation in the two systems.
2326 FRIIS AND NYHAGEN

Nomenclature
density of monomer
density of polymer
self-diffusion coefficient of monomer radicals in pure monomer
self-diffusion coefficient of monomer radicals in monomer-swollen
polymer
initiator efficiency factor
volume fraction of monomer in monomer-swollen polymer
initiator concentration
desorption rate constant
transfer rate constant t o nionomcr
decomposition rate constant of potassium persulfate
propagation rate constant
decomposition rate constant of a,a'-azoisobutyronitrile
termination rate constant in monomer-swollen polymer
monomer concentration in bulk polymerization
monomer concentration in polymer particles
Avogadro's number
total number of polymer particles
average radius of polymer particles
rate of polymerization
reaction time
volume of the polymer phase
critical free volume fraction
free volume fraction of monomer
free volume fraction of polymer
degree of conversion
ratio between volume fraction of polymer and monomer in monomer-
swollen polymer
degree of conversion when the separate monomer phase disappears
Greek Symbols
a! Vfn/VIm
P V*/Vf7n
[v J limiting viscosity number
The authors wish to thank Professor A. E. Hamielec and Peter Bo for valuable discus-
sions on emulsion polymerization and for their criticism of the manuscript.

References
1. W. V. Smith and R. H. Ewart, J . Chem. Phys., 16,592 (1948).
2. E. Vanzo, Ph.D. Thesis, State University College of Forestry a t Syracuse Univer-
sity, Syracuse, N.Y., 1962.
3. R. A. Patsiga, Ph.D. Thesis, State University College of Forestry a t Syracuse
University, Syracuse, N.Y., 1962.
4. V. Stannett, M. Litt, and R. Patsiga, J . Phys. Chem., 64,801 (1960).
5. V. Stannett, &I. Litt, and R. Patsiga, J . Polym. Sci. A-l,8,3607 (1970).
 VINYL ACETATE POLYMERIZATION 2327

6. D. Gershberg, paper presented at Joint Meeting of A.1.Ch.E. and 1.Ch.E. (En-

gland), London, June 14, 1965.
7. M. Nomura, M. Harada, K. Nakagawara, W. Eguchi, and S. Nagata, J . Chem.
Eng. (Japan) 4 (No. 2), 160 (1971).
8. D. H. Napper and A. G. Parts, J . Polym. Sci., 61, 113 (1962).
9. P. Harriott, J . Polym. Sci. A-1, 9 , 1153 (1971).
10. I). M. French, J . Polym. Sci., 32, 395 (1958).
11. G. E. Ham, Vinyl Polymerization, Vol. 1, Part I, Marcel Dekker, New York, 1967,
pp. 208-329.
12. F. Danusso, G. Pajaro, and D. Sianesi, Chim. Znd. (Milan), 41,1170 (1959).
13. J. Ugelstad, P. C. Mork, P. Dahl, and P. Itagnes, J . Polym. Sci. C, 27,49 (1939).
14. T. O’Neill, J. Pinkava, and J. Hoignd, paper presented at the 3rd Symposium on
Radiation Chemistry, Tihany, Hungary, May 10-15, 1971.
15. A. S. Dunn and P. A. Taylor, MakromoE. Chem., 83,207 (1965).
16. J. Ugelstad znd P. C. Mork. Br. Polym. J., 2 , 3 1 (1971).
17. A. Sheludko, Colloid Chemistry, Elsevier, Amsterdam, 1966, pp. 47-55.
18. A. Peterlin, H. Yasuda, and C. E. Lamaze, J . Polym. Sci. A d , 9,1117 (1971).
19. M. Nomura, M. Harada, W. Eguchi, and S. Nagata, J . Chem. Eng. (Japan), 4
(No. l ) , 54 (1971).
Received September 6, 1972
Revised December 20, 1972