Analytical Study of ABS Copolymers Using a Preparative Ultracentrifuge

15 Analytical Study of ABS Copolymers Using a Preparative Ultracentrifuge Β. CHAUVEL and J. C. DANIEL

Downloaded by FUDAN UNIV on December 18, 2016 | http://pubs.acs.org Publication Date: June 1, 1975 | doi: 10.1021/ba-1975-0142.ch015

Centre de Recherches Rhône-Progil, 93308 Aubervilliers, France

A quantitative method has been developed to separate free and graft copolymers in an ABS sample. The ABS powder is dispersed in MEK and then introduced into the cells of a preparative ultracentrifuge. After the reproducibility of the procedure was ascertained, the method was used to determine the grafting parameters of samples polymerized under specific conditions. This analytical technique is well suited to demonstrate how the grafting efficiency or grafting density is influenced by various polymerization conditions such as mercaptan content, monomer flow rate, emulsifier content, or polybutadiene content. The effects of other variables such as temperature, the initiator system, and characteristics of the polybutadiene latex can also be demonstrated.

W

hen a mixture of styrene and acrylonitrile is polymerized i n the presence of a polybutadiene latex by an emulsion radical process, an acrylonitrilebutadiene-styrene ( A B S ) copolymer is obtained. This A B S copolymer is actually a mixture of (a) a graft copolymer w h i c h contains some of the styrene/acrylonitrile ( S T / A N ) copolymer chemically bound to the polybuta diene backbone, and ( b ) a random copolymer, conventionally designated as a linear copolymer, w h i c h is not bound to the polybutadiene backbone but w h i c h consists of the portion of the styrene/acrylonitrile monomer that has poly merized separately. The separation of these two components enables us to study their macromolecular characteristics and to determine the quantity of S T / A N copolymer w h i c h is bound to the substrate. If it is sufficiently accurate, a quantitative separation technique provides a valuable source of information for correla tions between the properties of A B S copolymers and their structural charac teristics. It can also demonstrate how the A B S structure depends on different polymerization variables. Principle of the Analytical Method As far as we know, the first application of ultracentrifugation to phase separation of graft copolymers was made by Shashoua a n d V a n H o l d e ( 1 ). 159

Platzer; Copolymers, Polyblends, and Composites Advances in Chemistry; American Chemical Society: Washington, DC, 1975.

160

COPOLYMERS,

Table I.

POLYBLENDS,

AND

COMPOSITES

Reproducibility of the Separation Technique as Shown by Values of L ' a

Centrifugation Run Cell Number

1

2

3

1 2 3 4 5 6

76.0 76.6 77.8 77.2 76.7 76.9 76.9

76.2 76.3 76.9 76.2 76.1 76.1 76.3

76.4 76.7 76.7 76.4 76.2 76.0 76.4

Mean


a

In wt

%.

This method was subsequently used by Gesner ( 2 ) , by Moore and Frazer ( 3 ) , and by Huguet and Paxton (4) for A B S copolymers. The diameter of the polybutadiene latex particles used i n A B S synthesis usually ranges from 0.1/A to 1/x, and the polymer is always crosslinked within the particles. Consequently, after grafting, every polybutadiene particle be comes a particle of grafted and crosslinked polymer w h i c h is m u c h heavier than the linear copolymer molecules. Table II.

Analysis of Variance

Source of Variance

Sum of Squares

Degrees of Freedom

Variance

Between runs Retween cells Residue Total

1.0978 1.6644 1.0089 3.7711

2 5 10 17

0.549 0.333 0.101

The separation technique consists of dispersing an A B S powder i n an appropriate medium in which the linear copolymer is soluble and then centrifuging the dispersion. The graft copolymer microgels settle more rapidly than the free copolymer, despite the lower density of the S T / A N copolymer. S, the Svedberg sedimentation coefficient, depends on the molecular weight, according to the following equation (top of p. 162) : Table III.

Effect of

A ntioxidant Sample

1 2 3 4 5 6 7 8 9 10 11 12 13

TNPP,

DBPC,

TDM,

%

%

%

1.25 1.25 — 1.25 0.5 — 0.5 1.25 0.5 1.25 0.5 0.5 1.25

5 5 2 5 — 2 — 5 — 5 — — 5

0 0.1 0.1 0.2 0.2 0.3 0.3 0.4 0.4 0.5 0.5 0.6 0.6

ΙΛ

G,

%

%

%

52.1 67.8 64.3 70.3 72.1 74.5 74.3 76.0 74.7 79.8 76.5 80.3 80

47.1 36.8 37.1 30.2 30.3 27.5 28.0 23.5 26.6 21.6 25.4 23.8 19.5

99.2 104.6 101.4 100.5 102.4 102.0 102.3 99.5 101.3 101.4 101.9 104.2 99.5


L'+G,

15.

CHAUVEL AND DANIEL

Analytical Study of

ABS

G% e

-180

(A)

70


30 t /

60 50

20

40 30 20

10

10

^ 0,1 Figure 1.

(B)

0,5 T.D.M.conrenr. 4. o

Effect of TDM content on linear copolymer content (curve A) and on grafting efficiency (curve B)

the T D M Content

Ge,

%

40 19 18 16 10 9.5 7.2 8 6.5 2.5 4.5 0.1 2

G , d

%

160 75 68 65 39 38 29 32 27 10 18 1 9

Λ/w, X 10"

*(L') — 92 98 74 78 62 59 46 52 46 46 43 46

— 83.9 88.9 68.7 72.1 58.2 55.6 43.9 49.3 43.9 43.9 41.1 43.9

— 270 295 195 210 150 140 95 112 95 95 85 95

D , X 10 B

3

4

— 1.5 1.25 1.8 1.0 1.35 1.1 1.8 1.3 0.55 1.0 0.06 0.6


162

COPOLYMERS,

S

=

POLYBLENDS,

A N D COMPOSITES

Rf—

where M is the molecular weight, D the diffusion constant, p the solvent density, V the partial specific volume of the polymer, R the gas constant, and Τ the absolute temperature. }

g


Separation Technique A n A B S sample prepared i n our laboratory was used to develop the separation technique. Its composition was (wt % ) : butadiene, 20; acrylo nitrile, 21; and styrene, 59. T h e A B S powder was dispersed i n methyl ethyl ketone ( M E K ) at 1 wt % concentration. M E K was chosen because it is a good solvent for styrene/acrylonitrile copolymers, it is a poor solvent for polybuta diene, and its density is lower than that of polybutadiene and therefore lower than the density of the graft copolymer. Experiments w i t h an A n a l y t i c a l Ultracentrifuge. A t first the sedimenta tion phenomenon was studied with an analytical ultracentrifuge (Spinco model E ) equipped with a Philpott-Svenson optical device that enabled us to follow the moving boundaries of the different macromolecular species during a centrifugation run. W e observed that with progressive increase of the rotor speed, the graft copolymer settled immediately at 30,000 g whereas sedimenta tion of the linear copolymer was negligible. W i t h an acceleration of 200,000 g, the linear copolymer d i d not settle until 3 hrs later. This observation agrees w i t h those of Moore and Frazer (3) who reported good phase separation of an acetone suspension of A B S b y centrifugation at 40,000 and 50,000 g. Separation w i t h a Preparative Ultracentrifuge. A n M E K dispersion (20 ml) of A B S was centrifuged i n a preparative ultracentrifuge (Spinco model L ) . The type 30 rotor operating at 25,000 rpm provides 35,000 g at the top of the sample cell and 70,000 g at the bottom. T h e run lasted 45 m i n . Afterwards, the graft copolymer was recovered as a deposit adhering to the bottom of the cell. T w o successive, similar runs were made with the graft copolymer deposit being redispersed i n fresh M E K before each run. The two washing solutions were studied i n the analytical ultracentrifuge, and no linear copolymer was detected i n the second solution. Considering the sensitivity of the optical system, we estimate the quantity of linear copolymer collected i n the second washing solution to be less than 1 w t % of the initial A B S . I R spectroscopy detected only traces of polybutadiene i n the linear Table I V .

α

Sample

M onomer Introduction, hrs

1 2 3 4 5 6 7 8

0 0 0 3 3 5 8 20

DBPC,

% 2 2 2 2 2 2 2

a

Effect of

ΙΛ

G,

%

%

L'+G,

%

79.4 80.1 79 77.4 77.9 74.7 70 60.4

21.2 20.4 24.5 25.4 24.3 26.9 32.6 43.2

101.1 100.5 103.5 102.7 102.2 101.6 102.6 103.6

Stabilized by a mixture of hydroquinone, 0.1% and β naphthylamine, 0.1%.


15.

CHAUVEL

AND DANIEL

Analytical Study of

163

ABS

copolymer solution. T h e polybutadiene content of this solution was estimated at less than 0.5 wt % based on the initial A B S or less than 2 wt % based on the initial polybutadiene. These data i m p l y that all the sediment consists of microgels of graft copolymer. If a certain amount of polybutadiene were not grafted and not crosslinked, it w o u l d be insoluble i n M E K and then appear when the A B S sample is dispersed i n M E K . Although Gesner (2) detected soluble pure polybutadiene i n A B S compounds, this might be the result of different prepara tion techniques.


Determination of the Grafting Parameters in ABS Copolymers Our separation technique was used to determine the grafting parameters of an A B S sample prepared i n our laboratory to check the reproducibility of the method. Experimental. A 1 % dispersion of A B S powder i n M E K was prepared at 20 °C and introduced into six stainless-steel cells fitted on a type 30 rotor of an ultracentrifuge (Spinco model L ) . E a c h cell contained 20 m l of the dispersion. T h e ultracentrifuge was r u n for 45 m i n at 25,000 r p m and 2 0 ° C . Afterwards, aliquot s of the linear copolymer supernatant solutions were re moved from the cells b y syringe, placed i n flasks, and stored at 20°C. The graft copolymers, which were adhering to the bottom of the cells, were redispersed i n 20 m l fresh M E K . After 15 hrs, these dispersions were centrifugée! as above. T h e washing solution was then removed and replaced with fresh M E K for a second washing. After the third centrifugation, the washing solutions were added to the first linear copolymer solution. T h e graft copolymers were recovered by dispersing the remaining sediments i n M E K . Methanol was added to the various solutions or dispersions, and the con tents were then dried in vacuo at 50°C under air. It was then easy to deter mine G (wt % ) , the graft copolymer content, and U (wt %) = L, linear content, plus /, ingredient content. Emulsifiers a n d antioxidants which are soluble i n M E K are among the ingredients. Since J and Β (butadiene content) are known, it is possible to calculate: G , w t % of S T / A N grafts bound to 100 g polybutadiene d

G d

.

100 -(L>+B)

χ

1 0 0

G , the grafting efficiency, i.e., wt % of S T / A N grafts per 100 g of mono e

mer

"

100-(£+/)

X

°°

the M o n o m e r F l o w Rate G, e

%

1.5 0.6 2.0 4.1 3.4 7.5 13 24

Gd,

%

T7r(L')

6.1 2.5 8.2 16.5 13.9 30 54 98

75 83 63 85 88 48 43 —

Mw, X 10" 69.5 76.4 59.1 78.1 80.5 45.5 41.2 —

200 235 153 240 250 100 85 —

3

X 10 0.15 0.05 0.3 0.35 0.3 1.6 3.45 —


4

164

COPOLYMERS,

POLYBLENDS,

A N D COMPOSITES

L ' %

20

80 70 60


15

50 40

10

30

5L

20 10 5 10 15 Introduction Mme (hours)

Figure 2.

20

Effect of monomersflowrate on linear copolymer content (curve A) and on grafting efficiency (curve B)

η the reduced viscosity of the linear copolymer. (Viscosity was measured at 3 0 ° C w i t h 0.5 g copolymer/100 m l M E K . ) Discussion. T h e reproducibility of the experimental data were influenced by: (a) Antioxidant volatility. A n antioxidant is always added to the polymer to prevent oxidation during drying. Since most of the commonly used anti oxidants are volatile, the material balance is disturbed. The importance of this phenomenon depends on the nature of the antioxidant and on drying condi tions. T o minimize this factor, we chose a drying temperature below 5 0 ° C . (b) T h e tendency of the dry polymer to absorb moisture. Because of this, it is necessary to keep a l l copolymers i n a d r y atmosphere, over Ρ Ο , after centrifugation. (c) T h e copolymers ability to retain solvent. It is impossible to obtain polymers b y evaporating the solvent, a n d thus it is necessary to precipitate them with methanol and to dry them in vacuo. η

2

δ

Reproducibility. Three different centrifugation runs were carried out w i t h the 1 % A B S dispersion. Six cells were used i n each experiment for a total


15.

CHAUVEL A N D DANIEL

Analytical Study of

165

ABS

of 18. T h e L ' values are given i n Table I. F o r the sample studied, analysis of variance is presented i n Table II. T h e F ratio for the runs, 0.549/0.101 = 5.44, exceeds the tabulated 5 % critical of 4.1 but not the 1 % value of 7.56. Therefore, there are differences among the runs. T h e F ratio for the cells does not exceed the 5 % critical value. If w e consider the L ' value w i t h 15 degrees of freedom, we find an 0.18 variance. Hence the standard deviation of a single determination is 0.425, and the standard deviation of each Z? is 0 . 4 2 5 / V 6 = 0.175.


U s i n g the Fisher-Student function, the 9 5 % confidence limits for U (com puted for 15 degrees of freedom) are: 76.86 ± 0.37; 76.30 ± 0.37; a n d 76.40 ± 0.37. The best estimate for U is 76.41 which is the average obtained from the data i n Table I. T h e n 76.41 ± 0.26 are the 9 5 % confidence limits for the average (17 degrees of freedom). Application—Influence

of Polymerization Conditions on Grafting Parameters

Sample Preparation—Standard Recipe. A l l samples were prepared i n the same reactor with the same stirring conditions. A typical polymerization was performed according to the following recipe: Backbone latex: commercial F R S 2004 polybutadiene latex (Firestone Co.). Reactor feed ( b y w e i g h t ) : polybutadiene, 20; acrylonitrile, 2 1 ; styrene, 59; tertiary dodecyl mercaptan ( T D M ) (Aquitane-Total-Organico), 0.3; so dium Dresinate (Dresinate 731, Hercules Powder C o . ) , 1. Polymerization temperature, 60°C. Initiator system, potassium persulfate. A mixture of monomers and T D M is added with mixing to the polybuta diene substrate for 5 hrs i n the presence of the initiator system a n d the emul sifier. After polymerization, latexes were stabilized by adding 2,6-di-ferf-butylp-cresol ( D B P C ) or tri(nonylphenyl) phosphite ( T N P P ) as antioxidants. They were then coagulated with a calcium chloride solution and finally dried at 5 0 ° C for 15 hrs. Effect of Mercaptan Content. In the polymerization recipe, T D M is used as a transfer agent. It is interesting to know h o w each grafting parameter depends on this agent. T h e experimental results are presented i n Table III and Figure 1. A s the T D M content increased from 0.0 to 0.6, G decreased from 47 to 2 3 % , G decreased from 160 to nearly 0 % , G decreased from d

c

Table V .

Sample

1 2 3 4 5 6 7 8 9

B,

Effect of the Polybutadiene Content

Antioxidant, %

%

DBPC

5 10 15 20 20 25 30 30 50

5 d° d° d° d° d° d° 2 2

TNPP

1.25 d° d° d° d° d° d° 1 1

%

G,

%

L'+G,

G ,

G,

Monomer Flow Rate, ml/hr

93.4 86.4 81.2 71.1 72.5 68.4 53.4 56.1 33.8

6.7 14.3 20.1 29.5 30.3 33.5 48 46.3 66.7

100.1 100.7 101.3 100.6 102.8 101.9 101.4 102.4 100.5

40 43 34 56 45 35 65 51 37

2 5 6 14 11 13 28 22 39

31.5 30 28 26.5 26.5 25 23.5 23.5 16.5

ΙΛ

%

d

%

e

%



166

COPOLYMERS,

POLYBLENDS, A N D COMPOSITES

Butadiene content % Figure 3.

Effect of butadiene content on linear copolymer content (curve A) and on grafting efficiency (curve B)

40 to nearly 0 % , and η,, decreased from 1 to 0 . 4 % . [ 7 7 ] , the intrinsic viscosity, can be calculated from the value of 77,. using the equation given by Gerrens et al. (5) : [ η ]

=

1 + 0.21„

βρ

Table V I . Effect of Temperature Initiator: K2S2O8 Sample

1 2 3

T, °c 50 60 70

G,

L'+G,

G ,

G,

%

%

%

%

%

%

2 2 2

78.3 73.8 67.9

23.8 28.3 34.3

DBPC, ΙΛ

d

[-7]

e

102.1 8.5 2 8 102.1 31 16 102.2 60

x 10- x 10 200 0.2 120 1.4 3

77 53 42

71.3

50.2 40.2

83


4

3.9

15.

CHAUVEL AND DANIEL

Analytical Study of

Table V I I .

Effect of

167

ABS

Temperature

Initiator: Redox Catalyst Sample T , °C 1 2 3

60 50 40

TNPP,

L',

%

%

0.5 0.5 0.5

59.3 59.3 63.2

o,

L'+G,

G ,

Ge,

%

%

%

%

43.3 43.4 38.9

102.6 102.7 102.1

90 90 72

24 24 19

F r o m [ 7 7 ] we can derive M

d

W 44 44 69

X

42 42 64.3

M , w

10"

[η] = 3.6 Χ ΙΟ" Tï 2

w

X 10' 5.5 5.5 2.2

89 89 178

using the relationship of Shimura et al.

w

3

(6):

0 6 2

It is then easy to calculate D the grafting density, according to Dinges and Schuster ( 7 ) . D is the average number of grafts per monomer unit of the substrate. Calculations have been made assuming that M is the same for the graft and the free copolymer. This assumption is based on the data of (J9


g

w

/ J

40

I

1

50 60 ο Temperature C C)

1

70

Figure 4. Effect of temperature with different catalysts on grafting efficiency (curve A) and on linear copolymer content (curve B)


168

COPOLYMERS,

POLYBLENDS,

AND

COMPOSITES

Dinges and Schuster and has been verified by separating the grafts from the backbone by an ozonolysis technique. B y this method, we found that the re duced viscosity of the grafts is very close to that of the free copolymer. The data i n Table III show that D is nearly constant within 0.1-0.4% T D M . This means that the T D M content has a strong influence on the molecular weight of the grafts but practically no influence on the number of grafts. W h e n the T D M content is increased above 0.4%, however, the situation is apparently changed since D then decreases. g

g

Effect of the M o n o m e r F l o w Rate. The rate at w h i c h the monomer mix ture is introduced into the reactor greatly affects the grafting parameters. A s the time for introduction was increased from 0 to 20 hrs, we observed (see Table I V and Figure 2) that: G increased from 21 to 4 3 % , G increased from 5 to 100%, G increased from 1 to 2 5 % , and 77,. decreased from 0.75 to 0.43. W h e n the monomer flow rate decreases, the monomer concentration i n the polymer phase increases as does the probability for grafting. Thus, the varia tions in G , G and G can be explained easily. d


e

(h

e

It is surprising that 77,. is not constant when the monomer flow rate is varied. The decrease i n 77,. when decreasing monomer flow rate is probably related to a mercaptan influence; the mercaptan consumption rate is m u c h higher than the monomer consumption rate. Thus, if the introduction rate of the monomer mixture is high compared w i t h the polymerization rate, the mercaptan efficiency is poor, and the average molecular weight is high. O n the other hand, if the monomer introduction rate is low, mercaptan is efficient throughout the reaction, and the macromolecular chains are shorter. W h e n D is calculated, using the data of Table I V , it increases continuously when the monomer flow rate is decreased. (J

Effect of the Polybutadiene Content. A series of A B S samples was pre pared using 0.1 part T D M and varying the polybutadiene content from 5 to 5 0 % . The introduction time of the monomers was a constant 3 hrs. W h e n the polybutadiene content increased, the grafting efficiency increased from 2 to 4 0 % (see Table V and Figure 3 ) . One must note, however, that an increase in the polybutadiene content automatically decreases the monomer flow rate. It ensues that the curves i n Figure 3 result from the simultaneous effects of two parameters, the polybutadiene content and the monomer flow rate. Nevertheless, it seems reasonable to think that the concomitant varia tions of these two parameters produce effects i n the same direction. The polymer/monomer ratio increases when the polybutadiene content is increased and when the monomer flow rate is decreased. The grafting efficiency probably depends on this ratio, but the relative importance of the two parameters cannot be ascertained from our data. Effect of Temperature and the Initiator System. Polymerization tempera ture was varied from 50° to 70°C with K S O as initiator and from 40° to 60°C using a redox catalyst. The redox catalyst was (wt % ) : cumene hydroperoxide, 0.75; ferrous sulfate ( F e S 0 · 7 H 0 ) , 0.01; dextrose, 1; and sodium pyro phosphate, 0.5. Cumene hydroperoxide was mixed w i t h the monomers; the other ingredients were added into the emulsifier solution. The experimental data (see Tables V I and V I I and Figure 4) show that: 2

4

2

s

2

(a) The graft efficiency increases and the viscosity decreases w i t h increase in temperature. These results were expected since the transfer reaction con stants depend on temperature more than the propagation constants.


15.

Table V I I I . Sample

A A A A A


a

B,

L',

G,

L'+G,

G ,

G ,

6 7 8 9 9.5

35.8 28.2 26.6 21.5 19.8

56.8 63.1 67.6 71.6 71.9

45.2 39.1 34.7 31.2 31.1

102.0 102.2 102.3 102.8 103.0

21 31 22 32 42

12 12 8 9 10

b

b

Polymerization w i t h y - R a y s

Irradiation Time, hr

1 2 3 4 5 a

169

Analytical Study of ABS

CHAUVEL AND DANIEL

%

%

e

d

%

%

%

%

Monomers were placed i n the reactor for 9 hrs. η was 0.67. Γ

(b) W h e n the redox catalyst is used, the effect of temperature is lessened as it rises above 45°C. W e assume that this phenomenon is associated w i t h the shorter life time of the catalyst above this temperature. T h e very rapid decomposition of the catalyst yields a high instantaneous radical concentration; the radicals then destroy each other instead of attacking the polybutadiene backbone, according to Dinges and Schuster ( 7 ) . Table I X .

Effect of the Polybutadiene Backbone Grafting onto L P F 1351

Sample TNPP,

% 1 2

0.5 0.5

L',

G,

L'+G,

Gd,

G ,

%

%

%

%

%

54.6 56.0

46.5 46.1

101.1 102.1

145 140

M

e

33 31

54 60

57 64

M , X 10" w

131 159

3

X 10

4

6.0 4.5

(c) A s several workers have indicated (7, 8, 9, 1 0 ) , grafting strongly depends on the nature of the initiator system. U n d e r our polymerization con ditions, grafting is promoted when persulfate is replaced b y the redox catalyst. Thus, D increased by a factor of 3.5 when the operating temperature was 50°C. Polymerization w i t h γ-Rays. A few experiments were conducted using a C o radiation source. T h e data obtained in one such experiment are presented in Table V I I I . T h e radiation dose rate was 4000 rad/hr. T h e monomers were placed i n the reactor for 9 hrs. F r o m time to time, latex aliquots were removed to study the polymer characteristics as a function of time. T h e results d i d not show any important difference from a classical polymerization r u n carried out under the same conditions with K S , O used as catalyst. Effect of the Substrate Latex Characteristics. M a n y authors (4, 11, 12, 13) have noted the great influence of the latex substrate on the structure a n d properties of A B S . A m o n g the characteristics of the backbone latex, three have been reported as being very important. These are the degree of crosslinking of the polymer substrate (gel content, swelling i n d e x ) , the average g

6 0

2

Table X .

1

s

Influence of the Emulsifier Content

Antioxidant Sample

Soap,

TNPP,

DBPC,

L',

G,

L'+G,

Gd,

G ,

1 2 3 4 5 6

0 1 2 3 2 5

1.25 d° d° d° d° d°

5 d° d° d° d° d°

58.6 62.9 63 64.5 67.3 66.1

43.4 37.2 36.7 34.9 34.2 35.1

102 100.1 99.7 99.5 101.5 101.2

107 86 85 77 64 70

27 21.5 21 19 16 17.5

%

%

%

%

%

%

%


e

%


170

COPOLYMERS,

I

Figure 5.

ι

I

POLYBLENDS,

ι

I

AND

ι

1 2 3 4 5 Emulsifier content (wt % )

COMPOSITES

1

Effect of emulsifier content on linear copolymer content (curve A) and on grafting efficiency (curve B)

particle size of the rubber latex, and the nature of the polymer substrate (chemical composition, glass transition temperature). The data reported i n Table I X were obtained i n experiments i n which the polybutadiene latex used was Goodyear L P F 1351 rather than F R S 2004. This is a fine particle latex; the particle diameter ranges from 500 to 5000 A , the average being about 800 A . The degree of crosslinking of the polymer is very close to that of F R S 2004. W h e n using this latex w i t h a higher specific area, the standard recipe leads to G , G , and D values which are quadruple those obtained with A B S grafted onto F R S 2004. The average molecular weight of linear copolymer is not modified. Effect of Emulsifier Content. Table X and Figure 5 show that grafting efficiency decreases continuously when the emulsifier content varies from 0 to 5 % . This phenomenon probably results from an increase i n the micelle con centration w h i c h promotes the polymerization of a larger proportion of monomer in the water phase. d

e

g


15.

CHAUVEL AND DANIEL

Analytical Study of ABS

171

Conclusions Ultracentrifugation can be considered a sufficiently reproducible and effi cient technique for separating the two components i n an A B S graft copolymer. This method enabled us to show quantitatively h o w the grafting parameters are influenced b y polymerization conditions. Consequently, this technique is not only a valuable means for analyzing unknown A B S samples, but it is also particularly convenient for studying and developing a new process, for con trolling the sample characteristics when the process is extrapolated i n a pilot or an industrial plant, and then for checking the reproducibility of different batches.


Literature

Cited

1. Shashoua, V. E., Van Holde, Κ. E . , J. Polym. Sci. (1958) 18, 395. 2. Gesner, B. D., J. Polym. Sci. Part A (1965) 3, 3, 825. 3. Moore, L . D., Frazer, W. J., Amer. Chem.Soc.,Div. Polym. Chem., Prepr. 8, (2), 1482 (Chicago, September, 1967). 4. Huguet, M . G., Paxton, T. R., Amer. Chem. Soc., Div. Polym. Chem., Prepr. 11 (2), 548 (Chicago, September, 1970). 5. Gerrens, H., Ohlinger, H . , Fricker, R., Makromol. Chem. (1965) 87, 209. 6. Shimura, Y., Mita, I., Kambe, H., J. Polym. Sci. Part Β (1964) 2, 403. 7. Dinges, K., Schuster, H . , Makromol. Chem. (1967) 101 (2318), 200. 8. Allen, P. W., Merett, F . M . , J. Polym. Sci. (1956) 22, 193. 9. Allen, P. W., Ayrey, G., Moore, C. G., J. Polym. Sci. (1959) 36, 55. 10. Locatelli, J. L., Thèse, Université de Mulhouse, 1973. 11. Frazer, W. J., Chem. Ind. (1966) 33, 1399. 12. Farkas, G. Y., Crisan, T., Sirchis, I., Mater. Plast. (1970) 7 (7), 335. 13. Parsons, C. F., Suck, E . L . , A D V A N . C H E M . SER. (1971) 99, 340. RECEIVED April 3,

1974.


Analytical Study of ABS Copolymers Using a Preparative Ultracentrifuge

Recommend Documents