EFFECT OF POLYCARBONATE MOLECULAR-WEIGHT ON POLYMER BLENDS OF POLYCARBONATE AND ABS

(1)

Effect

of

Polycarbonate Molecular Weight on

Polymer Blends of Polycarbonate and

ABS

JIANN-SHINC WU,* SHU-CHEN SHEN, and FENG-CHIH CHANC

Institute of Applied Chemistry, National Chiao Tung University, Hsinchu, Taiwan, Republic of China

SYNOPSIS

The melt viscosity of polycarbonate/acrylonitrile-butadiene-styrene ( PC / ABS ) blends relative to PC is significantly lower, even lower than that of pure ABS in some compositions. Annealing above the TB of PC coalesces and coarsens phase structure in core and skin regions. Increase in the molecular weight of PC in PC /ABS blends results in low-temper- ature fracture toughness improvement but suffers from the disadvantage of higher melt viscosity. The selection of PC in PC/ABS blends must be a compromise between the toughness advantages of higher PC molecular weight and the disadvantages of higher melt viscosity. 0 1993 John Wiley & Sons, Inc.

I N T R O D U C T I O N

Polymer science and technology have undergone tremendous growth during the last two decades, particularly through chemical diversification. The concept of physically blending two or more existing polymers to create new products has attracted great interest and resulted in wide commercial utilization.

Polymer blends of polycarbonate ( PC ) with acry-

lonitrile-butadiene-styrene ( ABS ) provide a bal- ance of toughness, heat resistance, and ease of processing a t a cost lower than that of the high-perfor- mance thermoplastic PC. Since the appearance of the first PC/ABS patent nearly three decades ago,l blends or alloys of PC/ABS have been some of the most successful commercial polyblend products.

Numerous key patents have been specifi-

cally dealing with many aspects of properties, such as heat distortion temperature, impact strength, processability, or flame resistance. Fundamental studies of PC/ABS blends in terms of mechanical, rheological, morphological, and thermodynamic properties have been extensively reported in the open literature.'0-26 However, only limited work has been reported concerning, specifically, the effect of PC molecular weight in PC-ABS blends. Most lit-

* To whom correspondence should be addressed. Journal of Applied Polymer Science, Vol. 50, 1379-1389 (1993)

0 1993 John Wiley & Sons, Inc. CCC 0021-8995/93/081379-11

erature do not give reasons why the particular PC or ABS was selected in their studies. Molecular weight of the blend component is a n important variable, especially as a major component, that can significantly influence essentially all the above-mentioned properties of the resultant PC /ABS blends. The main objective of this study is to obtain a better understanding of the effect of changing the PC mo-

lecular weight and to provide better guidance for

proper selection of the starting materials in the PC/ ABS blends.

EXPERIMENTAL

Materials

Natural grades of PCs, Calibre*, from Dow Chemical

Co. with melt flow rates (MFR) of 3, 6, 10, 15, 20,

and 60 were used in this study (Table I). PC with

MFR = 15 (PC15) was considered as the general

purpose grade and was selected as the blend component in this PC/ABS polyblend studies when the PC molecular weight was not treated as a variable. ABS with 18% rubber content was obtained from Grand Pacific Petrochemical Corp. of Taiwan.

Melt Blending and Specimens Preparation

Melt blending was prepared by a corotating 30 mm-

diameter twin-screw extruder with a n L / D ratio of

29. The standard ASTM

4

in. specimens were pre-

(2)

Table I Melt Flow Rate and Molecular Weight of PCs MFR MIU Mn 3 6 10 15 20 60 35,900 31,500 28,200 25,100 24,100 18,100 13,300 11,600 10,600 9,700 9,600 8,300

Data extracted from Refs. 29 and 30.

pared using a n Arburg 3 oz injection-molding ma-

chine. The temperatures of extrusion and injection molding were adjusted to give the optimized processing conditions according to the blending compositions and the materials selected. Component

weight ratio, PC/ABS = 65/35, was close to most

commercially available products and was chosen to study the effect of PC MFR in this study.

Characterizations

The melt rheological properties were examined by a capillary rheometer, Model Rheograph 2001, Gottfert, Germany. Procedures for the mechanical properties-tensile, flexural, and standard notched Izod impact a t various temperatures-and the crit-

ical strain energy release rate ( G c ) were described

in our previous report^.^^-^' The injection-molded

specimens were annealed as a function of time at 175OC under atmosphere conditions. Phase morphologies of the blends have been characterized by

both transmission (TEM ) and scanning electron

microscopy (SEM). The fracture surfaces of the specimens were examined by SEM after etching with

30% KOH aqueous solution up to a maximum of 30

min, rinsing with water, and coating with gold. The TEM thin-layer specimens were cut by a microtome and stained with 2% OsOI solution for 6 h. The dif-

ferential scanning calorimetry (DSC ) measurements

were carried out using a Seiko differential scanning

calorimeter, Model SSC-5000. TB was taken as the

temperature at which the heat capacity reached one- half of the entire step change as observed on the thermogram.

RESULTS AND

DISCUSSION

h 0) v1 0 1000 a h 4 v1 0 0 v1 W .r(

3

4 \ PC15 PC15/ABS=80/20 PC15/ABS=65/35 pC15/ABS=50/50 PC15/ABS=35/65 PC15/ABS=20/80 100 I I 10

shi& rate (1/$oy

Figure 1

blends at 250°C.

Melt viscosity vs. shear rate of PC15/ABS

ing conditions such as melt extrusion or injection molding. Figures 1 and 2 show the apparent melt viscosity vs. shear rate of PC15/ABS blends a t 25OOC. The presence of ABS, even at as low as 20%, is able to reduce the melt viscosity of PC significantly. The presence of 35% or higher ABS in the blends results in viscosity close to or even lower than ABS. A similar phenomenon was also reported previously.21 The viscosity of polymeric materials changes greatly with temperature and the plots of

1 / T vs. viscosity a t constant shear rate ( 100 and

A

08880 shear rate 100 0-1

* shear rate 500 6-1 t ~ + +

-

shear rate 800 8-1

shear rate I100 6-1

0 2soo a

Melt Rheological Properties

The rheological properties of polyblend provide im-

(3)

EFFECT OF PC MW ON PC/ABS BLENDS 1381 h Q m 0 a .r( Y

24-

.r( d m -0 . 0 m -.d

1000 s-') for PC15 and ABS are shown in Figures

3 and 4. The viscosities of PC15 and ABS are fairly

close, 3700 vs. 2800 poise, a t 250°C and low shear

rate (100 s-l, Fig. 3). At a higher shear rate of 1000

s-l, the ABS viscosity is significantly lower than that of PC15,470 vs. 2400 poise, as shown in Figure 4. That means that the viscosity of ABS in the typ-

ical PC

/

ABS blend processing temperatures is much

more shear-sensitive than is that for PC. Figure 5 shows the effect of PC molecular weight on melt

viscosity for the blends of PC15/ABS = 65/35

composition. PC with greater viscosity, due to higher PC molecular weight, also results in higher viscosity of the PC15/ABS blends, as would be expected.

SEM and TEM Morphologies of the Blends

As mixtures of two polymers are deformed during flow, complex morphologies can develop. Pieces of one polymer may be drawn into filaments that may remain as filaments or break up into small drops or connect to each other to give an interconnected net- work. Under some conditions, ribbons or sheets of one polymer can be formed inside of the other poly-

mer. Thus, rheologically related morphologies of

polymer blends are very complex. By careful control of the etching condition, the features of phase do- mains can be obtained.

Figure 6 shows the SEM morphologies of the

cryogenic fractured surfaces perpendicular to flow direction of the PC15/ABS blends with various

component ratios. Figure 6 ( A ) shows the SEM

morphology of the blend containing 80% PC15.

Since PC15 is the major and continuous phase in the blend, the fine particles shown in Figure 6 ( A ) probably come from the ABS particles that have

P

ABS PC15

0 . o x 1

Figure 3

A B S at lower shear rate of 100 s-'.

Plots of melt viscosity vs. 1/T for PC15 and

1/T, K - l

Figure 4

ABS at higher shear rate of 1000 s-l.

Plots of melt viscosity vs. 1 / T for PC15 and

been etched out and recoated on the surface and this is certainly not the original morphology of the blend. The blends with 65 and 50% PC15 exhibit co-continuous morphologies as shown in Figure

6 ( B ) and ( C )

.

Because of the higher viscosity of

PC15 than that of ABS as mentioned earlier, the blend with 35% PC15 results in PC15 as a spherical

and dispersed phase, as shown in Figure 6 ( D )

.

Fig-

ure 6 ( E ) shows the morphology of the blend with

20% PC15 where the volume fraction of these

h Q, m .rl 0 a R ml000 0 0 m v

5 F

Qeeea PC3/ABS=65/35 w a e a PC6/ABS=65/35 ~&-&&6 PClO/ABS=65/35 Q++# PC15/ABS=65/35 m*+ PC20/ABS=65/35 U-H+ PC60/ABS=65/35 Figure 5 = 65/35 blends by varying PC MFR.

(4)

A. pC15/ABS=80/20 B. PC15/ABS=65/35

c.

PC15/ABS=50/50

D * pc15/ABS=35/65 E. pC15/ABSp20/80

Figure 6 Etched SEM micrographs of PC15/ABS blends by varying component ratio:

( A ) PC15/ABS = 80/20; ( B ) PC15/ABS = 65/35; (C) PC15/ABS = 50/50; ( D ) PC15/ ABS = 35/65; (E) PC15/ABS = 20/80.

spherical PC particle holes is estimated as less than

10%. This is a n indication that a significant amount

PC15 is dissolved in the ABS phase.

Figure 7 shows a few selected SEM morphologies

of PC/ABS = 65/35 blends, by varying the MFR

of PC, that were fractured in liquid nitrogen and under ambient conditions. The morphologies of the cryogenic fracture surfaces do not show significant

differences [Fig. 7 ( A ) , ( C )

,

and 7 ( E ) ] among these

blends. However, the more ductile blend (lower PC MFR) shows a much finer texture of the ambient

fracture surface [ compare Fig. 7 ( B )

,

( D ) and ( F ) ]

.

Greater surface shear yielding of the more ductile blend reflects in the resultant finer texture of the etched surface morphology. Besides the above-mentioned complex factors dictating the phase mor-

phology, the presence of rubber in this triple-phase blend was reported previously to promote compatibility between PC and SAN.22 Figure 8 shows the

TEM micrograph of the PC15/ABS = 50/50 blend

where the rubber particles distributed in the SAN phase are irregular. This ABS is the blended product from a higher rubber content ABS and SAN, and

this high-rubber ABS is the product directly from

the polymerization reactor.

Effect of Annealing

Figure 9 shows the skin layer SEM micrographs of

the PC15/ABS = 65/35 blend after annealing for

different times. The skin layer possesses an oriented sheetlike, multilayer composite structure before annealing because of high shear during injection mold-

(5)

EFFECT OF PC MW ON PC/ABS BLENDS 1383

A. PC3, c r y o g e n i c _{C. PC10,}_{c r y o g e n i c}

E. pC20, c r y o g e n i c

B. P C ~ , ambient D. PC10, ambient I?. PC20, ambient

Figure 7 Etched SEM micrographs of the cryogenic and ambient fracture surfaces of PC/ABS = 65/35 blends by varying PC M F R ( A ) PC3, cryogenic; ( B ) PC3, ambient;

( C ) PC10, cryogenic; ( D ) PC10, ambient; ( E ) PC20, cryogenic; ( F ) PC20, ambient.

ing [Fig. 9 ( A ) ]

.

After annealing, the multilayer-

like skin structure appears to have restructured and

creates large holes, as shown in Figure 9 ( B ) - ( D )

.

Figure 10 shows the annealing effect on the core region taken from the fracture surfaces perpendicular to the direction of injection molding. Upon annealing, the phases coarsen and transform into a nearly dispersed type structure. Since the annealing temperature is significantly lower than the typical

injection-molding temperature ( 175OC vs. 250°C)

,

we would expect a much higher rate of phase coarsening a t the melt-processing condition. Phase morphology coarsening of the PC/SAN system was previously reported.24 When annealing above the Tg of PC under zero-shear condition, interfacial tension provides the driving force to coarsen the phases in this partially miscible PC/ABS blending system.

The mechanisms contributing to the reduction in interfacial contact for this co-continuous system are considered to include coalescence, retraction, and thread breakup processes, in contrast to an occluded system that would require only coalescence of dis- crete particles.24

Tg of the Blends by DSC

Table I1 shows the Tg of the PC15/ABS with various

compositions. The partial miscibility between PC

and SAN has been well recognized. The Tg of the

PC-rich phase decreases with the increase of ABS

content, whereas the TB of the SAN-rich phase in-

creases with increase of PC content, as would be

expected. The difference between ATg(PC) in the PC-rich phase and AT,( ABS) in the ABS-rich phase is insignificant. The observed ATg(PC) and

(6)

Figure 8

blend.

TEM micrograph of PC15/ABS = 65/35

ATg(ABS) in the PC15/ABS = 50/50 blend are

11.4 and 9.5OC, respectively. Therefore, we expect the weight fraction of ABS dissolved in the PC-rich phase and the weight fraction of PC component dissolved in the ABS-rich phase to be comparable. Keitz et a1.16 also reported comparable AT'( PC) and

ATg(SAN), 7.OoC vs. 6.6"C, in the PC/SAN = 50/

50 blend with 26.7% AN content in SAN. Kim and Burns22 reported greater AT,(PC) ( 8 2 ° C ) in the PC-rich phase than ATg (ABS) (2.6OC) in the ABS-

rich phase from the blend of PC/ABS = 50/50 and

concluded that more of the ABS dissolved in the

PC-rich phase than did the PC dissolve in the ABS-

rich phase. Besides, the overall T, shifts observed

from our study are greater than those reported by Kim and Burns.22

The miscibility between PC and ABS depends on

the acrylonitrile (AN) content in the SAN phase l6

and the 38.3% AN content in SAN employed by Kim and Burns appears to be higher than the optimized range for best miscibility for the PC/SAN blending system. Paul et al.15 reported that 27% AN in SAN has the maximum adhesion between PC and SAN in their lap-shear tests. There is somewhat of a controversy on how much AN content provides the best compatibility between PC and SAN. How- ever, AN content of 38.3% is believed to be higher than the optimized value for best compatibility. The AN content in SAN in the ABS that we employed is approximately 28%. The difference in AN content in SAN can explain the observed miscibility difference between our results and results from Kim and Burns.22

Table I11 shows the Tg of PC/ABS blends by

varying PC MFR. The Tg shift of either the PC-rich

or SAN-rich phase depends on mutual solubility. Higher PC molecular weight in the blend shows higher T, and greater AT, (PC) in the PC-rich phase, probably due to better mixing, and, therefore, more of the SAN dissolved in the PC phase. The AT,( ABS) in the ABS-rich phase is at a maximum

with PC15, as shown in Table 111.

Tensile and Flexural Properties

The general trends obtained (data not shown) from the tensile and flexural properties of PC/ABS blends with different component ratios are very

Figure 9 Etched SEM skin morphologies of PC15/ABS = 65/35 blend annealed at

different times at 175OC: ( A ) before annealing; ( B ) annealed 5 min; (C) annealed 10 min;

(7)

EFFECT OF PC MW ON PC/ABS BLENDS 1385

(B) (C)

Figure 10 Etched SEM core morphologies of PC15/ABS = 65/35 blend annealed at different times at 175°C: ( A ) annealed 5 min; ( B ) annealed 10 min; ( C ) annealed 15 min.

similar to most literature previously r e p ~ r t e d . ' ~ * ' ~ , ~ ~ ~ ~ ' Modulus and yield strength are essentially additive,

but tensile elongation a t break has a pronounced minimum when plotted against blend composition.

Table IV shows the results of the blends with varying

PC molecular weight. Modulus and yield strength are essentially identical within experimental error, whereas the blend with PC3 shows an exceptionally high tensile elongation relative to other blends.

Izod Impact Properties

Figure 11 illustrates the Izod impact strength of the

PC15

/

ABS blends a t different temperatures. The

ductile-brittle transition temperature (DBTT) de-

creases with the increase of ABS content up to 50%

of ABS. This is a typical result similar to any rubber- toughening thermoplastic system. Relative to PC, the impact strength of a PC/ABS blend is lower when the fracture is in the ductile mode but is higher

Table I1

Component Ratios

T8 of PClB/ABS Blends with Various

PC-rich Phase ABS-rich Phase

PC/ABS T8 (PC) T, (ABS) 100/0 146.7 - 80/20 140.4 - 65/35 138.6 111.2 50/50 135.3 110.6 35/65 131.7 109.7 20/80 130.3 107.7 0/100 - 101.1

in the brittle mode, similar to the rubber-toughening

PCs that we reported p r e v i o ~ s l y . ~ ~ When the PC

content is less than 50%, the ductile-brittle tran-

sition phenomenon is not very clearly defined and

the toughening efficiency is poor. Figure 12 shows

similar plots for the PC/ABS blends by varying PC MFR. Higher PC molecular weight results in lower DBTT and greater fracture strength in both ductile and brittle modes. This result is not unexpected if we compare them with virgin PC with different molecular weights, as shown in Figure 13.

Critical Strain Energy Release Rate, Gc

A fracture mechanics approach to impact testing was developed independently by Brown31 and Mar- shall et al.32 by assuming the specimen to exhibit bulk linear elastic behavior. This method allows for small-scale yielding at the notch tip providing that it still possesses bulk linear elastic behavior. Some- times, a small correction may be made to the crack

length ( a

+

rp, where rp is the plastic zone radius)

to improve the accuracy of the calculation. There- fore, the experiments were purposely carried out a t low temperature (-40°C) and with a sharper notch

(0.125 mm notch radius ) to meet the criteria of this

method as closely as possible. The elastic energy,

Uc, absorbed as strain energy by the specimen at

fracture, can be expressed in terms of the compli-

ance, C:

(8)

Table I11 T, (PC), T, (ABS), AT, (PC), and AT, (ABS) of PC and PC/ABS = 65/35 Blends

AT, (PC) AT, (ABS)

Composition T g , PC TB, ABS in PC in SAN

-

PC3 153.2 PC3/ABS 141.3 107.7 11.9 6.6 PC6 149.4

-

PC6/ABS 139.0 108.6 10.4 7.5 PClO 149.0 PClO/ABS 138.9 110.3 10.1 9.2 PC15 146.7

-

PC15/ABS 138.6 111.2 8.1 10.1 PC20 148.5 PC2O/ABS 139.7 109.9 8.8 8.8 PC60 143.5 PCGO/ABS 135.8 108.6 7.7 7.5 ABS2 - 101.1

-

- -

-

- -

Table IV Tensile and Flexural Properties of PC/ABS = 66/35 Blends with Varying PC Melt Flowrate

PC

MFR

~~ ~

PC3 PC6 PClO PC15 PC20 PC60

Tensile

Modulus (MPa) 1880 1880 1840 1850 1870 1880

Yield stress (MPa) 62.6 60.3 59.2 63.6 61.8 62.0

Elongation (%) 88 33 31 35 35 26

Flexural

Modulus (MPa) 2050 2190 2020 2020 2170 2120

Yield stress (M.Pa) 88 91 89 84 91 89

( 2 )

P 2

dC

'

2B da

where B is the specimen thickness and a is the crack

tion. The critical strain energy release rate or frac-

ture energy, G,, derived from linear elastic fracture

mechanics, can be expressed by the following equation:

G

= L X - QQQ2.g PC15 n o o o o PClS/ABS=80/20 U PC15/ABS=50/50

-

PC15/ABS=35/65 W pC15/ABS=20/80 A U A PC15/ABS=65/35 %xXxx ABS o q , , , , I , . , I , , , I , , , I,,, I,,, I , , , , I , , , , , , I , , , , < , I

40

-bo -ko

-50

40 -io

b 1'0 do 3'0

40

510

t Temperature ( " C

)

Figure 11

ponent ratios.

(9)

EFFECT OF PC MW ON PC/ABS BLENDS 1387

-

"600:

E

:

2 :

z :

w

-3

600: h D - & -c , : m 400- 4 -0 : L d : $200

:

u -08860 PC3/ABSe65/35 eeeea PCb/ABS=65/35 bbbw PClO/ABS=65/35 W4+0 PC15/ABS=65/35 dw* pCZO/ABS=65/35 cccce PCbO/ABS=65/35 0 ( I ! I I I I I , , ( , , , , , , I , , I , , , , r , , , , l , , l , , , r , , ( , , , , , , , , , , , , , , , , , , , -60 -40 -20 0 20 40 Temp era

t

ur

e

('C

) Figure 12 PC MFR.

Izod impact strength vs. temperature of PC/ABS = 65/35 blends by varying

length. By substituting eq. ( 1) into eq. ( 2 ) and rear-

ranging, the following equation can be obtained ( 4 )

U, = G, X B X

D

X d) ( 5 )

Figures 14 and 15 show the plots of B X

D

X d) vs.

impact energies of various PC/ABS blends. The

corresponding Gc was calculated from the slope of

the graph and the results are summarized in Table V. The blends with co-continuous phase structure,

C

U, = G, X B X

-

d C / d a ( 3 )

Equation (3) can be expressed in the following

equation by introducing the specimen width, D

,

and

the dimensionless geometric factor, 4:

1000 PC 3 Notch radius: 10 m i l \ 0 PC 20 PC 60

-

700- 600- L 500- U 400- ?J r c3, c a, U VJ U iu

s

300- H 200- 100- 0 Temperature 7.c )

(10)

1.00 .( I ;;0.80

-

h

5

0.60 Q w 4J 0 0.40 B 0.20 0.00 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 B x D x $ (lo5 x ~ 2 )

Figure 14 Plots of impact energy vs. B X D X d for

PC15/ABS blends with different component ratios a t -40°C.

65 and 50% in PC15, show higher Gc values and this

range of the component ratio is of commercial in-

terest. For the blend with varying PC molecular

weight and a t a fixed component ratio, PC/ABS

= 6 5 / 3 5 , the resulting Gc increases with the increase of PC molecular weight. The blend with the lowest

molecular weight PC60 has the lowest Gc value.

Fracture toughness of PC is highly sensitive to its molecular weight, relative to other polymers, even above its critical entanglement molecular weight as

reported p r e v i o u ~ l y . ~ ~ These observed results are not

unexpected since the PC is the major component in

the PC/ABS = 6 5 / 3 5 blends.

CONCLUSIONS

1. Melt viscosities of PC/ABS blends relative

to pure PC are significantly lower, even lower

1.20 1 .oo c 2. 0.80 h I 4 w 4J 0.60

b

0.40 0.20 0.00 1.0 2.0 3.0 4.0 5.0 6.0 7.0 I B x D x

+

(lo5 x R - ~ ) Table V

Various PC/ABS Blends

Fracture Energies, Gc, of Composition Gc (kJ/mz) PC15 PC15/ABS = 80/20 PC15/ABS = 65/35 PC15/ABS = 50/50 PC15/ABS = 35/65 PC15/ABS = 20/80 ABS PC3/ABS = 65/35 PC6/ABS = 65/35 PClO/ABS = 65/35 PCPO/ABS = 65/35 PCGO/ABS = 65/35 3.68 4.59 8.91 7.15 6.78 5.09 6.53 13.30 9.86 8.75 8.03 5.32

than those of pure ABS in several compositions.

2. The observed

AT,(

P C ) in tbsPC-rich phase

and AT,(ABS) in the ABS-rich phase are

comparable, an indication of close mutual

solubility of the two phases.

3. Annealing above Tg of PC coarsens the phase

structure of the blend.

4. Higher PC molecular weight has the tendency

to raise the viscosity of the PC/ABS blend.

5. Higher PC molecular weight increases the

ductility of the blends in terms of lower

DBTT and higher impact strength.

6 . The critical strain energy release rate, Gc, of the blend has a similar trend as that of the observed impact strength.

7. The component ratio of PC/ABS = 6 5 / 3 5

can be considered as a compromise composition in terms of the product toughness, cost, and the processing melt viscosity.

8. The toughness advantages by increasing PC

molecular weight must be compensated by the

viscosity increase and this fact has to be taken into consideration in selecting the starting materials.

1. T. S. Grabowski, US. Pat. 3,130,177 (1964) (to Borg-

2. S. Furukawa et al., Jpn. Kokai 73,43,750 (1973) (to Warner Corp.).

Daicell Co. Ltd.)

.

Figure 15 Plots of impact energy vs. B X D X 4 for PC/ABS = 65/35 blends by varying PC MFR a t -4OOC.

3. R. Hasegawa et al., Jpn. Kokai 74,04,745 (1974) (to Teijin Ltd.)

.

(11)

EFFECT OF PC MW ON PC/ABS BLENDS 1389

4. E. Yonemitsu et al., Jpn. Kokai 75,85,651 (1975) ( t o

5. T. Kotani et al., Jpn. Kokai 76,128,356 (1976) (to

6. D. E. Henton, US. Pat. 4,218,544 (1980) ( t o Dow

7. D. Margotte et al., Ger. Offen. 3,002,662 (1981) ( t o

8. P. Y. Liu, US. Pat. 4,390,657 (1983) ( t o General

9. U. R. Grigo et al., U S . Pat. 4,472,554 (1984) (to Mo-

10. K. C. Rusch and R. H. Beck, J. Polym. Sci. Part C, 11. D. Stefan and H. L. Williams, J. Appl. Polym. Sci.,

12. V. Dobrescu and V. Cobxaru, J. Polym. Sci. Polym.

13. S. L. Cohen and N. R. Lazear, Plast. Eng., Aug., 23 14. T. Kusauchi and T. Ohta, J. Mater. Sci., 19, 1699 15. H. Suarez, J. W. Barlow, and D. R. Paul, J. Appl.

16. D. Keitz, J. W. Barlow, and D. R. Paul, J. Appl. Polym.

17. K. Koo, T. Inoue, and K. Miyasaka, Polym. Eng. Sci.,

18. G. Weber and J. Schocps, Angew. Makromol. Chem., Mitsubishi Gas Chemical Co.)

.

Japan Synthetic Rubber Co.). Chemical Co.)

.

Bayer A.G.)

.

Electric).

bay Chemical Co.).

3 0 , 4 4 7 (1970). 18,1451 (1974). Symp., 6 4 , 2 7 (1978). (1982). ( 1984). Polym. Sci., 29,3253 ( 1984). Sci., 29,3131 (1984). 25, 741 (1985). 1 3 6 , 4 5 (1985).

19. R. E. Skochdopolc, C. R. Finch, and J. Marshall, Po-

lym. Eng. Sci., 2 7 , 6 2 7 (1987).

20. W. Y. ChiangandD. S. Hwung, Polym. Eng. Sci., 27,

632 ( 1987).

21. W. K. Chin and J . L. Hwang, in Advances in Polymer

Blends and Alloys Technology, M. A. Kohudic, Ed., Technomic, Lancaster, 1988, Vol. 1, p. 154.

22. W. N. Kim and C. M. Burns, Polym. Eng. Sci., 28,

1115 (1988).

23. K. W. Mclaughlin, Polym. Eng. Sci., 29,1560 (1989). 24. D. Quintens, G. Groeninckx, M. Guest, and L. Aerts,

25. J. J. Herpels and L. Mascia, Eur. Polym. J., 26,997 26. M. Ishikawa and I. Chiba, Polymer, 3 1,1232 ( 1990). 27. F. C. Chang and M. Y. Yang, Polym. Eng. Sci., 30,

28. F. C. Chang and Y. C. Hwu, Polym. Eng. Sci., 31, 29. F. C. Chang, J. S. Wu, andL. H. Chu, J. Appl. Polym.

30. F. C. Chang and H. C. Hsu, J. Appl. Polym. Sci., 43,

31. H. R. Brown, J. Mater. Sci., 8 , 9 4 1 (1973).

32. G. P. Marshall, J. G. Williams, and E. E. Turner, J.

Polym. Eng. Sci., 30,1474,1484 (1990).

( 1990). 543 ( 1990). 1509 (1991). Sci., 4 4 , 4 9 1 (1992). 1025 ( 1991 )