Toughened Plastics II - American Chemical Society

13 Determination and Use of Phase Diagrams for Rubber- or Downloaded by UNIV OF CALIFORNIA SAN DIEGO on February 10, 2016 | http://pubs.acs.org Publication Date: May 5, 1996 | doi: 10.1021/ba-1996-0252.ch013

Thermoplastic-Modified Poly(cyanurate) Networks ZhiQiang Cao, Françoise Mechin, and Jean-Pierre Pascault* Laboratoire des Matériaux Macromoléculaires, UMR C N R S 5627, Institut National des Sciences Appliquées de Lyon, Bât. 403 20, Avenue A. Einstein, 69621 Villeurbanne Cedex, France

Blends of two cyanate ester monomers [4,4'-dicyanato-1,1-diphenylethane (DPEDC) or 4,4'-dicyanato-2,2-diphenylpropane (DCBA)] with several initially miscible reinforcing additives were studied as a function of cyanate conversion. The phase diagrams (temperature vs. conversion for the different transitions: phase sep aration, gelation, and vitrification) provide a good overview of the systems and allow easy comparisons. DCBA is a better solvent than DPEDC. Rubber systems based on butadiene-acrylonitrile have an upper critical solubility temperature (UCST) behavior, whereas poly (ether sulfone)s (PESs) induce a lower critical solubility tem perature (LCST) behavior. The acrylonitrile content of the rubbers and the molar mass of the PES additives also have a great influence on their miscibility. During an isothermal cure, phase separation al waysoccurs before vitrification; in rubber it generally occurs before gelation, and in PES it occurs together with gelation. The temperature and viscosity at which phase separation occurs are critical for the final morphology. Reactive additives accelerate the curing process and modify this morphology by inducing a complex ma trix-particle interface, and a substructure inside the dispersed par ticles;these modifications produce the best toughening effects.

B

R I T T L E T H E R M O S E T S A R E B E S T T O U G H E N E D b y the

introduction of a rub-

bery or thermoplastic dispersed phase (1-4). T h e dispersed phase can b e p r o d u c e d b y two methods: i n situ reaction a n d adding p r e f o r m e d particles. T h e first m e t h o d is used m u c h more because it is easy a n d it can create specif*Corresponding author. 0-8412-3151-6

© 1996 American Chemical Society 177

In Toughened Plastics II; Riew, C. Keith, et al.; Advances in Chemistry; American Chemical Society: Washington, DC, 1996.

178

T O U G H E N E D PLASTICS I I

ic morphologies. I n the first method, the initial mixture of monomer(s) a n d ad ditive is homogeneous at the isothermal c u r i n g temperature, T and the dis persed phase is obtained b y the phase separation i n d u c e d b y polycondensation o f the monomer(s). T h e m a i n factors that determine the degree of toughening are the final p o l y m e r b l e n d morphology and the adhesion between the two phases. F o r i n i tially miscible reactive systems, the first factor (including diameters, number, a n d volume fraction o f the dispersed spherical particles) is determined b y the competition between the rates o f phase separation and polycondensation. A d hesion between the two phases depends on the chemical and physical proper ties o f both the additive a n d the m o n o m e r (5-8). T h e phase-separation process i n d u c e d by the reaction and the formation of the morphology are intricate phenomena; for thermosetting systems, their study is focused mainly o n r u b ber-toughened polyepoxy matrices (I, 3). A n analysis of the experimental re sults leads to the following conclusions:

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on February 10, 2016 | http://pubs.acs.org Publication Date: May 5, 1996 | doi: 10.1021/ba-1996-0252.ch013

b

1. T h e concentration o f dispersed particles decreases w i t h T . 2. T h e volume fraction of dispersed phase, V , remains practically constant, goes through a m a x i m u m , or decreases w i t h T 3. T h e number-average diameter, D , of the spherical particles i n creases w i t h T 4. D and V increase w i t h the initial volume fraction o f the addi tive, φ. . 5. B u t V is always greater than φ . This fact means that the dis persed phase is not formed only of pure rubber additive, R, but also contains some epoxy-amine copolymer. It can be about 50 w t % . A second phase separation has been demonstrated to oc c u r inside the dispersed particles. (9) {

D

v

v

D

Λθ

D

Λ θ

O n the other hand, the initial miscibility o f the monomer(s) w i t h the addi tive has a great influence on both the phase separation and the final morpholo gies. M a n y experimental results (5-10) show that phase separation occurs w e l l before gelation time, £ i, or conversion, x j , and before vitrification ( i , x^). T h e better the initial miscibility, the higher the conversion at the c l o u d point, x , w h e r e the phase separation i n d u c e d b y polycondensation occurs at a given value o f T Initial miscibility and T control the "chemical q u e n c h " . T h e viscos ity at the c l o u d point, T| , obviously increases w i t h the conversion, x . T h e vis cosity, T} , can affect the nucleation and growth o f the dispersed-phase parti cles and consequently their average diameter (β, I I ) , and sometimes also the shape o f the particles, w h i c h is not necessarily spherical. ge

ge

vit

c p

v

{

CP

c p

CP

Cyanate ester ( C E ) resins are the key monomers for a n e w type o f h i g h performance polymer. T h e y were developed by H i - T e k Polymers d u r i n g the 1980s, then Rhône-Poulenc, and now Ciba-Geigy. T h e polycyclotrimerization



13.

C A O ET AL.

Determination and Use of Phase Diagrams

179

o f cyanates can take place by simple heating, or i t can b e catalyzed b y transition metal cations together w i t h labile hydrogen compounds, phenols, alcohols, or amines (12). L i k e epoxy systems, C E systems can b e toughened w i t h (J) engineering thermoplastics (13-18), i n c l u d i n g polysulfones o r poly(ether sulfone)s, polyimides, a n d polyesters o r polyarylates, a n d (2) rubbers, i n c l u d i n g butadiene-acrylonitrile copolymers (13), polysiloxanes (14), a n d p r e f o r m e d particles (15, 16). W e can conclude that the different concepts, rubber- o r thermoplastic-modified thermosets, work as w e l l w i t h C E as w i t h epoxy. B u t unlike epoxy, only a few results i n the literature give relations between the i n i tial miscibility o f m o n o m e r and additive, the c u r i n g conditions, a n d the m o r phologies a n d properties. This chapter examines the influence o f r u b b e r a n d thermoplastic additives (nature a n d e n d groups) o n polymerization kinetics and, mainly, on the evolution o f phase diagrams. W o r k on morphologies a n d properties is examined elsewhere (19).

Experimental Methods Monomers. T h e C E monomers, 4,4'-dicyanato-l,l-diphenylethane ( D P E D C ) and dicyanate of bisphenol A ( D C B A ) , were provided by Ciba-Geigy (Louisville, KY) i n reference Arocy L 1 0 and Arocy BIO, and were used as received. L i q u i d D P E D C contains 2 - 3 % impurities (trimer, monophenol-monocyanate, and ortho-para-substituted isomers). D C B A is a high-purity (>99.5%) crystalline powder (melting temperature 79 °C). The chemical structures of these monomers are shown i n Chart I. Additives. Acrylonitrile-butadiene rubbers were provided by B F G o o d rich (Brecksville, O H ) . T h e amino-terminated butadiene-acrylonitrile ( A T B N ) rubber was obtained by reacting earboxyl-terminated butadiene-acrylonitrile ( C T B N ) with an excess diamine, Unilink 4200 (from U O P , E l Dorado Hills, C A ) ; consequently, free diamine molecules always remained in the rubber. The rubbers have almost the same molar mass but different end groups, which have been characterized i n a previous work (20). Their structures are given i n Chart I, and they are described i n Table I. The two poly(ether sulfone)s (PESs) (Victrex, from I C I , United Kingdom) used in this study are described i n Table II. The two P E S additives have different molar masses and different concentrations o f phenolic end groups. B y H - N M R spectroscopy we have estimated that P E S 4 1 0 0 P is an unreactive oligomer, in contrast to PES5003P, which has approximately one O H per molecule. X

P r e p a r a t i o n o f the Blends. The C E blends, which contain 15 wt% additive, were prepared by manual stirring at room temperature for rubber additives, and with dichloromethane as a solvent for P E S additives. P o l y m e r i z a t i o n R e a c t i o n . T h e differential scanning calorimetry ( D S C ) pan, containing 10-15 m g of sample, was sealed under air. It was then


180


CH I

CH I

3

N . c o Q - ιç ^ } o a N

N s C

-c>Q^ç-Q.ac N S

CH

H


CH 3

- C

-(R) — C - C H

CN

3

2

R*NCOCH2CH - C-(R) 2

C H 4

9

3

-C-CH CH COOH

2

2

CN

2

CN

CTBN

NFBN

3

CH

3

HOOCCH CH —O(R)

CN

CH

3

D C B A , 278 g/mol

D P E D C , 264 g/mol

CH

3

CH

3

-C-CH^H^ONR*

CN

CN

C H 4

C4H9

9

ATBN

C4H9

Unilink4200

R = — C T B N > N F B N . This result is consistent w i t h the order o f the initial miscibilities of the rubbers w i t h D P E D C . c p

c p

c p

{


cp

c p

g

c p

g

c p

g e l

cp

cp

geJ

g

c p

b

To isolate these different effects, we d i d the same experiments w i t h 15w t % A T B N r u b b e r by replacing small amounts of A T B N 8 w i t h A T B N 3 1 , w h i c h contains only 1 0 % A N (compared w i t h 1 8 % A N i n the former) a n d seems to be i m m i s c i b l e w i t h D P E D C . F i g u r e 6 shows that the A N content o f A T B N h a d a drastic effect on the miscibility of A T B N w i t h D P E D C . Similarly,

Figure 4. Experimental phase diagram showing temperature versus conversion. Key: o, DPEDC-NFBN (15 wt%); ; DCBA-NFBN (15 wt%); a, phase separa tion; b, gelation fx » 0.6); and c, vitrification (T ) versus x. g


13.

C A O ET AL.


189

300n


200Vc . 100. 0-100 i

•

0.5

•—I

χ

1.0

Figure 5. Expérimental phase diagram showing temperature versus conversion for rubber (15 wt%)-DPEDC blends. Key: A, NFBN; +, CTBN; •, ATBN8. Curves (a), (b), and (c) are defined as in Figure 4.

C h e n et al. (20) f o u n d that the miseibility o f C T B N w i t h D G E B A decreased w i t h the decrease i n the A N content o f the rubber. F i g u r e 7 shows that i n contrast to r u b b e r - D P E D C blends, T decreases almost linearly as χ increases (once again i n the range of T studied, because i n fact it is a descending concave curve). This result confirms that the P E S - C E blends show L C S T behaviors. Moreover, the T versus χ curve o f the 1 5 % P E S 4 1 0 0 P - D P E D C b l e n d is to the left o f that of the 1 5 % P E S 5 0 0 3 P c p

x

c p

300

JOOJ

— ,

r-^r—,

05

1—. χ

1

1.0

Figure 6. Experimental phase diagram showing temperature versus conversion for rubber (15 wt%)~DPEDC blends. Key: •, ATBN8; and •, ATBN8-ATBN31 (5 wt% in the rubber blend). Regions (a), (b), and (c) are defined as in Figure 4.



190


D P E D C b l e n d . These results indicate that P E S 4 1 0 0 P is more miscible w i t h D P E D C than P E S 5 0 0 3 P , w h i c h is consistent w i t h the fact that P E S 4 1 0 0 P has a smaller molar mass. I n addition, the evolution o f T of the continuous phase o f the blends can be plotted i n F i g u r e 7 w i t h the T versus χ phase diagram. O n c e again, phase separation occurs before the vitrification o f the blends. F o r a sample c u r e d at 90 °C, phase separation is expected i n the same range as gelation (x « x j), and i n fact no phase separation is observed. Pellan and B l o c h (31 ) also report e d that this b l e n d was always clear w h e n p r e c u r e d at 90 °C, even w h e n this p r e c u r i n g u p to gelation was followed by a postcuring at 200 °C. W e think that this observation results from high viscosity d u r i n g polymerization a n d gelation o f the system. g

c p

cp

ge

M o r p h o l o g i e s a n d P r o p e r t i e s . T h e phase diagrams o f tempera ture versus conversion (or vs. time) are useful for controlling the c u r i n g o f a b l e n d . W e said i n the introduction that the morphologies are mainly controlled b y the temperature, T , or the viscosity, η , at w h i c h phase separation occurs. Generally, w h e n T decreases or r j (or x ) increases, the dispersed particles become smaller. I n F i g u r e 8 we show a series o f micrographs obtained by scanning electron microscopy o f the fracture surfaces o f different samples. Samples are blends o f D P E D C a n d different additives. T h e y were p r e c u r e d at Tj = 180 °C to control phase separation a n d gelation, then postcured at h i g h temperature (260 ° C ) i n order to reach full conversion (χ » 1). A s we saw i n the phase diagrams, the conversion (x ) or viscosity ( η ) at the c l o u d point depends on the additive used (Table III). c p

c p

ε ρ

c p

cp

ορ

cp

Table III shows the following: W i t h N F B N , x

c p

«

0.30 and the particles



191


C A O ET AL.

Figure 8. SEM images of the fully cured materiah. Part a: 15 wt% NFBNDPEDC. Part b: 15 wt% ATBN8-DPEDC. Part c: 15 wt% PES5003P-DPEDC.


192


are spherical and large (D « 7 μτη). W i t h A T B N , on the other hand, x ~ 0.47 and the particles are smaller (D « 1.7 μπι), but they also have like a b r a n c h substructure. I n this case, we expect to see not only an effect of the viscosity but certainly also o f the reaction between the rubber and cyanate functions be fore network formation. F u r t h e r m o r e , w i t h P E S 5 0 0 3 P , x « 0.40 and the particles are also smaller than w i t h N F B N . W e also see an effect o f the viscosity and of the reactive chain ends of the thenrioplastic. B u t because T is 180 °C, w h i c h is lower than the T o f the thermoplastic, one important effect is certainly also the vitrifica tion o f the dispersed particles w h e n phase separation occurs. T h e conse quence is that only w i t h thermoplastic additives is an evolution o f the mor phology observed between p r e c u r i n g and postcuring processes. c p


cp

{

g

W i t h the help o f these phase diagrams, it is now possible to define differ ent p r e c u r i n g a n d postcuring schedules. T h e effects of the c u r i n g process o n the final morphologies are presented elsewhere (19). T h e introduction of any additive improves the toughness of the material; i n particular, K o f A T B N 8 - D P E D C is more than twice that o f neat D P E D C . In contrast, different additives result i n different losses i n Youngs modulus, from 1 7 % to 5 0 % , and different losses i n the final material glass-transition temperature c o m p a r e d w i t h neat D P E D C (Table III). T h e fact that T is low er for A T B N 8 - or P E S 5 0 0 3 P - b a s e d materials than for those obtained w i t h N F B N obviously indicates that w i t h the latter, the continuous phase contains smaller amounts o f additive because of poorer miscibility. However, the resid ual fraction o f P E S i n the final matrix [ ( φ ) °°] calculated from Fox s equation for the P E S 5 0 0 3 P - D P E D C b l e n d is above the m a x i m u m theoretical value. T h u s , we believe that the decrease i n T for the P E S 5 0 0 3 P - D P E D C b l e n d re sults not only from the presence o f residual additive but also from the m o d i f i cation o f the poly(cyanurate) network because o f a possible reaction between P E S 5 0 0 3 P and D P E D C . I n fact, the system w i t h 15-wt% A T B N 8 has the best-balanced properties o f all the modified systems i n this study: This sample shows an excellent i m provement o f the toughness without too large of a loss i n T and i n the m o d u lus, w h i l e the system based on N F B N has worse mechanical properties. I c

g

Λ

π1

g

g

Conclusions Rubbers and P E S s are initially miscible w i t h cyanate ester monomers. Phase separation occurs d u r i n g the reaction. B y plotting the phase diagrams (temperature vs. conversion), it is possible to compare the effects of chain ends and A N content i n butadiene-acrylonitrile random copolymers and the effect of molar mass i n P E S . T h e cyanate ester m o n o m e r based o n bisphenol A is a better solvent than D P E D C , a n d both dicyanates are better solvents than DGEBA.



13.

C A O ET AL.

Determination

and Use of Phase Diagrams

193

W i t h the use o f phase diagrams, we are also able to control the temperature a n d the viscosity at w h i c h phase separation occurs. T h e final morphologies o f the three systems based o n the same dicyanate m o n o m e r a n d modified w i t h N F B N , A T B N , and P E S are quite different and have different interfaces. W h e n the additive can react w i t h the m o n o m e r before network formation, a two-level structure is observed: a primary structure (dispersed particles), and a substructure inside the dispersed particles. T h e complex morphology obtained i n this case gives the best toughening effect.

References 1. Rubber Modified Thermoset Resins; Riew, C. K.; Gillham, J. K., Eds.; Advances in Chemistry 208; American Chemical Society: Washington D C , 1984; and refer-ences therein. 2. Rubber-Toughened Thermosetting Polymers in Structural Adhesives—Develop-ments in Resins and Primers; Kinloch, A. J., Ed.; Applied Science Publishers: Lon-don,1980; Chapter 5, pp. 127-162. 3. Rubber-Toughened Plastics; Riew, C. K., Ed.; Advances in Chemistry 222, Ameri-canChemical Society: Washington D C , 1989; and references therein. 4. Toughened plastics I: Science and Engineering; Riew, C. K.; Kinloch, A. J., Eds.; Advances in Chemistry 233, American Chemical Society: Washington D C , 1993. 5. Williams, R. J. J.; Borrajo, J.; Ababbo, H . E.; Rojas, A. J. In Rubber Modified Ther-mosetResins; Riew, C. K.; Gillham, J. K., Eds.; Advances in Chemistry 208, Amer-icanChemical Society: Washington D C , 1994; pp. 195-213. 6. Montarnal, S.; Pascault, J. P.; Sautereau, H . In Rubber-Toughened Plastics; Riew, C. K., Ed.; Advances in Chemistry 222, American Chemical Society: Washington D C , 1989; pp. 193-224. 7. Verchère, D.; Sautereau, H . ; Pascault, J. P.; Moschiar, S. M . ; Riccardi, C. C.; Williams, R. J. J. In Toughened plastics I: Science and Engineering; Riew, C. K.; Kinloch, A. J., Eds.; Advances in Chemistry 233, American Chemical Society: Washington D C , 1993; pp. 335-363. 8. Rozenberg, B. A. Makromol. Chem. Macromol. Symp. 1991, 41, 165-177. 9. Chen, D.; Pascault, J. P.; Sautereau, H . ; Vigier, G. Polym. Int., 1993, 32, 369-379. 10. Manzione, L. T.; Gillham, J. K.; McPherson, C. C. J. Appl. Polym. Sci. 1981, 26, 889-907. 11. Ruseckaite, R. Α.; H u , L. J.; Riccardi, C. C.; Williams, R. J. J. Polym. Int. 1993, 30, 287. 12. Grigat, E.; Pütter, R. Angew. Chem. Int. Ed. 1967, 6(3), 206-218. 13. McConnell, V. P. Adv. Compos. 1992, May-June. 14. Arnold, C.; McKenzie, P.; Malhotra, V.; Pearson, D.; Chow, N.; Hearn, M . ; Robin-son, G. Proceedings of the 37th International SAMPE Symposium; Society for the Advancement of Material and Process Engineering: Covina, C A 1992; pp. 128136. 15. Yang, P. C.; Pickelman, D. M.;, Woo, E . P. Proceedings of the 35th International SAMPE Symposium; Society for the Advancement of Material and Process Engi-neering: Covina, CA 1990; p. 408. 16. Yang, P. C.; Woo, E . P; Laman, S. Α.; Jakylowsld, J. J.; Pickelman, D. M . ; Sue, H. J. Proceedings of the 36th International SAMPE Symposium; Society for the A d vancement of Material and Process Engineering: Covina, CA 1991, p. 437.



194


17. Shimp, D. Α.; Christenson, J. R. Plastic-Metal-Ceramics; Hornfeld, H . L., E d . ; So -ciety for the Advancement of Material and Process Engineering: Switzerland, 1990; pp. 81-93. 18. Srinivasan, S. Α.; McGrath, J. E . High Perform. Polym. 1993, 5, 259-274. 19. Cao, Z. Q.; Mechin, E ; Pascault, J . P. ACS Polym. Mater. Sci. Eng. Div. Prepr. 1994, 71, 752-753. 20. Chen, D.; Pascault, J. P.; Bertsch, R. J.; Drake, R. S.; Siebert, A. R. J. Appl. Polym. Sci. 1994, 51, 1959. 21. Verchère, D.; Sautereau, H . ; Pascault, J . P.; Moschiar, S. M . ; Riccardi, C. C.; Williams, R. J. J., Polymer 1989, 30, 107. 22. Cao, Z. Q.; Mechin, F.; Pascault, J. P. Polym. Int. 1994, 34, 41-48. 23. Verchère, D.; Sautereau, H . ; Pascault, J . P.; Moschiar, S. M . ; Riccardi, C. C.; Williams, R. J. J. J. Appl. Polym. Sci. 1990, 41, 467. 24. Georjon, O.; Galy, J.; Pascault, J. P. J. Appl. Polym. Sci. 1993, 44, 1441. 25. Gupta, A. M . Macromolecules 1991, 24, 3459-3461. 26. Simon, S. L.; Gillham, J. K. J. Appl. Polym. Sci. 1993, 47, 461. 27. Mirco, V.; Cao, Z. Q.; Mechin, F.; Pascault, J. P. ACS Polym. Mater. Sci. Eng. Div. Prepr. 1992, 66, 451. 28. Mirco, V.; Mechin, F.; Pascault, J. P. ACS Polym. Mater. Sci. Eng. Div. Prepr. 1994, 71, 688-689. 29. Mirco, V. P h D Thesis, INSA de Lyon, 1995. 30. Borrajo, J.; Riccardi, C. C.; Williams, R. J. J.; Cao, Z. Q.; Pascault, J. P. 31. Pellan, L.; Bloch, B. In C.-R.Journ. Natl. Comp. 1992, November, 161-172.


Toughened Plastics II - American Chemical Society

Recommend Documents