Diaminodiphenyl Sulfone - American Chemical Society

Downloaded by UNIV OF MASSACHUSETTS AMHERST on September 17, 2015 | http://pubs.acs.org Publication Date: May 5, 1996 | doi: 10.1021/ba-1996-0252.ch008

8 Effect of Rubber on Stress-Whitening in Epoxies Cured with 4,4'-Diaminodiphenyl Sulfone B u m Suk Oh , Ho Sung Kim *, a n d Pyo Ma 1

2

2

Department of Welding Technology, Cheon-An Junior College, Cheon-An City, Chung-Nam 330-240, South Korea Department of Mechanical Engineering, The University of Newcastle, Callaghan, Newcastle, N S W 2308, Australia 1

2

Stress-whitening of U-notched, four-point-bending specimens made from rubber-modified epoxies was studied. The epoxies were cured using 4,4 -diaminodiphenyl sulfone. Stress-whitening was observed to be a major response to deformation. The size of the stress-whitenedzone at the root of a notch decreased with increasing rub '

ber content. The stress-whitening is shown to be caused by hydro static stress. Also, two different species were found in the stress-whitened zone corresponding to two different shapes at the root of the U-notch. A circle-shaped species was reversible by heat ing below the glass-transition temperature and was deduced to be due to matrix cavitation, whereas a dendrite-shaped species was partly irreversible by heating and was due mainly to highly cavitat ed rubber particles and shear bands.

T

H E F R A C T U R E T O U G H N E S S O F B R I T T L E E P O X I E S can be i m p r o v e d b y the

ad-

dition o f r u b b e r particles. T h e modes o f deformation, i n the vicinity o f a crack, responsible for the improvement i n toughness have been identified. T h e y include shear-band formation between rubber particles, cavitation o f r u b b e r particles, a n d stretching o f rubber particles (1). T h e epoxy systems that can be effectively toughened, however, are confined to those w i t h a relatively l o w cross-link density. T h e molecular mobility o f these epoxy systems is higher than that o f those w i t h a high cross-link density, w h i c h results i n increased ductility, efficiently i m p r o v i n g the toughness (2). T h e molecular m o b i l i t y o f the highly cross-linked brittle epoxies is restricted, so that plastic deformation such as shear y i e l d i n g is l i m i t e d . Accordingly, toughening such epoxies b y a *Corresponding author.

0-8412-3151-6

© 1996 American Chemical Society 111

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

112

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

r u b b e r modification is ineffective (3). A l t h o u g h there has been little progress i n the toughening o f highly cross-linked brittle epoxies, epoxies w i t h a high cross-link density have advantages such as i m p r o v e d resistance to chemical at tack and high glass-transition temperatures (T s) (I).


g

T h e deformation responsible for the improvement o f toughness i n poly mers is generally accompanied by stress-whitening. T h e t e r m stress-whitening referred to here should be distinguished from that used i n describing fracture surfaces (4). H e r e , it describes subsurface discoloration o f the material i n re sponse to deformation. It is due to the scattering of light from free surfaces created d u r i n g deformation. A l t h o u g h stress-whitening is a well-known phe n o m e n o n i n polymers (5-7), its study i n rubber-modified epoxies w i t h high T s has not been w e l l advanced. g

This chapter examines the nature of stress-whitening i n such epoxies.

Experimental Details The material used was a diglycidyl ether of bisphenol A ( D G E B A ) based epoxy resin (Ciba-Geigy, GY250) cured using stoichiometric amounts of 4,4'-diaminodiphenyl sulfone ( D D S ) . The rubber used for the modifications was Hycar carboxy-teminated butadiene-acrylonitrile ( C T B N ) rubber (1300 χ 13). The curing schedule for all the rubber-modified e p o x y - D D S systems was as follows: first the rubber and then D D S were mixed with the epoxy resin and stirred at 135 °C until the D D S was dissolved; the systems were cured for 24 h at 120 °C and then pos tured for 4 h at 180 °C. The control epoxies were cured according to the same schedule. M o l d e d sheets of the epoxies were cut into double-notch four-point bend ( D N - 4 P B ) specimens i n accordance with the dimensions given i n Figure 1. Be cause more plastic deformation can be promoted i n a U-notch than in a crack, U notched specimens were used. The U-notched D N - 4 P B specimens were loaded at

Section Β

Figure 1. Shape and dimensions of test specimen.


8.

O H ET AL.

Effect of Rubber on

113

Stress-Whitening

Table I. Notation for R u b b e r - M o d i f i e d Epoxies


Designation

Rubber (phr")

RFO RF5 RF10 RF15 RHO RH15

0 5 10 15 0 15

DDS

(phr)

T (°C) g

222 2l7 195 216 76 106

32 32 32 32 16 16

b

b

e

e

e e

"phr denotes parts per hundred of resin by weight. ^Determined by differential thermal analysis at 20 °C/min using powder samples. Determined by differential thermal analysis at 20 °C/min using block samples. c

a crosshead speed of 0.5 mm/min until one of the notches fractured. The designa tions of the specimens and their compositions are given in Table I. F o r preparation of thin sections, the specimens were sectioned, and then the sectioned surfaces were progressively polished using different grades of abrasive. The polishing was finished with 0.05-μπι alumina powder abrasive. The next step involved mounting the polished specimens onto glass slides. The glass slides were cleaned and subsequently coated with a thin layer of an epoxy adhesive to bond with the polished surfaces. The mounted specimens were cut to 1.5 m m thick, and then a thinning procedure similar to the polishing was used. Thinning was carried out until the sections were =«35 μιτι thick.

Results and Discussion Two series of specimens, designated R F and R H , were used. T h e difference between the two series was i n the amount of c u r i n g agent (see Table I), w h i c h resulted i n high T s for the R F series and relatively low T s for the R H series. A s shown i n F i g u r e 2, the b e n d i n g force at the fracture, P , o f the R F series specimens decreased w i t h increasing rubber content, whereas the Pf of R H se ries specimens w i t h a rubber content of 15 p h r increased compared w i t h that o f control specimens. These results indicate that i n the R F series specimens, the fracture stress at the U-notch root decreases w i t h increasing rubber c o n tent, whereas i n specimen R H 1 5 it increases. Throughout this work, the surviving U-notches o f the D N - 4 P B rubberm o d i f i e d specimens were used for microscopic investigations of stress-whiten ing. T h e configurations of the sections, designated A and B , are shown i n F i g ure 1. T h e control specimens d i d not display any detectable stress-whitening. I n rubber-modified epoxies, stress-whitening at the root o f the U-notch oc c u r r e d at Section A , as shown i n Figures 3a and 3b, but not at the outer sur face. T h e stress-whitened zones of R F series specimens were circular (Figure 3a) whereas that o f specimen R H 1 5 was rather dendritic (Figure 3b). F u r t h e r more, contrary to expectations, the extension o f stress-whitened zones i n R F series decreased w i t h increasing rubber content. F i g u r e 4 shows measureg

g

f


114

T O U G H E N E D PLASTICS II


*

if

2

o •

RF Series RH Series

i

[3 0

5

10

15

Rubber (phr) Figure 2. Bending fracture force versus rubber content for U-notched specimens.

merits o f the vertical extent o f the stress-whitened zone from the root of the U notch for each rubber content. To obtain more information, stress-whitened zones were prepared at Sec tion B . F o r all the rubber-modified specimens, the size o f the stress-whitened zone increased from zero at the outer surface to a m a x i m u m at the midsection (Section A ) . F i g u r e 5 shows the whitened zone at Section Β of specimen R F 5 . T h e w h i t e n e d zone of R F series specimens can be three-dimensionally visual ized, as shown i n F i g u r e 6. T h e shape of the whitened zone is i n opposition to that o f the plastic zone ahead o f a crack tip i n dense materials, i n w h i c h the size of the plastic zone decreases to a m i n i m u m at the midsection because o f the state o f plane strain. A t the outer surface there w i l l always be plane stress, and hence the stress i n the thickness direction, σ~, is zero at the surface. C o n currently, plane strain prevails i n the interior, thus increasing the σ . i n the i n terior. It can accordingly be seen that the m a x i m u m hydrostatic stress is found at the midsection (Section A). Thus, stress-whitening appears to be due to the hydrostatic stress components rather than the deviatoric stress components. To identify the types o f damage occurring i n the stress-whitened zones, thin sections were examined under a bright field using a transmission optical microscope. F i g u r e 7a reveals details of the w h i t e n e d zone i n specimen R H 1 5 shown i n F i g u r e 3b. This type of damage is w e l l described elsewhere (4). T h e cavitated particles appear dark because o f light diffraction, and shear bands, w h i c h are biréfringent, have formed between these particles. I n R F series



8.

O H ET AL.

Effect of Rubber on

Stress-Whitening

115

Figure 3. Stress-whitened zone at the root of the U-notch of specimens after polishing. The specimens are viewed under a stereomicroscope. (a) Specimen RF5, Section A. (b) Specimen RH15, Section A.

specimens, the rubber content d i d not seem to affect the microscopic features of the w h i t e n e d zone, although it affects its size. Also, i n contrast to specimen R H 1 5 , no apparent shear bands were found i n the stress-whitened zones, and some cavitated rubber particles were present (see F i g u r e 7b). These cavitated r u b b e r particles were also found outside the stress-whitened zone, and thus no border, i n terms o f the density of rubber cavitations, was observed between the stress-whitened zone and the rest. T h e whitened zones of the R F series were not detectable under microscopic dark-field, bright-field, and cross-polarized light. This stress-whitening does not therefore appear to be due to cavitated r u b b e r particles. It can be deduced that the stress-whitening is due to matrix cavitation.




α ο

5

10

RH Series RF Series

15

Rubber (phr) Figure 4. Vertical extent of stress-whitening from the root of the U-notch, Section A.

Figure 5. Stress-whitened zone at the root of the U-notch of specimen RF5, Sec tion B, after being polished. The specimen is viewed under a stereomicroscope.



8.

O H ET AL.

Effect of Rubber on Stress-Whitening

117

Figure 6. Three-dimensional visualization of the stress-whitened zone of RF se ries specimens.

Figure 7. Bnght-field micrographs of thin sections of the whitened zone. The Unotch root is at the bottom. The scale bar represents 50 μπι. (a) Specimen RH15, Section A. (b) Specimen RF15, Section A.



118


Finally, to examine the reversibility o f stress-whitening, reactions o f sam ples to slow heating i n an oven were observed. T h e thin sections mounted o n glass slides were used as samples for fast reaction to heating. Stress-whitening vanished i n all R F series specimens w h e n the temperature rose from 130 to 150 °C (the duration was about 10 min), w h i c h is below the T^s of the samples. F o r specimen R H 1 5 , however, although the size of the stress-whitened zone was r e d u c e d noticeably w h e n the temperature rose from 100 to 130 °C, the rest d i d not change until the temperature rose further to 150 °C, w h i c h is above the T o f the sample. This observation indicates that the stress-whiten i n g i n specimen R H 1 5 has two components: one, w h i c h is similar to the w h i t e n i n g seen i n specimen R F 5 , is reversable b y heating, and the other is i r reversible b y heating. g

Conclusions This study o f stress-whitening i n rubber-modified epoxies showed that the size o f the w h i t e n e d zone at the root o f a notch decreases w i t h increasing rubber content. Stress-whitening has been shown to be caused by hydrostatic stress. Two different species of stress whitening were found. O n e is reversible by heating a n d is deduced to be due to matrix cavitation; the other is irreversible by heating a n d is due to highly cavitated rubber particles and shear bands.

Acknowledgment W e thank K. J . D o o l a n of the research laboratory at the B r o k e n H i l l P r o p r i etary Company, L t d . , for providing the glass-transition temperatures o f the samples.

References 1. 2. 3. 4. 5. 6. 7.

Garg, A. C.; Mai, Y. W. Compos. Sci. Technol. 1988, 31, 179. Pearson, R. Α.; Yee, A. F. J. Mater. Sci. 1989, 24, 2571. Sue, H. J. Polym. Eng. Sci. 1991, 31, 275. Pearson, R. Α.; Yee, A. F. J. Mater. Sci. 1986, 21, 2475. Breuer, H . ; Haaf, F.; Stabenow, J. J. Macromol. Sci. Phys. 1977, B14(3), 387. Smith, J. W.; Kaiser, T. J. Mater. Sci. 1988, 23, 3833. Lee, Y. W.; Kung, S. H. J. Appl. Polym. Sci. 1992, 46, 9.


Diaminodiphenyl Sulfone - American Chemical Society

Recommend Documents