T H E ROLE (3F S T R E S S IN L I O U I I ) M E D I A
DURING
FRACTURE
OF P O L Y M E R
MATERIALS
A. N. Tynnyi and A. I. Soshko F i z i k o - K h i m i c h e s k a y a Mekhanika Materialov, Vol. 8, No. 1, pp. 46-4!~, 1!)(;7 Polymer materials are widely used in technology because of their c h e m i c a l inertness with respect to various working m e d i a (acids, alkali, e t c . ) . The c h e m i c a l stability of polymers is usually determined on specimens or parts in the unstressed state. In practice, however, most parts made of polymer materials operate under the combined action of working m e d i a (liquid and gaseous) and m e c h a n i c a l stresses which activate various processes in the material. Consequently, the determination of the c h e m i c a l stability of these materials in the stressed state must involve studies of the changes in their p h y s i c o - m e c h a n i c a l properties (especially the l o n g - t i m e strength) under the influence of various m e d i a and investigations of the formation and growth of defects (cracks) which lead to macroscopic fracture. It is known [1] that the rate of crack growth is an exponential function of stress, i . e . , O ~- V 0 e ~,
where u0 and ~ are constants depending on the m a t e r i a l properties. This equation was derived by experiment under conditions in which the character of crack propagation remains unchanged. When this character varies under the influence of liquid s u r f a c e - a c t i v e media, this equation is no longer valid [2]. It was shown by several workers [ 8 - 5 ] that the strength and fracture of metals are substantially affected by surfaceactive media. The fact, discovered by P. A. Rebinder, that adsorption of surface-active m e d i a f~eilitates deformation indicates how necessary it is to determine the strength of materials under actual service conditions. As established in [6], the m a x i m u m effect of s u r f a c e - a c t i v e m e d i a is concentrated in the pre-fracture region and at the crack tip. Recent studies [ 7 - 9 ] showed that the strength of polymer materials is substantially affected by s u r f a c e - a c t i v e m e d i a and demonstrated the necessity of investigating adsorption phenomena at s o l i d / s u r f a c e - a c t i v e liquid interfaces. Since the stress applied to a part or a specimen determines the rate of crack propagation, two l i m i t i n g cases should be considered in an analysis of the effect of s u r f a c e - a c t i v e m e d i a on the strength and fracture of materials. 1. A solid is subjected to the action of high stresses which produce high strain rates; in this case, a surface-active medium acts only for a short t i m e in the i n i t i a l stages of crack propagation and has no effect in the later stages during which the rate of crack propagation is considerably faster than the rate of surface diffusion of the medium, 2. A m a t e r i a l is acted upon by low stresses which produce slow rates of strain; in this case, the rate of surface dif. fusion of the medium is equal to the rate of crack propagation so that the m e d i u m is always present at the crack tip. Under these conditions, t i m e becomes a significant factor in determining the character of the influence of surface-active m e d i a on solids. The aim of the present work was to study the effect of liquid m e d i a on the m e c h a n i c a l properties of some polymers under conditions of both high stresses and fast strain rates and low stresses and slow strain rates. The tests were carried out on amorphous glass-like polymers, block polystyrene (PS) and ST1 type plexiglas. These materials were selected for this study because they are widely used in various branches of industry, have high m e c h a n i cal properties, and are resistant to the action of acids, alkalis and other working media. To ensure that the e x p e r i m e n t a l specimens had properties i d e n t i c a l with those of industrial parts, they were made by methods generally adopted in industrial practice. The specimens were not h e a t - t r e a t e d because the removal of internal stresses in this m a t e r i a l is possible only at temperatures above the softening point, at which marked dimensional changes take place in the specimens. PS specimens were pressure-cast in molds at 200-220 ~ C, being held at the temperature for about 20 sec. ST1 specimens, corresponding to GOST 4649-55, were m a d e from one (10 m m thick) sheet of plexiglas. They were machined on a copying m i l l i n g m a c h i n e which ensured g e o m e t r i c a l similarity of specimens, and then polished. To remove stress concentrations, the ST1 specimens were annealed for 3 hr at 60 ~ C in a vacuum chamber. The UTS was determined on a tensile testing m a c h i n e at a constant strain rate of ] 0 m m / m i n ; the t i m e - t o - r u p t u r e tests were carried out on a four-position m a c h i n e of the leve type. The results (see table) show that most liquid m e d i a tested produce a reduction in the UTS of ST1. At the same t i m e , liquid working m e d i a (except benzene and kerosene) have no effect on the UTS of polysterene [10]. This is because crack nucleation in the surface layers of the latter m a t e r i a l is inhibited by large r e s i d u a l compressive stresses in these layers; as a result, crack nucleation in PS specimens 34
made by the method described above takes place in the specimen core, so that the working m e d i a do not interact with cracks and cannot affect their propagation. Regarding ST1, its UTS is affected to the greatest extent by ethyl alcohol, aviation grade benzene and water, less so by acids and their solutions, and almost not at all by such m e d i a as oil of various kinds. The character of fracture of STI tested in surface active m e d i a under stresses near to its UTS is the same as in air. In this case the effect of the working m e d i a is manifested only in the i n i t i a l stages of crack propagation since the rate of crack propagation in the later stages is much faster than the rate of surface diffusion of the working m e d i a . For instance, the rate of surface diffusion of oil and water is, respectively, about 8 x I 0 - 4 m m / s e c and g m m / s e c , which is only a small fraction of the rate of crack propagation at stresses near the UTS [11].
The UTS of ST1 Specimens in Various Working Media (Each Value is an Average of Six Test Results) UTS, k g / m m ~.
Medium Vacuum of 10 - 6 m m Hg Air Tap water Water + 3% NaC1 NazCO s solution (10 g/l) Ethyl alcohol Concentrated sulfuric acid Concentrated hydrochloric acid 50% nitric acid" Aviation g r a d e oil MS20 Castor o i l Transformer oil Aviation grade kerosene Aviation grade benzene
8.3 8.2 7.3 7.4 7.6 6.5 7.7 7,8 7.5 8.3 8.3 8,3 8.1 7.0
Consequently, the action of working m e d i a on ST1 under high stresses is of very short duration; however, it facilitates the formation of surface defects and crack propagation which leads to a reduction in the UTS. The results of l o n g - t i m e strength on ST1 specimens in various working m e d i a (air, oil, water, kerosene) are reproduced graphically (see figure) in the form of log r(o) curves, where o is the applied stress and r the t i m e - t o - r u p t u r e .
In general, the lower the applied stress, the more pronounced becomes the effect of the working m e d i a on r ; below a certain c r i t i c a l value of o, the effect of working m e d i a becomes m a r k e d l y weaker. The degree of reduction in r under the influence of water and kerosene is substantial; the effect of oil is rather small. Thus, the value of r recorded for ST1 specimens tested in oil at o > 6.5 k g / m m ~ is the same as in air; at o < 6.5 k g / m m 2, r of specimens tested in oil is slightly but significantly smaller than in air. The effect of w~ter and kerosene is noticeable already at o < 7.0 kg/mm2; at o < 3 - 4 k g / m m ~', there is a sharp increase on r of ST1 specimens tested in water and kerosene. The latter effect was observed in [8] on ST1 specimens stressed in water. It is attributable to changes produced in the character of crack propagation and in the mechanism of fracture by the interaction of molecules of liquid m e d i a with the solid m a t e r i a l at the crack tips, which becomes possible due to changes in the relative values of the rates of crack propagation and surface diffusion of the m e d i u m . It was shown that several secondary cracks, radiating from the main crack tip are formed under these conditions as a result of the action of a s u r f a c e - a c t i v e m e d i u m ; this leads to a reduction in the stress gradient and an increase in the breaking stress.
•l
w
oo~
J
4 O
4
o
3 2 .
2
.
.
.
3
.
.
.
.
i
.
.
.
~-
.
.
.
S
.
'
O", k
m 2---
T i m e - t o - r u p t u r e of ST1 in various m e d i a plotted against the applied stress: 1) vacuum and air; 2) water; 3) oil MS20; 4) kerosene. The data on the effect of kerosene and oil on t i m e - t o - r u p t u r e of ST1 obtained in the present study confirms t h e results of a previous investigation [8] and shows that differences in the properties of working m e d i a are reflected in the m e c h a n i c a l properties of ST1; it shows also that the applied stress level, 9 determines the crack propagation rate and the t i m e during which the m e d i u m can act on a crack tip, is a very important factor.
35
The differences in t i m e dependence of the strength of ST1 specimens under the influence of water, kerosene and oil must be attributed to differences in the properties of these media. Test results indicate that the t i m e - t o - r u p t u r e of ST1 in o[1 MS20 is longer than in water and kerosene. This may be explained by the fact that the crack propagation rate at o > 6.,5 k g / m m z is faster than the rate of surface diffusion of the working medium; as a result, c r a c k propagation, i. c . , the movement of a crack tip, is not affected by the medium and the t i m e - t o - r u p t u r e of ST1 in oil is the same as in vacuum or in air. Since the rate of surface diffusion of such m e d i a as water and kerosene is much faster than that of oil, the effect of these m e d i a on the kinetics of crack propagation is manifested in earlier stages, i . e . , at higher stress levels and faster crack propagation rates (see figure). REF ERENCES 1. S. N. Zhurkov and Z. E. Tomashevskii, ZhTF, no. 6, 1957. 2. A. N. Tynnyi and A. I. Soshko, FKhMM [Soviet Materials Science], no. 3, 1968. .q. V. h Likhtman, E. D. Shchukin, and P. A. Rebinder, e h y s i c o - C h e m i c a l Mechanics of Metals [in Russian], [zd. AN SSSR, Moscow, 1962. 4. V. I. Likhtman, P. A. Rebinder, and G. V, Karpenko, Effect of Surface-Active Media on Deformation of Metals [in Russian], lzd. AN SSSR, Moscow, 1954. 5. G. V. Karpenko, Effect of Active Liquid Media on the Fatigue of Steel [in Russian], lzd. AN USSR, Kicv, 1955. 6. P~ A. Rebinder, Jubilee Collection Dedicated to the 30th Anniversary of the Great October Socialist Revolution [in Russian], Izd. AN SSSR, 1947. 7. G. M. Bartenev and I. V. Razumovskaya, DAN SSSR, 150, no. 4, 1963. 8. A. I. Soshko, A. N. Tynnyi, and M. M. Gudimov, FKhMM [Soviet Materials Science], no. 5, 1965. 9. A. N. Tynnyi and A. L Soshko, FKhMM [Soviet Materials Science], no. 5, 196S. 10, A, I, Soshko and A. N. Tynnyi, FKhMM[Soviet Materials Science], no. 5, 196~5. 11. E. Shand, J. Am. Ceram. S o e . , 42, 474, 1959. 27 September 1966
36
Institute of Physics and Mechanics, AS UkrSSR/ L'vov