ATMOSPHERIC DURABILITY O F P O L Y M E R - F I B E R

COMPOSITES

IN COLD CLIMATES

V. N. Bulmanis, G. M. Gunyaev, V. V. Krivonos, G. P. Mashinskaya, V. M. Merkulova, G. I. Milyutin, A. A. Gerasimov, and S. A. Kuz'min

UDC 539.2:620.193:678.067

In evaluating the prospects for using composite materials in the fabrication of structures and end-items that are serviceable in regions with cold climates both the economic and technical aspects of the problem must be considered. In the foreign press, even at the start of the 1970s, it was argued that the savings from the use of composites would be particularly great in the far north

I1].

At the same time, we cannot ignore the fact that the natural conditions of a severe continental cold climate

are potentially very aggressive with respect to such heterogeneous systems as polymer composites. Atmospheric durability (climatic durability) is usually measured and characterized by a set of reversible and irreversible changes in the physicomechanical properties of materials when exposed to the action of the ambient medium. The main factors responsible for such changes in the service properties of composite materials in a cold climate are the low temperatures (down to -60~ the large daily temperature drops (as much as 40~ and crystallization of sorbed moisture. Results from Soviet studies of specific features of the action of cold-climate factors on composite materials are reflected in various publications: influence of low climatic temperatures [2]; influence of temperatUre drops and crystallization of sorbed moisture [3] and influence of natural exposure [4, 5]. Some experimental data have been published in industrial-branch reports. Information obtained in other countries has been summarized in the view 16] (no information is given on natural exposure). The present work has been aimed at analyzing the accumulated experimental information on the stability of the strength and elastic properties of three basic classes of polymer/fiber composites - plastics with glass, organic, or carbon fiber reinforcement - in the course of 2-3 years of natural exposure under cold-climate conditions (in the city of Yakutsk); the work has also been aimed at evaluating the influence of various structural, design, and methodological factors on the degree to which the mechanical properties of the materials are preserved (let us note that in only in application to certain grades of glass fiber-reinforced plastics).

I4, 5],

the influence of natural exposure was examined

The basic information on the composites that were investigated is presented in Table 1. The methods used in the experimental studies are summarized in Table 2. The specimens of the wound composites consisted of rings and segments with a thickness of 2-10 mm; the samples of the pressed composites consisted of bands and dumbbell testpieces with a thickness of 1.2-4.5 mm. The specimens, together with the plates of carbon fiber-reinforced plastics and organic fabric-reinforced Textolite, were mounted for exposure in the spring (April); specimens were removed for testing every 6 or 12 months. Before and after exposure, the materials were conditioned (held until they reached constant weight) in accordance with the standard [7]. After conditioning and cutting (if necessary), the specimens were subjected to mechanical tests. The rings were tested in tension (load on half-disks) and the segments in bending (three-point "convex upward" scheme). The length of the segments was selected so as to ensure failure of the specimens not only from the action of normal (compressive) stresses, but also tangential stresses. Specimens of the pressed composites were tested in tension (bands, dumbbells) and in compression and three-point bending (bands).

Institute of Physicotechnical Problems of the North, Siberian Branch, Academy of Sciences of the USSR, Yakutsk. Scientific-Industrial Association VIAM, Moscow. Central Scientific-Research Institute of Special Machinery Construction, Moscow Oblast. Translated from Mekhanika Kompozitnykh Materialov, No. 6, 1065-1073, November-December, 1991. Original article submitted February 5, 1990; revision submitted December 5, 1990.

698

0191-5665/91/2706-0698512.50 +1992 Plenum Publishing Corporation

T A B L E 1. Investigation of Composites Composite I Technology l'Reinf~

Organic fiber Winding Pressing

RBN cord VMPS fiber SVM cord SVM fabric

Carbon fiber

LU-P tape

Fiber glass

Winding

Pressing

Matrix

I

Winding ~ ' a nlay angle d

Polyester Epoxy

90~ ~ ~ RRCF* 90~(0o); 4-45~ 90~176 4-45~ ~phen~ lic --

Ep~ A Epoxy-phenol- 70=/90~ (l : I)** ic B i0~ ~ (2:3:t) [_+45~] Epoxy-phenol- [0~ ~ (1 : 1) ic C 10~176176 ( 2 : 3 : l)

LU-P t a p e

[__45 ~]

[00/9001 ( l : l ) [0~ 45~ ~] (1:6:9)

ELUR-P tape

[_+45 ~]

*RRCF denotes random reinforcement with chopped fiber. **Values shown in parentheses denote ratios of layers with different schemes of reinforcement.

TABLE 2. Experimental Procedures

Material

L Char- Temperature of Exposure Variant posure acter~ mechanitest samples of~lick#eondi- istic~ cal tests, ness tions ~ oC

| Glass fiber and Specimen polyester Glass fiber Specimen and epoxy Organic fiber Specimen and epoxy

i

OA, UB

I/-I;E

i

OA, UB

[I

20;--60

1

OA, UB

[I

20

Organie-fabric textolite

Specimen Plates,with specimens cut later

3 3

OA, UB [I OA, UB- [I

Carbon-fiberreinforced plastic

Plates, with specimens cut later

1

OA

I-[; E

20

20;--60 20

20;--60

Note. O A = open area; UB = unheated building; rI is the strength; E is the modulus of elasticity.

The mechanical tests were performed in INSTRON 1195 and FP-10 universal testers equipped with removable heat and cold chambers. The modulus of elasticity E in tension was determined with strain gauges in a TsTM-5 instrument; the modulus of elasticity in bending was found by measuring deflections of segments and bands. The testing program was not complete for all types of composites; the organic fabric-reinforced Textolite went through the most extensive testing program (see Table 2). The extent of the program was limited by the quantity of plates and specimens available. Nonetheless, in our opinion, the test results still enable us to arrive at conclusions regarding the influence of various factors on the degree to which the service properties of the composites are preserved. Particular attention was given to the selection of conditions for the mechanical tests of the composites after exposure. From the standpoint of service of materials and end-items, the basic interest is in the residual characteristics of "co-aged" composites with the most unfavorable combination of ambient temperature and humidity. For cold climatic conditions, testing 699

at -60~

in normal humidity is considered to be an analog of such a combination of unfavorable natural conditions (in the

winter, the humidity under natural cold-climate conditions is close to normal) [8]. In addition to the tests under these extreme conditions, other tests were performed at 20~ the normal temperature for tests in climatic research. Because the tests at -60~

are very complex and laborious, we also set ourselves the task of finding relationships that would make it possible in the

future to evaluate the serviceability of polymer composites under natural conditions, by means of exposure tests conducted only at 20~ In view of the large volume of information and the fact that the changes in mechanical characteristics of composites are rather small in most cases, we are presenting in quantitative form only those versions of the tests in which the deterioration of properties was statistically significant with a confidence probability p = 0.95 (as evaluated by the t-test [9]), this deterioration being either irreversible (as measured by test results on the co-aged composites at 20~ referred to the test results on the original composites at 20~ ites at 20~

or overall (results of tests on co-aged composites at -60~

referred to results on original compos-

The following results were obtained. FIBERGLASS- POLYESTER PLASTIC

The experimental results on this plastic are presented in Table 3. It must be noted that under severe continental climatic conditions (precipitation 175 mm/yr), the differences in the conditions prevailing in the open area and in the unheated building are due mainly to the comparatively high level of solar radiation (3.7 GJ/m2/yr). Since there is practically no adverse effect of ultraviolet radiation on composites [10-13], the unheated building exposure for these materials is very nearly equivalent to the open area exposure. Also, it is known [14-17] that in composites, as a result of climatic action, two processes may go forward simultaneously: an increase or growth of damaged areas and a strengthening of the material as a result of post-curing of the binder. Therefore, if the post-curing process (as a consequence of heating by radiation in the summer) is active, the unheated building storage may prove to be more severe than the open area storage. The fact that the strength properties of the fiberglass-reinforced plastic are more sensitive than the elastic properties to climatic effects can be explained on the basis that the damage that appears (delamination, etc.) has a very strong and specific effect on the strength indexes of composites. FIBERGLASS--EPOXY PLASTIC As can be seen from Table 3, no statistically significant losses of strength were observed. In a number of cases, in the tests for interlayer shear strength, the values for the fiberglass plastic increased by as much as 18% as a result of exposure. From these results we can conclude that in the epoxy-fiberglass plastic, the strengthening process prevails over the increase in degree of damage. ORGANIC FIBER-REINFORCED EPOXY PLASTIC Experimental results on this composite are presented in Table 4. Note the considerable decrease in strength characteristics of the composite with a winding angle +45 ~ i.e., the material with a large quantity of intersecting reinforcing fibers. Serious deterioration of properties was observed even in the first 6-12 months of exposure. Such behavior of the organic-fiber composite can be explained by specific features of its properties and also by the "unprotected edge effect." In carrying out natural exposure tests, the edge effect has several varieties: the absence of any protective coating in the form of binder and part of the surface, stress concentration close to the unprotected edge due to temperature drops, and stress concentration in the mechanical tests of the exposed specimens. In the case of the organic-fiber composited an additional aspect is particularly important: the capability of rapid saturation with moisture from the ends of the cut organic fibers. The results of tests on the composite with a 90 ~ winding angle, where comparatively little deterioration of strength properties were found, indicate that the behavior of the organic-fiber composite with ~o = ___45~ is almost entirely governed by the "unprotected edge effect." Let us note also that the decrease in strength characteristics of this composite in open area storage was greater than in the unheated building. This indicates that, in contrast to the fiberglass-reinforced plastics, the processes of post-curing the composite are of only secondary importance.

700

TABLE 3. Statistically Significant (p = 0.95) Loss of Mechanical Properties of Fiberglass-Reinforced Plastics Fiberglass plastic (and length of exposure)

Winding angle] Mode of r lay) I loading

Polyester (3 days)

90~ o) + RRCF

I

a~ct'-rl Expo" e sure !isti~ conditions

Bending

II E

90~ ~)

Epoxy

(2 years)

Interlayer shear Tension

H rl

OA UB OA UB OA b% OA U~

•

~

Interlayer shear Tension

H H

OA UB OA UB

Loss of strength, % :irrevers- [ overall ible ,None 16 None None None 18 Ndne N6ne None None None None

None None None None None None

Note. Dash indicates that no mechanical tests were performed (same notation in Tables 4 and 5).

TABLE 4. Statistically Significant (p = 0.95) Decrease in Strength Properties of Organic Fiber-Reinforced Plastics as a Result of 2-Year Exposure Organic-fiber [ Structure Mode of composite (and [ loading object exposed I , Epoxy, wound

90o(0 o) •

Epoxyphenolic, pressed (specimen)

~

Textolite

[Exposur~SpecimenILoss of strength, % condi- ithick- I irrevers-loverall ! tions_ ]ness, m ! i b l e j

interlayer shear Tension Tension Bending

Compression

Epoxyphenolic, pressed (plates)

Textolite

Tension

OA UB OA UB OA OA

lib

I0,0 3,0 2,2 2,5 2,2 2,5 4,5 2,5

OA

2,2 2,5 4,5

UB OA

2,5

Compression

OA

lib OA lIB

--

14

2,2 2,5

Bending

16

2,2 2,5 4,5 2,5 2,2 2,5 4,5 2,5

46 42 t9

None

26 20 None 14

19 16

None None None None None None None None None None None None

--

27

None None

None None None ---------------

ORGANIC FABRIC-REINFORCED E P O X Y - P H E N O L I C TEXTOLITE The results shown in Table 4 are extremely important. In testing specimens cut from the plates after exposure, in none of the 10 variants could we find any statistically significant loss of strength. Thus, the changes in strength that have been found in mechanical tests must be attributed completely to the "unprotected edge effect" (this also supports the view that the main reason for loss of strength of the wound organic fiber-reinforced plastic is the "unprotected edge effect"). The test results on the exposed specimens show that the aging effects are not only localized close to the unprotected edge, but are also nonuniform through the thickness of the specimen. This is indicated, in the first place, by the fact that the residual strength of the organic fabric-reinforced Textolite increases with increasing thickness of the specimen, and in the second place, by the fact that the greatest loss of strength takes place in the bending tests [3]. The same as in the case of the wound organic fiber-reinforced plastic, the open area storage proved to be slightly more "severe" than the unheated building storage. The atmospheric durability of organic fabric-reinforced Textolite under the conditions of a cold climate is greater than in a warm, moist climate [18]. Thus, the effect of alternating temperature drops and rises is less dangerous for the organic-reinforced plastics than is the action of moisture at above-freezing temperatures. This is consistent with the results of laboratory experi701

TABLE 5. Statistically Significant (p = 0.95) Loss of Mechanical Properties of Carbon Fiber-Reinforced E p o x y - P h e n o l i c Plastics as a Result of 3-Year Exposure in Open Areas

fiber

Type 1

[0o/90o] ( 1 : 1)

loading

~stic

Tension

H E I] E H

20 None 13 None 13

23 None None None

Tension

1-I E

13 None

17 None

Betiding Compression

H lIE

None N~ne

None None

Tens ion

I]

23

17

None

None

Tension

E I1

19

25

E

None None None iH None None None None None

None None None -None None None None --

None None None None None None None

None None None None None None --

Bending Compress ion

[0~ 45~176] (2 : 3 : 1)

[ _ 45o] Type 2

[0~ ~ (1 : 1)

Bend ing

[0o/+_45o/90o] (2 : 3 :

Type 3

Compress ion Tension

1)

I] E

H [I E

Bending

[]

Compression

E I]

[ -+45~

Tens ion

H

[0o/90~ (l : I)

Tension

[I E I]

Bending

E

E [0o/+_45o/90o] (I : 6 : 9)

[ _ 45~

'iirreversible I overall

Compression

[I

Tens ion

[I

25

E

None

Bending

II

Compression Tension

E 1] [I E

12 None

20 None None

--

30 None None None

-None None

ments that were cited previously [3, 19], indicating that, for organic-reinforced plastics, crystallization of sorbed water is not characteristic at temperatures even as low as -60~ CARBON FIBER-REINFORCED E P O X Y - P H E N O L I C PLASTICS Here w e must note that in none of the 30 variants of the determination of the modulus of elasticity E did we find any statistically significant deterioration of properties (Table 5). Thus, the relationship found for the polyester--fiberglass plastic has been confirmed: When exposed under climatic conditions, the rigidity of a composite does not change as much as the strength properties. The stability of the strength properties depends on the type of lay and also on the ratio of layers with different reinforcement patterns. The least stable is the I0~176

~ lay (1:6:9). From a comparison of the results obtained by different

methods of mechanical tests, it can be seen that the loss of strength properties is the least when measured by bending tests. Hence there are no grounds for believing that the aging of carbon fiber-reinforced plastics is nonuniform through the thickness of the exposed plates [3]. A noteworthy fact is that the percentage of irreversible changes in strength properties of the carbon fiber-reinforced plastics under cold climatic conditions is higher than in a warm, humid zone [18]. This means that the exposure to large temperature drops, including passage of the temperature through 0~ (i.e., the possibility of crystallization of sorbed moisture), in the case of the carbon-reinforced plastics, is more dangerous than the action of moisture at above-freezing temperatures. A comparison of the entire set of data reveals another important relationship. It is known [8] that the percentage of overall loss of strength of structural thermoplastics when they are aged under cold climatic conditions is greater than the percentage of irreversible loss of strength. As can be seen from Tables 3-5, for composites based on thermosetting matrices, 702

TABLE 6. Prediction of Loss of Strength in Storage of Composites in Open Area for 10 Years

I Structure

Material

Change i n strength, % irrevers- [ overall ible

....

Carbon fiber-reinforced p~astic

10~ ~]

[ _ 45~ [0~ ~ Type2 [0~176 ~] Type3 Organic fiber-reinforced ( 1 : 6 : 9 ) _+45~ epoxy plastic •

l

Organic fiber-reinforced epoxy-phenolic plastic (n = 2.2mm)

Textolite

42 38 34 51

45 34 38 53

54

48

42 57*

46 49"

*In bending tests; in all other cases, in tension tests.

there is no such adverse "effect of cold"; i.e., the overall loss of strength is little different from the irreversible loss, and in some cases may be even smaller. Let us examine the possibilities of calculating the strength indexes of exposed composites under low-temperature conditions (at -60~ with the actual experimental testing limited to 20~ By using the information presented in Tables 4 and 5, data on the low-temperature strength of the composites in the original state, and results from laboratory studies o f similar types of composites [20, 21], we have established the following. In tension and shear tests there is a definite relationship between the strength of the composites at different temperatures in the initial state and after a certain period of aging: At the temperature of mechanical testing where the strength of the composite in the initial state is higher, stronger measured effects of aging are also characteristic. Such behavior of composites can be explained as follows. Aging can be represented as an increase in the bulk damage, the influence of which becomes greater at lower temperatures if the fracture mechanism remains unchanged. Therefore, if the strength of the material H0(T ) increases with decreasing temperature (in comparison with H0(T)0 ), the conclusion is obvious; if the strength decreases with decreasing temperature, this indicates a change from a "bulk" to a "local" mechanism of fracture, Naturally, the accumulation of bulk damage has little effect on the strength of a composite that fails through a "local" fracture mechanism. These qualitative concepts can be put into the following form: no(T)

rq(L,)l-Io(To); ri0 (To) n(To)

rio(T)

Ho (To)


when

(1)

Ho(T)
On the basis of (1), we can propose the following approximate relationship: r i ( T ) ~ N (To)

1 + trio (T)]/[No (To) ] 2

(2)

In compression and bending tests, the measured effects of aging are practically independent of temperature: [I (V) -- n 0 ( r ) ~ n (7"0) --/;% (r~) or

n (r) ~ n (to) + [no (r) - no(to) I.

(3)

The service characteristics of composites that are of practical importance are those after natural exposure for longer periods than are feasible in testing (10-15 years). In order to obtain such information, we have used a model that was previously tested in [18, 22] for predicting (extrapolating) performance: H (To) ---i70(/'o) +~1 (1 - e -~t) - ~ ln(1 --1-•

(4)

703

where ~/ and fl are parameters of the material; 2 and • are parameters of the material and the ambient medium; t is the exposure time. Using Eq. (4), we have predicted irreversible changes in the strength of the composites after 10 years of aging under cold climatic conditions. In estimating the overall changes in strength, we used Eqs. (2) and (3). In Table 6 we present data for cases in which the predicted loss of strength is greater than 30%. In our opinion, such a loss of strength represents a serious deterioration of service properties of the composites. With regard to the organic fiber-reinforced plastics, we must note once more than the data presented here pertain to objects that have been damaged in the course of exposure by the "unprotected edge effect." CONCLUSIONS

1. From an analysis of experimental data obtained in two to three years of natural exposure of fiber-reinforced polymeric composites under cold climatic conditions, the following basic facts have been established: The atmospheric durability of composites based on glass and organic fibers under cold climatic conditions is generally higher than for materials based on carbon fillers; the atmospheric durability of composites with complex lays of the fiber decreases with increasing relative amount of transversal monolayers (monolayers that are oriented perpendicular to the direction of stress); the presence of an unprotected edge has practically no effect on the stability of the fiberglass-reinforced plastics, but has a very adverse effect on the atmospheric durability of the organic fiber-reinforced plastics - a difference that can be explained by the specific features of structure of the organic fibers; in exposure under cold climatic conditions, the strength characteristics of the plastics reinforced with glass or carbon fibers decrease to a greater degree than the elastic characteristics. 2. On the basis of a correlation of the factual information that has been obtained, relationships are proposed for finding the overall (irreversible + reversible) changes in the strength properties of composites as a result of climatic effects (with the experimental testing limited strictly to determinations of irreversible changes). 3. Using a previously tested semiempirical model, we have predicted changes in strength properties of these composites during 10 years of storage under cold climatic conditions; for certain carbon fiber-reinforced plastics and also for the organic fiber-reinforced plastics with unprotected edge, the predicted loss of strength is very serious - more than 30%. LITERATURE CITED 1. 2. 3.

J. C. Halpin and M. E. Waddoups, "Composite or steel pipe for North Slope applications," Eng. J., July-August, 32-36 (1972). S. A. Kuz'mln, V. N. Bulmanis, and A. S. Struchkov, "Experimental investigation of strength and deformability of fiberglass and organic fiber wound plastics at low climatic temperatures," Mekh. Kompozitn. Mater., No. 1, 57-61 (1989). V. N. Bulmanis, N. S. Popov, T. A. Starzhenetskaya, S. A. Kuz'min, G. I. Milyutin, and V. I. Polyakov, "Effect of alternating temperature cycling and moisture on strength of fiberglass and organic fiber wound plastics," Mekh. Kom-

5.

pozitn. Mater., No. 6, 1045-1051 (1988). N. S. Filatov, Climatic Durability of Polymeric Materials [in Russian], Moscow (1983). A. A. Gerasimov and V. N. Bulmanis, "Influence of static loads and cold-climate factors on deformability, strength, and

6.

life of fiberglass wound polyester plastic," Mekh. Kompozitn. Mater., No. 5, 862-867 (1988). H. W. Lord and P. K. Dutta, "On the design of polymeric composite structures for cold regions applications," J. Rein-

7.

forced Plast. Compos., 7, 435-458 (1988). GOST [All-Union State Standard] 9.707---81. ESZKS. Materials, Polymeric, Accelerated Test Methods for Climatic

8.

Aging [in Russian], effective January 1, 1983, Moscow (1982). Yu. S. Urzhuntsev and I. N. Cherskii, "Scientific principles of engineering climatology of polymeric and composite

4,

9.

10. 11. 12. 13. 704

materials," Mekh. Kompozitn. Mater., No. 4, 708-714 (1985). S. Brandt, Statistical and Computational Methods in Data Analysis, 2nd ed., Elsevier, New York (1976). A. V. Doos, "Influence of atmospheric exposure on mechanical properties of fiberglass-reinforced plastics SVAM," in: Physical Chemistry and Mechanics of Oriented Fiberglass-Reinforced Plastics [in Russian], Moscow (1967), pp. 234-239. N. N. Pavlov, Aging of Plastics Under Natural and Artificial Conditions [in Russian], Moscow (1982). P. Meelroy and D. Roylance, "Weatherability of sheet molding compound," Hi-Tech. Rev. 1984, 16th Nat. SAMPE Tech. Conf. Albuquerque, N.M., October 9-11, 1984, Vol. 16, Covina, Calif. (1984), pp. 505-512. S. Royland, "Weathering of fiber-reinforced epoxy composites," Polym. Eng., 18, No. 4, 249-254 (1978).

14.

P. G. Polezhaeva, M. Yu. Pashkevich, and E. P. Bogorodskaya, "Atmospheric durability of fiberglass-reinforced polyester

i5.

plastic," Plast. Massy, No. 5, 57-58 (1986). O. V. Startsev, Yu. M. Vapirov, I. S. Deer, V. A. Yartsev, V. V. Krivonos, E. A. Mitrofanova, and M. A. Chubarova, "Effect of extended atmospheric aging on properties and structure of carbon-reinforced plastic," Mekh. Kompozitn. Mater., No. 4, 637-642 (1986).

16.

O. V. Startsev, V. P. Meletov, B. V. Perov, and G. P. Mashinskaya, "Investigation of mechanism of aging of organic fiber-reinforced laminated plastic in a subtropical climate," Mekh. Kompozitn. Mater., No. 3, 462-467 (1986).

17.

J. P. Sandifer, "Effects of corrosive environments on graphite/epoxy composites," Proceedings of 4th International Conference on Composite Materials, Tokyo (1982), pp. 979-986.

18.

V. N. Bulmanis and O. V. Startsev, "Prediction of change in strength of fiber-polymer composites as a result of climatic

19.

action," Preprint, Yakutsk Branch, Siberian Branch, Academy of Sciences of the USSR, Yakutsk (1988). A. N. Aniskevich and N. E. Khramenkov, "Investigation of the effect of moisture on the properties of organic fiber-rein-

2{.}.

forced plastic by thermoanalytical methods," Mekh. Kompozim. Mater., No. 5, 911-916 (1989). N. S. Popov, "Effect of moisture and alternation of above-freezing and subfreezing temperatures on mechanical proper-

21.

ties of wound polymeric composites," Candidate's Dissertation, Yakutsk (1989). E. A. Eskin, V. K. Fedchuk, A. S. Petrov, and A. V. Kuleznev, "Strength and features of fracture of fiberglass-reinforced plastics in the 70-400 K temperature interval," Probl. Prochn., No. 8, 97-101 (1988).

22.

G. I. Milyutin, V. N. Bulmanis, T. S. Grakova, N. S. Popov, and A. M. Zakrzhevskii, "Investigation and prediction of strength characteristics of organic fiber-wound epoxy plastic under various combination of ambient conditions," Mekh. Kompozitn. Mater., No. 2, 247-253 (1989).

INFLUENCE ON THEIR

OF STRUCTURE FAILURE

OF POLYETHYLENE

OXIDE FILMS

IN CYCLIC BENDING

A. G. Efimov, I. V. Grushetskii, and S. V. Kharitonov

UDC 620.17:539.43:678.01

In a study of the static fatigue of polymhylene oxide (PEO) films [1], it was found that the curves have a complex character: Beginning at a certain stress level, a further increase in the stress does not shorten the life, but instead increases it [2]. This behavior is due to differences in the fracture mechanism. At low stresses, a quasibrittle fracture mechanism is realized, in the course of which, on the background of 3-5% strains, microdefects are accumulated, and a major crack is initiated and propagated through the material. Alter fracture in this interval of stresses, small residual strains remain. Also possible is a basically different mechanism in which the fracture is preceded by a repeated orientational drawing, after which we are dealing with an entirely new material. Since PEO films in service are presumably in an unoriented state, we need fatigue test methods in which fracture is effected on a background of small strains. Bending resistance tests will meet this condition. As objects of investigation we took four grades of high-molecular-weight PEO from Soviet and foreign production (Table 1). Film materials were prepared from the PEO samples by milling and pressing. The original powders were milled with a roll temperature of 120~ for 3 min, pressed at t = 120~ and p = 150 MPa for 2 min, and then cooled to room temperature in the press. This processing gave films with a thickness of 0.25 _ 0.5 mm,* density 1.81, 1.215, 1.230, and 1.223 for the samples 1, 2, 3, and 4, respectively. In the cyclic tests for which the results are shown in Table 2, we used a procedure developed at the Institute of Polymer Mechanics, Latvian Academy of Sciences [3], which is based on passing the specimen around cylindrical *As in Russian original -- Translator. Ivanovo Scientific-Research Institute of Film Materials and Artificial Leather for Industrial Applications. Institute of Polymer Mechanics, Latvian Academy of Sciences, Riga. Translated from Mekhanika Kompozitnykh Materialov, No. 6, 1{)74-1080, November-December, 1991. Original article submitted June 10, 1991. 0191-5665/91/2706-0705512.50 9

Plenum Publishing Corporation

705