UDC E F F E C T OF THE S T R U C T U R A L T E F L O N IN L I Q U I D M E D I A
STATE
ON THE S T R E N G T H
620.172:678.6'74:677.251.001.5
OF
N. G. Kalinin, A. N. Kogut, A. I. Soshko, and A. N. Tynnyi Fiziko-I
Fig. 1. X-ray diffraction patterns of teflon F4D: a) 40% crystallinity; b) 80% crystallinity.
F4 specimens were made from powder by pressing [1]; F4D specimens were cut from tubes obtained by extruding ~lubricated pastes. ~ The long-term strength of teflon specimens was studied by determining their time-to-rupture under constant loads. t h e tests were carried out on specimens differing in the degree of crystallinity (40 and 80%) obtained by subjecting them to appropriate heat treatments. The specimens were heated above the crystalline melting point and rapidly cooled. Under these conditions crystal nuclei cannot grow to a large size and a large proportion of the material passes into an amorphous phase, especially in its surface layers. Due to the low thermal conductivity of teflon, however, this process takes place not in the entire specimen volume but only in its surface layers. This is especially noticeable in the case of F4D specimens which after quenching become brighter and have thin transparent surface layers consisting of the maximally amorphous material; the speci-~ men core is cooled at a slower rate and is characterized by a higher degree of crystallinity. A high degree of crystallinity was produced by heating the specimens above the crystalline melting point and cooling them slowly in a furnace. The heat treatment was carried out in a vertical tube furnace with a hinged bottom. After 4 hr at 880 ~ C the specimens were instantaneously transferred into a vessel with liquid nitrogen. This ensured a sufficiently large t e m p e r a ture drop necessary to obtain specimens characterized by the lowest degree of crystallinity. The degree of crystallinity was determined by X-ray diffraction analysis of specimens cut from F4D tubes or from specimens for mechanical tests. The optimum thickness of specimens for X-ray measurements was determined from the Kratki-Krauss formula
d---
2 l k - -1 cos 20
1
"5' 59
where p is the linear absorption coefficient. The density of our specimens was about 2.17 g / c m 3. Starting from published data on the mass-absorption coefficients of substances appearing in the structural formula of teflon, we calculated the linear-absorption coefficient p which for Cu K a radiation was equal to 14.6; hence the optimum specimen thickness was calculated at approximately 0 . 9 - 1 . 0 ram. It has been postulated [4.5] that the intensity observed on X-ray patterns is due to two effects which can be separated by drawing a continuous line tangent to the minima of the experimental intensity curve. The general equation of intensity
~ 1 (s) dVs = ~ p2 (r) dV (r) may, when applied to crystalline and amorphous phases, be written in the form
* ( I ($)am @ I ($)r
where the vector
~=
dVs : S p~(r)(dVam+dVcr) "-- Vam-}-Vcr "" e %am']-$ %cr,
2 Sin 0 , p is the electron density of the substance, ~ is a vector characterizing the position
i of electrons in the volume, dV s and dV(r) denote elements of reciprocal space and volume, O is the diffraction angle, and Vam and Vcr denote the volumes of amorphous and crystalline phases, respectively. In the case of disordered specimens* Icr and Iam have practically the same distribution with respect to values ~ so that
S lcr (S) dVsmfflcr (8) dO
%,
/am(S)dVs , S [am(O)dlg"eam . The above formulas give results close to the real values. The degree of crystatRnity was determined by a method developed by Urbanchuk [5] using lines corresponding to Bragg angles of 8.6, 18, .and 18.6 ~
% 45
5
a.s
4
2.5
3 I
I
0.5
~0
~.5
2.0
2.5
32
3.5
cr,kg/mrn 2
Fig. 2. Dependence of the long-term strength of teflon F4D on its degree of crystallinity: 1, 2)40~ crystallinity, in air and kerosene, respectively; 8,4)80~ crystallinity, in air and kerosene, respectively.
ZO
I
I
25 32 tr, kg/mm2
/,G
Fig. 3. Dependence of the longterm strength of teflon F4 on its degree of crystallinity: 1, 2) 40% crystallinity, in air and kerosene, respectively; 3) 80% crystallinity, in kerosene.
* X-ray diffraction patterns on flat plates revealed the absence of orientation in starting material specimens.
6O
When an ionization method was used for recording, the X-ray diffraction measurements were carried out on a URS-80tM camera, the distribution being recorded on graph paper. In the case of high-crystalline specimens, the crystalline regions may reach large linear dimensions so that reflections from different specimen regions may differ in intensity. To obtain averaged diffraction intensityvalues, the specimens were rotated (about an axis normal to the specimen surface) with the aid of a special attachment .-e to a GUR-4 goniometer. To eliminate incoherent radiation and other reflections distorting the amorphous back~ ground, the differential filtration was carried out on Xrays diffracted from the specimen. 8 ta X-ray diffraction photographs were taken using Ni and Co filters under identical conditions and the resulting two curves were used to obtain the difference curve. fo determine the degree of crystallinity by this method, X-ray diffraction patterns were obtained for both standard and experimental specimens. The standard specimens were obtained by a long annealing treatment at 880 ~ C. The degree of crystallinity was calculated from the following formulas: F__~A = X a .
ha _
FB
h~
Xo
l -
X a
4
A
08
0
O
7
2
c~,kg/mm2
$
~0
!#
ta e, kg/mrn2
22
Fig. 4. Kinetics of swelling of(a) F4D and (b) F4 specimens in relation to applied stress : 1) 40 % crystallinity; 2) 80% crystallinity.
1 -- X b
where FA, FB and h a, h b denote the areas of crystallinity peaks and amorphous background levels on intensity curves of specimens A and B. Ihe strongest reflection is at a small angle to the primary beam so that both its profile and height are strongly dis~ totted by the diffuse scattering of the primary beam in air. The Bragg angles of the next strong reflections are 18 and 18.5 ~ but their intensities are too Iow for measurements using the ionization method. The determination of the degree of crystallinity from these diffraction lines was done with the aid of X-ray diffraction patterns (Fig. 1) obtained in a camera 114 mm in diameter. The results obtained by these two methods coincided. Although calculating the degree of crystallinity from X-ray diffraction patterns on films is more tedious, this method makes it possible to eliminate mistakes due to the influence of the primary beam. Specimens of teflon F4D before and after long-term strength tests ( i . e . , after the), had adsorbed certain quantities of the working medium) were examined with a UEM-100 electron microscope by the replica method. The replicas were made by vacuum deposition of chromium or titanium on carefully polished polymer specimen surfaces from which they were separated with the aid of gelatin. The results of long-term strength tests carried out on F4D and F4 specimens (Figs. 2,3) showed that their t i m e - t o rupture is substantially reduced under the influence of liquid media, especially kerosene, and that their long-term strength decreases with increasing degree of crystallinity. At the same time, kerosene has no effect on the breaking stress of teflon in short-term tests; this may be attributed to the relatively short time of interaction between the medium and polymer specimens at fast loading rates [2]. The specimens tested for long-term strength were weighed before and after tests; the results of these measurements are reproduced in Fig. 4. It will be seen that the swelling characteristics of teflon are strongly dependent on its degree of crystallinity. It should be noted that the extent of swelling of teflon specimens with an amorphous structure is negligible. The results of previous investigations showed that a combined action of applied loads and working media leads to a limited swelling of plastics, while no swelling is observed in the absence of external loads. Generally speaking, the effect of working media on plastics (cracking, loss of strength and gas tightness) may be attributed to the onset of swelling whose intensity is related to the stress state of the material. The application of an external load weakens the intermolecular interaction forces; since the molecular chains or molecular aggregates are not uniformly stressed because of the absence of structural order, the most intense weakening of intermotecular forces is observed in regions of the maximum stress. A sharp increase in the diffusion of working media into these regions takes place; eventually this leads to the destruction of bonds between chain molecules and their separation in space. In the case of crystalline polymers (such as teflon F4 and F4D) the most intensely stressed load-carrying regions are boundaries of the supermolecutar structures.
61
Electron-microscope examination showed that the micmstructure of teflon F4D (Fig. 5a) in its initial state consists of long unoriented fibrillar formations. As shown by the photograph in Fig. 5b, straining this polymer leads to the fibrous formations becoming oriented in the direction of the applied load, The microstructures of teflon specimens deformed in air and in a liquid medium are identical, the same applying to specimens examined before and after the exposure to the action of a liquid medium.
F.ig. 5. Microstructure of F4D and F4 specimens containing 4 0 9 crystalline phase: a) initial state; b) after deformation to rupture in air or liquid working media. Thus, nonuniform stress distribution in a polymer promotes selective action of liquid working media, i . e . , accelerates the permeation of these media into the maximum stress regions. Naturally, liquid medium molecules reduce the intensity of molecular interaction and, as a result, facilitate the "slip" or "unplaiting" of molecular chains or molecular formations, thereby reducing the energy required to separate the material into fragments. The intensity of the permeation of liquid molecules into, and their interaction with, a solid poIymer determines in the final analysis the degree to which the medium affects the long-term strength, permeability, swelling characteristics, and other poIymer properties. The extent to which stressed teflon is affected by liquid media depends on the retative contents of the crystalline and amorphous phases, on the character of interaction between the liquid and the solid material, and on the stress state at the polymer phase boundaries. It may be concluded from the results obtained that heat treatment under optimum conditions leads to the formation of a large quantity of fine crystaIs with a large proportion of an amorphous phase being formed, especially in polymer surface layers; this Should reduce the possibility of a stress gradient being formed under the influence of applied loads and therefore inhibit the diffusion of liquid media into polymers of this kind. REFERENCES 11 D. D. Chegodaev, Z. K. Naumova, and Ts. S. Dunaevskaya, Teflon [in Russian], Goskhimizdat, Leningrad, 1960. 2. 3. 4. 5.
A. N. Tynnyi and A. I. S0shko, Ft
25 March 1968
62
Institute of Physics and Mechanics AS UkrSSR, L'vov