No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
Why do plastics stress‐crack?
Abstract
Possible applicability of Griffith's theory to stress‐cracking in microcrystalline organic polymers is considered. Although the time‐dependent nature of the phenomenon and the plastico‐elastico‐viscous character of the medium make such application debatable, it is found that this approach in combination with the contributions of others notably Rebinder and his associates, can provide rationalization for many of the empirical facts. The observed “case‐hardening” action of surfactants on specimens under stress suggests that the mobility of polymer chain segments in surface layers may in fact be restricted under these conditions as theorized, facilitating local concentration of stresses to a level exceeding the strength of the material.
The environmental stress cracking resistance (ESCR) was studied for pipes made of bimodal poly(ethylene-co-1-butene) copolymer. Ductile or brittle failures of the pipe were observed after pipe exposure to different temperatures and hydrostatic pressures. Changes in the phase composition, size of crystalline and amorphous domains and molecular mobility were studied in failed test pipes by DSC, SAXS, solid-state ¹³C NMR spectroscopy and time-domain ¹H NMR T2 relaxometry. The ESCR test at 20 °C causes a decrease in crystallinity, which suggests lamellar fragmentation, whereas a slight crystallinity increase is observed at a test temperature of 70 °C. At both temperatures, molecular mobility in the amorphous phase decreases, suggesting stress-induced elongation of chain segments in the amorphous phase. ESCR tests at 95 and 110 °C led to crystallinity increases and lamellar thickening owing to partial melting and recrystallization during long-term pipe exposure to high temperatures. The test causes an increase in molecular mobility in the amorphous phase due to a change in partitioning of ethyl chain branches between the crystal-amorphous interface and the amorphous phase. The role of molecular mobility in the amorphous phase on the type of failure and on reorganization of semicrystalline morphology during the ESCR test are discussed.
The abstract for this document is available on CSA Illumina.To view the Abstract, click the Abstract button above the document title.
Besides other facets of product development, it is imperative for medical device manufacturers to take great efforts through proper evaluation and consideration of material properties under practical conditions to prevent product failure at the end-uses. The environmental stress crack (ESCR) induced by chemical agents plays a significant role on material performances. In this contribution, in-depth studies have been carried out on different medical plastic materials, such as polycarbonate, copolyesters, ABS, acrylics, rigid thermoplastic polyurethane and their blends. More attention will be focused on a copolyester material for its unique ESCR behavior. Variation of chemical agents (such as different types of hospital disinfection solutions) have great impacts on physical and functional properties. Various plastics shows distinct environmental stress cracking phenomena under different conditions. Mechanisms of ESCR phenomenon under different environments have been explored. Fibril reinforcement by cold crystallization and chain session by hydrolysis of the copolyester may have contributed to its excellent chemical resistance against a wide range of chemicals and its catastrophic failure in acidic or basic environment. In addition, appropriate definition of product failures is also critical in making materials decisions.
The effects of an isopropanol environment on the deformation characteristics of copoly (acrylonitrile-butadiene-styrene) (ABS) and poly (vinyl chloride) (PVC) have been investigated. Creep tests were performed in tension and torsion both in air and in isopropanol. It was noted that accelerated rates of creep were encountered during the tensile tests in isopropanol as compared with air but no differences were noted during the torsion tests. An explanation is proposed that the greatly increased creep strains encountered in tension result from diffusion of the fluid into the polymer. The rate of diffusion is controlled by changes in the free volume which occur on application of uniaxial tensile stresses. The existence of the fluid molecules in the polymer matrix reduce its local creep resistance and produce swelling, both of which are manifested as increased creep strains. The fact that no effects were noted during torsion tests (where the free volume remains constant) or when an unstressed specimen was immersed in the fluid indicates that no significant absorption occurs at the equilibrium (unstressed) free volume.
The stress-cracking of polycarbonate by a gaseous or liquid agent results from the diffusion of this agent into the polymer. The low molecular weight polymer fractions and the chain ends within the bulk of the polymer become more ordered during the diffusion and swelling process by their partial solubility in the crazing agent, causing crystallization. The creation of interfaces at areas of order-disorder causes high shearing forces at this boundary and voids within the bulk of the polymer. These voids are then propagated as crazes or cracks at stresses much lower than the tensile strength of the polymer. Therefore, a stress-cracking agent need not diffuse rapidly, but must be an effective environment for swelling and/or crystallization. Data from diffusion, density, thermal and molecular analyses are presented to support this mechanism.
Single crystals of polypropylene consisting of lamellas have been grown from the melt. Electron diffraction patterns indicate that the molecules are folded within the lamellas and have crystallized in a hexagonal unit cell.
A criterion for the transition in fracture mode from ductile to cleavage is calculated for fracture at a notch. It is concluded that the temperature-dependence of this transition probably arises more from that of the Peierls-Nabarro stress required to move a free dislocation in α-iron than from the temperature-dependence of the locking of e dislocation source. The transition temperature is found to be a function of grain size, the friction on a free dislocation, the strength of the dislocation locking, the degree of triaxiality of the applied stress and the elastic constants.
The morphology and orientation of a number of polymers as crystallized from solution was studied with the electron microscope combined with selected area electron diffraction. In the course of this work single crystals could be prepared which were most striking in the case of straight chain polyethylenes. There is evidence among others, that the fibrillar (sheaf-type, spherulitic) crystallization, characteristic of polymers, develops through formation of flat single crystals. The single crystals contain screw dislocations and grow by spiral terraces. The observed orientation and the minimum thickness of the crystals leads to the inescapable conclusion that the molecules must bend sharply back on themselves forming a regular folded configuration. Accordingly the strength of the dislocation, consequently the height of growth step, would correspond to the distance between successive bends. Various other observations are consistent with this new concept.
The stresses round a piled-up group of dislocations are investigated with reference to the initiation of a crack. A crack should form when the group consists of about 1000 dislocations piled up under a stress of the magnitude occurring in a cold-worked metal. The stress system due to the crack is obtained. The crack will have a stable equilibrium length, but this length is likely to be determined by the amount of plastic flow which takes place round the tip of the crack.
By means of a constant stress test, the environmental stress-cracking behavior of linear polyethylene has been studied on a macro and micro scale in an effort to determine the mechanism of the process. Upon the application of stress, linear polyethylene develops a network of very fine, elliptical fissures, the edges of which are connected by cold-drawn material. In the absence of an active environment, these fissures slowly grow and interconnect, resulting ultimately in the formation of a “neck.” When exposed to an active environment, however, the cold-drawn material ruptures as it is formed at the tips of the fissures. Unsupported, these fissures grow rapidly and interconnect resulting in sample failure. Fissures form both around and through the centers of spherulites with less cold drawing occurring at the interspherulite boundaries. Macroscopic studies confirmed the observation that active environments attack stressed polyethylene specifically at microzones of cold drawing. The effect of low molecular weight hydrocarbon species on the stress-crack resistance of linear polyethylene was evaluated. The role of flaws in the process is also discussed. Attempts have been made to establish a criterion of environmental activity. All of the active stress-cracking agents studied were found to reach similar levels of absorption in polyethylene; however, the specific chemical nature of the environment and not merely its level of absorption determines its ability to cause stress cracking.
The slow growth and recovery of tensile microfractures in films of organic glasses create an apparent tensile creep-enhancement of the coefficient of penetrant diffusion. These effects, although of a greater magnitude, are strikingly similar to those found by McAfee with Pyrex glass. This situation encourages the notion of progressive submicrofracture in inorganic glasses. The advent of increased porosity under tension suggests a role of localized vapor absorption in the environmental-stress crazing and in the “anomalous” vapor diffusion in organic glasses. It becomes increasingly apparent that progressive breakdown is a critical consideration in the mechanical behavior of viscoelastic solids.
Isotactic polypropylene was melted and crystallized at 125°C. Crystallization was observed dynamically on a microscopic heating stage. Spherulites, when first formed, appear to have a lower order of birefringence than after some growth. Between 4–4.5 min. of crystallization cracking occurs between spherulitic boundaries. In a parallel experiment samples crystallized at 125°C. were quenched after different degrees of spherulization. The effect of degree of spherulization on stress-strain properties was studied and the necks and fractures of the samples studied. The completely spherulitic samples broke abruptly at spherulitic boundaries without necking. The rate of elongation affected the nature of the fracture of completely spherulitic samples. At low rates of elongation the spherulites have a tendency to “pull out” at the fracture. At high rates of elongation the spherulites break more sharply at their boundaries.
The thickness of single crystal lamellae of polyethylene, crystallized from solution, is found to increase greatly during annealing at temperatures above 110°C. The change is observed with small-angle x-ray diffraction and electron microscopy. The increase in thickness of the lamellae takes place at the expense of their lateral perfection; holes develop within the lamellae. All evidence indicates that a major refolding of the molecules occurs; this emphasizes the need for a new concept of the amount of motion and freedom that polymer molecules can have in the solid state. A similar process apparently occurs during the annealing of bulk samples crystallized from the melt.
A surface-etching technique has been devised which reveals the spherulites in molded Polyethylene for study with the electron microscope. By means of this technique it is possible to observe the effect of various heating cycles and polymer structures on the size and perfection of the spherulites. The etching technique shows that two different types of large spherulites may be formed in polyethylene. One, which may be formed on molding, contains a relatively low amount of internal order. The other, which forms on long heat aging or annealing, contains a high degree of internal order. Heat aging at temperatures well below the melting point will induce the poorly ordered type of spherulite to break up and regrow in the highly ordered form. Spherulite structure was observed to depend not only on heat treatment, but also on chain branching and molecular weight. Increased linearity induces the formation of a more regular, sheetlike or laminated structure within the spherulites, while lower molecular weight causes the lamellae and spherulites as well to be larger in size. Extremely high molecular weight linear resins showed no spherulitic structure as such, but were fibrillar in appearance. Annealing or long heat aging of these resins induced the formation of a row type aggregate structure rather than the typical spherulites observed in normal molecular weight resins. Electron microscope studies of stress-cracked surfaces indicate that a high degree of internal spherulitic order such as is formed on long heat aging may well be the primary cause of stress-cracking in polyethylene. This is further indicated by resins such as the ethylene/propylene copolymers, in which highly ordered spherulites do not form with heat aging. These resins are very resistant to stress cracking. Structural studies show that the ethylene/propylene copolymers are very nonrandom, and it is believed that this interferes with the formation of perfect, and consequently crack-prone, spherulites.
The general aspects of the problem of fracture and strength are briefly reviewed. The discrepancy between the observed strength of materials and the theoretical value calculated from the molecular structure is usually ascribed to the presence of adventitious flaws present in all samples tested. The Griffith theory of brittle fracture gives the relation between the tensile strength T and the size of the flaw in the sample c as T = A(Eγ/c)1/2. The tensile strengths of samples of poly(methyl methacrylate) containing artificially induced cracks of known size were determined. The results could be represented over a wide range of experimental conditions by the form of the Griffith theory. The value of the specific surface energy obtained from the position of the theoretical curve was ∼3 × 105 ergs/cm.2. The value calculated from the molecular structure of the material is ∼450 ergs/cm.2. The discrepancy is attributed to the energy consumed in a viscous flow process in a thin layer of material at the fracture surface. As a result of this process it is believed that in the surface layer polymer molecules are oriented normally to the fracture plane. This mechanism accounts for the colors observed on the fracture surface and for their lability on exposure to solvents or heating. The Griffith theory postulates a strong dependence of tensile strength on surface energy. In poly(methyl methacrylate) the largest contribution to the surface energy appears to arise from a viscous flow process. Hence for polymers the strength may depend much more on the intermolecular (van der Waals) forces than on then ature of the intramolecular primary valence bonds.
Lamellae are basic structural elements of the acetal resin, polyoxymethylene. Lamellae, similar to those observed after precipitation of polymer single crystals from solution, are found both in the interior and on the surface of samples crystallized from the melt. Lamellae crystallized from the melt are organized into spherulites or hedrites. The latter are hexagonal structures, several microns thick, grown by slow crystallization of thin films. The thickness of the hedrite depends directly on the crystallization temperature and only indirectly on the thickness of the original film. New lamellae originate at screw dislocations on previously established lamellae at the centers of both spherulites and hedrites, as well as on the periphery of growing spherulites. A mechanism based on these observations is postulated for the crystallization of polyoxymethylene from the melt.
The extinction behavior of spherulites in many polymers and nonpolymers is shown to result from a characteristic twisting of crystal orientation acquired during growth. Extinction patterns are calculated for model spherulites consisting of radial fibers in which uniaxial or biaxial crystalline fibers twist regularly in a right- or left-handed sense about the fiber axis. The results are in good agreement with observed extinction patterns in all of the following cases: diametral sections of three-dimensional spherulites, diametral sections tilted on a Federow universal stage, and two-dimensional spherulites grown in thin films. Details of this correlation will be given in Part II. Rules are given for (a) detecting the existence of twisting orientation, (b) determining whether it is of right- or left-handed sense, and (c) determining the crystallographic axis about which twisting occurs.
The stress cracking of polyethylene was studied under various experimental conditions, and the importance of the effect of both crystallinity and molecular weight of the polymer was established. Many of the variables were investigated, such as the thermal pretreatment of the sample, temperature of the sample, surface tension, and concentration of the stress-cracking agent, and the magnitude of the stress. The importance of this test procedure and the relationship that exists between the stress cracking of polymers and between similar problems affecting nonpolymeric systems have been analyzed. Some microscopic studies of stress cracking have also been made.
The spherulitic crystallizations of polypropylene were investigated by a polarizing microscope. At high temperature, crystallization proceeds through the process of dendritic mechanism. Long needles grow rapidly at first from the nucleus to the supercooled melt, and then they fan out around each needle by branching. The rates of these two processes depend on the polymer and the condition of crystallization. In large well-grown spherulites, cracks are formed both between radial fibers and at boundaries during heat aging. The cracks are void and single spherulites are separated. The boundary cracks are also caused by thermal decomposition of molecules and the radial cracks by stresses due to the rearrangement of molecules or the perfection of spherulites.
Die Hrteverminderung H einer Kristallflche bei Adsorption grenzflchenaktiver polarer Stoffe aus Lsungen ist experimentell nachgewiesen und untersucht. — H ist mit einer Abnahme der freien Grenzflchenenergie verbunden. Die H Konzentrationskurven verlaufen den entsprechenden Adsorptionsisothermen parallel.H erreicht einen Minimalwert bei Sttigung der Adsorptionsschicht mit orientierten polaren Moleklen. In einer homologen Reihe nimmt der H-Effekt mit der Orientierungsfhigkeit der Molekle des Hrteherabsetzers zu (entsprechend der Traubeschen Adsorptionsregel). Der Einflu der Polaritt des umgebenden Mediums und des festen Krpers wurde ebenfalls untersucht. — Es ist auf die. Bedeutung des H-Effekts zur Erleichterung der Ribildung bzw. Rierweiterung und des mechanischen Dispergierens hingewiesen worden).
A systematic study of impurity segregation in spherulitic crystallization is described. The experiments deal principally with high polymers, in which the role of ``impurities'' is fulfilled by stereoirregular molecules or by molecules of low molecular weight. It is shown that these species are rejected preferentially by growing crystals and that their diffusion plays a vital part in governing over‐all morphology. In particular, openness of texture is related to the concentration of impurity present; and coarseness of texture, which is a measure of the ``diameters'' of crystalline fibers between which impurities become concentrated during crystallization, is determined by δ=D/G, where D is the diffusion coefficient in the melt and G is the radial growth rate of the spherulite. Results provide substantial support for a theory of spherulitic crystallization proposed by the authors.
To account for spherulitic crystallization from the melt, one must explain the origins (i) of fibrous crystal habits in the absence of appreciable temperature gradients and (ii) of profuse noncrystallographic branching. Attention is drawn to properties held in common by spherulite‐forming melts of various types and, in particular, to the facts that (a) they are multicomponent systems, (b) they exhibit small coefficients of self‐diffusion, and (c) they crystallize slowly. It is shown that a consequence of these properties is that a plane crystal face cannot grow without suffering an instability of profile. Analagous instabilities lead in metal crystals to a cellular interface but, because of unusual growth kinetics, instability in spherulite‐forming melts gives rise to a drastic modification of crystal habit. Bundles of discrete fibers are formed whose widths are determined by δ = D/G, D being the coefficient of self‐diffusion and G being the growth rate. δ is generally small in these systems and commensurate with the scale of crystalline disorder in the fibers. It is this circumstance that allows noncrystallographic branching to occur.
The expression V=xV c +(1-x)V a , for the specific volume of a semicrystalline polymer, is a highly successful approximation; here x is the degree of crystallinity, V c is the specific volume for the crystalline regions, and V a is the specific volume for the amorphous regions. In specimens for which x is near unity, however, this equation demands values for the parameters which are inconsistent with other experimental evaluations. We suggest that the discrepancy arises because of the presence of voids. Thus in addition to crystalline and noncrystalline regions we propose regions of zero density. The average volume of such voids is not known, but their existence can be inferred from the significant, permanent density enhancement which can be brought about by pressure‐induced crystallization.
A survey has been made of spherulites formed by the isothermal crystallization of polypropylene from the melt in the temperature range 110°–148°C. Four types of spherulite have been distinguished and their structural morphology, optical properties, melting behavior, and growth rates have been examined.
Polyethylene of such molecular weight and structure that it readily fibers or cold draws to 300–600 percent elongation by usual uniaxial, tensile stressing may react quite differently under biaxial tension. When biaxial tension in 1:1 ratio is applied to a diaphragm, some polymers show brittle fracture with ≪20 percent elongation at break. However, if the average molecular weight of such polyethylenes is shifted upward by crude fractionation, or an initially higher average is used, the polymers orient under complex stresses. Then, they usually elongate several hundred percent before rupture. Variations in crystallinity are also significant, although most technical polyethylene soon attains at room temperature enough crystallinity so that this factor does not cause big differences. Although the whole study is so far preliminary, x‐ray scattering of stressed samples suggests that preferred glide on certain crystallite planes tends to occur as the yield point approaches. These are such as to inhibit smooth alignment of the long chain axis in the direction of stressing. This could lead to brittleness. Apparatus for complex stressing of sheets and tubes is described. Strains are taken from coordinates printed on the sample by the silk screen process. High speed stressing was also observed. The speed of retraction of amorphous polyethylene chains approaches that of rubber.
- Hittmair