摘要:

Transverse Crack Propagation in Fibre Reinforced Composites BY J. G. MORLEY Wolfson Institute of Interfacial Technology, University of Nottingham Received 12th June, 1972 The energy absorbed in propagating a crack transversely to the direction of alignment of the reinforcing elements in a fibrous composite is dependent on the nature of the fibre/matrix interface. Weak interfaces can debond allowing a matrix crack to propagate through the composite without immediately causing fibre failure. Energy can then be absorbed in various ways by the fibres left bridging the crack until they fail sequentually as the crack propagates. When the shear strength of the interface between the reinforcing elements and the rest of the composite structure is self adjusting, so that it is controlled by the local tensile stress in the reinforcing element, failure of the reinforcing elements can be prevented whatever the magnitude of the tensile strain developed in the composite structure.Reinforcing elements of this type, therefore, remain intact over the entire length of the matrix crack so that the energy absorbed during crack propagation can increase with increasing crack length. Preliminary studies of transverse crack propagation have been made using a model system incorporating self-adjusting interfaces of this type and the results obtained are discussed. The nature of the interface between the matrix and fibres is the most important single factor influencing the form of crack propagation in polymers reinforced with strong stiff brittle fibres such as glass and carbon.Cracks can propagate easily across the interface with little energy absorption if the fibres are very strongly bonded to the matrix. If the interface is weaker, the matrix crack may not cause immediate failure of the fibre which is then left bridging the tip of the propagating matrix crack. In these circumstances, the interface will debond for some appreciable distance on each side of the crack but residual frictional effects between fibre and matrix will control the extent of the debonded length. The crack-bridging fibres carry an extra stress, following the failure of the surrounding matrix, and this excess stress is transferred back to the composite through shear stresses acting at the fibre/matrix interface as a result of the residual frictional effects.The stress carried by the bridging fibre is therefore a maximum at the crack surface and decreases progressively for increasing distances from the primary matrix crack. The more nearly uniform the strengths of the fibres, the more likely they are to break at, or near, the surface of the primary crack, but if the fibres contain severe flaws there is a possibility that the fibre can fail at a point some appreciable distance away from the matrix crack, and if this occurs, additional energy can be absorbed by frictional losses as the fibre is pulled out of the matrix.' Energy can therefore be absorbed in the region around the tip of the pro- pagating transverse crack by a local increase in the elastic energy stored in the bridging fibres, by the debonding of the fibre/matrix interface, and by pulling fibres out of the matrix against residual frictional effects.However, the volume of the composite which can take part in these energy-absorption processes is limited to a small region on each side of the primary matrix crack and this is controlled by the residual stress transfer between the debonded fibre and the rest of the composite system. The bridging fibres will fail when the local strain carried by them exceeds some critical limiting value which is usually in the region of 1 or 2 %. As the width of the crack 109110 CRACK PROPAGATION increases, therefore, the bridging fibres fail sequentially so that at any one time the energy absorption processes are confined to a relatively small region around the tip of the advancing transverse crack.Theoretically, very high work of fracture values can be obtained if the shear strength of the fibrelmatrix interface, following debonding, is not constant but diminishes as the local tensile stress applied to the fibre increases.2 In such a system the maximum tensile stress which can be developed in the fibre by shear stresses resulting from frictional effects at the fibrelmatrix interface is limited to the stress at which the shear strength of the interface becomes zero. If this stress (amax) is less than the ultimate breaking strength of the fibre (of), then it is not possible to break the fibre by tensile deformation of the composite whatever the fibre length. When such a fibre is bridging a matrix crack, the extent of the stress transfer length, over which the extra stress carried by the fibre is transferred back to the matrix, is no longer constant but increases as the stress carried by the bridging fibre increases.The volume of the composite which can take part in energy absorption processes is now no longer limited to a relatively small zone on each side of the primary transverse crack but can extend to distances comparable with the lengths of the individual fibres or to the dimensions of the total composite structure if the fibres are completely continuous throughout the material. If the shear strength (7) of the interface is a linear function of the local fibre tensile stress o so that z = zo (1 - a/oma), where zo is the shear strength of the interface when the tensile stress of the fibre is zero, the stress distribution along the bridging fibre takes a simple exponential form ox = omax (1 -exp(-270~/a,,,r)), where ox is the stress on the fibre at a distance x from the free end and r is the radius of the fibre.The fibre tensile stress therefore approaches the limiting value of omax asymptotically and when the fibre is long compared with the stress transfer length, fibre pull-out will take place at a sensibly constant stress value. Experimental reinforcing elements of this type can be constructed straightfor- wardly using steel hypodermic tubes and steel wire.3 If the wire core is crimped into an approximately sinusoidal wave form having an amplitude greater than the internal bore of the tube, there will be a force exerted by the wire on the walls of the tube at the point of contact of each convolution.There is thus a frictional force resisting longitudinal displacement of the wire within the tube. Since the wire is convoluted, a tensile stress applied to the wire has a resolved component tending to straighten the wire and reduce the force exerted by the wire on the tube due to the initial elastic deformation. When these two forces are in balance, the local frictional force resisting longitudinal displacement is effectively zero. The stress level (amax) at which this occurs can be controlled by the geometry of the system. An experimental arrangement, simulating a metal matrix reinforced with elements of this type, has been constructed to enable an initial study to be made of the force required to be applied at the mouth of a matrix crack as a function of crack length in order to propagate the crack in a direction perpendicular to that of the reinforcing elements.The apparatus consists of two rectangular sectioned steel bars each having a width of 1.9 x m mounted horizontally one above the other. Two similar sheets of thin metal are clamped in a vertical position to the sides of the steel bars and one end of each bar is attached to the upper and lower jaw of a tensile test machine. A long narrow notch is cut to the same length in each of the two thin metal sheets to form an initial crack which can then be caused to prop- agate by the force applied by the tensile test machine. The further ends of each beam are attached to each other by a yoke into which each beam is pivoted, the two beams being parallel with each other at the start of the experiment.A series of vertical holes m and a height of 3.17 xJ . G . MORLEY 111 spaced 4.1 x m between centres are drilled along the centre lines of both beams to accommodate hypodermic tube steel wire reinforcing elements of the type described above. The wires thus bridge the upper and lower beams and the force required to pull the individual wires through the tubes remains constant for increasing separation of the steel beams. m (2 x in.) in thickness having a width of 6.35 x ni (2.5 in.). The stainless steel wire had a diameter of 6.1 x m mandrel and pre-tensioned by a load of 35 kg to form a helix having a wavelength of 6 x m and an outside diameter of 8.7 x m.The hypodermic tube used had an internal diameter of 8.25 x m and an external diameter of 1.06 x m. The initial slot length in the brass sheets measured from the point of application of a load to the steel beams was 1.2 x 10-1 m. Experimental results were first obtained for crack propagation in the metal sheets alone and, subsequently, the effect on crack propagation of the presence of the wires was examined. It was assumed that the steel bars would be held rigidly together in their initial parallel position by the still unfractured portions of the metal sheets so that the bars would deffect as simple cantilevers whose length would correspond to the crack length measured at any time from the crack tip to the point of application of the load at the end of the beams.This assumption was found to be in reasonable agree- ment with the experimental data since a linear relationship was obtained between the deflection of the point of the application of the load and f Z3 (where f is the force applied and I the crack length). The slope of this line approximated to the expected value calculated from the dimensions and the elastic properties of the beams. The condition controlling crack propagation of the metal shests as the crack length is increased by a small amount, is that the change in stored elastic energy in the steel beams should equal the work done in fracturing the metal sheets in extending the crack. The relationship between applied load and crack length for a simple rectang- ular bar of material containing a crack propagating in a plane equally spaced between two opposite faces (so that a double cantilever specimen is formed when a load is applied at the mouth of the crack perpendicularly to the direction of the crack prop- agation) has been calculated by Gilman.4 He derived the surface energy y of a crack face as The metal sheets used were brass shim material nominally 5.08 x m and was close wound on a 1.04 x y = (6f 2Z2)/(E~2h3), (1) where h is the distance between the crack and the opposite faces of the rectangular block parallel to the crack (i.e., the thickness of the cantilever beams), w is the width of the block, fthe crack opening force and Z the crack length.E is Young’s modulus of the material. Eqn (1) has been improved by a correction factor to take account of the shear deflection of the cantilever beams and of the compliance of the uncracked portion of the specimen by multiplying the right-hand side of eqn (1) by a factor of the form (1 +A(h/Z)+B(h/Z)2), where A and B are constants depending on the conditions at the fixed ends of the cantilever beams.The various equations which have been proposed have been discussed by Shockley and Groves who applied them to the study of apparent surface energy of MgO. The equations describing crack propagation in a rectangular block would be expected to apply to crack propagation in the thin metal sheets (in the absence of the bridging wires) if allowance is made for the crack being generated only in the thin metal sheets rather than across the whole width of the cantilever beam.To a first approximation, therefore, a linear relationship between the reciprocal of the square of the crack opening forcefplotted against the square of the crack length Z would be112 CRACK PROPAGATION expected; and this is found to be the case. Fig. 1 shows the results obtained for three sets of sheet metal specimens. Measurements were obtained, subsequently, of the crack-opening force for in- creasing crack lengths with bridging wires in position. The force required to prop- agate the crack in the metal sheets is now increased by the uniformly distributed load on the two cantilever beams generated by the bridging fibres. If the force per unit length of beam generated by the bridging fibres is given asfw, then the total force which we would expect to have to apply to the beam to propagate the crack would be increased tof+dfJ) x 3/8.The contribution to the crack opening force developed by the wires would therefore be expected to increase uniformly with increasing crack length. In the series of experiments with bridging wires, the steel beams were initially 0.14 0.12 0.1 0 0.0 8 N E \ N 4. 0.06 / / / /' x 5 % I R I I 1 I I 1.0 2.0 3.0 4.0 5.0 6.0 1 /f2/kN-" FIG. 1 .-Relationship between crack opening forcefand crack length I for three sheet brass specimens. arranged to be non-parallel, with their separation decreasing towards the end at which the load was being applied. The beams were then pulled into a horizontal position which ensured that all the wires were stressed to their upper limited value (omax) and from the limiting force applied to the beams and the geometry of the arrangement the average value of om,, for each wire could be calculated.The thin metal sheets containing notches as before were then clamped into position and the load increased to propagate the crack in the metal sheets, the results being illustrated in fig. 2. Curve (1) shown in fig. 2 is a transposition of the best straight line drawn through the experimental points shown in fig. 1. These values were then subtracted from the experimentally observed values of crack-opening force at various crack lengths for the combined wire and metal sheet system to give the contribution to the crack- opening force generated by the wires themselves ; these are shown in fig. 2 for wiresJ . G. MORLEY 113 set at 8.2 mm (curve 3) and 16.4 mm (curve 2) intervals.The slopes of the straight lines shown in fig. 2 are drawn to correspond with the expected increase in crack- opening force with increasing crack length calculated from the average experimental value of c, for the wires. These straight lines are drawn through the averagefand I values for the two sets of observations. They are in tolerable agreement with the experimental observations of the loads developed by the bridging wires for different crack lengths and also with the points shown (A) which correspond to the initial loads on the assembly of wires when the metal sheets were first clamped into position. The experimental arrangement therefore again approximates to two simple cantilever beams whose length at any time corresponds to the crack length in the metal sheets.5.0 4.0- - P 0. I 0 . 2 0.3 Ilm FIG. 2.-Relationship between crack opening force fand crack length 1 for sheet brass specimens (1) and crack bridging wires (2), (3). Fig. 2 illustrates the considerable change in the relationship between the crack- opening force against crack length for the unreinforced metal sheets and the metal sheets reinforced by crack bridging fibres of the type discussed above. Instead of the crack-opening force diminishing with increasing crack length, which is the normal situation, we now have a condition in which the crack-opening force is increasing with increasing crack length. The bridging fibres therefore provide a powerful energy sink which increases in magnitude as a transverse crack progresses.The relative volume fraction of wires in the " composite " is approximately 20 and 40 % by volume for the two systems studied. For orthodox fibre reinforced systems, the energy being absorbed during transverse crack propagation is confined at any one time to a rela- tively small region about the crack tip, and the fracture process can be considered as114 CRACK PROPAGATION being broadly similar to that of crack propagation under similar conditions for ortho- dox materials, such as ductile metals or elastic brittle solids. The addition of rein- forcing elements of the self-adjusting interface type discussed here enables the energy- absorbing processes to extend over the whole length of the crack. At the same time, the extent of the material on each side of the crack, in which energy absorbing processes are occurring, can be increased substantially.When the fibres are continuous throughout the material, this dimension can extend to the limits of the material in which the crack is propagating. Measurements of the work of fracture of unreinforced sheet brass specimens and similar specimens containing crack bridging reinforcing elements have been made using the apparatus described above. For the reinforced system the energy absorbed by the bridging fibres was almost completely due to frictional losses as the core elements were withdrawn from the sheath elements which were themselves fixed to the steel beams. Owing to the short length of the crack bridging fibres only a negligible amount of energy was absorbed, by comparison, in their elastic extention.The numerical value of the average work of fracture of the three experimental samples of unreinforced sheet brass plates calculated from the slope of graph of Z2 against llf” shown in fig. 1 is 2.54 x lo7 ergs cm-”. The work of fracture was also calculated from the area under the force displacement curve shown diagramati- cally in fig. 3. The increased separation of the beams at the point of application f 0 Q S b FIG. 3.-Relationship between the crack opening force f and the width of the mouth of the crack S (diagramatic). of the loadfis denoted by S. The triangular area OZla represents the elastic strain energy in the system when the crack is about to extend from its initial length ZI. When the crack has extended to a distance Z2 the stored elastic energy in the system is given by the triangular area OZ2b.The area bounded by 0Z1Z2 is therefore a measure of the work done in propagating the crack over a distance Z2 - Zl and thus the work of fracture can be calculated. The value of the work of fracture obtained from the experimental force displacement curve for an additional brass sheet specimen of theJ . G. MORLEY 115 same nominal thickness as the specimens included in fig. 1 was 2.90 x lo7 ergs cm-2 which is in reasonable agreement with the previous values. The second method has been used to compare the work of fracture values of the reinforced and unreinforced metal sheets and one set of experimental observations is illustrated in fig. 4. The upper curve in fig. 4 refers to a composite system in which the bridging fibres are placed at intervals of 32.8 mm (0.0328 m) along the steel beams.The lower curve is drawn through experimental points obtained for un- reinforced metal plates. Values of works of fracture were calculated for increases in crack length from an initial value of 0.12 to 0.243 m and then to 0.314 m for both the reinforced and unreinforced brass plates. (Force, deflection) curves, obtained as the systems were unloaded between incremental increases in crack length, are shown in fig. 4. The areas bounded by these curves indicate the work done in propagating the crack over the incremental distances mentioned above. 0 . 0 0 2 0.004 0.006 S/m FIG. 4.-Relationship between the crack opening forcefand the distance of separation of the ends of the steel beams for a reinforced and unreinforced sheet metal specimen.In all cases the work of fracture was calculated by dividing the work done in propagating the crack by the cross-sectional area of that part of the specimen through which the crack had propagated. The alternative argument that the work of fracture should relate to the two new surfaces created (-twice the cross-sectional area) was not followed since for the reinforced specimens complete fracture does not occur- the crack faces still remaining bridged by the reinforcing fibres. Both the crack-opening force, and the work done in propagating the crack a given distance, are increased substantially for the reinforced system compared with the unreinforced system. The work of fracture of the unreinforced material is116 CRACK PROPAGATION constant but from fig.4 the work of fracture of the reinforced system increases from an average value of 8.6 x lo7 ergs cm-2 (as the matrix crack length increases from 0.12 to 0.243 m) to an average value of 1.55 x lo8 ergs cm-2 (as the crack increases from 0.243 to 0.314 m). The energy absorbed by the bridging fibres is given by PnZ2, where P is the pull-out load on an individual fibre, n the number of fibres per unit length of crack, I the crack length and 2 the average increase in distance of separation of the steel beams from their initial position. In other words, the work which has to be done against the bridging fibres in propagating a matrix crack from length zero to length I is, pro- portional to the area between the crack faces.For a semi-infinite reinforced plate containing a matrix edge crack of length I, extending in a direction perpendicular to the edge of the plate, the width of the mouth of the crack, (if the matrix is allowed to relax completely), will be given approximately by 281, where I is the crack length and E the general tensile strain in the plate prior to the propagation of the crack. The values of works of fracture obtained by use of the apparatus described above can therefore be compared with those expected to occur in a real system providing the distance of separation of the ends of the steel beams is comparable with the maximum expected width of a matrix crack. From fig. 4 the separation of the ends of the beams, when the crack length is 0.314 m, is 5.5 x m which would correspond to an elastic strain in the matrix of a composite of about 7.0 x A working strain of this magnitude is comparable with values which would be expected in high strength structural materials and therefore the data shown in fig.4 can be taken to give a reasonable guide to the values of works of fracture which should be attainable for high-duty composites reinforced with duplex fibres. In these systems the work of fracture will increase with increasing crack length over the values obtained from the data in fig. 4. Also, the volume fraction of reinforcing wires in the sample to which fig. 4 refers represents only about 0.09 of the cross-section of the " composite " and the steel reinforcing wires themselves were stressed at less than 0.3 of their breaking stress. For cracks of appreciable size therefore it seems that works of fracture values in excess of lo9 ergs cm-2 should be obtainable in composites of this type containing higher volume fractions of reinforcing fibres each operating at appreciably higher pull out stress levels. Even higher specific works of fracture are, in principle, obtainable if the reinforcing elements consist of strong lightweight materials such as resin bonded glass or carbon fibres. Estimates of works of fracture based on this model assume that the energy would be absorbed entirely by frictional losses. However, where the fibres are of appreciable length some of the energy will be absorbed as increased strain energy in the bridging fibres. For maximum effective- ness with either mechanism it is clearly necessary for the reinforcing fibres to be appreciably stiffer than the matrix in order that the maximum amount of energy can be absorbed for a given extent of elastic relaxation in the matrix. The work described forms part of a programme of research on advanced composite materials which is supported by the Science Research Council and the Wolfson Institute of Interfacial Technology. A. Kelly, Proc. Roy. SOC. A, 1970, 319, 95. J. G. Morley, Composites, 1971, 2, 80. J. G. Morley, The Properties of Fibre Composites (Conf. Proc. National Physical Laboratory, 4th November 1971) I.P.C. Science and Technology Press Limited, I.P.C. House, 32 High Street, Guildford, Surrey, 1971, pp. 33-35. J. J. Gilman, J. Appl. Phys., 1960, 31, 2208. D. A. Shockley and G. W. Groves, J. Amer. Ceramic Soc., 1968, 51, 299.

ISSN:0370-9302    

DOI:10.1039/S19720200109

出版商:RSC    

年代:1972

数据来源: RSC

摘要:

Some Problems of Design in Fibre Reinforced Materials BY W. D. BIGGS Engineering Department, University of Cambridge Received 12 June, 1972 Designing in composites involves a number of problems-the more important are associated with the anisotropy of the material and with the problems of joining. Some areas needing study are reviewed. The conceptual phase of design is the most critical and it relies heavily upon a combination of experience, judgement and creativity. Most designs involve some combination of analysis and synthesis and depend upon the selection of the material as a necessary, early stage. The number of factors involved in materials selection is, in theory, very considerable but, it is probably fair to say that, providing there is no single overriding need (such as, for instance, corrosion resistance), four factors dominate in practice. These are : (a) Economics-this is generally the most important though the relative importance of prime cost, fabrication cost and maintenance cost are not always considered fully.(b) Mathematical convenience. The designer will always tend to select a material which permits him to utilize the basic simplifying assumptions of linear elasticity and isotropy (especially if these carry with them the advantage of ductility). On the whole, a designer fends to stick ta the material which has served well enough in the past and to treat material variations as special cases. This has the merit of requiring no deep thought and only comes unstuck when situations arise which are outside of his experience or his competence.(d) Ease of joining. Probably the largest group of structural designs are those which are conceived, analyzed and built in terms of a set of discrete simple members which must be joined together. This again tends to reinforce the importance of (b)- the materials tend to be chosen with familiar joining processes in mind. Where the above approach has failed it has been, as often as not, ascribable to one, or both, of two factors. Either the loading conditions and/or the response of the material were not readily covered by the conventional assumptions, or else there were variations from the original design arising from manufacturing or fabricating variables and which could not be considered at the design stage. It is clear that none of the above four factors apply to fibre reinforced materials at the present time and it is the purpose of this paper to examine some of the needs.Specific problems will not be detailed, though a few which are in urgent need of solution are listed in an appendix. (c) Familiarity. This follows from (b). ASSESSMENT OF MATERIAL The most important factor for the engineer is the realization that, in using fibre reinforced materials, he now has the problem (and the responsibility) of designing the material as well as the configuration. Thus, the stepwise process of analysis and synthesis contains one extra correlation-that of relating, at each step in the process, the integration between the structure and the material. 117118 DESIGN IN FIBRE REINFORCED MATERIALS This leads to the necessity for parametric design studies in order that the interaction between material and configuration can be separated and considered.The use of the “ structure loading index ” lm3 forms a convenient starting point but the methods have, so far, been limited ta relatively simple cases and have been used, mostly, for research purposes. Since this type of approach must involve the material properties it must, necessarily, be capable of taking into account the inevitable variations which accrue from the raw material itself and from the process of manufacture. A vital stage here would be the attempt to combine in some form of “lumped” index a measure of the statistical variation likely to be expected.l* The only other approach would seem to involve prohibitively expensive quality control methods.A further problem is that the assumptions of linear elasticity and of isotropy do not, necessarily apply. These need not, in principle, be overwhelming problems given access to adequate computational facilities 5 * 6-though, even with these, the problems of handling analyses on a material which may in the extreme case require the manipulation of 8 1 elastic constants are formidable enough. But, mathematical inconvenience apart, the anisotrapy associated with fibre systems has a deeper consequence. Since anisotropy is inseparable from maximiza- tion of mechanical properties-and, indeed as the material becomes more and more isotropic it becomes less and less attractive-a fully isotropic carbon fibre composite is, weight for weight, less attractive to the designer than titanium or even a heat-treated aluminium alloy.Thus, the designer’s emphasis must, necessarily be towards the maximization of anisotropy-and from this follows anisotropy of shape-long slender members possessing the maximum strength or stiffness in one preferred direction are the natural consequence. This must lead to a perpetuation of what the author calls “ Meccano ” engineering-with the problem of jointing as an overriding consideration and returns us (albeit somewhat circuitously) to the question of fam- iliarity. BEHAVIOUR OF STRUCTURAL ELEMENTS We have noted already that the conventional processes of design involve conven- tional assumptions about the behaviour of the material-any divergencies being treated as special cases to be assessed, in relation to the conventionally determined properties, by making simple modifications to the established rules.The fundamental structural members are beams, columns and membranes. We may separate these into two groups, those which carry a stress in the thickness direction, and those which do not. Provided that attention is limited to simple beams and trusses (which may be considered as essentially two-dimensional), the conventional techniques of elastic analysis are probably adequate providing that suitable attention is paid to aniso- t r ~ p y . ~ - ~ Some work has been done in this area and, to a large extent, the results have been validated by experimental work. The problems of elastic and post-elastic buckling in compression have received a fair amount of attenti~n,~. lo but cannot be satisfactorily handled until a coherent theory of the micromechanics of composites in compression is further developed.This requires among other things, that materials scientists stop thinking about properties defined only by the conventional tensile test. Prediction of the failure criterion is especially important in membrane structures. At present, failure criteria can only be predicted in those structures where the thickness is so small that “ through ” stresses can be neglected, i.e., in conditions approximating to plane stress. One of the major problems here is that of transmitting “ in-plane ” shear-thus in a simple column the coupling between axial, transverse and twistingW . D. BIGGS 119 modes is difficult and conventional theory is of little help.Plate elements can be analyzed for in-plane deflections using the assumption that each ply behaves inde- pendently-but the inhomogeneities which exist in the thickness direction of a thick plate make the assumptions of linearity doubtful, and finite element techniques are needed to determine the magnitude of the coupling between membrane and plate behaviour. In most cases the published work suggests that the response of the material is non-linear-even in those cases where the conventional stress-strain relations have been shown (or have been assumed) to be linear. Since these types of non-linear behaviour are likely to be greatly influenced by initial imperfections the effect of these needs to be critically studied.The inherent discontinuity of structure in the thickness direction means that one of the basic techniques of elastic analysis must be reconsidered. It is generally assumed that, if the deflection of a plate has been determined, then the strain over the plan and over the thickness are both known. But, even in plane stress situations with no bending, the stress distribution depends upon the first derivatives of the displacement variables, or upon the second derivatives of the stress function. Thus, if the distribution of the displacements is only known approximately, the knowledge of the stress distribution will be even less accurate. Experimental validation of stress and strain functions for fibre reinforced materials are badly needed. Lastly (though by no means exhaustively) we have the problems which arise when the principal directions of elasticity are not collinear with the structural axes.These have barely been touched at the moment. One of the effects would be, for instance, to distort a normal buckling mode into a skew mode with consequent variation also of the vibrational characteristics. Simple problems of this sort have been tackled- but the computational difficulties render the solutions totally impractical for design applications. We mention it here as a particular-almost a classic example of a deflection mode which never arises in a conventional homogeneous material. But if one had to single out a particular effort demanding attention one would, unhesitatingly, select the analysis of joints, fixings and stress transmission.Even in isotropic homogeneous materials the average engineering approach can only be described as one of ignorance-in some cases indeed it is not going too far to call it criminally casual. The approach arises -almost entirely-from the " metal men- tality " which assumes that there is always an adequate reserve of ductility to take care of infelicities in design. This assumption is manifestly untrue of composite materials and vastly increased effort into joints and fixings is needed if composites are to be used without a very large weight or economic penalty. It is true that the literature contains many examples of specific joint analysis 11* 12-but most of these are based upon finite element methods and make use of modifications to standard elastic solutions.But they remain just as empirical as the methods used for analyzing joints in metal structures and, with less experimental justification, their validity is doubtful. LIMITING BEHAVIOUR In view of the principal topic of the present Discussion, it is, perhaps, appropriate to say a few words about failure. Most of the " standard " failure criteria (maximum shear stress, maximum shear strain energy etc.) only work within limits when applied to fibre composites and, moreover, they are only true for particular cases. Again, we must suggest that, all too often, the search for suitable failure criteria has been borrowed too facilely from design processes in conventional materials and it is neces- sary that the whole topic be re-assessed with reference to fibre composites.But it is in the field of those types of failures which have their origins in the micro-120 DESIGN I N FIBRE REINFORCED MATERIALS mechanical behaviour that a more serious assessment needs to be made. We refer, especially to the fields of fatigue, creep and brittle fracture, and it is the authors’ opinion that it is time that a partial moratorium should be called on work which appears, at the moment, to offer diminishing returns when examined in relation to the macroscopic behaviour on which the designer must, perforce, base his calculations. As an example, fracture mechanics which, despite some 20 years of work, remains an interesting postmortem technique of little use in the conceptual stage of design. This is not to say that all work should cease-but rather that the macroscopic behav- iour is deserving of more immediate attention.And, to the argument that the macro- scopic behaviour can, eventually, only be understood by understanding the micro- scopic behaviour, I would reply that several decades of metallurgical research have not yet put us in that happy position-even for metals. QUALITY CONTROL Throughout the above we have stressed the need for a full and complete under- standing of the variations which may normally be expected to occur-especially as fibre reinforced materials possess certain characteristics which make them more susceptible than metals to the presence of chance defects. The author’s experience in the application of new (or, at least unfamiliar) materials leads to the suggestion that consistent properties are always more important than maximum values, and while it is always desirable to strive for the best attainable, it is often little more than an academic exercise.If design predictions are to be made with some certainty that the final product will justify them, it is essential that the material display predictable properties. These need not, in fact be the “ best ” properties- indeed, any design based upon the best properties will almost certainly fail on a basis of statistical variation. The important statement above is that theJinaZproduct must be right. Here again we come up against a divergence from established “ metal ” practice. In a fibre reinforced material the method of fabrication is inherent to the product-one cannot cast a large block of composite and machine away the bad bits.Thus, it is probably less important to check the quality of the raw material than it is to check and control the fabrication procedure and to test the final product, for there is no reason for assuming that the results of tests on the starting elements will necessarily be repre- sentative of the finished artefact. HIGH PERFORMANCE COMPOSITES Most of the previous comments are general in the sense that they do not refer specifically to any particular fibre/matrix combination. But, in view of the present interest, it is pertinent here to make a few comments about high performance fibres- especially carbon-and some of the particular problems involved. One of the more important problems concerns the fracture toughness.By comparison with metals, this is low and there is little evidence, from theoretical considerations, that it can be raised to a value which is competitive with metals with- out, at the same time, incurring quite severe penalties in terms of strength, stiffness or weight. The same is true of the fatigue properties and designers must learn to distinguish between (and allow for) that fatigue which arises from crazing of the resin and that which arises as a consequence of low fracture toughness. In practice, it seems probable that both mechanisms will exert an influence in an unspecified (and probably unspecifiable) ratio. Carbon has a negligible coefficient of thermal expansion when compared to, say,W. D. BIGGS 121 glass. Any thermal stresses generated during fatigue loading may, therefore, be expected to affect the resin matrix so that the fatigue behaviour of such composites may well be significantly affected by such factors as rate of loading, dwell time at peak load, resin cure etc.But the low expansion also means that any redundant structure must be designed so that it will accommodate thermal strains and the stresses which result. The behaviour of structures in which CFRP is bonded to metal does not appear to have been studied to any great extent-one major problem here is likely to be " shakedown " as a result of plastic deformation in the metal. Some of these difficulties might be ameliorated (though not necessarily overcome) by the use of metal matrices. Apart from manufacturing problems there is the probable difficulty of compatibility between fibre and matrix over very long periods of loading at temperatures below those at which creep becomes a dominating mechan- ism.The effect of long-term incompatibility is difficult to predict and there appears, at the moment, no experimental reason for believing that the normal extrapolation processes based upon standard rate laws will apply. Since, by its nature, such work must be very long term it is desirable that much effort be devoted to it now even though the fibres used may not, in the event, prove to be the final solution. But to wait until a commercially viable material is available in quantity before starting such work will ensure that the test data are about ten years too late to be useful to the designer.SUMMARY AND CONCLUSIONS The author firmly believes that there is a grave danger that too much academic curiosity can easily blind us to the needs of the user-in this case the designer. All too often, in the past, a " new " material has not been so widely different from the familiar one and, as a result, the designer has not found it necessary to state his needs too explicitly and he has been able to modify existing practice. This approach came badly unstuck when polymers entered the market and the designers approach was purely substitutive, ending, as often as not, in disaster. We are now tackling a new material-and the designer is being compelled to rethink his whole design philosophy. In so doing he must state his needs as precisely as possible and express them in terms of the problem which he must solve. But there is little point in his doing so until, or unless, materials scientists start to take some note of the needs and supply the information which he wants and not the information which they think he ought to need.And, in my mind,.there is no doubt that investigations of the macroscopic continuum behaviour are much more urgently needed than studies of the micromechanics, unless the latter can manifestly contribute directly to making predictive assumptions about continuum behaviour. In the context of the present Discussion such a remark may seem an impertinence- and, indeed, it is. But it is hoped that the paper as a whole may provide a background against which the more detailed studies may be compared.Hopefully, also, it will help to stress the need for coherent interdisciplinary groups which are able, and qualified, to study the conceptual phase of design in relation to the design of the material and the mode of fabrication. APPENDIX Several problems have been mentioned in the paper-the problem ofjoining is undoubtedly outstanding. But, among more general problems which should be tackled on an inter- disciplinary basis, we may list the following. (a) A critical re-assessment of the assumptions of elastic stress analysis and of any modifications which may be needed to apply these assumptions to composites. It will be122 DESIGN I N FIBRE REINFORCED MATERIALS necessary to note here that a deficiency in the theory must not be assumed to stem from a deficiency in the material-the theory must be examined in the light of the material available.(b) A critical study of the location of points of large deflection. Composites are discon- tinuous in the thickness direction so that the assumption that the largest deflection will occur at the extreme fibre is not necessarily valid. (c) The consequences of non-linearity-especially at points of critically large deflection. ( d ) A critical evaluation of the failure laws for composites. Theories originating in the micro-mechanical behaviour are of some help here but cannot be usefully exploited unless they are capable of extension to describe the continuum behaviour. Problems in pulsating and dynamic loading are at least as important as those in static loading. (e) Since attempts to confirm theoretical predictions are dependent upon the reliability of the product the questions of quality control and non-destructive testing must form an integral part of the design exercise. N. F. Dow et al., Design Criteria and Concepts for Fibrous Composite Structures, NASA Report N. F. Dow et al., Studies of Mechanics of Filamentary Composites NASA Report CR-492 (June, 1966). G. H . Hayward, Conceptual Design of Boron Composite Structures AFML Composites Meeting, Denver (September, 1967). N . F. Dow et al., Evaluation of the Potential of Advanced Composite Materials for Aircraft Structures AFML-TR-66-144 (May, 1966). R. Tetlow, Design Charts for Carbon Fibre Composites. W. H . Wittrik, Aero. Quart., 1952, 4, 54. AFML-TR-66-3 13 (September, 1966). S. G. Lekhnitski, Theory of Elasticity of an Anisotropic Elastic Body (Holden-Day, San Fran- cisco, 1963). lo J. Singer, Buckling of Unstifened Orthotropic Conical Shells 7th Aero. Congress, Paris (June, 1965). K. R. Berg, Mechanics of Composite Materials, ed. Wendt (Pergamon Press, London, 1970). l2 F. J . Filippi et al., Application of. . . Composite Materials to Helicopter Rotor Blades AFML- NASW-1377 July 1967. Cranfield Memo No. 9. ' C. W. Rogers, Application of Advanced Fibrous Reinforced Composites to Airframe Structures ' T. L. Chao, Properties Report No. 3, Case Institute of Technology, (February, 1967). TR-66-209 (August, 1966).

ISSN:0370-9302    

DOI:10.1039/S19720200117

出版商:RSC    

年代:1972

数据来源: RSC

摘要:

GENERAL DISCUSSION Prof. R. Sh. Mikhail (Ain Shams University, Egypt and University of Salford) said : I congratulate Proctor on his interesting review but would mention a significant exception to the general phenomenon discussed in his paper, namely, that, in general, interaction between the fibre reinforcement and the matrix leads to loss of strength. In our work on glass-fibre-reinforced cement pastes, the idea was to prepare a glass fibre which should be alkali resistant and which should withstand corrosion at pH values around or higher than 12, which was brought about by saturated lime solution resulting from the hydration of Portland Cement. When alkali-resistant glass was used, the engineering properties of the composite did not greatly improve. Trials are now being carried out in which the fibre core should be alkali resistant, whereas the fibre surface should be reactive, in order to allow for a chemical reaction to take place, and give products, which are in essence very similar to the hydration products of Portland Cement itself, and which would establish a stronger bond with the glass surface.This example presents a situation where chemical interaction between the reinforcement and the matrix is desirable, provided that it terminates at a certain depth in the fibre. Dr. B. A. Proctor (Pilkington Bros.) said: Mikhail raises an interesting point but one which does not invalidate the general remarks made in my paper. Almost any interaction between fibre and matrix will weaken a high-strength reinforcement but the effect may not be noticed initially if the fibres have already been damaged and weakened prior to, or during, incorporation in the matrix.Also, the effective utilisation of fibre strength in a composite depends on a degree of fibre/matrix bonding and this may be increased by fibre/matrix reaction. Hence controlled amounts of fibre/matrix interaction may lead to an optimum in composite strengths. This interaction must be carefully limited if the effects on fibre strength are not to be noticed and if composite embrittlement due to too great a bond strength are to be avoided as described in my paper. Prof. W. C. Wake (City University) said: Mention by Proctor of the influence of the additives, in particular the size, applied to glass filaments, recalls work with which I was associated a few years ag0.l A single glass filament taken from a commercially-treated roving was drawn through a water drop held on the tip of a hypodermic syringe needle and the contour of the drop photographed at random points along the fibre. A histogram of the frequency of given contact angles against the contact angle showed a bi-modal distribution with one mode approximately at the contact angle for polyvinylacetate (PVA), the main constituent of the size, and the other almost 90, typical for a siloxane resin.The two materials had been applied for the same emulsion, and substantial areas of glass had a PVA rather than a siloxane cover. It is doubtful if this dissolves in styrene during application of the resin and, if it does, it leaves bare glass.Water absorption is, in any case, encouraged at these points and in all the discussion about realizing maximum strength there has not yet been any comment about the fact that in use, little of the maximum dry strength can be used for design purposes. It is a lamentable fact that, in water, glass-reinforced W. C. Wake, 23rd Congr. Pure Appl. Chem., (Boston, 1971). I23124 GENERAL DISCUSSION plastics rapidly drop strength until the long-term load bearing efficiency is but a frac- tion of the manufactured strength. Prof. D. D. Eley (Nottingham University) said : The logarithmic dependence of fracture time on applied stress for silica depicted in fig. 8 of Proctor's paper points to an activated process (" flow-process ") underlying rupture, as in organic polymers.Temperature variation experiments should lead to a value of the activation energy which may throw light on the nature of the " flow-unit "; is there any evidence on this? A general point is whether the structure of natural composites, such as wood and bone, give any clues in developing synthetic materials. Dr. M. J. Owen (Nottingham University) said: The work of McGarry and Mandell seems to imply that the use of toughened matrices, e.g., those involving rubber additives, is unlikely to have a significant effect on the toughness of composites. McGarry has reported that toughened epoxy resins led to improved fatigue per- formance in the laminate, although I believe his tests were confined to about 100 cycles. It has been our experience with polyester resins that the use of more flexible resins can suppress resin cracking in the tensile test and can delay resin cracking under fatigue conditions for something like 1000 cycles, but thereafter resin cracking develops in the normal way, and there is then no improvement in fatigue life for lives of the order of lo6 cyc1es.l Dr.M. J. Owen and P. T. Bishop (Nottingham University) said: In the last section of their paper, McGarry and Mandell refer to the K,, value for randomly oriented chopped fibre reinforcements. They have found that a reasonably constant KIc B- 90" X - 50 75 100 125 150 specimen width/mm FIG. 1 .-Variation of uncorrected KC value with crack length. M. J. Owen and R. G. Rose, paper presented to 7th Int. Reinf. Plastics Conf. Brit. Plastics Fedn, 1970.GENERAL DISCUSSION 125 10 T E z 8 - E 1 6 - u c - 2 - value could be obtained if an adjustment was made for the size of the damage zone.At the University of Nottingham we have used a similar technique to analyze fracture toughness tests on chopped strand mat, unidirectional fabric and balanced weave fabric reinforced polyester resin, hereafter designated A, B - 0" (major principal direction), B-90" (minor principal direction) and C . The effect of crack length was investigated by testing geometrically similar double-edge-notch specimens 75, 100 and 150 mm wide in tension to failure. To obtain K,, values from the results the load - 'Average uncorrected values 01 1 I I I 50 75 100 125 I50 specimen width/mm FIG. 2.-The effect of the equivalent yield zone correction factor on Kc for material A.at which crack propagation occurred was required. This critical load was deter- mined from a plot of load against displacement obtained from a clip gauge which recorded the opening of the faces of one of the notches. Several tests had shown that crack propagation occurred when the slope of this load deflection curve first became zero. The variation of KIc with crack length is shown in fig. 1. The authors' values of KIc for materials similar to A and C were slightly lower, viz., 9.27 MN m-3 (max. value) and 18.5 MN m-+ respectively. Apart from material B-90", there is a distinct increase in Kc with crack length. To determine a value of Kc which is a measure of the fracture toughness of the material an approach similar to the plastic zone correction factor proposed by Irwin was used.This consisted of adding the quantity r,, given by eqn (1) to the crack length : where gEy = equivalent yield stress. The equivalent yield stress was determined by trial and was taken as the value which produced a KIc value independent of crack length. This is shown in fig. 2. This approach produced constant values of K,, for both material A and C but not for material B-0". This was not surprising since126 GENERAL DISCUSSION for this material side grooves had to be machined in the specimen to prevent shear cracking parallel to the fibres. Dr. M. J. Owen (Nottingham University) said: In relation to this paper by Biggs, the glass-reinforced plastics industry already exists and has been expanding its output steadily year by year for more than 20 years.This apparently steady growth is to be expected because design and development costs far outweigh the cost of new materials themselves. These costs normally have to be met from individual products. Thus, if new products are to stand a chance in the market place, development costs have to be kept to a minimum and progress takes place in small incremental steps from one design to the next. New materials will remain as scientific curiosities unless they possess favourable combinations of properties which permit either completely new developments, substantial improvements in performance, or significant reductions in costs. There are many examples such as fume stacks, chemical process vessels, and the Ministry of Defence's mine-sweeper project, where advantage has been taken of favourable combinations of properties.One area of considerable importance which Biggs has not mentioned is that of cost reduction. The majority of GRP constructions involve labour-intensive processes. It is likely that future developments will take place through the introduction of mechanical, and preferably fully automated, processes based on injection moulding or spray-up techniques. Inevitably, this means the use of short fibres rather than highly oriented continuous fibre systems. In this connection, some test results were obtained a number of years ago when it was found that a chopped strand mat pressure vessel containing real features such as branch connections, had superior performance to a filament-wound vessel because the strength was governed by fibre-orientation difficulties at the branch connection.' Prof. D. D. Eley (Nottingham University) said: With regard to the paper by Biggs, it should be possible to change a polymer matrix from a viscoselastic to a dilatant or shear-hardening medium by incorporation of a sufficient concentration of spherical particles, if this would offer any advantage. Prof. W. C. Wake (City University) said: The compression failure mode and its prevention as illustrated by Biggs recalls some plant structures. I was recently looking in my garden at Bocconia Cordata (syn. Macleaya) and considering the stems which, unlike bamboo, have no external nodes but only internal septa. Would it be possible to imitate this structure in GPR? The septa which in Bocconia seem to delay Euler collapse are not fibrous like the hollow stem and, indeed, do not appear particularly substantial. My second point is that I should not have expected multiple bending in wave form under compression but only Eulerian behaviour. I. W. Reid, Proc. Inst. Med. Eng., 1964-5, 179, 313, 89.

ISSN:0370-9302    

DOI:10.1039/S19720200123

出版商:RSC    

年代:1972

数据来源: RSC

摘要:

Morphology and Mechanical Properties of a Rubber Reinforced Composite BY G. ALLEN,* D. J. BLUNDELL,t M. J. BOWDEN," F. G. HUTCHINSON 7 AND G. M. JEFFST Received 26th June, 1972 The properties of interstitial polymers composed of glassy domains dispersed in a continuous rubber matrix are described. The effects of the relative composition of glass and rubber and the domain size of the glassy polymers on the bulk properties is described in detail for a composite in which polymethyl methacrylate is the glassy phase and a polyurethane is the elastomeric component. The observed moduli are always higher than values calculated on the basis of theories which assume no interfacial interaction between the two phases. Furthermore, the loss peak for the rubber process is least well defined when the surface area of the glassy domains is largest.These effects are inter- preted as evidence for intermolecular interactions between the two components in an interfacial layer. Inclusion of a small proportion of a rubber phase considerably improves the impact resistance of a glassy polymer at the expense of only a relatively small reduction in modulus. The most common commercial materials of this type, e.g., high impact polystyrene (HIPS) and acrylonitrile-butadiene-styrene (ABS) incorporate an elasto- mer as a component part of spherical inclusions dispersed in a continuous matrix of the glassy polymer. The rubber-modified composites described in this paper have similar impact-properties, but the morphology is essentially reversed, i.e., domains of glassy polymer dispersed in a continuous rubber phase.These composites are made by a casting technique which involves the polymerization of a glassy vinyl monomer inside a swollen, cross-linked elastomer. This particular system is very flexible because not only the volume composition of glass and elastomer can be controlled, but also the size of the plastic domains can be readily varied, and the chemical nature of the two components can also be changed. Chemical and physical properties of these two-phase composites have been the subject of an extensive fundamental study.l-s In this paper we report that there is indirect evidence for the existence of considerable molecular interaction at the phase boundaries and that this interaction has an effect on the overall modulus of the mater- ial.The paper describes morphological and mechanical relaxation studies together with a comparison of the observed modulus and moduli predicted by various theor- etical models. Most of the work is concerned with composites in which the glassy phase is polymethyl methacrylate (PMMA) and the elastomer is a polyetherurethane (PU) made by reacting an organic di-isocyanate with a mixture of polypropylene oxide triols and diols. Full details of the preparation of these composites are published elsewhere. EXPERIMENTAL MECHANICAL PROPERTIES Measurements of the tensile strength were performed with a Hounsfield Tensometer (type E) using a strain rate of 5 mm/min.6 The charpy notched impact strength was meas- * Chemistry Department, University of Manchester, Manchester M13 9PL.ICI Limited, Corporate Laboratory-Runcorn, The Heath, Runcorn, Cheshire. 127128 RUBBER REINFORCED COMPOSITES ured with a Hounsfield Plastic Impact Tester, using test samples which were notched by a cutter of 0.25 mm radius. Dynamic mechanical measurements were made using a torsion pendulum in which the operating frequency was maintained near to 1 Hz. Using this equipment the shear modulus G' and mechanical loss tangent, tan 6, weie measured over a temperature range from - 140 to + 150°C. For a given run the temperature was maintained constant to kO.25"C. The sample dimensions were 80 x 7 x 3 mm. DIELECTRIC MEASUREMENTS Dielectric loss measurements were made on samples in the form of discs 50 mm in dia- meter and of thickness 3 mm, which were held in a standard 3-terminal cell.The measuring equipment was a General Radio model 161 5-A transformer ratio-arm bridge which operated in the frequency range 60 Hz-10 kHz. The temperature of the cell was maintained constant to kO.25"C during the run and measurements were performed in the temperature range - 196 to + 180°C. ELECTRON MICROSCOPY Thin sections of the composite materials were cut at room temperature with a LKB Ultra Microtome. In fig. 2 the structure was enhanced by shadowing the sections with vacuum- deposited chromium. The staining in fig. 3 was achieved using the technique of Kato.' RESULTS AND DISCUSSION GENERAL PROPERTIES The cast sheets have a clear appearance and when the preparative conditions are optimized the transmittance of the products is very close to that of PMMA acrylic sheets (i.e., 90 % at 600 nm through 3 mm sheets).The mechanical properties such 0 10 2 0 3 0 4 0 5 0 6 0 composition FIG. 1.-Effect of wt. fraction of PU elastomer on (0) shear modulus, (0) tensile strength and (e) notched impact strength (N.I.S.).FIG. 2.-Electron micrograph of a 80 % PMMA/20 ”/, PU composite section shadowed with chromium. FIG. 3.-Electron micrograph of 80 % PMMAj20 % P. Butadiene urethane composite stained with [To face page 128 osmium tetroxide.ALLEN, BLUNDELL, BOWDEN, HUTCHINSON AND JEFFS 129 as modulus, ultimate strength and toughness depend primarily on the overall com- position. There are, however, several subsidiary effects on properties which depend on variables such as (i) cross-link density of the PU, (ii) the mol ratio of isocyanate to hydroxy groups in the PU formulation, (iii) the time between the gelation of the PU and the initiation of polymerization of the polyvinyl phase.Fig. 1 shows the dependence of shear modulus, tensile strength and impact strength of PMMA/PU composites on overall composition. The modulus and tensile strengths fall from the limiting value for PMMA as the amount of PU in the composite is increased. At compositions below 10 % w/w of PU the composites display brittle tensile properties and break at small elongations. Compositions containing more than 10 % w/w PU show ductile behaviour. The difference in impact strength with composition reflects this behaviour and it will be noted from fig. 1 that the toughness increases significantly when the proportion of PU rises above 10 % w/w.MORPHOLOGY The appearance of two distinct maxima in the tan 6, curves (corresponding to the glass-rubber transition region of each component) clearly indicates the presence of two distinct phases in the composite. The actual morphology can then be deduced from the electron micrographs shown in fig. 2, which were obtained for a material containing 80 % PMMA. The micrograph indicates that to a first approximation the PMMA phase has separated into circular regions and that the PU phase is in fact the matrix material. Since this micrograph is only representative of the structure in a random two-dimensional plane, it is, in principle, possible that the circular regions represent sections through a three-dimensional structure comprising either an open-cell or a closed-cell PMMA phase structure.There are two main reasons for believing that the closed-cell model is more realistic. First, even where the PMMA phase structure is large (- lpm) and hence much larger than the thickness of the electron microscope sections ( - 0.1 pm), the PMMA domains still appear to be unconnected. Secondly, except in rare instances, the outline of the PMMA domains always appear as near-circular and never display the changes of curvature that would be expected with the joining of neighbouring cells. This is particularly well illus- trated in fig. 3 which shows the micrograph of a special composite in which the PU phase was made from a polybutadiene diol so that the elastomer phase could then be directly stained with osmium tetroxide. Thus the general conclusion is that the composite consists of near spherical domains of PMMA closely packed and embedded in a continuous matrix of PU.There does not appear to be geometric continuity to the glassy PMMA phase between domains, and, to a first approximation, any interaction between the domains can only occur at the molecular level. When all other parameters are held constant, there is a systematic variation in the size of the PMMA domains with composition, which ranges from a diameter of > 1 pm for 90 % PMMA to < 50 nm for 50 % PMMA. The smaller domains are difficult to detect and usually only appear in the electron micrographs as small shaded bumps. Variation in domain size at constant composition can be achieved by allowing the PMMA phase to polymerize at different points during the formation of the PU network.Careful investigation into the two chemical processes has established that the average size to which domains grow depends on the relative completion of the PU 2-E130 RUBBER REINFORCED COMPOSITES network at the time when vinyl polymerization is initiated. Irrespective of whether the network formation is controlled either by chemical or physical means, the more complete the network at a given composition, the smaller are the final PMMA domains. One convenient method of varying the gel cure without affecting the chem- istry or the overall composition is to vary the period between the physical gel point of the PU and the onset of the vinyl polymerization so that polymerization occurs after varying stages of completion of the urethane reaction.An illustration of the variation of domain sizes achieved in this way are given in table 1. TABLE VARIATION OF DOMAIN SIZE WITH POST GELATION PERIOD FOR 80 % PMMA/ 20 % Pu COMPOSITES post gelation period/h domain sizelnm 0 180 12 88 24 70 48 65 Composites were also made using polyacrylonitrile in place of PMMA and the same domain morphology was observed. The domain size of 80 % PAN/20 % PU is generally.slightly smaller than the corresponding PMMA system, and the size of the domains is less sensitive to changes in the gel structure. MECHANICAL AND DIELECTRIC RELAXATION The mechanical loss properties of PMMA/PU composites are illustrated in fig. 4 where tan 6, and G are plotted as functions of temperature for composites containing progressively increasing amounts of the PU species.Also shown are the correspond- ing curves for PMMA.8 The principal feature to note in fig. 4 is the PU-a (T,) process which, for the 80 %PMMA/20 %PU sample occurs as a prominent peak in tan 6, and a discreet fall in G‘ near -50°C. At high temperatures the increase in tan 6, and the fall in G’ denote the onset of the glass-rubber transition process in PMMA. The PU-a loss peak and fall in G’ increase in magnitude as the amount of PU in the composites increases and, in addition, the process moves to higher temper- atures. Similar effects are evident in PAN/PU composites. These measurements reflect the morphological observations described above in that special features distinc- tive of the separate species are observed, i.e., the polyvinyl and the PU exist in largely separate phases.However, the size of the polyvinyl domains decreases as the amount of PU in the composites is increased and, as described above, the PU-a process moves to higher temperatures. These features indicate some interaction between the phases since the PU-a process is increasingly hindered as the size of the domains decreases and hence the interfacial area between the phases increases. Further information can be gained from a study of the dielectric relaxation phen- omena of the composites. Fig. 5 shows the variation of tan 6, (measured at 1 kHz and 10 kHz) with temperature for 40 % PMMA/GO % PU and 40 % PAN/60 % PU composites.The peaks in tan 6, for the PMMA/PU composites are at higher temperatures than the corresponding peaks for PAN/PU composites. In addition the activation energy AH for the process increases from 118 kJ/mol in PAN/PU to 151 kJ/mol in PMMA/PU composite. From mechanical loss measurements it is observed that, for composites containing equal amounts of the PU phase, the lossALLEN, BLUNDELL, BOWDEN, HUTCHINSON AND JEFFS - 1 0 I N I E 2; 0.1 9 u 0.01 150 -100 - 5 0 0 5 0 I00 d,, I 5 0 -0 - 4 - 131 T/"C FIG. 4.-Temperature dependence of shear modulus G' and loss tangent, tan a,, for (0) 100 % PMMA, 80/20 (0), 60/40 (a), and 40/60 (0). peaks are rather better resolved in PAN/PU than in PMMA/PU composites. These composites have similar morphologies and the major difference between PMMA and PAN on a molecular scale lie in the size and polarity of the respective side groups, and the energies of interaction with the polyurethane molecules. The PAN/PU com- posites have the lower temperatures for the PU-a process, the lower activation energy 0 .2 0. I cg 3 Y 0.01 0 . 0 2 - ; 5 0 -100 -50 0 5 0 100 T/"C FIG. 5.-Dielectric loss tan 6~ as a function of temperature for 40 % PMMA/60 % PU (0), 40 % PAN/60 % PU (0) composites.132 RUBBER REINFORCED COMPOSITES and better-resolved loss peaks. By comparison, PMMA/PU composites have higher temperatures for the PU-a process, higher activation energies and less well-resolved loss peaks. An explanation for these observations is that phase separation is more complete in the PAN composite because PAN will be less compatible with the poly- urethane than will be PMMA, but in each case there will be some degree of mixing at the molecular level.Intimate mixing probably occurs at the interface boundary and might extend through the predominantly PU phase, mixing being more extensive in the PMMA composite. Further evidence of the interaction between the polyvinyl and the PU species is found by changing the composite preparation. The effect of the variation of the post-gelation period between the gel point of the PU and the initiation of the polymer- ization of the vinyl is shown in fig. 6. Tan 6, and G' are plotted as functions of i c I N I E z 2 0. 0. ( E (.o d c >.I - 100 -50 0 5 0 TIT FIG. 6.-Shear modulus and loss tangent as functions of temperature for 80 % PMMA/20 % PU composites prepared with post gelation periods of 0 h (O), 12 h (O), 24 h (a) and 48 h ( 0).temperature for a series of composites containing 80 % PMMA/20 % PU and progressively increasing post-gelation periods as listed in table 1. The tan 6, curves show that the PU-a peak at about - 50°C decreases in magnitude and the correspond- ing change in G' decreases as the post-gelation period increases. These results also suggest that there is an increase in the interaction between the PMMA and the PU at the molecular level as the post-gelation time increases. This correlates with theALLEN, BLUNDELL, BOWDEN, HUTCHINSON AND JEFFS 133 fall of domain size as shown in table 1, and the associated increase in interfacial surface area.For a given composition there is always a good correlation between the domain size and the definition, size and temperature of the PU-a process. The observed correlation is consistent with increase in molecular interaction at the phase boundaries as the domain size decreases. These dynamic-mechanical and dielectric studies, whilst supporting the morpho- logical evidence of a two-phase structure, have also indicated that there is some interaction between the phases. This interaction could occur in either or both of the following situations. The molecules of the separate species could interact at the interphase boundaries, or as the result of molecules of the polyvinyl linking one domain with another. MODULUS/COMPOSITION RELATIONSHIPS The morphological studies have established a geometry consisting to a first approximation of spherical polyvinyl domains, embedded in a continuous PU matrix. Several theories have been formulated to predict the composite moduli of such cases.These theories are strictly for two-phase systems in which there are sharp boundaries between phases. They assume perfect adhesion between matrix and inclusions and that the inclusions are uniformly distributed. One of the earliest was formulated by Einstein and modifications of his theory have been proposed by Guth,l0 Smallwood and Mooney.12 Kerner l3 has set up a more sophisticated theory which, when simplified to the case where the shear modulus G2 of the inclusions is much greater than G1 of the matrix, yields the equation for the modulus of the composite where Pi = 2(4- 50i>/5(1- GI), and 41 and 4, are the volume fractions of matrix and filler and o1 is Poissons ratio for the matrix. A corresponding equation can be obtained for the inverted case of low modulus elastomer inclusions in a high modulus matrix.Hashin and Shtrikman l4 have formulated a general theory which allows arbitrary phase geometry and predicts bounds for the composite modulus, when G, % G1 of These bounds, in fact, coincide with the two extreme cases accounted for by Kerner’s theory. Kerner’s theory in particular does not consider any direct interaction between domains in the form of a continuity of the included phase. However, two other theories by Budiansky and Davies l6 have been shown to be appropriate to cases where there is such an interacti0n.l’ In particular, Davies proposed the relationship, and has suggested that it is best suited to morphologies where there is continuity in both phases for all 0 < 42 < 1.134 RUBBER REINFORCED COMPOSITES The results of shear modulus measurements on a series of PMMA/PU composites of variable composition are shown in fig.7. Also plotted are the curves predicted by the above theories. The theories based on the Einstein equation all predict composite moduli which are lower than experimental values. The bounds of Hashin and Shtrikman are too far apart to be of predictive value. Kerner’s theory should be I I I I 0.2 0.4 0.6 0.8 1 .o 4 2 FIG. 7.-Modulus/volume fraction composition curves. Numbers in parenthesis are references. appropriate to hard polyvinyl domains in an elastomer matrix but, in fact, it predicts moduli significantly lower than the observed values.However, at high &, the inverse case of elastomer domains in a PMMA matrix is in fair agreement with experi- ment. The theory which is most successful in predicting the observed modulus is that of Davies in which a continuity in the PMMA is inferred. Similar results are found with PAN/PU composites. The theories have been tested further using model composites consisting of pre- formed spherical PMMA particles (Diakon moulding powders) embedded in a PU matrix. The results are shown in fig. 8. Here the non-interacting theories based on isolated non-continuous inclusions agree fairly well with the observations. The theories which do not consider interactions between phases are adequate therefore to describe the modulus of composites where the filler is in the form of spherical bodies with sharp boundaries and is dispersed in a continuous matrix.But only those theories appropriate to continuity in the polyvinyl phase agree well with the results for the PMMA/PU and PAN/PU systems described in this paper. In fact, Kerner’s theory for hard inclusions, which would appear to be a suitable model for the observed morphology is one of the least successful. The best fit is given by the Davies’ equation which, in terms of the physical model involved, should be among the least appropriate since it implies a continuity in both phases. The situation is resolved by takingALLEN, BLUNDELL, BOWDEN, HUTCHINSON A N D JEFFS 135 account of the interfacial molecular interaction detected in the relaxation studies.The interaction correlates with domain size and could occur through intermixing of the two species at the domain boundaries. Such a diffuse interfacial boundary was not envisaged in the above theories, It is also possible that some interaction could result from molecules of polyvinyl connecting between domains. Since the inter- domain region is narrow (5-20 nm) the intermixing could involve a large proportion of O J ’ I I 1 I I 6’ 0 0. I 0.2 0.3 0.4 0 . 5 FIG. 8.-Shear modulus G as a function of volume fraction +2 of PMMA particles in a PU matrix. (0) PU, (0) Diakon DV400 (< 50 pm), ( 0 ) Diakon MGlOl(270 pm), (M) Diakon MG102 (600 pm). the PU phase. The interaction could be sufficient to reduce the mobility of the PU and bring the local effective modulus closer to that of polyvinyl.Thus, although the domains and matrix may be chemically different, the local modulus between domains could in places be physically equivalent to a continuity of polyvinyl such as is envisaged in the Davies’ equation. This work formed part of the programme of the Manchester University/ICI Joint Laboratory. We thank ICI Limited for support and encouragement. G. Allen, M. J. Bowden, D. J. Blundell, F. G. Hutchinson, G. M. Jeffs and J. Vyvoda, to be published. G. Allen, M. J. Bowden, D. J. Blundell, G. M. Jeffs and J. Vyvoda, to be published. G. Allen, M. J. Bowden, G. Lewis, D. J. Blundell and G. M. Jeffs, to be published. G. Allen, M. J. Bowden, G. Lewis, D. J. Blundell, G. M. Jeffs and J. Vyvoda, to be published. G. Allen, M. J. Bowden, S. M. Todd, D. J. Blundell, G. M. Jeffs and W. E. A. Davies, to be published. K. Kato, Polymer Eng. Sci., 1967, 7 , 38. Solids (Wiley, London, 1967). A. Einstein, Ann. Phys., 1905, 19, 549. ASTM D 636-67T. * N. G. McCrum, B. E. Read and G. Williams, Anelastic and Dielectric Eflects in Polymeric lo E. Guth, J. Appl. Phys., 1945, 16, 20.136 RUBBER REINFORCED COMPOSITES l1 H. M. Smallwood, J. Appl. Phys., 1944, 15, 758. l2 M. Mooney, J. Colloid Sci., 1951, 6, 162. l 3 E. M. Kerner, Proc. Phys. Soc., 1956, 69, 808. l4 Z. Hashin and S. Shtrikman, J. Mech. Phys. Solids, 1963, 11, 127. l5 B. Budiansky, J. Mech. Phys. Solids, 1965, 13, 223. l6 W. E. A. Davies, J. Phys. D (Appl. Phys.), 1971, 4, 1176. I7W. E. A. Davies, J. Phys. D. (Appl. Phys), 1971, 4, 1325

ISSN:0370-9302    

DOI:10.1039/S19720200127

出版商:RSC    

年代:1972

数据来源: RSC

摘要:

Interface Morphology and Mechanical Properties of Uni- directional Fibre Reinforced Nylon 6 BY T. BESSELL, D. HULL AND J. B. SHORTALL Department of Metallurgy and Materials Science, University of Liverpool, Liverpool L69 3BX Received 23rd June 1972 Unidirectional glass and carbon fibre reinforced Nylon 6 composites have been prepared by the in situ anionic polymerization of caprolactam on the fibres. In this way, it is possible to control the solidification and crystallization characteristics of the nylon. The morphology of the nylon around the fibres has been determined using sections of 0.15 Vf (volume fraction) composites and also thin films containing single fibres. The fibres nucleate directional crystallization (columnar growth) of the nylon around the fibres. The amount and form of the columnar growth depends on the type of fibre and the polymerization conditions.Preliminary studies on the mechanical properties and fracture characteristics of 0.15 Vf composites have been made and the results are discussed in terms of the matrix and interface morphology. One of the most efficient ways of making use of the properties of thermoplastics as engineering materials is by reinforcing them with fibres. With nylon 6 this has resulted in the production of a useful and well established engineering material. Reinforcement by short random fibres imparts an increase in rigidity and tensile strength with increased dimensional stability under load. These properties, together with the toughness, abrasion resistance and ductility of the nylon itself has meant that fibre reinforced nylon composites have found many applications particularly where components are subjected to high stresses.In the conventional approach, these composites are prepared by incorporating the fibres into the nylon by dry mixing and then either injection or compression moulding the mixture into the desired shape. The mechanical properties of a composite are a function of the aspect ratio of the fibre. Damage to the fibres during processing greatly reduces the aspect ratio so that the optimum improvement in mechanical properties which should be achieved with random fibre reinforcement is seldom reached. The composites then show a less than two fold increase in strength over that of the nylon alone. However, by using uniaxially aligned continuous fibres (in which the fibre loadings may be increased to over 70 % by volume), maximum benefit can be obtained from the high modulus fibres when the load is applied in the fibre direction.In a recent paper we have outlined a method of preparing uniaxial aligned fibrous composites involving the in situ anionic polymerization of caprolactam directly on the fibres. This method overcomes the difficult problem of incorporating uniaxially aligned fibres in a thermoplastic matrix, and preliminary work has shown that the composite has interesting fracture characteristics associated with the interfacial properties. EXPERIMENTAL The anionic polymerization of nylon 6 was carried out using dry, high purity 8-caprolactam with sodium hydride as the catalyst and acetyl caprolactam as the initiator.For polymer- ization, 0.01 molar proportions of both catalyst and initiator were used. To prepare samples 137138 FIBRE REINFORCED NYLON 6 of nylon, the polymerizing solution was cast into a steel mould having a rectangular cross section 4 x 8 mm. The nylon was polymerized at the desired temperature, usually between 410 and 480 K, for one hour in an atmosphere of nitrogen. The mould was then allowed to cool slowly in air. Full details of the methods used are given elsewhere.2 Nylon reinforced composites were prepared using either glass or carbon fibre. The glass fibre was supplied by Pilkington Brothers in the form of roving, in which the individual fibres had a mean diameter of 12 pm and a fibre tensile strength of between 0.75 and 3.4 GN m-2. The carbon fibres, used to make tensile specimens, were supplied by Le Carbone Ltd., and Courtaulds Ltd., and had diameters of 11 and 8.2 ,urn and tensile strengths of 1.5-2.2 and 1.72-2.41 GN m-2 respectively.The Courtaulds fibre had a surface treatment to give an increased interlaminar shear strength. Unidirectional continuous fibre composites with a nylon 6 matrix were fabricated using a modified leaky mould technique. The mould, which measured 140 mm long, consisted of a lower portion containing two 8 mm wide channels. Into this fitted a top platen which rested on stops leaving rectangular channels 4mm deep. The lower portion of the mould was supported by a base plate which held fibre tensioning attachments. The mould was sprayed with a P.T.F.E.mould release agent prior to assembly. Reweighed tows, of either glass or carbon fibres, giving the required fibre volume fraction were placed through the mould, tensioned and held in position. The mould and support plate with the fibres in position was placed in a vacuum oven maintained at the nylon polymerization temperature required. 8-caprolactam and sodium hydride (catalyst) were reacted under nitrogen at 393 K to produce sodium caprolactam. This was then heated to the desired polymerization temp- erature with the requisite amount of initiator. The solution was then poured onto the fibres in the mould, and the top platen placed in position. The in situ polymerization was allowed to continue under nitrogen for one hour. After this period the mould was cooled in air before disassembly. The resultant composite samples, measuring 140 x 8 x 4 mm, were stored under vacuum until tensile specimens were prepared.Rectangular specimens, 140 mm long with an 8 x 4 mm cross section, were prepared for modulus determinations according to Courtaulds’ specification^.^ The specimen ends were tabbed with aluminium tabs measuring 4Ox 8 x 4 mm, again prepared according to Court- aulds’ specification. The tabs were bonded to the composites using an Araldite adhesive which was allowed to cure at room temperature for seven days. Curved neck specimens with a radius of about 55 mm and having a minimum width of 4 mm were used for the fracture studies. These specimens were also tabbed. The surfaces of these specimens were polished on emery paper commencing with 600 grade and finishing with grade 410.The final polish was achieved with alumina (0.3 pm) powder. RESULTS MORPHOLOGY To establish the effect of the reinforcing fibres on the morphology of Nylon 6, it was first necessary to study the relationship between the spherulite structure of anionically polymerized nylon and variables such as catalyst and initiator concentra- tions and the temperature of polymerization. Morphological examinations were performed on microtomed sections, 5 ,urn thick, mounted in MS 550 silicone fluid. When viewed between crossed polars these sections showed the characteristic birefringence pattern associated with a spherulitic structure. The morphology was also studied by examining the surfaces of bulk specimens after polishing and etching in a 5 % m-cresol/methanol solution.The spherulite diameter was found to vary with polymerization temperature, as illustrated in fig. 1. The material polymerized at 473 K had a mean spherulite diameter of 60 pm and the spherulite boundaries were clearly defined. The spherulite size decreased with decreasing polymerization temperature, having a mean value of 50 pm at 453 K and 30 pm at 433 K. The spherulitic structure and spheruliteFIG. l ( a and b).-Spherulitic morphology of Nylon 6. (a) polymerized at 473 K, (b) polymerized at 413 K. Specimens etched in 2 % m-cresol in methanol. [To face page 138(4 FIG. 2(a-e).-Optical transmission micrographs using crossed polars of thin films of Nylon 6 showing columnar growth around carbon fibres.a, Grafil HM-S (type 1) ; b, Grafil A (type 11) ; c, Hitron HMG-50s surface treated carbon fibre ; d, Hitron HMG-50 untreated fibre ; e, Hitron HMG-50s fibre in matrix of small spherulite size.FIG. 3.-Section of a 0.15 Vf carbon fibre composite microtomed perpendicular to fibre axis showing the nucleatirig effect of the fibres. Optical transmission micrograph between fibre. crossed polars. FIG. 4.-Optical transmission micrograph using crossed polars of a thin film of Nylon 6 showing the very limited columnar growth around a glass FIG. 5(u and b).-Fracture specimens showing extent of pull out in (a) 0.15 Vf glass fibre composite, (6) 0.15 Vf carbon fibre composite.FIG. 6.-Transverse matrix cracks leaving the fibres intact FIG. 7.-Redirected crack in a fractured 0.15 P'f glass fibre/nylon composite. The tensile axis is parallel to the fibre axis.T .BESSELL, D. HULL AND J . B . SHORTALL 139 boundaries of nylon polymerized at 413 K were ill-defined, although this material showed birefringence effects when viewed through crossed polars. of the effect of catalyst and initiator concentrations on induction time and morphology of nylon 6, we have shown that the spherulite dia- meter is dependent predominantly on the polymerization temperature and moulding and cooling conditions and not on catalyst or initiator concentrations. These concentrations are known to have an effect on molecular eight,^ relative viscosity and polymerization conversion. 7-9 The moulding conditions will also have an effect on the resultant spherulitic structure, since it is known that the crystallization process of this polymer is lo on the morphology of nylon, which was polymerized in the presence of glass and carbon fibres, were made by preparing thin films of nylon containing single fibres.A small drop of liquid monomer, catalyst and initiator was placed onto a glass slide and the fibre laid across it. A second glass slide was then positioned on top to disperse the solution into a film. Polymerization was allowed to take place under nitrogen in an oven held at the required temperature. In the preliminary work,l we established that a columnar crystallization structure can be produced at the fibre interface. This work has been extended to determine the effect of different types of fibre on the columnar structure.Carbon fibres from different manufacturers were used and a range of columnar structures obtained are illustrated in fig. 2. Fig. 2a and 2b show the difference in structure arising from the type of carbon fibre. The crystalline structure around Grafil HM-S, Type 1 (Court- aulds Ltd.) fibre is much finer than the corresponding structure around Grafil A, Type 11. Two zones can be identified in the columnar region; an inner zone with a fine random structure and an outer zone with a fibrillar structure extending to the spherulitic region. The inner structure is attributed to rapid nucleation of the nylon on the carbon fibre surface and is more pronounced on Type I fibres. A similar effect has been observed on Modmor I and Modmor I1 (Morganite Ltd.) and Rigilor AG (Type I) and Rigilor AC (Type 11) (Le Carbone Ltd.) carbon fibres.No marked differences in the columnar growth were observed with surface treatment. Fig. 2c and 2d show a surface treated and an untreated Hitron HMG-50 (Hitco) carbon fibre. The inner zone is well defined around these fibres and is wider than the fibre diameter. The width of the columnar region is related to the spherulite size as illustrated in fig. 2e, which shows a surface treated Hitron HMG-50s carbon fibre in a matrix of small spherulites. The morphology of nylon in the composites was examined in thin films prepared by microtomy and in sections of bulk specimens prepared by mechanical polishing followed by etching. Unfortunately, it was only possible to produce thin sections of carbon fibre composites, microtomy of glass fibre composites being impossible. The columnar growth around carbon fibres in aO.15 V, uniaxially aligned composite was similar to that found in the thin films described above (fig. 2), although the width of the columnar region tended to be more uniform and to extend to about the fibre diameter into the matrix. In these aligned composites, the columnar crystalline region around adjacent fibres tend to impinge on each other resulting in a modification of the spherulitic structure. Microtomed sections, cut perpendicular to the fibre axis (fig. 3) show the nucleating effect of the carbon fibres. When observed from this direction the columnar growth has the characteristic spherulitic appearance with the carbon fibres acting as nucleating sites.As in the single fibre observations (fig. 2c and 24 it was found that the columnar growth was unaffected by surface treatment of the fibres, since both treated and untreated fibres gave rise to the same morphology. However, as indicated later, the In the investigation Preliminary experiments140 FIBRE REINFORCED NYLON 6 strength of the bond between the fibre and the columnar region is increased by surface treatment. The nucleating effect of the surface coated glass fibres was found to be considerably less than that of the carbon fibres. Fig. 4 shows a thin film specimen containing a glass fibre. The extent of directional crystallization normal to the fibre is very limited and at some points along the fibre length it is non-existent.Sections of specimens polished perpendicular to the fibre axis did not show a correlation between the posi- tions of the glass fibres and the spherulite nuclei, again indicating that the glass fibres do not exhibit the same nucleating effect as carbon fibres. MECHANICAL PROPERTIES The mechanical properties reported here form part of an initial evaluation of these composites. Only a limited number of specimens were tested and thus no confidence limits are presented for the values given. The tensile modulus (E) and the tensile strength determinations of both 0.15 Vf carbon and glass fibre composites were carried out on an Instron testing machine using tabbed specimens. The rectangular cross section specimens used for the modulus determinations failed prematurely at about half the fracture stress of the necked tensile specimens.This premature failure occurred at the aluminium tabbed area due to stress concentrations associated with gripping. In the case of glass fibre/nylon composites, the modulus agrees fairly well with that predicted from the simple law of mixtures, viz. The moduli and fracture stresses of the composites are given in table 1. where c, f and m refer to the composite, fibre and matrix respectively. This calcula- tion is based on the reported modulus of undamaged glass fibres. The values of modulus and tensile strength of the nylon 6 have been taken as 1 GN m-2 and 75 MN m-2 respectively. TABLE ELASTIC MODULUS AND TENSILE STRENGTH OF GLASS AND CARBON FIBRE NYLON COMPOSITES composite system elastic modulus/GN m-2 tensile strength/MN m-2 observed calculated observed calculated 0.15 Vf unidirectional glass fibre/ 8.2 12.1 450 320-575 nylon 0.15 Vf unidirectional carbon fibre 0.15 Vf unidirectional carbon fibre (Le Carbone) Nylon 6 30.7 56.3-62.3 3 80 290-397 (Courtaulds) Nylon 6 - - 300 3 2 6-429 The difference between the theoretical value of 12.1 GN m-2 and the observed value of 8.2 GN m-2 may be due to slight fibre misalignment which is known to cause an apparent reduction in modulus and to the additional deformation of the tab joints.The fracture strengths of the necked specimens were found to agree very well with the values calculated from the law of mixtures. The composites reinforced with surface treated Courtaulds fibres having a slightly lower fracture stress than that predicted.T . BESSELL, D.HULL AND J . B . SHORTALL 141 FRACTOGRAPHY Examination of the glass fibre composites after fracture showed that a large amount of pullout had occurred (fig. 5). It was also noted that in these specimens many cracks propagated transverse to the fibres, leaving the fibres bridging the gap between them as illustrated in the polished section in fig. 6. In some cases, the trans- verse cracks had been redirected parallel to the fibres at distance of about 2pm from the fibre (fig. 7). This indicates that there is a surface of weakness some distance from the fibre surface which may be related to the columnar region around the fibre (fig. 4). This suggestion is supported by the fact that, on examination of the fracture surfaces in the scanning electron microscope, some fibres were surrounded by sheaths of nylon (fig, 8).X-ray studies confirmed the presence of nylon on these fibres. In contrast, the fractured carbon fibre composites did not show the extensive pull out (fig. 56) that had been observed with the glass fibre composites. Some evidence for matrix cracking perpendicular to the fibres was apparent but on a very limited scale and restricted to an area very close to the actual fracture surface. Optical microscopy revealed that these matrix cracks are not redirected by the fibres to the same extent as in the case of glass fibre composites. Stereoscan microscopy of fractured 0.15 Vf surface treated carbon fibre specimens, fig. 9 and 10, showed that a limited amount of pull out of fibres had occurred, although there were many areas on the fracture surface in which the fibre fracture had occurred in the failure plane of the matrix (fig.9). In contrast, carbon fibres, which had not been surface treated, exhibit a much greater degree of fibre pull out on the fracture surface (fig. 11). This suggests that the bond strength between the surface treated fibre and the nylon matrix is improved by the surface treatment. No evidence for the sheathed type pull out has been observed with carbon fibre composites. DISCUSSION In composites prepared by anionic in situ polymerization of caprolactam on uniaxially aligned graphite and glass fibres, a system has been developed in which there is a layer between fibre and the bulk nylon. In thermosetting systems it has been shown that layers with an intermediate modulus can markedly affect the mechanical properties due to improvements in the stress transfer between the fibre and the resin.Thus, the elongation l2 and toughness l3 of composites have been increased. The changes in properties are sensitive to the thickness of the layer. Although such layers can be introduced, with some difficulty, into thermosetting systems this has not yet been demonstrated in nylon filled systems and the in situ polymerization route described here offers one possibility. The most significant difference in the fracture modes of the glass and carbon fibre composites is the propagation of longitudinal modes along surfaces some distance from the fibre surface in the glass fibre reinforced material.The limited evidence available in fig. 4,7 and 8 suggests that this is related to the columnar growth structure around the glass which results in a weak interface between the columnar structure and the main spherulitic structure. No corresponding effects were observed in carbon fibre composites, and it will be noted that the columnar region has a width in excess of one fibre diameter. Failure of these composites is brittle in nature and usually occurs by the catastrophic propagation of a single crack. The reasons for these fracture modes in nylon/glass and nylon/carbon systems can be understood by consideration of the energies associated with crack propagation. For the glass/nylon system, the fractographic studies revealed that some of the pulled142 FIBRE REINFORCED NYLON 6 out glass fibres were coated with a sheath of nylon.Energetically the removal of a fibre plus a sheath would be possible, provided the bond strength between the fibre and the sheath is much higher than between the sheath and bulk matrix. It is considered that the bond strength in a glass fibre nylon system is increased by the chemical bonding between physisorbed water (which invariably exists on the surface of glass fibres 14) and the fibre. This allows for the formation of a strong hydrogen bond between the fibre and the reactive -C=O and -N-H groups of the polyamide. Thus it is possible that less energy is required to pull the fibre with the columnar sheath from the matrix than to sever the strong chemical bond between fibre and polymer, even allowing for the increased frictional forces which have to be overcome in pulling the irregular surfaced sheath from the bulk matrix.The columnar growth in the carbon fibre system has been shown to be cylindrical in nature, the fibres acting as nucleating sites for row spherulites in the bulk material (fig. 3). With nylon/glass specimens, the fibres do not act as nucleants to the same extent and matrix cracking on a large scale, crack redirection and pull out is observed, The reason for the difference in nucleating activity between the carbon and glass fibres cannot yet be explained. Nucleation of polymers against surfaces is still not clearly understood and such consideration as wetting of the fibres, surface energies and super- cooling may be involved to a greater or lesser extent.It has been suggested l 5 that columnar growth against fibres from a thermoplastic crystalline polymer matrix may be determined to some extent by low energy sites on the surface of the carbon fibre and by coherency between the two crystal structures. In the glass fibre composites, the columnar growth is limited and localized, and in some areas is not apparent at all. The evidence in fig. 7 suggests that the glass fibre polymer interface is relatively weak in shear, so that transverse cracks can circumvent the reinforcement leaving the fibre bridging the gap. Fibre pull out and gross matrix and interface cracking will then occur. In the carbon fibre composites, the columnar crystalline region is large, and pull out of a sheath is not possible. The extent of pull out will be determined by the strength of the bond between the fibre and columnar region and the distribution of weak points along the fibres.16 By increasing the shear strength of the fibre/matrix bond by surface treatments fracture surface pull out is reduced.This is illustrated by comparison of fig. 9 and 11. In both these cases the extent of the columnar region appeared to be unaffected by the surface treatment. As a result of this preliminary work, several modifications to the nylonlfibre composites are being studied with the aim of optimizing the fracture mode. Control of the extent of the columnar growth on both graphite and glass surfaces is being investigated. Also by variation of the volume fraction of the fibres the columnar region will be restricted in growth by impingement on adjacent columnar regions and a system consisting of cylindrical rows of spherulites can be developed, nucleated around fibre cores. The authors thank the Science Research Council for the award of a studentship (T. B.). T. Bessell, D. Hull and J. B. Shortall, Nature Phys. Sci., 1971, 232, 127. T. Bessell and J. B. Shortall, Europ. Polymer J., in press. Carbon Fibres, Design Engineering Series (Morgan-Grampion, ed. E. A. Smith). G. Stea and G. B. Gechele, Europ. Polymer J., 1965, 1, 213. G. B. Gechele and G. Stea, Europ. Polymer J., 1965, 6, 233. G. Stea and G. B. Gechele. Europ. Polymer J., 1970, 6, 233. ’ Takaya Yasumoto, J. Polymer Sci. A , 1965, 3, 3301.T . BESSELL, D. HULL AND J . B . SHORTALL 143 * 0. Wichterle, J. Sebenda and J. Kralicek, Fortschr. Hochpolym. Forsch., 1961, 2, 578. l o 0. Wichterle, J. Sebenda and J. Tomka, J. Polymer Sci., 1962, 57, 785. l1 R. H. Knibbs and J. B. Morris, U.K.A.E.A. Research Group Report No. AERE-R6926, 1971. l 2 A. S. Kenyon and H. J. Duffy, Polymer Eng. Sci., 1967,7, 189. l 3 A. S. Kenyon, J. Colloid Interface Sci., 1968, 27, 761. l4 E. P. Plueddermann, 25th Annual Conference, 1970, Reinforced Plastics/Composites Division l 5 S. Y. Hobbs, Nature Phys. Sci., 1971, 234, 12. l6 G. A. Cooper and A. Kelly, Interfaces in Composites, ASTM, STP 452, 1969. Takaya Yasumoto, J. Polymer Sci. A , 1965, 3, 3877. S.P.1.

ISSN:0370-9302    

DOI:10.1039/S19720200137

出版商:RSC    

年代:1972

数据来源: RSC

摘要:

Some Interfacial Problems in Metal Matrix- Carbon Fibre Composites BY S. V. BARNETT, S. J. HARRIS AND J. V. WEAVER Department of Metallurgy and Materials Science, University of Nottingham, Nottingham, NG7 2RD Received 14th July, 1972 Metals reinforced with continuous graphite (Type I) fibres are prone to compatibility and oxida- tion problems when they are subjected to high temperature treatments. To prevent such interactions, the feasibility of using evaporated barrier layers has been examined in a nickel matrix. A number of techniques, e.g., tensile testing, thermal-balance determinations and microscopy have been employed to test the usefulness of this approach. Results have shown that 1 pm thick metal coatings either react vigorously with the fibre or allow diffusion of oxygen or carbon to take place sufficiently quickly so as to delay fibre degradation by only a short time interval.Whilst chemically stable non- metallic coatings tend to prevent diffusion of both carbon and oxygen, they are extremely susceptible to brittle cracking due to differential thermal expansion effects. In this respect, zirconium carbide and boron nitride seem to be affected to a lesser degree than other carbides, nitrides and oxides, and can provide protection over a limited temperature range. The demand for higher specific strength materials which have the capacity to operate in hostile environments, e.g., oxidizing atmospheres above 523 K has en- couraged work on metal matrix composites. Observations 1-4 made on carbon fibre reinforced metals have indicated that a number of fundamental problems exist, each of which requires a solution before such materials can develop their full potential.To a large extent these problems are associated with metal-metal, metal-fibre or TABLE ST STANDARD FREE ENERGIES OF FORMATION OF CARBIDES AND OXIDES 5 9 compound (carbides) ALC3 B4C cr23c6 C0,C Fe3C Ni3C Sic TdC Tic wc ZrC standard free energies of formation at 1300 K/kJ mol-1 - 147 - 43 - 461 + 5 -2 + 25 - 44 - 154 - 167 - 49 - 170 compound (ox ides) A1203 B203 co coo CUO FeO Fe203 Ir02 NiO Si02 Ta205 TiOz wo3 ZrOz Cr203 standard free energies of formation at 1300 K/kJ mol- standard free energies of formation of other compounds at other 1 temperatures/kJ mol-1 - 1686 - 957 BN - 137 at 1300 K - 173 at 900 K - 227 CO -192at 900K - 798 - 143 - 41 - 179 - 487 - 0.04 - 125 Ir07 -70 at 900 K Ni,C +27 at 900 K NiO - 159 at 900 K - 646 - 1480 - 710 -518 - 850 144S .V . BARNETT, S. J . HARRIS A N D f . V . WEAVER 145 composite-environment interfaces, inasmuch as they affect the chemical, physical and mechanical properties of the composite. The chemical stability of carbon fibres in a number of metal matrices may be simply assessed on the basis of the free energy data quoted in table 1. Metals divide themselves into two broad groupings, those that form stable carbides, e.g., Cr, W, V, Zr, etc. and those that either do not form carbides, e.g., Cu, Ir, Ag, etc. or form relatively unstable carbides, e.g., Ni, Co, etc. Principal interest in fabricating carbon fibre composites has centred around aluminium and nickel.Aluminium is an interesting case because on one hand the free energy of formation of its carbide is large and negative but on the other it does not rapidly form this compound at the fibre-metal interface until the temperature is raised close to the melting point of the metal.7 This slow reaction considerably helps the development of a bond between the fibre and matrix, during fabrication, whilst nickel in the absence of any direct chemical reaction with carbon does not establish anything other than a poor bond. TABLE 2.-sOLUBILITY OF CARBON AND OXYGEN IN METALS solubility of solubility of solubility of carbon at 1273 K carbon at 873 K oxygen at 1273 K element atomic % atomic % atomic % Ag A1 Au c o Cr Fe Ir Mn Mo Ni Ti V c u <0.001 (S) < 0.050 (S) <0.30 (S) -0.50 (S) < 0.026 (S) 0.04 (S) 2.00 (S) <0.05 (S) 0.40 (S) - 0.45 (S) <0.001 (L) - < 0.050 (L) - <0.30 (L) - <1.00 (S) - <0.05 (S)* - 10.00 (S) - ~ 0 .0 5 (S) - 1.20 (S) 0.05 1 1.80 (S) 1.00 (S) - 1.60 (S) 0.032 6.70 ( S ) - - - 0.020 * estimated on the basis of evidence given in ref. (16) The direct and excessive formation of carbides at the fibre surface produces a severe reduction in strength. This is not the only compatibility problem associated with metal matrix composites, for metals which do not form stable carbides can degrade carbon fibres under certain conditions, particularly where limited solid solutions exist, see table 2. Contrary evidence exists of the mechanisms by which the solubility aspect manifests itself.Jackson and Majoram have shown, chiefly by X-ray techniques, that both carbonized and graphitized fibres recrystallize to form a three-dimensional graphite when such fibres are encased in a nickel matrix and subjected to heat treatment at temperatures > 1273 K in a vacuum of - 10-2N m-2. Barclay and Bonfield on the other hand have provided evidence of fibres remaining intact and not recrystallized at 1373 K provided the heat treatment was carried out at pressures below 10-4N m-*. Although the explanations of these observations are in dispute, it is apparent that the stability of carbon fibres in nickel is dependent on the surface condition of the fibres prior to coating, the gas content of the coating and the vacuum pressure of any subsequent heat-treatment process.Oxidation proteetion at elevated temperatures is a major problem with continuous fibre systems, for even if the metal provides some protection for the fibres, it may be146 METAL MATRIX COMPOSITES difficult to stop local attack where fibres come to the surface, e.g., fibre ends. Stud- ies lo completed on aluminium-matrix composites have indicated that an acceptable oxidation rate was obtained up to 673 K for graphitized fibre reinforcements, provided the fibres were well bonded and fibre ends not exposed. The case with nickel is quite different, and indeed for most metals which have small negative free energies of formation for their oxides (table l), since oxygen effectively ‘‘ diffuses ” to the metal- fibre interface, resulting in oxidation of the carbon.on nickel- graphite fibre composites have shown that within a period of one hour at 873 K the fibres have been attacked severely. Within the present programme of work, attempts have been made to examine methods of controlling some of the reactions that cause degradation of carbon fibres in a nickel matrix. In principle, an assessment of the value of using a barrier layer interleaved between fibre and matrix has been attempted. Such an assessment has involved tensile testing, microscopy and thermobalance determinations on graphitic (Type I) fibres which have been coated with selected metals and compounds and then overcoated with nickel. The selection of barrier layer materials was made on the basis of diffusion data and oxidation resistance.Results obtained TABLE 3.-cARBON DIFFUSION DATA IN METALS AND CARBIDES D at 1273 K/mmz s-1 ref. material DO/IXIIII~ s-1 U/kJ mol-1 HfC Tic0.60 Tic Ta2C NbC M o ~ C W Ta Nb V Zr Cr Fe Ti Ni 5.ox lo6 3.3 x 104 1 .OX 103 1 . 9 ~ 104 1 .OX 105 8 . o ~ 103 1 x 10-3 7 . 0 ~ lo2 1.2x lo2 1.6 1.8 5x 10-1 5x 10-1 3 x 1.7 3 . 2 ~ 10-1 536 477 435 460 347 372 33 1 381 167 163 134 113 109 100 84 79 9x 10-l6 8x 10-l6 4x 10-l6 4x 3 x 10-l2 1 x 10-l1 4 x 10-l2 1 x 4~ 10-15 3 x 10-7 4~ 10-7 2~ 10-5 2~ 10-5 2~ 10-4 3~ 10-5 2x 6x 12 12 12 12 12 12 12 12 13 14 14 14 14 14 14 13 17 Since carbon goes into solid solution in most metals interstitially, its diffusion coefficient tends to be much larger than those found for elements in substitutional solid solutions. An analysis l1 of the diffusion of fibre material into a finite coating has shown that a 1 pm thick barrier layer with a diffusion coefficient of carbon of mm2/s would saturate in approximately one day, whilst a similar thickness of material of diffusion coefficient of mm2/s would saturate in -300 years.Available experimentally determined diffusion coefficients in metals and compounds are given in table 3. It can be seen that only the refractory metal carbides, e.g., HfC, ZrC and TIC have values in the desired range, i.e., to 10-l8 mm2/s. To provide a comparison, iridium was selected as a ductile metallic barrier layer. Al- though the diffusion coefficient for carbon in iridium at 1273 K is not known, Criscione, Volk and Smith l5 obtained a certain amount of data from which it may be obtained if the solid solubility of carbon in iridium is estimated.If the solubility is -0.05 atomic %, then the diffusion coefficient at 1273 K is - mm2/s. ThisS . V . BARNETT, S . J. HARRIS AND J. V . WEAVER 147 value is low when compared with other values of carbon diffusion coefficients in metals. Another criterion for barrier materials was resistance to oxidation. On these grounds, iridium again appears to be an attractive metal, since it has been shown l6 that oxygen does not effectively diffuse through it, and that below 1395 K a protective film of iridium dioxide forms. Alumina was chosen as another material for pre- venting fibre oxidation and carbon diffusion because of its high stability and melting point. Provided alumina is of high purity there is little reaction with nickel at temperatures up to 1373 K.Boron nitride has been used as a means of protecting boron fibres from oxidation in a nickel matrix, although the free energy data (table 1) indicate that the nitride is not highly stable. However, the nitride is more stable than the carbide and therefore should not interact with the fibre. From the oxidation standpoint, the same argument can be applied to the carbides of Ti and Zr, the oxides of these metals being highly stable. EXPERIMENTAL PREPARATION AND HEAT TREATMENT OF COATED FIBRES Small amounts of graphitized fibre (prepared by pyrolysis up to -2973 K) were coated by physical evaporation; the evaporant was heated by a focussed and accelerated beam of electrons. This process was used to apply coats of nickel, aluminium oxide, iridium, zirconium and titanium carbides and boron nitride.Small groups of fibres were spread and fixed to metal frames (150 mmx 60 mm) by using a conductive thermosetting silver preparation. These frames were mounted on a jig which allowed them to rotate within the evacuated bell-jar (vacuum pressure < This rotation was necessary to obtain complete coating on the fibres and thus counteract the line-of-sight nature of the deposition process. Attempts were made to minimize interaction between crucible and evaporant, co-evaporation of crucible material, and contamination of the evaporant beam by the bell-jar atmosphere which could all have affected the properties and composition of the coating. To minimize surface contamination of the carbon fibres prior to coating of high tension voltage was applied across them in an attempt to bake off adsorbed species, etc.Table 4 gives a list of the evaporants together with their chemical composition, physical states and operating conditions for evaporation. N m-2). TABLE 4.-EVAPORATION CONDITIONS FOR COATINGS physical coating electron electron state of thickness gun beam evaporant pm potential /kV current /mA comments material nickel iridium alumina boron nitride zirconium carbide titanium carbide sheet 0.5-1.0 3.0 60 good smooth coatings wire 0.5-1.0 3.2 80 good even coatings granules 0.5-1.0 2.5 50 easy operation to powder - 1.0 3 .O 80 charge did not melt but produce even coatings turned from white to grey carbides was difficult, coating rates were required powder - 1.0 3.5 110 evaporation of both powder - 1.0 3.4 100 particularly where fast Boron nitride coatings were also applied by the passage of the fibres through an aqueous solution,* containing boron salts.Coating was achieved by immersing bundles of fifteen fibres into the solution. After immersion, the fibres were transferred to an alumina holder148 METAL MATRIX COMPOSITES ready for treatment in the furnace. The specimens were introduced slowly into the hot zone (1273 K) of the nitriding furnace to allow slow drying of the fibre and coating. Electrodeposited overcoats of nickel were obtained from a sulphamate solution. To facilitate plating on to non-conducting fibre coatings a thin layer of physically evaporated nickel was put down prior to immersion in the electrolyte.Specimens were subjected to heat treatment in bundles, each of which contained indi- vidually coated fibres. The bundles were inserted into alumina tubes which were in turn supported in a graphitic carbon block; the whole assembly was placed in a tube furnace, which was controlled to k 2 K . The presence of the carbon block acted as a getter for residual oxygen in the furnace tube. Sufficient time was allowed for outgassing at temper- atures below 623 K, whilst the vacuum was maintained at a pressure below 5 x 10-3N m-2. Heat treatment in oxidizing conditions was carried out in a similar furnace arrangement, with the exception that the furnace tube remained open to the atmosphere to enable a free supply of air to flow over the specimen.COATED FIBRE STRENGTH TESTS Tensile tests were carried out on micro-composites; a technique used by Jackson and Majomn2 Each composite was prepared from approximately fifteen fibres after coating and heat treatment by applying a few drops of Araldite MY753/HY951 resin and curing. The failure load of each composite was measured on a straining frame with a load cell of 9 N capacity. Throughout this work, strengths are quoted as bundle breaking loads rather than stress values. This is done for three reasons. First, the strength changes are large and are immediately apparent from the bundle breaking load. Second, there is the problem of counting and measuring the diameter of the fibres. Although attempts were made to keep the bundles at fifteen fibres, the handling problems resulted in an inaccuracy of & 1 fibre, which must in part account for the spread in results obtained.The mean fibre diameter was determined as 8.8 pm for the batch of fibres, with a standard deviation of rf: 0.40 pm. Finally, it must be remembered that a composite was being tested and some allowance would have to be made for the strength of the matrix. This would be of particular importance in testing of coated fibres. However, the neglect of this factor is reasonable since the coatings were thin, i.e., 0.1-0.5 pm, and no evidence was found of abnormally high strengths with as-coated fibres. OXIDATION KINETICS Experiments were performed using an automatic recording thermobalance which allowed continuous monitoring of the weight of the sample.Initial tests using as-received graphite fibres established a rate dependence on the initial sample weight, for this reason a minimum sample weight of 50mg was imposed. The thermobalance measured to an accuracy of k0.l mg. MICROSCOPY Coated fibres, before and after heat treatment, were examined in the scanning and trans- mission electron microscopes (S.E.M. and T.E.M.). Small pieces of fibre were trapped between two grids before insertion in the T.E.M., fibre ends, edges and coatings could then be examined in shadow. To examine the fibre surfaces of nickel overcoated specimens after treatment, cross and taper sections were prepared for optical microscopy. It became necessary to remove the thick nickel electrodeposit for S.E.M. observation and this was achieved by etching the bundles in a dilute nitric-sulphuric acid mixture.The fibres were recovered by filtration. RESULTS TENSILE TESTS AFTER HEAT TREATMENT IN VACUO The bundle breaking loads obtained for plain carbon fibres after heat treatment for one day at temperatures up to 1423 K in uacuo fall in the range 100 gf to 190 gf * Boric acid and urea (19 g of each) were melted at 415 K in air ; the resultant liquor was made up to 250 ml with distilled water.S . V . BARNETT, S . J . HARRIS A N D J . V . WEAVER 149 (equivalent to an individual fibre breaking stress range of 1070-2040 MN/m2). With iridium coatings there was no obvious reduction in breaking load even when the fibres had been held for 36 h at 1433 K, see fig. 1. However, the imposition of an overcoat of electrodeposited nickel over the iridium significantly changed the strength of the mini-composite at temperatures in excess of 1273 K, see fig.2. After a one day treatment at 1373 K a complete loss in strength had occurred. For comparison purposes, a series of specimens in which the fibres had been directly coated with nickel (by physical vapour deposition) were tested and are plotted on fig. 2. 1401 8 100 z 8. 8. 8. 691 m 3 6 hours El e temperature/"C FIG. 1 .-Breaking load for mini-composites containing iridium coated fibres after vacuum heat treatment. The horizontal lines refer to the loads obtained on uncoated Type I graphite fibres tested under similar conditions. The tensile behaviour of alumina coated samples are shown in fig. 3, together with samples which had been coated with nickel.The single alumina coating does not appear to reduce the bundle breaking load when such values are compared with those associated with uncoated fibres. However, the nickel electrodeposited samples became more and more difficult to handle as the heat treatment temperature increased. A definite strength reduction had occurred in samples treated at 1323 K and a number of " no-load " failures occurred with fibres heated to 1373 K.150 METAL MATRIX COMPOSITES 200 L, 8 180 - 160 - 140 - 8 s g 8 0 - a 60- 40 - 20 - I I *. 800 900 I000 1 roo temperature/"C FIG. 2.-Breaking load for mini-composites containing fibres which have been (a) nickel coated and (b) iridium coated, followed by an overcoat of nickel, and vacuum heat treated for 24 h.TENSILE TESTS AFTER OXIDATION TREATMENT Tests carried out on plain carbon fibres after four-hour periods of treatment showed that there was a marked degradation at 773 K and a complete loss of strength at 873 K. Iridium coated samples gave almost exactly similar results, except that some fibres were at least capable of being tested after the 873 K treatment, see fig. 4. Nickel coated fibres appeared to be degraded at lower temperatures than either the uncoated or iridium coated samples, see fig. 5. However, statistically there was no significant difference between results, i.e., a " t " significance test showed that there was a greater than 70 % probability that the differences in the 773 K test results occurred by chance. The plating of an alumina coating on the fibres did not prevent the breaking load of the fifteen fibre bundles from falling after they had been treated at temperatures in the range 673 to 873 K, see fig.5. Again there was evidence that some strength could be retained after 4 h at 873 K ; the range of the breaking loads were somewhat higher than those obtained on the iridium coated samples.S . V . BARNETT, S . J . HARRIS AND J . V . WEAVER 151 The tensile test results obtained on boron nitride coated fibres are shown in fig. 6. The fibres in this series of experiments had been coated by immersion in the boron salts solution (held at 323 K) and nitrided at 1273 K. To remove any contribution ------- .--- 4 : I, .-\- - - -7- - - -- - - - ------ - - '. .. Uncoated. D-43 Alumina-coated - Nickel - c o o f e d 1 i.1 . . \ ! 24-h vacuum heat-treatment temperature/"C FIG. 3 .-Comparison of breaking loads obtained on mini-composites containing (a) uncoated, (b) nickel coated, and (c) alumina coated fibres after vacuum heat treatment for 24 h. from the -0.1 pm thick boron nitride, the heat treated fibres were ultrasonically cleaned. The appearance of a limited number of testable specimens after treatment for 4 h at 873 K was encouraging; it was noted that three control samples of un- coated fibres did have sufficient strength to be tested in this particular set of experi- ments. OXIDATION BEHAVIOUR USING THE THERMOBALANCE The results of the thermobalance oxidation (at 868 K) of fibres coated with boron nitride from a continuously stirred and heated solution are summarized in fig.7. It would appear from these results that there is no advantage in increasing the thickness of the boron nitride coating, i.e., from 1 vol % (0.02 pm thick) to 19 vol % (0.5 pm thick). The points on the graph represent the calculated carbon loss assuming that152 METAL MATRIX COMPOSITES I 40t -I I I 1 I m 200 300 400 500 600 temperature/"(= FIG. 4.-Breaking load for mini-composites containing iridium coated fibres after heat-treatment in air for 4 h. the boron nitride weight remains constant. Standard uncoated fibres suffered complete loss in weight in -20 h whilst fibres with 1 vol % boron nitride coatings have lost only some 10 % of their weight at this stage. OPTICAL AND ELECTRON MICROSCOPY PLAIN CARBON FIBRES Scanning electron microscope examination showed the fibres to be essentially circular in cross-section with regular ridges on the surface.Shadow micrographs of fibre edges and fracture ends in the transmission electron microscope (T.E.M.) showed the edges to be essentially smooth and the fibrilar substructure to be in evi- dence at thin projections on the fracture surface. Electron diffraction patterns obtained from fibre edges and ends showed a limited angle of arcing on the 002 and 004 rings, and streaking in the 10 and 11 bands in the (001) direction, see fig. 8. After treatment in air for 4 h at 673 K, slight pitting of the fibre surface was observed in the S.E.M. As the temperature increased up to 873 K the incidence and depth of pitting increased markedly. COATED FIBRES SUBJECTED TO VACUUM HEAT TREATMENT Prior to heat treatment, all types of coating were examined in the S.E.M.to When fibres were first coated with ensure that they were of a continuous nature.S . V . BARNETT, S . J . HARRIS AND J . V . WEAVER 0 153 f I , I I I i a 0 ? I I I I I I I 1 I , I I I I I I i 7 I 1 d l ? t j I I I I I I I I I I I I I I I 1 I I I I I t I I I I I I I A )--. Uncooted. c)----o Alumina-coated &---A Nickel-coated I I 1 1 300 400 500 600 4-h air heat-treatment temperature/"(= FIG. 5.-Comparison of breaking loads obtained on mini-composites containing (a) uncoated, (b) nickel coated, and (c) alumina coated fibres after heat treatment in air for 4 h. alumina and subsequently with nickel, the two layers were discernable in the S.E.M., see fig.9a. The nickel and alumina thickness in this particular case were 0.2 and 0.5 pm respectively. On occasions when nickel was electrodeposited around an entire fifteen fibre bundle, cross sections of this type of micro-composite were examined by optical microscopy, see fig. 96. During treatment in vacuum for one day from 1123 K and upwards, agglomeration of thin vapour-deposited nickel coatings occurred, and at 1273 K there was complete spheroidization, see fig. 10a. In these specimens surface pits were observed on carbon fibre surfaces. Transmission electron micrographs of the surface regions between the nickel spheroids showed that a number of growths existed, see fig. lob. At higher magnification, fig. lOc, a fine grained layer was observed to be attached in places to the fibre surface together with a number of separate particles or flakes which were not in general associated with the fibre surface.Diffraction patterns were obtained partic- ularly from the fine grained material, see fig. lOd, and calculated " d " spacing indi- cated that it was graphite. The diffraction rings obtained were unlike those associated154 200 180 160 ru M 3 140- 9 8 \ 4 2 120- z w ru 0 100- 4 - .g 80- 24 9 2 60 40 20 METAL MATRIX COMPOSITES m rn As-coated . o BN-coated (0.lymthick ) ultrasonically cleaned of coating after oxidation 0 Uncoated standards - - - 00 0 0 0 - 0 - 0 0 - 0 I I I I NONE 400 500 600 0 B. BB 00 0 m 0 0 0 0 0 0 0 0 O n 00 0 0 0 0 0 0 0 heat treatment temperature/"C FIG. 6.-Breaking loads obtained on mini-composites containing boron nitride coated fibres after heat treatment in air for 4 h.test time in hours 10 x ) 30 40 50 60 80 100 120 I40 0 Uncoated fibres standard. I. vol.percent cmting . 0 4 . - - - - 19. " " . CO.lp4 FIG. 7.-Oxidation tests on boron nitride coated fibres in the thermo-balance at 868 K represented as percentage carbon loss from the fibre.FIG. 8.-Electron diffraction pattern from very thin area of fibre surface showing 002 and 004 arcs. The fibre axis parallel to the 002 arcs. [To face page 154FKG. 9.--(a) As deposited duplex nickel- FIG. 9.-(h) Specimens with electrodeposited aIuminia coating on a gr-aphite fibre, 0.2 p.m nickel overcoatings, sho\iing the distribution and-0.5 pin thick respectively. Magnification of a bundle of fifteen fibres.Magnification ;< 6000. x 600. FIG. 10.-(a) Nickel coated fibre heat treated in vacuum for one day at 1273 K. Magni- FIG. lO.-(b) Graphite film associated with nickel globules (as in fig. I O U ) as observed in fication x 12 500. transmission electron niicruscope. Yagni- fication x 1500. FIG. 10.--(c) Higher magnification picture of a FIG. lO.-(d) F,lectron diffraction pattern regicn shown in fig. JOh showing particle or flak: graphite within the film. Magnification x 20 000. obtained from a region shown in fig. 1Oc.FIG. 11 .-Iridium coated fibre after heat FIG. 12.-Scanning electron micrograph of treatment in vacuum for one day at 1373 K. zirconium carbide coatings after heat treat- Magnification x 8000. ment in vacuum at 1273 K. The nickel overcoat has been etched away.(a) Showing longitudinal cracks. Magnification x 3000. FIG. 12.-(b) Showing circumferential cracks. Magnification x 6100. FIG. 12.-(c) Showing absence of attack on the fibre surface where the coating has been re- moved. Magnification x 7200. FIG. 13.-Scanning electron micrograph of pits on a carbon fibre surface after heat treatment in air under a poor carbide coating and nickel FIG. 14.-Scanning electron micrograph of car- bon fibre surface after heat treatment at 873 Kin air under a boron nitride coating, indicating ab- overcoat. Magnification x 6100. sence of surfacebutting. Magnification x 6000.S . V . BARNETT, S . J . HARRIS AND J . V. WEAVER 155 with the original fibre, i.e., there was no preferred orientation, suggesting that the graphite was of three-dimensional form.Iridium also agglomerated in vacuum at temperatures in excess of 1273 K, see fig. 11. The size of the agglomerates was much smaller than with nickel and there was no apparent disruption or pitting of the fibre surface, i.e., the ridges originally present could easily be seen. Electron diffraction evidence of three-dimensional graphite on the surface could not be found. However, fibres coated with iridium and nickel showed a similar tendency towards coating agglomeration, after treatment at 1323 K for one day. Growths were observed between the agglomerates and electron diffraction from these regions gave three-dimensional graphite patterns. Examination of vacuum heat-treated alumina coated fibres in the S.E.M. showed that there was a separation of the coating from the fibre.Various degrees of spalling resulting from circumferential and longitudinal cracking of the alumina coating were noted. Electron diffraction showed that the preferred orientation of the fibre was retained even with specimens heated for 24 h at 1373 K ; the remains of the coating gave rise to diffraction patterns which were indexed as y-alumina. A series of specimens was prepared with zirconium carbide coatings which were in turn overcoated with nickel in the form of a mini-composite ; these were subjected to heat treatment for 96 h in the temperature range 1173 K to 1323 K. After this treat- ment, taper samples were cut from the mini-composite and examined in the optical microscope. It was noted that some of the fibres had seLerely degraded and others had not been affected at all. These observations may be explained in terms of the variable stoichiometry of the original zirconium carbide coatings.In a number of cases, where the nickel was etched away leaving the carbide coating on the fibre, the S.E.M. showed that both longitudinal and circumferential cracks, fig. 12 a-c, have occurred in the coating, but in the main the coating is still adherent to the fibre surface. Where sufficient coating has been removed, the fibre surface was not observed to be damaged. Titanium carbide coatings did not prevent degradation of the fibres after treatment in excess of 1273 K. OXIDATION HEAT TREATMENT After four hours at 673 K, the nickel coat had cracked; longitudinal and radial cracks were observed.At 773 K only radial cracks were observed, and X-ray diffraction analysis on the coating showed that it consisted almost entirely of nickel oxide. Iridium coated fibres suffered no visible damage after treatment at 673 K, at 773 K slight pitting has occurred, but at 873 K the coating had cracked radially and longitudinally. Beneath the regions where the coating had exfoliated, the fibre surface was badly pitted. Electron diffraction patterns obtained on the specimens were indexed, the " d " spacings closely corresponded to those quoted for iridium dioxide. Fibres which had been coated with alumina showed evidence of large scale cracking and spalling after treatment at 873 K, where such disruption had occurred pitting of the fibre surface had taken place, see fig.13. Similar behaviour was noted where a nickel overcoat had been used on top of the alumina ; nickel oxide had formed which had behaved in a brittle manner in conjunction with the alumina. Boron nitride coatings were shown to be rather uneven in thickness ; undoubtedly this was due to the chemical means by which the material had been applied. After treatment for four hours at 773 K and 873 K, there was no evidence of fibre attack on a large scale. When evaporated boron nitride coatings existed under a nickel overcoat for four hours at 773 K, no evidence could be found of attack on the fibre surface, i.e., after the coatings had been removed by ultrasonic agitation, see fig. 14.156 METAL MATRIX COMPOSITES DISCUSSION Barrier layers may be broadly characterized into those that agglomerate, i.e., ductile metals, and those that tend to crack and spall, i.e., brittle non-metals.METALLIC BARRIER LAYERS The results obtained on the nickel coated graphite fibres support the findings of Jackson and Marjoram,' indicating that graphite fibres recrystallize into a three dimensional form after heat treatment above 1273 K and under a pressure of 5 x N m-2. Both electrodeposited and evaporated nickel coats appear to bring about this fibre degradation process. The electron microscope technique shows two types of graphite growing between the nickel spheroids, a fine grained film and flakes. The fine grained film appears to have been left behind as the nickel lowered its surface energy and agglomerated; it is possible that a surface diffusion process has taken place on the nickel as it begins to form the spheroids. The flake graphite was not in the main associated with the fibre surface and must have gone through a process involving dissolution, diffusion and precipitation in the nickel phase, i.e., rather than an in-situ recrystallization process on the fibre.Iridium has been shown to be compatible with carbon fibres for up to 36 h at 1373 K. However, agglomeration is observed to take place, which would not commend it as a barrier layer. No evidence could be found of three-dimensional graphite growths or of fibre weakening. When iridium was interleaved between the nickel and the fibre, the fibres were appreciably weakened and three-dimensional diffraction patterns for graphite were once again observed.Such a process could have taken place either by diffusion of nickel through iridium, or diffusion of carbon through the nickel after iridium agglomeration had taken place. Data are not available on nickel solubility and diffusion in iridium. However, using the data given in tables 2 and 3 together with an assumed diffusion coefficient of - 10-lo mm2/s for Ni in Ir at 1373 K, it may be shown that 99.9 % carbon saturation of the 1 pm iridium layer would have taken place in -20 s, whilst a 50/50 atomic % Ni/Ir alloy would have resulted in -200 s. The carbon solubility of the Ni/Ir alloy may increase markedly with rising nickel content, i.e., if note is taken of the solubility levels of this element in the two metals concerned, see table 2.Since the carbon solubility in iridium is at least an order of magnitude smaller than that of carbon in nickel, it could be simply assumed that simple carbon dissolution was the rate controlling degradation process in nickel-carbon fibre composites. However, Barclay and Bonfield have shown that this process does not degrade fibres at temperatures < 1353 K, and that the impurity content of the fibre-nickel coating and the furnace atmosphere are all important. It is known that oxygen effectively diffuses through nickel and its oxide to attack carbon fibres in the oxidation studies. Therefore, small concentrations of oxygen at the nickel-fibre interface could effectively " catalyze " the dissolution process. Iridium forms a protective oxide and could provide a barrier to oxygen diffusion, hence preventing oxygen from reaching the fibre surface ; thus the damaging aspects of the dissolution process never effectively begin.When nickel is placed on top of the iridium and interdiffusion takes place, the possibility of oxygen permeating the coatings may once again increase, thus leading to fibre degradation. If oxygen does play some role in the dissolution process it is not clearly defined at present, since on thermodynamic and experimental evi- dence l 8 oxygen dissolved in nickel and/or nickel oxide in the presence of carbon in solid solution reacts to form carbon monoxide above 723 K. In the solid state theS. V . BARNETT, S. J . HARRIS AND J . V . WEAVER 157 difficult step in this process is the formation of gas bubbles; however, no evidence has yet been found of such bubbles in coated fibres.Assuming dissolution of carbon takes place, such a process would cease after the limit of solid solubility has been reached, and precipitation would not normally take place, unless the unstable carbide of nickel forms. However, a driving force for precipitation may result from the following : (a) Differences in solubilities between carbonized fibre, graphite fibre and three dimensional graphite, in general a more stable phase has a lower solubility than a less stable phase, i.e., stability of three dimensional graphite > graphitized fibres > carbonized fibre. (b) High local concentrations of impurities, e.g., oxygen, sulphur, etc., in the coating changing the solubilities of carbon in nickel.Finally impurities, in particular compounds occluded as boundary films, etc., may act as heterogeneous nucleation sites for three-dimensional graphite, otherwise nucleation away from the fibre surface would be difficult without a large precipitation driving force. Treatment of nickel coated fibres in an oxidizing atmosphere results in a significant drop in strength at 773 K. The weakening of the fibres is associated with local pitting of the fibre surface, suggesting that some surface sites are more vulnerable to attack. Above 673 K clear evidence exists of nickel oxide formation. The oxide cracks circumferentially and allows attack to take place directly on the fibre surface. Spalling of the iridium coating was observed after treatment at 873 K and evidence was found of iridium dioxide.Some degree of protection was afforded by iridium up to 873 K and, where failures had occurred, the thickness of the iridium may have been insufficient to prevent oxide formation through the section. NON-METALLIC COATINGS The observations made on alumina and the carbides together with nickel and iridium oxides, indicate that crack formation, followed by spalling, are the initial steps in the failure of these brittle barrier layers. Some coatings appear to be more susceptible to this type of failure, e.g., alumina > titanium carbide> zirconium carbide> boron nitride. Three possible reasons for this type of failure are : (i) Oxygen ingress through pores in the coating resulting in gaseous products that expand and lift the coating away.(ii) Volume changes during oxidation of the coating (not alumina) producing sufficient internal stressing to cause coating failure. (iii) Excessive stressing due to thermal mismatch between coating and fibre during cooling. Mechanism (iii) appears to be the most important because of the cracks which have been observed in vacuum heat treated samples, although mechanisms (i) and (ii) may still be relevant in oxidizing conditions. Thermal mismatch results from differences between the moduli of elasticity, Poisson’s ratios, and thermal expansion coefficients of the fibre and coating. Coating thickness also may play a significant part in the process. Longitudinal and circumferential cracks have been observed in both alumina and carbide coatings.An examination of the values of the linear coefficients of expansion of alumina, titanium carbide, zirconium carbide and carbon fibres along the fibre axis, indicates that compressive longitudinal stresses will be set up in virtually all coating materials on heating, resulting in a brittle failure around the circumference of the coated fibre. Alumina has the greatest mismatch with respect to the fibre, followed by titanium carbide and then zirconium carbide. Of the other physical properties to be considered, Poisson’s ratio and elastic modulus do not differ greatly from one coating material to another.158 METAL MATRIX COMPOSITES Boron nitride may be unique in that having a graphite-like structure, it also possesses similar anisotropic properties to the fibres.Accepting this, the stress condition in a boron nitride coating will be dependent upon there being an epitaxial relationship between coating and fibre. APPLICATION OF BARRIER LAYERS The desirable properties of an effective barrier layer for a carbon fibre reinforced nickel may be summarized as a material that possesses a low carbon diffusion rate, a low diffusion rate in the nickel matrix, a low solubility in the matrix, resistance to oxidation, and low oxygen permeability, together with thermal expansion, Poisson ratio and elastic modulus values that are compatible with the fibre. On the basis of diffusion and solubility data, the majority, if not all, metals and alloys may be ex- cluded. The non-metals give some encouragement, particularly in respect of the degree of protection afforded by zirconium carbide under vacuum conditions at 1373 K and boron nitride at > 873 K in air.However, there are appreciable diffi- culties in applying these coating materials to - 8 pm carbon fibres ; obviously evaporation techniques are not ideally suited to coating multi-fibre tows. Consider- ation may be given to the use of such coatings on a duplex system of fibres proposed by M0r1ey.l~ Here, the barrier layer would act as an outer tube on a core prepared from a bundle of carbon fibres, thus reducing the total surface area for coating in any given composite. By controlling the properties of the fibre core/barrier layer and the barrier/metal matrix interfaces it may be possible to introduce a system of weak and strong bonded interfaces which could lead to higher works of fracture in these materials without impairing transverse strength. We are grateful to the late Dr. E. Holmes and Dr. A. A. Baker for many helpful discussions and to the Ministry of Defence for providing a research contract under which part of this work was financed. One of the authors (SVB) gratefully acknow- ledges the award of a studentship by the Science Research Council. The authors are also indebted to Prof. J. S. L1. Leach for the provision of laboratory facilities. P. W. Jackson, Met. Eng. Quart., 1969, 9, 22. P. W. Jackson and J. R. Marjoram, J. Mat. Sci., 1970, 5,9. P. W. Jackson, D. M. Braddick and P. J. Walker, J. Mat. Sci., 1971, 6,419. J. M. Evans and D. M. Braddick, Corrosion Sci., 1971, 11, 611. 0. Kubaschewski, E. L1. Evans and C. B. Alcock, Metallurgical Thermochemistry (Pergamon, London, 4th Edn, 1967), p. 421. J. F. Elliot and M. Gleiser, Thermochemistry for Steelmaking (Addison Wesley, Massachusetts, 1960), p. 131. G. Blankenburgs, Int. Conf. on Interfaces, Melbourne, 1969. * M. Hansen, Constitution of Binary Alloys (McGraw-Hill, New York, 1958). R. B. Barclay and W. Bonfield, J. Mat. Sci., 1971, 6, 1076. lo A. A. Baker, P. W. Jackson and C. Shipman, Fibre Sci. Techn., 1972, 5, 285. l1 S. V. Barnett, PhB. Thesis (University of Nottingham, 1970). l 2 R. A. Andrievsky, V. S. Eremeev, V. N. Zagryazkin and A. Panov, Zzvest. Akad. Nauk. SSSR, l 3 I. I. Kovenskii in Diflusion in bcc Materials (A.S.M., Ohio, 1965). l4 L. V. Pavlinov and V. N. Bykov, Phys. Metal Metallogr., 1965, 19, 73. lS J. M. Criscione, H. F. Volk and A. F. Smith, A.Z.A.A. J., 1966, 4, 1791. J. C. Chaston, Platinum Met. Rev., 1965, 9, 51. l7 J. J. Lander, H. E. Kern and A. L. Beach, J. Appl. Phys., 1952, 23, 1305. D. M. Braddick and S . J. Harris, Trans. Znst. Met. Finishing, 1972, 50, 46. l9 J. G. Morley, Proc. Roy. Soc. A , 1970, 319, 117. Neorg. Materialy, 1967, 3, 2158.

ISSN:0370-9302    

DOI:10.1039/S19720200144

出版商:RSC    

年代:1972

数据来源: RSC

摘要:

Effect of Surface Treatments on the Interfacial Bond Strength in Glass Fibre-Polyester Resin Systems BY J. B. SHORTALL AND H. W. C. YIP Department of Metallurgy and Materials Science, The University of Liverpool Received 3rd July, 1972 The effect of fibre surface treatments on the bond strength of 30 pm E-glass fibres, embedded in a polyester resin, have been studied by mechanical rupture of the interfacial bond using a shear de- bonding specimen. The surface treatments included water lubrication, silanes of different reactivity towards the resin, boundary lubricant, film forming polymer, silicone resin and blends of these components. The results of these studies suggest that the measured bond strengths are closely related to appropriate wetting of the fibre by the resin. An attempt has been made to interpret the bond strength in terms of the physical and chemical forces operating in the interface region.The ability to control the interfacial bond strength in fibre-reinforced materials is of the greatest importance if the interface is to play its dual role of stress transmission between the two phases (in which case a strong bond which resists failure is required) and of increasing the fracture toughness by deflecting crack growth and delocalizing stress at the crack tip (in which case a weaker bond is needed). + O m m + FIG. 1 .-Shear debonding specimen geometry. The object of the present investigation is to gain an understanding of the relative contributions to the bond strength of the various coating materials commonly employed with glass fibres which are used in conjunction with unsaturated polyester resins. The purpose of the size coating is to improve the performance of the fibre in the composite and to protect the glass during the processing steps which lead up to its incorporation in the resin.The size consists of a silane coupling agent which pro- motes adhesion between the glass and the resin matrix and which improves the 159160 INTERFACIAL BOND STRENGTH resistance of the glass to attack by aqueous environments. Lubricants are incor- porated into the size to reduce friction and abrasion during processing ; film forming polymers such as poly vinyl acetate are used to promote adhesion between filaments. In earlier work we have studied the feasibility of various methods available for evaluating interfacial bond strength, with the conclusion that the most useful method is one using a shear debonding specimen.2 This involves the uniaxial compression testing of a rectangular block of resin containing a single discontinuous glass fibre, Using this test method, the variation of the bond strength with surface treatment is shown to range from an extremely weak bond to a strong bond which can only be ruptured by unstable flow of the matrix.The debonding and subsequent crack growth in the specimen were followed optically, the interface separation giving rise to a strong reflection of incident light. Under dark field illumination one can follow these processes visually. Microscopic examination of thin sections indicate that adhesive failure had occurred rather than cohesive failure of the resin.fig. 1. MATERIALS The E-glass fibres, with a nominal diameter of 30pm, were supplied by Pilkingtons Research and Development Laboratories. The polyester resin employed was Crystic 195 manufactured by Scott Bader Company Ltd. This resin uses methylmethacrylate as the cross linking agent. Used in conjunction with the resin were catalyst M and accelerator type E. The different fibre coatings used were: (a) Midland silicone resin system 2540/2541 ; (b) a reactive silane A174; (c) an unreactive silane A153 (both silanes were manufactured by Union Carbide Co.) ; ( d ) boundary lubricant (Morpan TPB) ; (e) polyvinyl acetate ; cf) a mixture of (b) and ( d ) ; (9) a mixture of (b) and (e), and (h) a mixture of (b), ( d ) and (e).All these constituents were added to the fibre from an aqueous system. In addition, fibres were also supplied coated only with a water lubricant applied immed- The silicone resin was also used uncured as a liquid lubricant for the compression test and iately after drawing. in the cured form as the mould release agent. EXPERIMENTAL EQUIPMENT The equipment used included an Instron machine, a Zeiss Jena Stereomicroscope with magnification from x 4 to x 100 times, a Reichert Zetopan polarizing microscope, a Zeiss Tessovar photomicrography system and an Isomet diamond cutting saw. SPECIMEN PREPARATION Because specimen preparation is a critical preliminary to the measurement of any mech- anical property, considerable care was taken both with moulding and surface finish of the specimens. The actual specimen preparation methods used are described fully elsewhere. The volume ratio of unsaturated polyester to accelerator to catalyst used was 100 : 2 : 2.Gel time was about 20 min and as the Crystic 195 is a room temperature curing resin it was cured at room temperature and atmospheric pressure for 168 h. The dimensions of the specimens were height 30 mm, depth 10 mm and width 10 mm. Prior to mechanical testing, the interface region was examined with a stereomicroscope so that any debonded area could be detected and the specimen rejected. Specimens were further examined between crossed-polarizers to ensure correct fibre alignment and a similar distribution of built-in stresses (due to curing shrinkage). One out of three specimens made was rejected for one or other of the above reasons.The test assembly and manner of testing is described fully e1sewhere.lFIG. 2.-Fibre debonding. (a) Initial debonding at fibre ends ; (6) propagation of interfacial crack. To face page 1 6 1 ]J . B . SHORTALL A N D H . W . C . Y I P 161 The cross head speed was fixed at 0.02cm/min, which was found to give sufficient time for manoeuvre and to yield reproducible results. In compression tests the slenderness ratio of the specimen is important ; a value of three was chosen as it was believed that the end constraint effect could be minimized when the fibre length equals the width of the specimen. RESULTS The results of compression tests on fibres with different coatings are listed in table 1.A 10 % coefficient of variation was obtained between specimens. In most cases, interfacial fracture appeared to have occurred in two distinct phases. At first, because of the difference in fibre end geometry, the two ends debond at different stresses and, if the average value is taken, reasonably consistent TABLE 1 .-EFFECT OF SURFACE TREATMENT ON INTERFACIAL SHEAR STRENGTH surface treatment debonding stress stress for crack (mean value)/ propagation/ no. of specimens MN m-2 MN m-2 tested silicone resin type 2540/2541 unreactive silane (phenyltrimethoxy silane) 22.2 boundary lubricant (tetra-decylpyridin- ium bromide) 25.2 polyvinylacetate 37.1 water lubricated fibre 34.5 reactive silane (y-methacryloxy- propyltrimethoxysilane) 52.3 reactive silane plus boundary lubricant 56.7 reactive silane plus boundary lubricant plus polyvinyl acetate 65 .O no bonding obtained 48.0 37.9 50.5 ill-defined 90.9 70.7 yield stress limited 10 8 8 10 10 8 7 10 values are obtained for the shear debonding strength.Then, as the stress is increased, a point is reached when the stable cracks jump a long distance towards the centre of the specimen.l The fibres coated with silicone resin did not bond at all. In this case, a separate interface was observed after curing which was similar to that occurring in other speci- mens after debonding. The fibres coated with a water lubricant debonded in a very gradual but continuous manner over a range of stress levels, there being no sharp distinction between crack initiation and propagation. The initial debonding stress was moderately high, but the subsequent interfacial separation occurred in an ill-defined manner.The majority of the fibre types examined failed in this way, fig. 2. DISCUSSION The different stress levels required for crack initiation and propagation are difficult to understand; no attempt has been made to rationalize this difference in this paper. The relative behaviour of the size components has therefore been interpreted solely in respect of the stress required to initiate debonding in a particular system. The mechanisms available for adhesion in single fibre-resin systems involve both chemical and physical bonding; the constituents of the size applied to glass fibres utilize some of these mechanisms to bond either with the glass surface or the resin or both.Chemical bonds have been shown to be the strongest forces involved in pro- ducing adhesion, but their formation may be limited by steric requirements and they 2-F162 INTERFACIAL BOND STRENGTH may be limited to only a portion of the bonded area.3 The chemical bonds can act as anchor points which prevent the incremental breakdown of dispersion forces in adhesion failure and are stable enough to withstand attack by water. The results listed in table 1 should reflect the relative contribution of the different types of bond to the adhesive strength. To attempt to distinguish these contributions it is necessary to consider the role of each constituent in the size and its method of bonding to resin and to the glass. In view of the unexpectedly high bond strength for the water lubricated fibre system, the role of the adsorbed water on the fibre surface must first be evaluated.It is known that E-glass is easily hydrated by water and its surface saturated with hydroxyl groups whereupon complete wetting (0” contact angle) is ~btained.~ Also, free water exists on the glass surface. This can restrict chemical reaction with the glass surface by an insulating action and has also been shown to interfere with the chemisorption of silanes on the glass.6 I I I L- -1 -o-si””””x o i(/ FIG. 3.-Interphase region. The wetting of a high surface tension substrate with a low surface tension adherent should result in good adhesion, and is therefore one of the main reasons for the strong bond between the high surface energy silicate glass and the resin.’ Perfect wetting of the glass by the resin would result in an adhesive bond stronger than the cohesive strength of the matrix resin, but this is not obtained, possibly due to the presence of the water layer, as is confirmed by the fact that the observed separation was by adhesive failure rather than cohesive failure of the resin.Some bonding may occur through hydrogen bonds formed between hydroxyl groups in the glass surfaceJ . B . SHORTALL AND H . W . C . YIP 163 and carbonyl groups which exist in the resin, fig. 3. In the light of these observations, it may be assumed that the markedly high bond strength in the water lubricated fibre- resin system is mainly a reflection of the adhesive bonding, which is impaired to some extent by the presence of the water.The interpretation of the effects of the various size components on bond strength must now be considered, bearing in mind the bonding between fibre and resin alone. The silicone resin coating acted as an effective barrier to any bond between the glass surface and the resin. This is to be expected, as the silicone resin has no functional groups capable of bonding with either the glass or the resin. Very low surface wetting was achieved, as is revealed by examination of specimens containing the silicone coated fibres which showed that they differed in no way from specimens vhich had been compressed to the stage of crack merging. The bond strength measurements on specimens containing glass-fibres coated with a silane, reactive only towards the glass, revealed relatively weak bonding.This can be accounted for by the non-reactivity of the silane towards the resin, bonding only occurring between the silane and the glass. This takes place as the methoxy groups of the silane are hydrolyzed by water to form silanol groups which undergo conden- sation with similar groups on the glass surface, or between adjacent molecules, to form siloxane linkages and yielding a multi-layer coverage of the fibre surface,6* fig. 3. The reaction of the silane with the glass surface is an energetically favourable process which leaves the unreactive phenyl groups aligned toward the polyester molecule. In this case, the force between the two would be of the weak dipole-induced dipole type, since the benzene molecule is easily polarizable and the carbonyl group in the polyester acts as the permanent dipole.Some hydrogen bonding may still take place between the glass surface and the resin as there is evidence that when a silane reacts with a glass surface some hydroxyl groups remain ~nreacted.~ The low bond strength in this system can be attributed to the fact that the critical surface tension of glass is lowered by treatment with silane, the wetting of the glass by the resin is severely impaired and the strength of the adhesive bond reduced. The high bond strengths obtained using the reactive silane can be readily under- stood. The silane provides functional groups which are reactive towards the glass surface as discussed earlier. The group X, fig. 3, is a reactive organic group, CH2 = C(CH3)C00w designed to match the reactivity of the polyester system and which establishes strong covalent bonds with the resin, the vinyl group in the molecule acting in the same way as the styrene molecule, forming a bridge between the unsaturated polyester chains.Thus, the bridging of the interface by the bifunctional coupling agent gives good adhesion through covalent bonds to both the resin and the glass. This chemical coupling may be considered to yield the ultimate in wetting in the form of molecular continuity from resin to glass.1° This is reflected in the high stress level required for initial debonding in this system. There may alsa be some contribution from hydrogen bonding between resin and glass. The boundary lubricant used was a quaternary ammonium salt containing a long aliphatic chain.At the glass surface, there exist metallic cations which diffuse into the aqueous phase because its high permittivity lowers the attractive forces between ionic species on the surface. This leaves an anionic surface which is capable of orienting cations. Thus, physical adsorption of an oriented monolayer of the long chain aliphatic cations of the lubricant occurs through ion exchange reactions with the glass surface and the lubricant is uniformly and strongly adsorbed, fig. 3. The long chain and small cations are not likely to give strong bonding with the resin.164 INTERFACIAL BOND STRENGTH In this case, the adhesion of the resin to the glass is reduced by the presence of the lubricant and no chemical coupling can take place across the interface. The polyvinyl acetate coating on the glass fibres may function in a similar way to the chemically adsorbed water in reducing resin-glass adhesion.However, it contains the polar acetyl groups which give rise to the possibility of hydrogen bonding to the surface hydroxyl groups of the glass and to the reactive carbonyl groups in the resin. The intermediate bond strength in this system may thus be a result of a reduction in the surface tension of the glass by the polyvinyl acetate impairing adhesive bonding, plus a contribution to bond strength from hydrogen bonding. The high bond strength in the reactive silane +lubricant system must reflect the bond strength due to the reactive silane, chemical coupling of the silane with the surface occurring before the ion exchange reaction of the lubricant with the surface can take place.Finally, the high bond strength of the complete size must be considered a syner- gistic effect based on all the separate contributions. Crack propagation in this case occurred in the plastic region of the stress-strain curve indicating that the yield stress of the resin was less than the crack propagation stress. In this case, therefore, the stress criterion is inadequate. In view of the lack of collaborative techniques for direct investigation into the structure of the interface region and its immediate vicinity it must be realized that these suggestions, put forward to account for the relative contribution to bonding of the size constituents, are to some extent tentative; considerably more work needs to be done using chemical and physical techniques to gain a better understanding of bonding in glass fibre-polyester resin systems.Coupling agents have generally been evaluated by their effect in improving the composite flexure strength.1° However, the mechanism of failure in such tests may be complicated and may not involve interfacial bond failure. Therefore, the method reported here offers considerable advantages. Future work will involve an attempt to try to understand the different stress levels required for crack initiation and propagation in these systems and a suitable model system is currently being investigated. The authors wish to acknowledge the award of a Pilkington Brothers Research Studentship (H. W. C. Y.). J. B. Shortall and H. W. C. Yip, J. Phys. D., in press, K. Gutfreund, L. J. Broutman and E. H. Jaffe, Advanced Fibrous Reinforced Composites, 10, R. B. Dean, Chemical Adhesion, Official Digest of the Federation Society Paint Technol. (June 1964). 0. Johannson, F. Stark, R. Barry et al., Investigation of the Physical-Chemical Nature of the Matrix-Reinforcement Interface, AFML-TR-65-303, Pt. 1 , Sept. 1965. J. A. Laird and F. W. Nelson, The Efect of Glass Surface Chemistry on Glass-Epoxy Systems, SOC. Plastics Engrs. Trans. (April 1964). 0. K. Johannson et al., Fundamental aspects of Fibre Reinforced Plastic Composites, ed. R. T. Schwartz and H. S. Schwartz (Interscience Publishers, N.Y. 1968). ' W. A. Zisman, Surface Chemistry of Glass Fibre Reinforced Plastics, 19th Annual Meeting, S.P.I. (Feb. 1964). J. L. Koenig and P. T. K. Shih, J. Colloid Interface Sci., 1971, 36, 247. R. Evans and T. E. White, Fundamental Aspects of Fibre Reinforced Plastic Composites, ed. R. T. Schwartz and H. S. Schwartz (Interscience Publishers, N.Y. 1968). lo S. Sternson and J. G. Marsden, Fundamental Aspects of Fibre Reinforced Plastic Composites, ed. R. T. Schwartz and H. S. Schwartz (Interscience Publishers, N.Y. 1968). 5 2 5 (1966).

ISSN:0370-9302    

DOI:10.1039/S19720200159

出版商:RSC    

年代:1972

数据来源: RSC

摘要:

Reinforcement of Thermoplastics using Carbon Fibres BY W. H. BOWYER AND M. G. BADER Department of Metallurgy and Materials Technology, University of Surrey, Stag Hill, Guildford, Surrey. Received 15th September, 1972 The tensile mechanical properties of carbon fibre reinforced nylon 6.6 and polypropylene have been studied. Two distinct ranges of fibre length and a range of fibre volume fraction from 0 to 0.2 have been examined in each matrix. Longer fibres are more effective in improving strength and stiffness than are equivalent fractions of short fibres. The fibre matrix bond strengths are higher than the yield strength of unfilled matrix for both systems and it is suggested that this is due to a modifica- tion in matrix structure occurring at the fibre/matrix interface. A critical fibre length exists such that fibres of critical length or longer strain by the same amount as the matrix at their centres.The value of critical length is related to the strain in the matrix and the fibre matrix bond strength. Fibres which are subcritical in length carry a constant stress, and fibres which become subcritical by fracture continue to carry stress at the level appropriate to their new length as strain increases. Some speci- mens of each material failed at strains which exceed the mean fibre failure strain. In these cases considerable multiple fracture of the fibres occurred. In specimens failing below the mean fibre fracture strain less fibre breakdown occurred. It is suggested that failure in these composites may be initiated in the matrix and not by fibre fracture.NOMENCLATURE Ef = modulus of fibre, Em = modulus of matrix V , = volume fraction of fibre, o, = stress in the composite, E, = strain in the composite, Euc = mean fracture strain, T = shear strength of the fibre matrix bond, L = fibre length, E = mean effective fibre length in the composite, L, = critical fibre length at a composite strain of E , L, = critical fibre length at fibre fracture, r = fibre radius, C = orientation factor, q = reinforcement efficiency. In previous papers 1* we have reported investigations into the tensile properties of glass fibre reinforced nylon and glass fibre reinforced polypropylene. The stress- strain relationship for these materials may be represented by an equation of the following type L,, = critical fibre length at a composite strain of 0.01, where L, = EfEcr/T.165166 REINFORCED THERMOPLASTICS By fitting experimental data to this equation it was possible to determine values for C and z for each system. It was also shown that when fibres became shorter than L, they continued to support load. This led to an increase in the ultimate strength of composites containing very low volume fractions of fibres. In a series of carbon fibre reinforced nylon composites, specimens containing low fibre volume fractions showed a decrease in ultimate strength. This paper reports work in which the properties of carbon fibre reinforced nylon and polypropylene matrices have been examined as a function of fibre volume frac- tion. For each matrix composites containing (a) a high proportion of fibres which are expected to be shorter than L, at E = 0.01 and (b) a high proportion of fibres which are expected to be longer than 5 L, when E = 0.01, have been examined.EXPERIMENTAL The stock materials used in this study were all prepared from polymer granules and (1) Nylon 6,6 (ICI Maranyl A 100 carbon fibre (Modmor type I1 surface treated) blend (2) Nylon 6.6/carbon fibre blend containing fibres ranging from 0.01 to 0.6 mm in length. (3) Polypropylene (ICI PXC 8639)/carbon fibre blend containing fibres of length 6 rnm. (4) Polypropylene/carbon fibre blend containing fibres in the range from 0.01 to 0.6 mm in In each case (1)-(4) above the basic compounding operation was used to produce a material of the highest volume fraction of fibres tested.This was then diluted with more polymer at the final moulding stage in order to produce mouldings containing various volume fractions of fibre up to the maximum. Tensile specimens were produced from all the compounds by injection moulding on a screw preplasticising machine into a mould with an end gate. Moulding conditions were arranged so as to cause a minimum of fibre breakdown during the moulding operations. The test piece used had a cross-section of 5 mm x 2 mm and a total length approx. 75 mm. Before testing all specimens were stored in an air-conditioned laboratory at 20°C and 50 % relative humidity for one week. Tensile testing was carried out on an Instron testing machine using a sensitive strain-gauge extensometer. The fibre volume fraction in each specimen was deter- mined by burning off polymer from a weighed part of the specimen and weighing the residual fibre.An experiment using a thermobalance has indicated that the corrections required to take account of oxidation of the fibres during this burn off procedure is insignificant. The fibre length distribution in specimens containing short fibres was determined by burning off polymer from a dummy specimen, examining the fibres under a microscope and counting the numbers of fibres of various lengths. It was not convenient to apply this method to speci- mens containing long fibres due to the wide range of fibre lengths present. continuous tows of carbon fibre in our laboratories and are listed below : containing fibres of length -6 mm. length. RESULTS In fig.1A and 1B the nominal stress-nominal strain curves for typical composites from each batch of material at comparable fibre volume fractions are compared with the curves for the unfilled polymer. Both nylon and polypropylene in the absence of filler show a well-defined yield point and exhibit cold drawing behaviour. Cold drawing was inhibited in the fibre filled materials. Only polypropylene specimens containing a Vf of less than 0.1 of short carbon fibres or less than 0.05 of long carbon fibres showed any tendency to draw. All other specimens failed at strains below the yield strain of the unfilled matrix. It was a characteristic of reinforced nylon speci- mens that failure was catastrophic and complete fracture of the specimen occurred. In the polypropylene specimens which did not draw fracture occurred in two stages.The first stage was a sharp reduction in the stress on the specimen and the secondW. H . BOWYER AND M. G . BADER 167 I I I I I I 0.01 0.02 0.03 0.04 0.05 0 0.1 0 . 2 0.3 0.4 nominal strain nominal strain FIG. 1 A.-Typical stress-strain curves for FIG. lB.-Stress-strain curves for unfilled materials 1-4. Material 1, Vf = 0.14; mater- nylon 66 and polypropylene. 1, Nylon 66. ial 2, Vj- = 0.18; material 3, V’= 0.16; 2, polypropylene. material 4, VJ = 0.15. stage was a more gradual reduction in stress which preceded final separation. The extent of the reduction in stress which occurred in the first stage of failure was roughly proportional to the fibre volume fraction. Failure strain has been plotted as a function of V’ for the four materials in fig. 2 and 3 ; in all cases the failure strain has been taken as the strain at which the stress on @e\ I A LONG FIBRES SHORT FIBRES I 1 I 10 2 0 3 0 vf I i / A A SHORT FlBAES I J I 0.1 0.2 0.3 v FIG.2.-Strain at fracture against yf for fibre FIG. 3.-Strain at fracture against Vf for fibre reinforced nylon composites. A, long fibres ; reinforced polypropylene composites. 0, 0, short fibres. long fibres; A, short fibres.168 REINFORCED THERMOPLASTICS 4 cj 3 0.8- - 1.0 h d 8 J 0 . 6 - - 2 $ 0.4- cu .e 0.2- 8 & Y I SHORT FIBRE REINFORCED POLYPROPYLENE I 0.2 0.4 0.6 0.8 1.0 I I I I !,,NYLON SHORT FIBRE REINFORCED 3 0.8- L 0 . 6 - v1 0 0.4 - cj u .s 0.2 - L I .o 4 fibre length/L (mm) fibre IengthlL (mm) FIG.4A. Fibre length distribution curve for FIG. 4B.-Fibre length distribution curve for material 2; short fibre reinforced nylon. material 4; short fibre reinforced poly- propylene. the specimen fell to 98% of the maximum stress. The cumulative fibre length distributions for materials (1) and (4) are shown in fig. 4A and 4B. Fibres in mat- erials (2) and (3) ranged in length from 0.01 to 6 mm. The proportion of fibres which were shorter than 1 mm was small (less than 10% by volume in both cases). There I I I 0 0. I 0.2 Vf FIG. 5.-Stress at strains of 0.005 and 0.01 against Vf for material (1) (long carbon fibres in nylon). n “E 1 1 3 z Y z a .- E 5 3 Y .- Y v1 Y 2 v1 I I L----.i Vf 3 0 0. I 0.2 FIG. 6.-Stress at strains of 0.005 and 0.01 against Vffor material (2) (short carbon fibres in nylon).W.H . BOWYER AND M . G . BADER 169 103) I I FIG. 7.-Stress at strains of 0.005 and 0.01 FIG. 8.-Stress at strains of 0.005 and 0.01 against Vf for material (3) (long carbon fibres against vf for material (4) (short carbon fibres in polypropylene). in polypropylene). vf Yf FIG. 9.-Maximum stress (fracture stress) FIG. 10.-Maximum stress (fracture stress) against V f for material (1) (long carbon fibres in against Vffor material(2) (short carbon fibres in nylon). nylon).170 REINFORCED THERMOPLASTICS was a tendency for some of the fibres to remain in clumps of length 6 mm in these materials and the effect was more marked for materials of type (4) than of type (3). Fig. 5-8 show the variation in nominal stress borne by the composite at two levels of strain (i.e., stiffness) as a function of Vf for the four materials.The solid lines have been fitted to the data by a least squares analysis. Fig. 9-12 show the relation- ship of maximum stress with Vf for the four materials. The solid lines in these figures are drawn from calculated values of fracture stress. 2 0.1 0.2 Vf FIG. 11 .-Maximum stress (fracture stress) against Vf for material (3) (long carbon fibres in polypropylene). Vf FIG. 12.-Maximum stress (fracture stress) against Vf for material (4) (short carbon fibres in polypropylene). DISCUSSION In order to calculate fibre-matrix bond strengths it is necessary to solve eqn (1). This has been done for both systems by making use of the data in fig. 3,5 and 8. The results are compared with the yield strengths and true fracture strengths of the unfilled polymers in table 1.(A value of 220 GNjm2 has been used for Young’s modulus of the fibres.) TABLE 1 yield stress/ true fracture shear strength of matrix MN/m2 stress MN(m2) fibre matrix bond 7/(MN/m2) nylon 65 260-300 130 polypropylene 27 120-1 50 29 Taking these calculated values of z we have calculated the values of the orientation factor C for materials (2) and (4) by substitution in eqn (1). For materials (1) and (3) where we did not measure the full fibre length distribution an alternative method of determining C has been employed. The value of z for materials (2) and (4) indicate that the critical fibre lengths at e = 0.01 are 0.12 mm for the nylon +carbon system and 0.26 mm for the polypropyl-W .H . BOWYER AND M . G . BADER 171 ene+carbon system. Since only a small proportion of the fibres in unstrained specimens of materials (1) and (3) were shorter than 1 mm, it has been assumed that the materials approximate to ones in which all fibres are longer than 5L,. Such materials approximate in their behaviour to ones in which all the fibres have the same length.4 This has been termed the mean effective fibre length E. Values of C and E/LE~ have been obtained by solving eqn (1) using the observed values of stress at two levels of strain. For the nylon+carbon material, only results from specimens which failed at strains in excess of 0.012 have been used in this calculation to avoid errors due to strain concentrations which immediately precede fracture.A knowledge of C enables the fibre reinforcement efficiencies to be calculated by substitution in eqn (3) : Values of C and the reinforcement efficiencies at two levels of strain for each material are given in table 2. The higher reinforcement efficiencies for material (2) compared with material (4) is a result of the higher bond strength in the nylon based material. In spite of the incomplete dispersion of fibres in material (1) much higher reinforcement efficiencies are achieved in this material than in the short fibre reinforced nylon. The orientation factor for this material is also higher and this could be due to long fibres having a greater tendency to align themselves along the direction of flow during moulding. TABLE 2 ‘I reinforcement effciency at strains material C E = 0.005 E = 0.01 (1) 0.57 97 94 (3) 0.35 78 57 (2) 0.32 72 61 (4) 0.33 60 40 The long fibre reinforced polypropylene (3) shows less of an improvement over the short fibre reinforced oplypropylene (4) than does material (1) over material (2).This is due to a combination of the lower bond strength between polypropylene and carbon and the less complete dispersion which was observed in this case. We conclude that the poorer dispersion of fibres and the lower orientation factor are both a result of the lower fluidity of polypropylene compared with nylon at their moulding temperatures. In fig. 9-12 the lines marked I represent the component of composite fracture stress which the matrix would support at the observed fracture strains (i.e., Em&,).Clearly, considerable strength reinforcement has been achieved in all specimens. The mean fracture stress for type I1 fibres has been measured as 308 MN/m2 and this corresponds to a fibre fracture strain of 0.014. Values for L, calculated at an assumed fibre fracture strain of 0.014 are 0.36 mm for the nylon+carbon system and 0.17 mm for the polypropylene +carbon system. Less than 20 % by volume of the fibres in material (ii) and less than 10% by volume of the fibres in material (iv) were longer than L, (fig. 4A and 4B). In materials (1) and (3) almost all fibres were longer than L,. We expect therefore that during testing of the short fibre reinforced specimens an increasing proportion of fibres will become subcritical in length as strain increases ;172 REINFORCED THERMOPLASTICS fibres longer than L, will fail when their fracture strain is reached.Since eqn (1) is derived by assuming a linear build up in stress from each end of the fibre and L, is that length of fibre which is just long enough to be loaded to its fracture stress at the centre, fibres will break into lengths of between L,/2 and L,. In the materials containing long fibres, multiple fracture of fibres should occur in specimens which strain beyond the fibre fracture strain before failure. Single failure of the fibres would occur in specimens which fail below the fibre fracture strain. It has been confirmed by observation of fibres from burned off specimens that consider- able fibre breakdown occurred in specimens falling at strains above 0.014, whilst the fibre breakdown in specimens which failed at lower strains was relatively slight.The fact that reinforcement was achieved in specimens where multiple fracture occurred indicates that fibre fragments continued to support load. A large decrease in composite failure strain is observed when Vf is increased in the range 0-0.1 for all four materials. This means also that there is a reduction in the stress which can be supported by the matrix up to the point of failure (i.e., Em&,). The increase in load supported by fibres as Vf is increased may be lower than, equal to or greater than this reduction in load supported by the matrix at fracture (depending on the fibre length distribution and the fibre matrix bond strength). If it is lower fracture stress will show an initial decrease with increasing Vf. If it is equal then there will be an initial plateau in the fracture stress Vf relationship and if it is higher then there will be an initial region where the fracture stress against Vf relationship has a gradually increasing slope. The full lines on fig.9 and 11 have been calculated from eqn (1) at the observed fracture strains. For specimens which failed at strains below 0.014 the calculated values of L/Lc1 and C have been employed. For specimens which strained by more than 0.014 before fracture it has been assumed that fibres were broken into lengths between L, and E,/2 with a mean 3L,/4. These lines fit the data well. Several specimens of material (2) (short carbon fibres in nylon) failed at strains close to 0.014.The line marked I1 on fig. 10 is calculated from eqn (1) using the observed fibre length distribution assuming that all fibres failed at their fracture strain ; the line marked I11 has been calculated by assuming that no fibre failure occurred before composite failure. At low Vf the data points lie close to line I1 but as Vf increases they are closer to line 111. This is consistent with multiple fibre fracture at low Vf and a gradual transition from multiple fibre fracture to single fibre fracture as Vf is increased. Since the fracture strain of the composite is much higher than the fracture strain of the fibres at low Vf and approaches the fracture strain of the fibres at higher Vf, this is completely reasonable. In material (4) the fracture strain never falls as low as 0.014, and it may be assumed that fibre fracture occurs as with the long fibre specimens; again the cal- culated line fits the observations well.Since these tests are carried out at constant extension rate rather than constant loading rate a mechanism involving transfer of stress from the fractured fibres to the surrounding matrix may not be invoked. In all cases the fracture strain decreases as Vf increases (i.e., matrix volume fraction decreases), and in materials reinforced with long fibres the rate of decrease in fracture strain with increasing Vf is higher than in the materials reinforced with short fibres, The values of fibre matrix bond strength determined are higher than the flow stress of the unfilled matrix. This indicates that the presence of the fibres modifies the matrix properties, at least in the region close to the fibres, and the proportion of the matrix so affected will increase as Vf increases.This together with the stress concentration which occurs at fibre ends may limit the ability of the matrix The mechanism of failure in these materials is still unclear.W. H. BOWYER AND M . G . BADER 173 to flow and so lead to a matrix failure rather than a fibre failure controlled fracture process. CONCLUSIONS The interfacial bond strengths in carbon fibre reinforced nylon 6.6 and carbon fibre reinforced polypropylene are higher than the yield strengths of the unfilled matrices. This indicates that a local modification to matrix structure occurs close to fibres. The modification in matrix structure close to fibres and the stress concentrat- ing effects of fibre ends may limit the ability of the matrix to draw and lead to a matrix failure dominated fracture process.In all cases, nylon/long fibres, nylon/short fibres, polypropylene/long fibres and polypropylene/short fibres, linear relationships exist between stress at strains of 0.005 or 0.01 and Vf. The reinforcement efficiencies were higher when the longer fibres were used in both matrices. The fracture stress against volume fraction relationships for the four materials were highly dependent on fibre length distribution and the fracture strain against volume fraction relationship. In the polypropylene based materials where drawing of the matrix occurred in specimens containing low fibre volume fractions, a continuous increase in fracture stress occurs for increasing Vf. In the nylon based materials where no drawing occurs at low volume fractions an initial plateau occurs in the strength against Vf relationship. This plateau extends to a higher level of Vf in the short fibre reinforced material than it does when long fibres are used. The stress-strain relationships of all four materials may be represented by eqn (1) up to the stress at which failure occurs or when fibre fracture starts to occur. When the failure strain of the composite exceeds the fibre fracture strain, multiple fibre fracture occurs and the equation of the stress-strain relationship must be modified to take account of this. When fibres become subcritical in length due to increased composite strain they continue to support stress at a constant level. When fibres fracture, the fragments continue to reinforce but at a lower level depending on their length. This work was carried out in the Department of Metallurgy and Materials Tech- nology, University of Surrey. The authors thank members of the technical staff for their valuable assistance, the Science Research Council for financial support, I.C.I. Ltd. (Plastics Division) and Morganite Modmor Ltd., for donation of materials. W. H. Bowyer and M. G. Bader, J . Mat. Sci, 1972,7, 1315. M. G. Bader and I. Bell, unpublished work. A. Kelly, Strong Solids (Oxford, 1966), p. 141. * M. G. Bader and W. H. Bowyer, J. Phys. D. (Appl. Phys.), 1972, 5,2215.

ISSN:0370-9302    

DOI:10.1039/S19720200165

出版商:RSC    

年代:1972

数据来源: RSC

摘要:

GENERAL DISCUSSION Prof. W. C. Wake (City University) said: I would ask Allen et al., to investigate the interfacial region by constructing a large scale model with one polymer adhering to the other in the form of a butt joint. There was some Russian work in which this was done and a transverse section examined by fluorescence u.-v. microscopy. This showed a definite interfacial region such as the authors are postulating. Prof. M. W. Roberts (University of Bradford) said: In relation to the paper by Barnett et al., although thermodynamic data can provide valuable guide-lines as to the surface phases with which we might be dealing, there is need for caution. For example, in the reaction of N2(9) with Fe(s), the pressure of N2(9) in equilibrium with the nitride at - 300°C is - lo3 atm.We would argue that no nitride could form at, say, 1 atm. On the other hand, if a trace (e.g., Torr) of NH3(g) is present, then dissociation of the molecule to give chemisorbed nitrogen and then nitride (eqn (2)) occurs : 4Fe+4N2 -+ Fe,N (1) Fe (2) 2NH3 + Nads NaCls+3H&) - Fe4N Similarly, in the reduction of an oxide such as NiO, the equilibrium constant is well over on the H20(g) side at, say, 400°C and Ni would be formed. But in the presence of traces of H2S(g), say, 20p.p.m., then a stable sulphide layer would be formed since the equilibrium shown in eqn (3) (3) requires only about 1 in lo5 parts of H2S to be present in the H&) at 1 atm and 3Ni + 2H2S + Ni3S2 + 2H2 - 300°C. Dr. R. G. Linford (Berkeley Nuclear Lab.) said: Auger electron spectroscopy (AES) has recently been much used for studying segregation of impurities to metal surfaces? Such segregation means that the metal-matrix interface has unexpected components present that may influence the behaviour of the interface in unsuspected ways.Also, the carbon-metal interface has been much studied by AES.2 Do Barnett et al. believe that AES might have application in the study of metal-fibre composites ? Prof. W. C. Wake (City University) said: The surface treatment of carbon fibres by nitric acid should not be regarded merely as a cleaning and etching treatment to remove amorphous material. Substantial carboxylic acid activity exists on the surface after such treatment and must interact with epoxy resin or with metal matrices. see, e.g., T. M.Hass, J. T. Grant and G. J. Dooley 111, J. Vac. Sci. Techn., 1970, 7,43 ; L. A. Harris, J. Appl. Phys., 1968,39,1428 ; J. Ferrante and D. H. Buckley, ASLE Trans., 1972,15,18. see, e.g., J. T. Grant and T. W. Haas, Surface Sci., 1971,24, 332 ; J. P. Coad and J. C. Riviere, Surface Sci., 1971, 25, 609. J. H. Herrick and A. T. Laskaris, 23rd Znt. Congr. Pure AppI. Chem. (Boston, 1971). 174GENERAL DISCUSSION 175 Dr. S. J. Harris (Nottingham University) said: It is appreciated that small quantities of surface impurities, for example oxygen, sulphur, and nitrogen, can have a major effect on the kinetics of processes taking place at the carbon fibre/metal interface. The fibres used in the present work were not surface-treated, and attempts were made to minimize surface contamination prior to metal coating by baking in vacuum N/m3) at temperatures close to 900 K.Auger electron spectroscopy measure- ments on nickel surfaces have shown that carbon and sulphur were present whilst the metal was held at 1273 K (no evidence of oxygen was found). Two distinct types of fine structure were observed with the carbon peak, these have been interpreted in terms of an irreversible transition from Ni3C to graphite on heating through the temperature range 673-873 K. have indicated that carbon monoxide adsorbs on clean nickel surfaces, but as soon as contamination from carbon arises the process becomes more difficult. All of these observations have relevance to the problems associated with the fibre/nickel interface during vacuum heat treatment.However, further studies, including Auger spectroscopy, are required on this interface before the kinetics of carbon fibre degradation are elucidated. In addition, LEED studies Prof. R. Sh. Mikhail (Ain Shams University, Egypt and University of Salford) said: As Shortall and Yip have correctly pointed out, E-glass is easily hydrated by water so that its surface becomes saturated with hydroxyl groups. In a very short time after its preparation it also picks up water vapour and possibly other vapours from the ambient atmosphere and becomes completely saturated with them. We found that even free water exists in a solution adhering to the glass surface. Cleaning of the glass surface, prior to any reliable surface measurements, is now well known to be a difficult task involving ion bombardment. I certainly believe that the glass surface used by Shortall and Yip was a contaminated surface and under such circumstances, a displacement mechanism should precede the physical and/or chemical interaction with the various lubricants used and this might even take place incompletely.I should welcome their comments regarding the initial state of their glass surface and the treatment they used to clean it from pre-adsorbed matter. Mr. R. Leveson (Imperial College) said: The observation of Bowyer and Bader that interfacial bond-strength can be higher than that of the matrix material demon- strates an effect which has significance in many mechanical interface problems, including those relevant to friction. Instead of an actual modification of matrix structure in the vicinity of a high modulus fibre, a transfer of stiffness can occur due simply to applied mechanics considerations alone.Classical elasticity theory is inadequate for treating these situations and it is necessary to apply the modern theories of micro-elasticity. A simple example is to be found in the work of Sadowsky, Hsu and HussainY3 who have derived an expression for the apparent shear modulus as a function of the distance y from a plane interface between two media of absolute shear moduli GI and Gz where J. P. Coad and J. C. Riviere, Surface Sci., 1971, 25, 609. M. Onchi and H. E. Farnsworth, Surface Sci., 1968,11,203. M. A. Sadowsky, Y. C. Hsu, M. A. Hussain, 1963, Technical Report on Benet R and E Labora- tories, Watervilet Arsenal, Watervilet, N.Y., U.S.A. AD 424879.176 GENERAL DISCUSSION I being a structure-dependent modulus of micro-elastic theory (the couple-stress constant). Finkin has recently applied these ideas to the friction of very thin solid films, and has proposed that micro-elastic stiffening can inhibit frictional shear. The expression (where A is constant, and I is the micro-elastic modulus) for the coefficient of friction f as a function of the film thickness h, derived by Finkin, appears to have wide applicability. E. F. Finkin, Wear, 1971, 18, no. 3.

ISSN:0370-9302    

DOI:10.1039/S19720200174

出版商:RSC    

年代:1972

数据来源: RSC

摘要:

Electrodeposited and other Coatings Solid-Solid Interfaces-Electrodeposited and Dynamic Coatings BY J. P. G. FARR AND G. W. ROWE Department of Industrial Metallurgy, University of Birmingham, P.O. Box 363, Birmingham B15 2TT Received 26th June, 1912 Observation of the first stages in the electrodeposition of oxide and of metals on to metallic substrates is considered. The paper is concerned with applications of electron and ion microscopy and with the possibility of correlating microscopic observations with electrochemical experiments. Mechanical effects of the electrochemical treatment are mentioned briefly. The solid-solid interface in electrodeposition is rather inaccessible, and most recent research has concentrated on the metal/solution interface. Even these studies leave poorly understood the structure of the electrical double layer at solid-electrode/ electrolyte surfaces, through which both the initial and subsequent growth processes occur.Furthermore, despite a number of careful kinetic experiments on metal/metal ion exchange, there are undecided matters relevant to the initial stages in the electro- deposition of one metal on to another. A model of electrocrystallization that has gained general acceptance involves electron transfer to give " adatoms " or " adions ", followed by surface diffusion of adspecies to the various points of incorporation in the There is another possibility, that direct deposition occurs at growth sites involving diffusion zones in the electrolyteY4* and in practice mechanisms may be mixed. Bewick and Thirsk have re-emphasized that to extract kinetic constants for actual crystal growth requires information on the topography of the growing surface ; the growth site (step-line) spacing must be measured independently.If surface diffusion contributes significantly, some estimate of adion concentrations and mobilities should be possible. Adion concentrations ri at the reversible potential have been found to range over two orders of magnitude on a variety of metals; agreement between workers is not precise, the derived concentrations involve estimates of the " true " atomic surface area compared with the " apparent " macroscopic areas and on other electrode parameters.6 Surface self-diffusivities D, could be obtained from ri and the adatom flux Vi, knowing the growth site spacing I, since D, = VgZ2/rgb.In fact, I has normally been estimated rather than measured. Typical values have been cmY1 > 10-5*5 cm.2 Recent electron microscopy suggests that on copper, lead and nickel electrodes I may be less than 50 A (see fig. 1). We may conclude that kinetic experiments do not give unequivocal evidence for adspecies, but that in some electrocrystallization systems they probably are intermediates although they are not yet well characterized. The position is more obscure for the first stages of deposition of a new phase on to a foreign substrate. It is convenient to consider anodic and cathodic processes separately, and to deal first with anodic processes, for it may be necessary to remove anodic artefacts before a substrate is in an appropriate condition for cathodic deposition experiments.177178 SOLID-SOLID INTERFACES ANODIC PROCESSES Kinetics and electron microscopy are well reviewed elsewhere. Attention is drawn to a brief survey by Redhead and Armstrong of techniques for analysis of the structure and composition of adsorbed layers, which includes electron impact desorp- tion, applied by Roberts (this Discussion) to the study of polymer formation on metal substrates. Ellipsometry is capable of detecting the formation of a fraction of a monolayer equivalent of surface film in electrochemical experiments,lO* but it may not be possible to distinguish between very thin optically uniaxial films and isotropic films l 2 (see also P a ~ h l e y ~ ~ ) . of the iridium surface with examination of a fine tip of this metal in the field-ion microscope.l 4 The distinctions between layer formation and localized growth, and between phase formation (e.g., formation of phase oxide) and chemisorption can be observed directly. In particular, for iridium there is close similarity between surfaces allowed to oxidize in the air and those anodized in dilute sulphuric acid to a potential less than that required for oxygen evolution. Hoare has described the general features of the anodic oxidation of iridium. Bold and Breiter l6 performed potential sweep experi- ments which our own experiments confirm. There is good agreement with earlier work l7 that oxygen coverage is highly reversible and that an IrO, layer does not thicken appreciably because this oxide is conductive, thus allowing oxygen evolution to occur on its surface.Fortes and Ralph l8 first achieved the imaging of phase IrO, in the field ion microscope; they describe the limitations and requirements of the technique for this type of application; and they describe the structure of the interface between iridium and the oxide formed at temperatures above 500°C. Schubert and Ralph l 9 have extended this work to the examination of anodic films on iridium. They report a relatively uniform, disordered film less than five monolayers thick formed below 1.0 V with respect to hydrogen. Around 1.5 V, deep pits formed in the { 100) regions. Tyson, Southworth and Farr applying triangular sweeps, limited by the onset of hydrogen evolution cathodically and of oxygen evolution anodically, find a similar thin uniform disordered film, but after as many as 20 sweeps (730 mV/s) corrosion of { 100) regions is apparent (see fig.2). Nanis and Javet 2o also found that treating an iridium tip with 2 N H2S04 for 10 min followed by 10 min in air produced a surface perturbation extending to only 3-atom layers. A similar observation was made by Rendulic and Miiller 21 on a previously characterized tip exposed to the atmosphere for 15 min. There is therefore good agreement in the results from anodizing experiments and from ion microscopy on the oxidation of the iridium surface. It is unfortunate that few metals form satisfactory specimens for ion microscopy. Schubert and Ralph l9 have studied platinum specimens, again finding only " amorphous " anodic films up to 2.25 V (against hydrogen). The air-formed film was indistinguishable from the anodic film even within the range 0.95-1.2 V where Pt (OH), has been postulated 22 or partial Pt-0 coverage may occur.23 A slight thickening of this film occurred between 1.2 and 2.25 V but there was no localized pitting as with iridium.Phase oxide has therefore not been imaged on anodized specimens so far, even though in its anodic irreversibility platinum is very different from iridium (cf. Stonehart, Kozlowska and Conway 24). We have found promising a combination of a potentiodynamic treatment CATHODIC PROCESSES FIELD ION MICROSCOPY The ready reduction of the thin oxygen coverage suggests that the interface between an iridium substrate and a metallic deposit may be accessible in the ion-(6) FIG.1 .-Electron microscope replica images of the fine structure on copper, lead and nickel electrodes (J. P. G. Farr, G. W. Greene, A. Loong and A. McNeil). (a) Electropolished surface of copper (near (100)). x 200,000. Size of fine features is about 50 A. (b) Cu electrodeposited on to copper [To face page 178 (near (loo)), for 30 s at 10 mA/cm from acidified copper sulphate solution. x 200,000.(4 FIG. 1 .-(c) Electropolished surface of Ni (between (100) and (310)). x 150,000 (cf. electropolished Ni (lll), ref. (48)); (d) the surface of a lead electrode prepared for exchange measurements by electropolishing, followed by lightly etching in 10 % HN03. x 150,000 (1.5 mm = lOOA).(6) FIG. 2.-Field ion microscope images of an iridium tip.(J. P. G. Farr, H. N. Southworth and A. Tyson.) (a) Field-evaporated iridium tip, before electrolytic treatment. A { loo} region lies centrally, SO that along the equator there is a progression from (002) through (113) to the (111) poles. (b) The Same tip, after subjection to 50 potential sweeps from the onset of hydrogen evolution to that of oxygm in 1 MH,SO,. Between 2 and 3 monolayers werz field-evaporated from the (1 11) poles in the microscope before recording this image.(4 FIG. 2.-(c) The same tip, after the field evaporation of another 2-3 monolayers from the (1 11) poles. The (002) image is now restored, showing that the result of electrolytic treatment was a pit, some 5-6 atomic planes deep, in this region.J . P .G . FARR A N D G. W. ROWE 179 microscope. Nanis and Javet 2o performed the simplest experiment, comparing their acid treated and aerially oxidized iridium tip with tips immersed in 2N H2S04 solutions containing M iridium ions. They found much deeper perturbation than in experiments without iridium ions in the electrolyte. For comparatively short immersion times (less than 5 min) an order of relative exchange currents was suggested io(l 11) < io(lOO) < i0(210). For times longer than 5 min the treatment produced features protruding some 30 atomic layers, centred on [3 101 directions. These features were interpreted as spikes of electrodeposited iridium and although their electrochemical conditions were not well defined, Nanis and Javet were able to esti- mate an upper limit, (Tyson, South- worth and Farr have observed similar spikes in multiple potential sweep experiments in solutions containing trace amounts of iridium).Probably, local cells are estab- lished, the cathodic spikes supporting oxidation elsewhere ; there was no evidence that the localization of growth was due to the emergence of screw dislocations. More control was exercized by Rendulic and Muller 21 in electrodepositing platin- um from hot H2Pt(N02)2S04 on to iridium and tungsten ion microscope tips. Platinum deposits some 500 A thick on tungsten were not epitaxial ; they were poly- crystalline with dimensions 50-500 A and showed heavy lattice distortion. The deposition on iridium was more interesting. Epitaxy was not obtained at high current densities which resulted in [l 1 11 oriented crystals ; at the highest current densities the deposit consisted of small heavily deformed crystals.However, at lower current densities the expected epitaxy was found. The first nucleation of platinum on iridium occurred near (012) and (135) areas, where islands of about 40 atoms of platinum repeated the iridium structure. Only part of the surface grew epitaxially, the remaind- er being covered by [l 1 11 oriented crystals. Within the epitaxial layer there were cracks some 20-lOOA long at the interface and it was suggested that these were a result of lattice misfit (2.18 %). Lattice defects in the substrate sometimes continued in the epitaxial coating. Rendulic and Muller obtained poycrystalline deposits of copper or iridium and tungsten. These few experiments suggest that the ion microscope may be useful in the study of nucleation and epitaxy on characterized substrates.It is possible that electroplating may be a useful preparative technique whereby metals that would rupture under the field stress in the microscope may be supported on adequately strong substrates.21 A cm-2 for the exchange current density. TRANSMISSION ELECTRON MICROSCOPY Advances in the study of vapour-phase thin-film epitaxy have been reviewed by Pashley 2 5 and by Matthews.26 The nucleation of electrodeposited gold films on (1 11) silver substrates was studied in detail by Dickson, Jacobs and P a ~ h l e y . ~ ~ They found considerable similarity with films formed by evaporation on to similar substrates and concluded that the mechanism of growth is similar.Significant difference was observed in the influence of substrate surface irregularities on nucleation and growth, and it was suggested that this was due to local electric field effects in electrodeposition that did not apply to evaporated deposits. These authors found that islands of gold were nucleated randomly on smooth silver substrates and that the islands subsequently coalesced in a “liquid-like” way. Double positioning (i.e., two twin related orienta- tions) can occur in the (1 l l) silver substrate which therefore contains non-coherent twin boundaries perpendicular to the plane of the film. The double-positioning structure is continued in the electrodeposited gold, the boundaries remaining per- pendicular to the plane of the film until the deposit becomes a complete film.At this stage, overlapping of one orientation by the other tends to occur, giving stepped180 SOLID-SOLID INTERFACES boundaries through the gold deposit. Dislocation structures result near the double- positioning boundaries. Probably, the growth of gold on gold occurs by the lateral spread of layers no more than a few atoms thick, a process deduced for the electro- crystallization of other metals from light and electron microscopy.2* 28 There appears to be evidence for the surface migration of gold over the substrate and over the gold islands. It is particularly interesting that Dickson, Jacobs and Pashley agree with Bassett and Pashley 29 that the first lOA of deposit make no contribution to island growth.They suggest that some alloying occurs between gold and silver during the initial stages of deposition. The misfit between gold and silver is only 0.18 % and even if alloying should increase the strain, Matthews 30 has demonstrated by the absence of misfit dislocations in the early stages of growth that pseudomorphism occurs in this system. Pseudomorphism is, however, no longer regarded as necessary for epitaxy. There have been some other, less detailed, studies of electrodeposited films by transmission electron microscopy (see, e.g., ref. (31-33) and work cited in ref. (27)). However, the experiments surveyed here show how the microstructure of electro- deposits may change markedly during the post-nucleation stage, by coalescence. Caution seems appropriate in deducing the nature of the initial stages of growth from observations of the structure and topography of thickened deposits.Even in the type of experiment made by Dickson et al., care was required to avoid producing artefacts when stripping the gold by dissolving away the silver substrate, and possibly some electrochemical control would have been desirable. The substrates used by Dickson 27 and by Ives et aZ.33 were better characterized than those in many other studies on the electrocrystallization of metals. Using bulk substrates, nucleation may be governed by segregated impurity 34-36 ; the presence of substructure in single crystal substrates and the technique of annealing adopted may affect the perfection of epitaxy 36 ; slight misorientation from a low-index substrate orientation may outweigh the influence of emergent screw dislocations on the initial 38 All these effects require examination in greater detail at the earliest stages of growth and at high resolution.It is suggested that the observations of Verma and Wilman (this Discussion) using the powerful technique of grazing incidence high-energy electron diffraction 9* 2 5 on the nickel + copper system are compatible with the transmission electron microscopy described and with the transmission electron diffraction of Sullivan and Oxley (this Discussion) for the equivalent substrate orientations. There is also agreement with the findings of Keen and Farr 28 (Zu, Cu), Farr and Cliffe 39 (Co, Ni) and Browns- word and Farr 35 (Cu, Ni) concerned with later stages of epitaxial growth.ELECTROCHEMICAL EXPERIMENTS A variety of electrochemical experiments on the deposition of small amounts of metal on inert substrates is relevant to our preoccupation with the interface structure at the commencement of the formation of a new phase. It has been shown that sub- monolayer amounts of silver will deposit on gold and platinum at appreciable under- voltages. Sandoz and co-workers 40 survey a number of voltage sweep experiments for these and other metals; their own results agree with data obtained by Rogers et d 4 1 using a tracer technique. Rogers et al. have presented evidence for the deposi- tion of copper at electrode potentials anodic to the equilibrium potential on a number of Others have demonstrated this for copper on smooth 4 3 9 45 and platinized platinum 44 and for thallium on smooth platinum.46 In potential-sweep experiments, there is fair agreement that the first current peaksJ .P . G . FARR AND G . W. ROWE 181 in the cathodic excursion are due to monolayer and sub-monolayer phenomena; subsequent peaks correspond to crystallization of the new phase. (In cases of gross lattice misfit, only one peak may OCCU~.~') The first peaks are ascribed to adsorptive effects, the local bonds between, e.g., Cu and Pt, being stronger than those between Cu and Cu. These results, again, are qualitatively not incompatible with the type of observations that have been made in both the field ion microscope and by electron microscopy although equivalent systems have not yet been examined, nor have the techniques been combined.Clearly, electrochemical experiments of this type would usefully supplement experiments such as those by Menzies and Stirling (this Discussion) on alloy deposition. ELECTROCHEMISTRY AND MACHINING APPLIED TO KINETIC FRICTION STUDIES We consider dynamic coating primarily in the context of tribology and suggest that a profitable inter-relationship with electrochemistry can be established. Electro- chemical techniques permit accurate kinetic investigations, and friction is a sensitive mechanical parameter indicating surface changes. As shown by Tabor (this Discussion), intimate contact between clean surfaces -t 2 0 0 - I00 I I 1 I I I I - 2 0 4 I V (against N.H.E.) FIG. 3.-The friction of molybdenum sliders in M/10 H&04 containing M/10 thioacetamide : (a) coefficient of friction ; (6) current flowing.182 SOLID-SOLID INTERFACES results in cohesion across an interface that is comparable with that across a grain boundary.Any adsorbed surface films greatly reduce the surface forces and the friction. Probably the most important adsorbate is oxygen, either from the air or from gaseous solution in a lubricant. One way of studying such effects is by metal cutting experiments, for cutting provides large quantities of nascent surface without the necessity of high-vacuum cleaning. The time constants of adsorptive processes have been estimated from capacitance measurements after cutting aluminium in an aqueous chloride An effect attributed to ionization of the metal occurs in & .d .I c, 0 3- 0 - 4 0.6 0 .2 Oa8:! 9 - - 0 . 8 - 0 . 7 -0.6 - 0 . 5 V (against Ptd Pt) FIG. 4.-The friction of an iron-chromium alloy in molten KClfLiCI eutectic at 400°C. The friction falls as the potential is made less negative, but remains at a low value for all potentials once the film is formed. (a) Coefficient of friction ; (b) current flowing. about s, and finally molecular film formation taking several seconds. Similar experiments, involving scraping various metal surfaces at frequencies up to 500 times/s, have enabled Anderson and co- workers to distinguish between double-layer charging and faradaic currents.50 An alternative technique of scouring metal surfaces has been used, e.g., by Tomashov and Ver~hinina,~~ to ascertain the presence of oxide and the extent of adsorption.s, followed by hydrogen discharge over ELECTROCHEMICAL LUBRICATION Early studies of platinum in H2S04 showed a strong dependence of friction upon electrode potential 5 2 * 53 but this was easily obscured by small traces of sulphideJ . P . G. FARR A N D G . W . ROWE 183 impurity. Fig. 3 shows that controlled low friction can be obtained by formation of a film on molybdenum from a suitable S-containing ele~trolyte.~~ At a potential of +0.3 V, relative to a normal hydrogen electrode, the friction falls steadily over several minutes and a visible film is formed. Electron diffraction confirms the pres- ence of MoS, in this film, together with some oxides of molybdenum. The same principle can be applied for high-temperature lubrication from a molten- salt bath 5 5 (fig.4). The kinetics of the film formation can be studied by applying the galvanostatic pulse technique, following the potential developed as a function of time.55 EXAMINATION OF THE REHBINDER EFFECT Metal cutting has been used by Barlow 56 to examine the Rehbinder effect,57 i.e., the change in mechanical properties due to a surface active agent. A very small electrode was traversed along the surface of a workpiece just ahead of a cutting tool so that the stock material was anodically etched away. When the etching region coincided exactly with the zone of intense shearing at the root of the swarf, the cutting force was appreciably reduced. This was explained 5 8 as the removal of the oxide film which otherwise imposed a barrier to the emergence of dislocations. A similar effect was obtained by the use of reactive lubricants containing chlorine or iodine, which formed surface layers that were weaker than the oxide. A series of papers by Westwood 5 9 has shown the important connections between surface layers, surface potential and dislocation movement, especially in semi- conductor material.We are grateful to (the late) Prof. E. C. Rollason for the provision of laboratory facilities and for his interest; we thank Messrs Joseph Lucas Limited, Climax Molybdenum Co. Ltd. and W. Canning & Co. Ltd., for financial support. R. S. Perkins and T. N. Anderson, in Modern Aspects of Electrochemistry (Butterworths, London, 1969), vol. 5, p. 203. N. A. Hampson, in Electrochemistry IZ (Spec. Report Chem.SOC., 1972), p. 117. Reviews of metal electrodeposition have been made, e.g., by A. Bewick and H. R. Thirsk, in Modern Aspects of Electrochemistry (Butterworths, London, 1969), vol. 3, p. 327 et seq. ; J. O’M. Bockris and A. Damjanovic, Modern Aspects of Electrochemistry (Butterworths, London, 1964), vol. 3. J. O’M. Bockris and A. G. Razumney, The Electrodeposition of Metals (Plenum Press, New York, 1966) ; E. Raub and K. Miiller, Fundamentals of Metal Deposition (Elsevier Publishing Co., London, 1967) ; H. Fischer, Angew Chem., 1969, 81, 101. Reference to anodic crystallization is made by A. Bewick and H. R. Thirsk, loc. cit ; see also R. D. Armstrong, J. A. Harrison and H. R. Thirsk, Corrosion Sci., 1970, 10,679 ; M. Fleisch- mann and H. R. Thirsk, in Adu. in Electrochemistry (Interscience, New York, 1963), vol.3 ; M. Fleischmann and H. R. Thirsk, Electrochim. Actu, 1960,2,22 ; M. Fleischmann and J. A. Harrison, Electrochim. Acta 1966,11,749 ; S. K. Rangarajan, Cunud. J. Chem., 1965,43,1052. D. Vermilyea, J. Chem. Phys., 1957, 27, 819. D. J. Astley, J. A. Harrison and H. R. Thirsk, Truns. Furuduy SOC., 1968, 64, 192. C. Schnittler, Electrochim. Actu, 1970, 15, 1699. H. Kita, M. Enyo, J. O’M. Bockris, Can. J. Chem., 1961, 39, 1670. P. A. Redhead and R. A. Armstrong, Nederlunds Tijdschrift Vucuumtechn., 1970,8, 144. Electrochem., 1963, 6, 61. Symp. Faruduy SOC., 1970. ’ M. Fleischmann and H. R. Thirsk, Electrochim. Actu, 1960, 2,22. lo A. K. N. Reddy, M. A. J. Devanathan and J. O’M. Bockris, J. Electrounal. Chem. Znterfuc.l2 M. J. Dignam, M. Moskovits and R. W. Stobie, Truns. Faruduy Soc., 1971, 67, 3306, l3 F. G. Will and C. A. Knorr, 2. Elektrochem., 1960, 64, 258. l4 A. Tyson, H. N. Southworth and J. P. G . Farr, unpublished. J. P. Hoare, J. Electrounal. Chem. Interfuc. Electrochem., 1968, 18, 251.184 SOLID-SOLID INTERFACES l6 W. Bold and M. Breiter, Electrochim. Acta, 1961, 5, 169. l7 J. A. V. Butler and G. Drever, Trans. Faraday Soc., 1936,32,427. l8 M. A. Fortes and B. Ralph, Proc. Roy, Soc. A, 1968,307,431. 2o L. Nanis and P. Javet, J. Electrochem. SOC., 1968, 115, 509. 21 K. D. Rendulic and E. W. Muller, J. AppI. Phys., 1967, 38, 550. 22 J. S. Mayell and S. H. Langer, J. Electrochem. Soc., 1964, 111, 438. 23 J. P. Hoare, Electrochim. Acta., 1972, 17, 593. 24 P. Stonehart, H.A. Kozlowska and B. E. Conway, Proc. Roy. Soc. A, 1969,310, 541. 2 5 D. W. Pashley, Recent Prog. Surface Sci. (Academic Press, London, 1970), vol. 3, p. 23 et seq. 26 J . W. Matthews, Phys. Thin Films, 1967, 4, 137. ” E. W. Dickson, M. H. Jacobs and D. W. Pashley, Phil. Mag., 1965, 11, 575. 28 J. M. Keen and J. P. G. Farr, J. Electrochem. Soc., 1962, 109, 668. 29 G. A. Bassett and D. W. Pashley, J. Inst. Met., 1959, 87, 449. 30 J. W. Matthews, Phil. Mag., 1966, 13, 1207. 31 R. Weil and H. J. Rhead, J. Appl. Phys,, 1950, 21, 1068 ; Trans. Inst. Metal Finishing, 1955, 53, 60; R. Weil and H. C. Cook, J. Electrochem. Soc., 1962, 109, 295. 32 L. Reimer, Z. Metallkund., 1957,48, 390 ; 2. Phys., 1958,150,99 ; Electrochim. Acta, 1960,2, 215; J. K. Dennis and J.J. Fuggle, Trans. Inst. Metal Finishing, 1967, 45, 217. 33 E. R. Thompson and K. R. Lawless, Appl. Phys. Letters, 1966, 9, 138; A. G. Ives, J. W. Edington and G. P. Rothwell, Electrochim. Acta, 1970, 15, 1797. 34 T. B. Vaughan and H. J. Pick, Electrochim. Acta, 1960, 2, 179. 35 R. Brownsword and J. P. G. Farr, Electrochim. Acta, 1971, 16, 845. 36 P. E. Lighty, D. Shanefield, S. Weissmann and A. Shreir, J. Appl. Phys., 1963, 34, 2233. 37 L. H. Jenkins, J. Electrochem. Soc., 1970, 117, 630. 38 E. Budewski, W. Bostanoff, T. Vitanoff, Z. Stoinoff, A. Kotzewa and R. Kaishev, Electrochim. 39 D. R. Cliffe and J. P. G. Farr, J. Electrochem. SOC., 1964, 111, 299. 40 D. P. Sandoz, R. M. Peekema, H. Freund and C. F. Morrison, J. Electroanal. Interfac. Electvo- 41 L. B. Rogers, D. P. Krause, J. C. Griess and D. B. Ehrlinger, J. Electrochem. Soc., 1949,95,33. 42 R. C. DeGeiso and L. B. Rogers, J. Electrochem. Soc., 1959, 106, 433. 43 M. W. Breiter, C. A. Knorr and W. Volkl, 2. Elektrochem., 1955, 59, 681. 44 M. W. Breiter, Trans. Faraday SOC., 1969, 65, 2197. 45 B. J. Bowles, Electrochim. Acta, 1970, 15, 589. 46 B. J. Bowles, Electrochim. Ada, 1965, 10, 717, 731. 47 D, J. Astley, J. A. Harrison and H. R. Thirsk, J. Electroanal. Interfac. Electrochem., 1968, 19, 48 R. C. Pitkethly, Aspects of Adhesion (University of London Press, 1969), vol. 5, p. 248. 49 T. Hagyard and J. R. Williams, Trans. Faraday SOC., 1961, 57, 2288. 50 T. N. Anderson, J. L. Anderson, H. Eyring, J. Phys. Chem., 1969, 73, 3562. 51 N. D. Tomashov and L. P. Vershinina, Electrochim. Acta, 1970, 15, 501. 5 2 F. P. Bowden and L. Young, Research, 1950,3, 235. 53 J. O’M. Bockris and R. K. Sen, Surface Sci., 1972, 30, 237. 54 Brit. Pat. Appl. 55605169 (1969). 5 5 D. R. Atkins, J. P. G. Farr and G. W. Rowe, Trans. A.S.L.E., 1971. 56 P. Barlow, Proc. Inst. Mech. Eng., 1966, 181, 687. 57 W. I. Lichtman, P. A. Rehbinder and G. W. Karpenko, The Influence of Surface-active Material upon the Deformation of Metals (Chemical Publishing Co., New York, 1960). 5 8 E. N. da C. Andrade and R. F. V. Randall, Proc. Phys. SOC. B, 1952, 65,445. 5 9 A. R. G. Westwood, Phil, Mag., 1967, 16, 505. C. C. Schubert and B. Ralph, 18th Field Emission Symp, (Eindhoven, 1971), p. 36. Acta, 1966, 11, 1697 ; Phys. Stat. Sol., 1966, 13, 577. chem., 1970,24,165. 325.

ISSN:0370-9302    

DOI:10.1039/S19720200177

出版商:RSC    

年代:1972

数据来源: RSC