Mechanical Behavior of Liquid Route Processed SiCf/Ti Composites Under Longitudinal and Transverse Loadings

Due to the high melting point and strong chemical reactivity of titanium alloys, titanium matrix composites (TMCs) are usually processed through solid-state routes such as the foil-fiber-foil technique. An alternative method consists in the deposition of the matrix on the fibers. However, techniques such as physical vapor deposition lead to a very low deposition rate, contrary to liquid route processing using a levitating liquid alloy sphere held in a cold crucible. In order to investigate the effects of the resulting thermal shock on carbon-coated SiC fibers, and select an appropriate fiber, fibers are subjected to a pure thermal shock using a laser bench facility. These fibers are then tensile tested to failure in order to evaluate the resulting fiber strength degradation and, thus, the maximum acceptable temperature. Mechanical characterization of the liquid route processed TMC is then investigated through longitudinal and transverse tensile and creep tests at temperatures representative of aeronautical applications. The specimens, unbroken after long-duration creep tests, are then subjected to tensile loading to failure: conditions representative of service, i.e., short-time overspeeding of a gas turbine. Finally, interpretation of the mechanical tests through micrographical and microfractographical examinations is focused on the identification of the deformation and failure mechanisms specific to the liquid route processed composite, e.g., nucleation, under either longitudinal or transverse loadings, of internal cracks in the α-phase of the titanium-based matrix, explained through a physical model involving a high shear stress and normal stress combination, leading to cleavage.

     

Strength degradation of SiC fiber during manufacture of titanium matrix composites by plasma spraying and hot pressing

Titanium matrix composites (TMCs) reinforced with Sigma 1140+ SiC fiber have been manufactured by a combination of low pressure plasma spraying (LPPS spray/wind) and simultaneous fiber winding, followed by vacuum hot pressing (VHP). Fiber damage during TMC manufacture has been evaluated by measuring fiber tensile strength after fiber extraction from the TMCs at various processing stages, followed by fitting of these data to a Weibull distribution function. The LPPS spray/wind processing caused a decrease in mean fiber strength and Weibull modulus in comparison with as-received fibers. A number of fiber surface flaws, primarily in the outer C layer of the fiber, formed as a result of mechanical impact of poorly melted particles from the plasma spray. Coarse feedstock powders promoted an increase in the population of fiber surface flaws, leading to significant reduction in fiber strength. The VHP consolidation promoted further development of fiber surface flaws by fiber bending and stress localization because of nonuniform matrix shrinkage, resulting in further degradation in fiber strength. In the extreme case of fibers touching, the stress concentration on the fibers was sufficient to cause fiber cracking. Fractographic studies revealed that low strength fibers failed by surface flaw induced failure and contained a large fracture mirror zone. Compared with the more widely investigated foil-fiber-foil route to manufacture TMCs, LPPS/VHP resulted in less degradation in fiber strength for Sigma 1140+ fiber. Preliminary results for Textron SCS-6 fiber indicated a much greater tolerance to LPPS/VHP damage.     

On creep of unidirectional fiber composites with fiber damage

The roles of broken fibers in the creep of continuous fiber reinforced composites are studied theoretically. The unidirectional fiber composite is modeled using a cylindrical cell consisting of a single broken fiber and a shell of the fiber material embedded in an elastic-power law creeping matrix. The time-dependent creep behavior of the composite is calculated using finite elements; both longitudinal and transverse loads are considered. It is shown that when fibers are broken, the increase in the overall creep strain of the composite and the axial stress in the intact fibers can be significant. It is also demonstrated that applied transverse tension can reduce the composite creep strain and the normal stress in the fibers; the opposite is true with applied transverse compression. Matrix plasticity is found to have very limited effect on the creep behavior of the composite. The accuracy of the McLean formula for undamaged composite is also examined.     

Creep of Geotextiles Using Time–Temperature Superposition Methods

A temperature-accelerated tensile testing program was conducted in this study to characterize a woven polypropylene geotextile regarding its long-term stress–strain response, creep failure, and tensile strength remaining after sustained creep loading. Specimens were tested in a load frame that allowed control of multistage load paths. Consistent with current standards for rapid loading of geotextiles, roller-type grips capable of accommodating wide-width (200-mm) specimens were used in this study. The test program included: (i) Rapid loading tensile tests at room and elevated temperatures; (ii) conventional and temperature-accelerated creep tests; and (iii) rapid loading tensile tests conducted after sustained creep loading. Creep strain data for periods beyond 100 years were collected at various load levels using 8-h long tests involving the stepped isothermal method. The creep–failure curve, traditionally defined as time to rupture for sustained creep loading at various load levels, was defined in this study as the deviation of the creep curve from linear behavior in a semilogarithmic scale. A new approach was implemented to quantify and reference the residual tensile strength obtained from rapid loading at elevated temperatures of specimens that had been subjected to sustained creep. In spite of the significant slope in the creep-failure curve of the geosynthetic tested in this study, the residual tensile strength exceeds 90% of the ultimate tensile strength. An alternative to the current design approach, which involves use of creep-failure curves to define creep reduction factors is proposed. This involves use of creep-induced tensile strength loss, creep failure, and creep strains in the design of reinforced soil structures.     

Creep deformation and damage in a continuous fiber-reinforced Ti-6Al-4V composite

Mechanisms of longitudinal creep deformation and damage were studied in an eight-ply unidirectional-reinforced SCS-6/Ti-6Al-4V composite. The composite was creep tested in air under constant tensile load at temperatures from 427 °C to 650 °C and stresses from 621 to 1380 MPa.In situ acoustic emission (AE) monitoring and post-test metallographic evaluation were used to study fiber fracture and damage during creep. At low creep stresses, creep rates continuously decreased to near-zero values. This was attributed to a mechanism of matrix relaxation and the time-dependent redistribution of load from the ductile matrix to the elastic fibers. At higher stresses, progressive fiber overload occurred during creep loading. In this case, the composite exhibited a stage of decreasing creep rate (due primarily to matrix relaxation), followed by a secondary stage of nearly constant creep rate due to fiber fracture. The results indicate that interfacial oxidation damage and inefficient load transfer at elevated temperatures significantly decreased the capability of broken fibers to carry load. As a result, additional time-dependent stress redistribution occurred in the composite, which was responsible for the secondary creep stage.     

Degradation of residual strength in SCS-6/Ti-15-3 Due to fully reversed fatigue

Little attention has been given to residual strength degradation in titanium matrix composites (TMCs) after exposure to fatigue loading. To address this problem, fatigue tests on SCS-6/Ti-15-3 were performed to investigate the fatigue life and residual strength behavior of TMCs with different fiber volume fractions. Results indicate that fiber volume fraction seems to have an effect on both of these quantities. Lower fiber percentages result in a material where the characteristics of the matrix, such as hardening or cracking, play a much larger role in the composite response. Fatigue lives were not affected by fiber volume fraction at higher strain ranges, but lower fiber volume fractions resulted in shorter fatigue lives at lower strain values. Also, a slight increase in residual strength occurred up to 75 pct of fatigue life, for the lower-fiber volume fraction material. Despite these distinctions between specimens with different fiber contents, all specimens tested retained the majority of their strength prior to failure. This article is based on a presentation made in the symposium “Fatigue and Creep of Composite Materials” presented at the TMS Fall Meeting in Indianapolis, Indiana, September 14–18, 1997, under the auspices of the TMS/ASM Composite Materials Committee.     

Effects of thermal residual stresses and fiber packing on deformation of metal-matrix composites

The combined effects of thermal residual stresses anmd fiber spatial distribution on the deformation of a 6061 aluminum alloy containing a fixed concentration unidirectional boron fibers have been analyzed using detailed finite element models. The geometrical structure includes perfectly periodic, uniformly spaced fiber arrangements in square and hexagonal cells, as well as different cells in which either 30 or 60 fibers are randomly placed in the ductile matrix. The model involves an elastic-plastic matrix, elastic fibers, and mechanically bonded interfaces. The results indicate that both fiber packing and thermal residual stresses can have a significant effect on the stress-strain characteristics of the composite. The thermal residual stresses cause pronounced matrix yielding which also influences the apparent overall stiffness of the composite during the initial stages of subsequent far-field loading along the axial and transverse direction. Furthermore, the thermal residual stresses apparently elevate the flow stress of the composite during transverse tension. Such effects can be traced back to the level of constraint imposed on the matrix by local fiber spacing. The implications of the present results to the processing of the composites are also briefly addressed.     

Cyclic creep-rupture behavior of three high-temperature alloys

Some important characteristics of the cyclic creep-rupture curves have been studied for the titanium alloy 6Al-2Sn-4Zr-2Mo at 900° and 1100°F (755 and 865 K), the cobalt-base alloy L-605 at 1180°F <910 K), and for two hardness levels of 316 stainless steel at 1300°F (980 K). Such curves have been used successfully in a previous investigation for the evaluation and prediction of strain-cycling life at elevated temperatures within the creep range. The cyclic creep-rupture curve relates tensile stress and tensile time-to-rupture for strain-limited cyclic loading and has been found to be independent of the total strain range and the level of compressive stress employed in the cyclic creep-rupture tests. The cyclic creep-rupture curve was always found to be above and to the right of the conventional (constant load) monotonic creep-rupture curve by factors ranging from 2 to 10 in time-to-rupture. This factor tends to be greatest when the creep ductility is large. Cyclic creep acceleration was observed in every cyclic creep-rupture test conducted. The phenomenon was most pronounced at the highest stress levels and when the tensile and compressive stresses were completely reversed. In general, creep rates were found to be lower in compression than in tension for equal true stresses. The differences, however, were strongly material dependent.     

Transverse creep and stress-rupture of borsic-aluminum composites and borsic-aluminum composites containing stainless steel and titanium

The transverse creep and stress rupture behavior of a number of Borsic®-aluminum composites was investigated at temperatures from 200° to 400°C. The cpmposites studied consisted of nominally 50 vol pct Borsic fiber and included matrices of 6061, 2024, 2219, and 5052 aluminum alloys. The effect of heat treatment was studied in the heat-treatable alloys. Where transverse composite behavior differed from matrix alloy behavior, the difference was found to be due primarily to a change in fracture mode at higher matrix strength levels from matrix failure to one which involves longitudinal fiber splitting. Of the four basic matrix alloys tested, the best creep resistance was obtained with the 2024 matrix. Additional improvement of transverse creep and stress rupture resistance was realized by incorporating transverse reinforcements such as SAP alloy foil, titanium alloy foil, and 0.002 in. stainless steel wire in the composites. These reinforcements made possible good transverse properties at 400°C with density increases of ≤15 pct. The two best additions were 21 pctβ III titanium foil and 6 pct AFC-77 stainless steel wire. A transverse fracture mode incorporating longitudinal fiber splitting was documented and characterized, and its effect on composite behavior determined. The use of nonsplitting fibers such as 5.6 mil B and 5.7 mil Borsic in preventing this fracture mode was investigated.     

Fatigue in selectively fiber-reinforced titanium matrix composites

Many applications of the Ti alloy matrix composites (TMCs) reinforced with SiC fibers are expected to use the selective reinforcement concept in order to optimize the processing and increase the cost-effectiveness. In this work, unnotched fatigue behavior of a Ti-6Al-4V matrix selectively reinforced with SCS-6 SiC fibers has been examined. Experiments have been conducted on two different model panels. Results show that the fatigue life of the selectively reinforced composites is far inferior to that of the all-TMC panel. The fatigue life decreases with the decreasing effective fiber volume fraction. Suppression of multiple matrix cracking in the selectively reinforced panels was identified as the reason for their lack of fatigue resistance. Fatigue endurance limit as a function of the clad thickness was calculated using the modified Smith-Watson-Topper (SWT) parameter and the effective fiber volume fraction approach. The regime over which multiple matrix cracking occurs is identified using the bridging fiber fracture criterion. A fatigue failure map for the selectively reinforced TMCs is constructed on the basis of the observed damage mechanisms. Possible applications of such maps are discussed.     

Transverse and cyclic thermal loading of the fiber reinforced metal-matrix composite SCS6/Ti-15-3

The transverse properties of a SiC fiber reinforced Ti alloy matrix composite subjected to transverse mechanical and cyclic thermal loading have been investigated. Fibers and matric have a mismatch in the coefficients of thermal expansion that induces thermal stresses in addition to those caused by mechanical loading. When fluctuations occur in the operating temperature the thermal stresses change and this could cause an incremental accumulation of plastic strain or increase in creep rate. The composite under consideration has a modest mismatch and it was found that the strain accumulation is caused by creep deformation in the matrix at the high temperature portion of the thermal cycles. In the early stages of the deformation for low transverse loading the interface is in compressive contact and the creep rate is accelerated by the cyclic thermal stresses. After debonding has occurred the cyclic thermal stress component is diminished and the creep rate is given by a matrix with holes.     

Transverse elastic moduli of unidirectional fiber composites with interfacial debonding

A mathematical model is developed to predict the transverse elastic moduli of unidirectional fiber composites. Two cases of fiber/matrix interfacial bondings are investigated: perfect bonding and complete debonding,i.e., no transferring of tensile stress on surfaces where the fiber and matrix materials are separated. Fibers are assumed to exhibit transverse isotropy, and the elastic interactions among fibers are also considered. In the analysis of the completely debonded case, the cavity formation model is adopted and the original fiber and surrounding cavities are replaced by an imaginary anisotropic inclusion. Closed form solutions of effective elastic moduli are obtained for transverse tensile and compressive moduli, and transverse shear moduli. The limiting case of fiber-like voids is also studied. Numerical results are presented for two metal matrix composites, where the optimal wetting between the fiber and matrix is difficult to be obtained and hence the interfacial bonding and transverse property have been major concerns. Formerly Visiting Associate Professor at The University of Delaware     

Transverse creep of SiC/Ti-6Al-4V fiber-reinforced metal matrix composites

The transverse creep response of an 8-ply SiC (SCS-6)/Ti-6Al-4V composite was measured at 482 °C from 69 to 276 MPa. Creep samples with fibers exposed at the edges as well as specimens with fully embedded fibers were tested under stepped loading conditions with increasing load. The response of each sample geometry was compared with creep data from the unreinforced matrix (‘neat’ material). The samples with exposed fiber ends exhibited minimum creep rates that were higher than those of the neat material at all stresses, and the stress exponent was slightly large than the neat material. The embedded fiber samples possessed minimum creep rates that were smaller than the neat material at low stresses (<115 MPa), but became equivalent to the exposed fiber data at the highest stress (276 MPa). The apparent stress exponent for the embedded fiber composite was significantly larger than the neat material. The exposed fiber test data were well represented by a standard Crossman analysis, where the fibers were considered to have completely debonded. A stress singularity in the interfacial region at the sample edge is responsible for this behavior. The Crossman model was modified to incorporate the effect of a finite interface strength (120 MPa), and this was used to describe the response of the samples with embedded fibers. A reasonable fit to this representation was obtained. However, the measured minimum creep rate at the lowest stress was significantly lower than that predicted by the Crossman analysis for fully bonded fibers. This article is based on a presentation made in the symposium “Fatigue and Creep of Composite Materials” presented at the TMS Fall Meeting in Indianapolis, Indiana, September 14–18, 1997, under the auspices of the TMS/ASM Composite Materials Committee.     

Micromechanical modeling of unidirectional continuous sigma fiber-reinforced Ti-6Al-4V subjected to transverse tensile loading

A micromodeling analysis of unidirectionally reinforced Ti-6-4/SM1140+ composites subjected to transverse tensile loading has been performed using the finite-element method (FEM). The composite is assumed to the infinite and regular, with either hexagonal or rectangular arrays of fibers in an elastic-plastic matrix. Unit cells of these arrays are applied in this modeling analysis. Factors affecting transverse properties of the composites, such as thermal residual stresses caused by cooling from the composite processing temperature, fiber-matrix interface conditions, fiber volume fraction, fiber spacing, fiber packing, and test temperature are discussed. Predictions of stress-strain curves are compared with experimental results. A hexagonal fiber-packing model with a weak fiber-matrix interfacial strength predicts the transverse tensile behavior of the composite Ti-6-4/SM1140+ most accurately.     

The transverse creep deformation and failure characteristics of SCS-6/Ti-6AI-4V metal matrix composites at 482 °C

While continuous fiber, unidirectional composites are primarily evaluated for their longitudinal properties, the behavior transverse to the fibers often limits their application. In this study, the tensile and creep behaviors of SCS-6/Ti-6Al-4V composites in the transverse direction at 482 °C were evaluated. Creep tests were performed in air and argon environments over the stress range of 103 to 276 MPa. The composite was less creep resistant than the matrix when tested at stress values larger than 150 MPa. Below 150 MPa, the composite was more creep resistant than the unreinforced matrix. Failure of the composite occurred by the ductile propagation of cracks emanating from separated fiber interfaces. The environment in which the test was performed affected the creep behavior. At 103 MPa, the creep rate in argon was 4 times slower than the creep rate in air. The SCS-6 silicon-carbide fiber’s graphite coating oxidized in the air environment and encouraged the separation of the fiber-matrix interface. However, at higher stress levels, the difference in behavior between air- and argon-tested specimens was small. At these stresses, separation of the interface occurred during the initial loading of the composite and the subsequent degradation of the interface did not affect the creep behavior. Finally, the enrichment of the composite’s surface by molybdenum during fabrication resulted in an alloyed surface layer that failed in a brittle fashion during specimen elongation. Although this embrittled layer did not appear to degrade the properties of the composite, the existence of a similar layer on a composite with a more brittle matrix might be very detrimental.     

Thermal conductivity and thermal expansion of graphite fiber-reinforced copper matrix composites

The high specific conductivity of graphite fiber/copper matrix (Gr/Cu) composites offers great potential for high heat flux structures operating at elevated temperatures. To determine the feasibility of applying Gr/Cu composites to high heat flux structures, composite plates were fabricated using unidirectional and cross-plied pitch-based P-100 graphite fibers in a pure copper matrix. Thermal conductivity of the composites was measured from room temperature to 1073 K, and thermal expansion was measured from room temperature to 1050 K. The longitudinal thermal conductivity, parallel to the fiber direction, was comparable to pure copper. The transverse thermal conductivity, normal to the fiber direction, was less than that of pure copper and decreased with increasing fiber content. The longitudinal thermal expansion decreased with increasing fiber content. The transverse thermal expansion was greater than pure copper and nearly independent of fiber content. formerly with NASA Lewis Research Center, is retired David L. McDanels, This article is based on a presentation made in the symposium “High Performance Copper-Base Materials” as part of the 1991 TMS Annual Meeting, February 17–21, 1991, New Orleans, LA, under the auspices of the TMS Structural Materials Committee.     

Creep of Polymer Matrix Composites. II: Monkman-Grant Failure Relationship for Transverse Isotropy

This research concerns polymer matrix composite (PMC) materials having long or continuous reinforcement fibers embedded in a polymer matrix. The objective is to develop comparatively simple, designer friendly constitutive equations intended to serve as the basis of a structural design methodology for this class of PMC. Here (Part II), the focus is on extending the damage/failure model of an anisotropic deformation/damage theory presented earlier. A companion paper (Part I) by the writers deals with creep deformation of the same class of PMC. The extension of the damage model leads to a generalization of the well known Monkman/Grant relationship to transverse isotropy. The usefulness of this relationship is that it permits estimates of (long term) creep rupture life on (short term) estimates of creep deformation rate. The current extension also allows estimates of failure time for various fiber orientations. Supporting exploratory experiments are defined and conducted on thin-walled specimens fabricated from a model PMC. A primary assumption in the damage model is that the stress dependence of damage evolution is on the transverse tensile and longitudinal shear traction acting at the fiber/matrix interface. We conjecture that a supplemental mechanism of failure is the extensional strain in the fiber itself. The two postulated mechanisms used in conjunction suggest that an optimal fiber angle may exist in this class of PMC, maximizing the time to creep failure.     

Effects of Loading Rate and Duration on Axial Behavior of Concrete Confined by Fiber-Reinforced Polymer Sheets

In this study, 18 concrete cylinder specimens were tested either under uniaxial compression at different loading rates or exposed to sustained axial stresses after being jacketed externally with carbon-fiber-reinforced polymer (CFRP) sheets. The specimens were cast using medium strength concrete. All the specimens had identical dimensions and level of confinement. Loading rate and applied sustained stress level were the main test parameters. Applied loading rate varied between 0.0002 and 0.04 strain/min. Four stress levels between 0.52 and 0.85fcc′ (0.90 and 1.46fco′) were used in short-term creep tests. Test results showed that the stress-strain behavior of CFRP confined concrete was influenced by the change in loading rate, and CFRP confinement provided considerable increase in the creep performance of concrete. The strength enhancement was more pronounced for specimens loaded at higher strain rates, while specimens loaded at slower strain rates exhibited better deformability. Results obtained from short-term monotonic loading tests were also compared with the results of two analytical approaches originally developed for plain concrete. None of the specimens failed during the short-term creep tests. However, the lifetime of the specimen, which was subjected to 0.85fcc′ (1.46fco′) sustained axial stress, was predicted as 20 days. Results of residual strength tests showed that specimens did not have any strength loss due to sustained loading.     

Effect of normalization temperature on the creep strength of modified 9Cr-1Mo steel

The effect of normalization temperature from 850 °C to 1050 °C on the structure and creep-rupture properties of modified 9Cr-1Mo steel was studied. Normalization at temperatures below 925 °C resulted in structures containing significant polygonized, recovered ferrite. The ferrite structures had poor creep-rupture strength: roughly two orders of magnitude increase in minimum creep rate or decrease in rupture life for 850 °C compared to 1050 °C normalization at test conditions of 600 °C and 145 MPa. Room-temperature strength and hardness were also reduced. The microstructure after normalization at the standard 1050 °C temperature consisted of tempered martensite with fine M₂₃C₆ carbide along prior austenite and lath boundaries and fine MX carbonitride precipitates within the laths. Normalization at temperatures between 925 °C and 1000 °C also resulted in reduced creep strength in comparison with 1050 °C normalization, even though tempered martensite microstructures were formed and little change in room-temperature strength was observed; the reduction was attributed to subtle differences in the MX precipitates. The effect of reduced normalization temperature was more pronounced for higher-temperature, lower-stress creep-rupture conditions.     

Fiber fracture during the consolidation of metal matrix composites

A detailed study of conditions leading to fiber fracture during the consolidation of Ti14wt%Al21wt%Nb/SiC (SCS-6) composite monotapes has been conducted. For this continuous fiber reinforced composite system, the incidence of fracture increases with consolidation rate at higher process temperatures. Increasing consolidation temperature at a fixed pressure reduces the number of breaks per unit length of fiber. Examination of partially densified compacts has revealed the existence of significant fiber bending and ultimately fracture due to monotape surface roughness (asperities) which places the fibers in three point bending. A representative volume element has been defined for the consolidating lay-up and its response analyzed to predict the fiber deflection (and hence probability of failure) when the surface asperities deform either by plasticity or by steady state creep. The relationships between fiber fracture and process conditions predicted using the volume element are similar to those observed experimentally. The cell analysis suggests that fiber fracture is decreased by increases in fiber stiffness, strength, and diameter and by decreases in matrix yield and creep strength and monotape surface roughness.