It is well known that the mechanical properties of laminated composites depend on the stacking sequences. As the bending modulus of elasticity is affected remarkably by the stacking sequences, the stress and strain analyses for laminated composite structures require a special technique. Therefore, a new analytical method has been proposed in order to calculate the mechanical behavior of laminated structures in this paper. The proposed method has been applied for a laminated composite structure as a numerical example. As a result, it has been recognized that the numerical behavior of the composite structure with anisotropic properties under mixed loads of bending and tension can be analyzed by the proposed method, even if it has two different moduli of elasticity of bending and tension. In addition, CPU time of FEM based on the proposed method can be reduced remarkably as compared with ordinary FEM.
When structures are designed, the stress analysis using FEM is carried out at first, and the most desirable structure is obtained from the computational results. When the effect on human sense is considered for constructing composite structures as in the cases of tennis racket, golf shaft and fishing rod, the distribution of curvature under stress has to be kept in mind because the deformation at use is an important design parameter. A new method is suggested to realize the deformation with a uniform distribution of curvature in this paper. In order to evaluate this proposed method, the computational results were compared with the experimental ones. It was recognized that these results agreed very well. As an example, the new method was applied to design a composite chair made from a stampable sheet. As the result, an optimum chair design with a uniform distribution of curvatures was obtained.
A fracture criterion based on the concept of linear notch mechanics for predicting strength in static load is subjected to further theoretical and experimental scrutiny. An experimental program is presented which examines the effects of notch-root radius and fiber orientation on the fracture behavior of FRP plates. This is accomplished by obtaining experimental data in tension and bending tests on a glass cloth/epoxy laminate (JIS: EL-GEM) containing notches with a wide range of notch-root radii. To examine the effect of fiber orientation on the fracture behavior, the stress distributions near the notch root of notched FRP plates were determined by finite element analysis. The experiment showed that the nominal stress at fracture decreased with decreasing notch-root radius for a constant notch depth. It has been verified that the maximum elastic stress at the notch root when the specimen fails, σmax, c, is governed by the notch-root radius ρ only and independent of type of static load. In other words, the one-to-one relation curve between σmax, c and ρ for notched FRP plates in tension tests agrees well with the curve in bending tests. The fracture characteristics mentioned above was independent of fiber orientation. On the basis of the concept of linear notch mechanics, the experimental results can be clearly explained.
Influence of water on the shear strength of adhesive-bonded single lap joint of CFRP was investigated. A paste type of epoxy adhesive and a film type adhesive were used. The lap joints were immersed in an acidic water of pH 3, a salt water simulated seawater and a distilled water at temperatures from 30°C to 80°C for a period from 210 hours to 3000 hours. After the immersion, the adhesive shear strength was measured by tensile test. It has become clear that, the shear strength of the film-adhesive joint which had higher shear strength before the immersion decreased more than that of the paste-adhesive joint. When the adhesive was the same, the degree of deterioration in the shear strength of CFRP was not affected by the kind of water. The apparent activation energy (ΔEa) for 15, 25 and 50% deterioration of the shear strength by immersion in water was calculated. Its values were 28-36kJ/mol for the paste-adhesived joint and 45-65kJ/mol for the film-adhesived joint.
Tensile and flexural fatigue properties of glass/aramid and glass/carbon fiber hybrid laminates in polyester and vinylester resins have been evaluated systematically and presented as basic design data in this paper. Seventeen kinds of laminates including basic laminates and combined hybrid laminates were fabricated by hand-lay-up technique and comparatively evaluated. The general hybrid effect was characterized systematically from the viewpoint of hybridization merit of FRP when aramid fiber, carbon fiber, glass/aramid fiber or glass/carbon fiber hybrid reinforcements were substituted for glass fiber reinforcements in the standard glass fiber laminate constitutions for marine use. The test data indicated that the S-N curves of hybrid FRP laminates were not influenced very much by the difference in fracture strain of matrix. It was shown that fatigue strength at 106 cycles decreased approximately 40 to 75 percent of initial static strength. It was also shown that the S-N curves combined laminates could be predicted from the data of basic laminates utilizing a simple rule of mixture.
Although several methods have been proposed to evaluate interfacial bonding properties of composite material, they have some disadvantages and a useful evaluation method has not been established. The purpose of this paper is to propose and show an evaluation method of bonding properties and damage of the interface of composites. From the point of view of the rheological properties of composites in which interfacial bonding largely affects, the analyses were done by using a standard solid model. It was shown that the difference in interfacial bonding corresponded to the coefficients of elements in the model. Two types of treatment on fibers, which gave different interfacial bonding properties, reflected on the coefficients of elements. The evaluation was made during the fatigue process of composites, and gave useful informations on fatigue mechanism other than simple decrease in modulus. This method was also proved to give fatigue life prediction.
The subperforation flatwise impact damage resistance of carbon fiber/PEEK (CF/PEEK) laminated composites (APC-2/AS4) was evaluated under low and high velocity impact tests. The stacking sequences of the specimens were [0°/30°/0°/-30°]s, [0°/60°/0°/-60°]s and [0°/45°/90°/-45°]s. The drop weight test system as the low velocity impact test and the air gun test system as the high impact test were used to consider the effect of impact velocity on damage area. The damage areas after impact were measured with the ultrasonic C-Scan. The impact damage resistance was assessed as the ratio of the damage area (DA) and the impact energy (IE). To estimate the effect of the stacking sequence related to the impact resistance (DA/IE), the stacking parameter β, which indicates the difference in in-plane stiffness between the adjacent laminae, is proposed. From the experimental results, it is obtained that the stacking parameter β is valid to select better stacking sequence of composite laminate for impact resistance.
A pyrolysis method was applied for recycling of carbon fiber in CFRP. From the results of thermo-gra-vimetry and differential thermal analysis, carbon fiber in CFRP began to be oxidized after almost all the resin in CFRP disappeared by pyrolysis. Consequently, it was possible to recover carbon fiber with an almost 100% yield as a residue by pyrolyzing CFRP under appropriate conditions. The recovered carbon fiber has good mechanical properties and a surface condition as those of the original, enough to be used again.
The effects of hydrogen on cyclic superelasticity and other mechanical properties of Ti-Ni-Cu alloy were studied in comparison with Ti-Ni alloy. In both alloys hydrogen did not affect the accomodation process of martensite crystals, but affected only the plastic deformation process by interacting with dislocations. It seems that hydrogen which affects superelasticity is in a dissolved and/or weakly trapped state. In Ti-Ni-Cu alloy, the critical hydrogen content which lowers cyclic superelasticity was higher than that of Ti-Ni alloy. This was attributed to the difference in surface properties of these alloys. In conclusion, Ti-Ni-Cu alloy can be used under a higher hydrogen absorbing condition compared to Ti-Ni alloy.
The effects of the modified austemper (designated MA) on the microstructure and mechanical properties of silicon-modified 4340 steel have been studied with the aim of developing a MA steel for ultrahigh strength applications. The microstructure of the MA steel consisted of triple phases of carbide-free upper bainite, light-tempered martensite and retained austenite. This microstructure was produced by a partial transformation at either 593 or 623K for a required time, followed by oil quenching and subsequent tempering at 473K after austenitization at 1173K (designated 593K-MA, 623K-MA, respectively). Compared to the quenched and tempered steel, the 593K-MA treatment improved the fracture toughness (KIC) at high strength level of 2000MPa, owing to an increase in total elongation and also improved the Charpy impact energy. According 623K-MA treatment, its KIC was dramatically developed at strength level of 1700MPa, owing to a remarkable increase in total elongation and Charpy impact energy, as compared to the QT steel. Compared to the corresponding conventional austemper treatment, its KIC becomes significantly increased with a remarkable increase in strength.
It has been pointed out by the authors that in conventional fracture toughness KI tests of crack-rate sensitive, perfectly brittle materials, a crack begins to extend before a maximum load (fracture load, Pmax) appears in a load-deflection curve. The amount of the extension depends upon the compliance of the testing machine, initial crack length x0 and specimen geometry. Therefore, the obtained KI-values (an apparent KI; KIi) are always underestimated as far as the values are evaluated by using Pmax and x0. The present paper aims to show how the KIi-value depends on x0 and specimen geometry. Both numerical simulation and experimental data on a glass and an alumina ceramic show that the KIi-value increases with x0 and also the value depends upon specimen geometry (3 point bending, CT and single edge cracked tensile specimens).
The effect of crack-length on the fracture strength σ0 of rate-sensitive, perfectly brittle materials was studied by numerical simulations in 3 point bending (3PB) and single edge cracked tensile (SECT) specimens with a small crack. Simulations including rate effect show that c-value in σ0∝x0-c relation (x0; non-dimensional initial crack length) is smaller than the value (c=0.5) which is reduced from linear fracture mechanics without any rate effect. The difference in σ0 between 3PB and SECT specimens and the size effect on σ0 are also studied. All of the simulated results show quantitatively good coincidence with the extensive experimental works reported previously on brittle materials such as glasses and ceramics.
It is desirable to design a structure and its members with a reasonable reliability. This study is intended to propose a method of estimating the reliability of the flexural design strength of double reinforced concrete member with a rectangular section by using the probabilistic method, considering variations in strength of reinforcing bars and concrete. This method is an advanced one from the method proposed previously for the strength of single reinforced flextural member with a specified reliability. By applying this method, the strength values provided in the ACI and JSCE Codes for various combinations of both materials used in general were evaluated. It has been proved that the strength defined by the ACI Code is partly insufficient to obtain a specified reliability in case of high steel ratio, whereas that by the JSCE Code is satisfactory for any comoration.
Static compression tests of unidirectional AFRP with a circular hole are carried out. The experimental result shows that the compressive strength of unidirectional AFRP containing a circular hole depends on the hole diameter. As the hole diameter increases, the compressive strength tends to be lower in spite of a decrease in the stress concentration factor αT. A good agreement between the experimental and the predicted compressive strength is observed by the application of two failure criteria. For the specimen with small hole diameter, the strength can be predicted by the Soutis' theory based on the stress distribution around the hole and linear fracture mechanics. For the specimen with large hole diameter, the strength can be predicted by the point-stress criterion (d0=1.5mm). Finally, the failure mechanism of the unidirectional AFRP subjected to compressive loads is discussed.
In-plane Mode II fracture toughness tests of unidirectional GFRPs are conducted using a four-point shear loading test method. In the previous paper, the effect of the pre-crack length on the fracture toughness was investigated, and the advantage of this test method was also discussed. In this study, the same test method is used to evaluate the fracture toughness of the unidirectional GFRP. The normalized stress intensity factor concerning an orthotropic body of the unidirectional GFRP is calculated by a finite element method. The mode II fracture toughness is found to be constant regardless of pre-crack length. Using the finite element method, the propagation of the damage zone at the initial failure process is obtained. The analytical mode is shown to be quite effective and the initial failure process is well simulated by this modelization.
The bending fatigue strength of Ti-6Al-4V alloy was investigated in the temperature range of 20°C to 450°C. A small artificial defect was introduced in the form of a small hole drilled in the surface of an unnotched specimen. The diameter of the hole was about 0.1mm and its depth was virtually the same as the diameter. The fatigue strength σF of the unnotched specimen was compared with the σFN of specimens with a small hole. The following results were obtained. (1) The fatigue strength is remarkably decreased even by such a small hole. The decrements of the fatigue strength caused by the hole, (σF-σFN)/σF are about 40%, 32%, 27% and 47% at 20°C, 250°C, 350°C and 450°C, respectively. The values of σFN predicted by the Murakami's √area parameter model are about 3%-17% smaller than the experimental results. The prediction seems to give a conservative estimation. (2) In the temperature range of 350°C to 450°C, both σF and σFN are increased by solution hardening, dynamic strain aging and so on. Since the static yield strength is hardly influenced by such a strengthening factor, the temperature dependence of the yield strength may be used as the reference for evaluating the increments in the fatigue strength. It is concluded that the maximum increments of σF and σFN come to about 60%.
In order to examine the influence of acidic rain on the CFRP for outdoor use, three kinds of CFRP were immersed in an acidic water of pH 3 at various temperatures for different immersion durations. The weight gain and interlaminar shear strength (ILSS) were measured. The progress of deterioration was about the same or a little faster in an acidic water compared with that in a distilled water. Even if the matrix was the same, the degree of deterioration in ILSS varied among CFRPs containing different kinds of fibers. The apparent activation energy (ΔEa) for 50-75% deterioration of ILSS in an acidic water was 57-60kJ/mol, while that in a distilled water was 41-62kJ/mol.
A study has been made on the thermal and rheological properties of the blends of Polycarbonate (PC) and p-Quarterphenyl derivatives (QP). QP, known as a thermotropic liquid crystalline material, does not decompose thermally at the normal molding temperatures for engineering polymers. The aim of this study is to clarify how the phase structures of these blends affect their rheological properties. The results obtained are as follows: (1) PC/QP system is well represented by an UCST phase diagram. The phase containing an excess amount of QP consists of crystals. (2) In the incompatible region, the existing crystals cause long relaxation behavior. When the crystals melt at high temperature, the viscosity of the system decreases considerably. (3) The degree of decrease in viscosity of the system is dependent on the chain length of the substituent of QP.
Recently, many attentions have been paid to an optical fiber as a sensor for intelligent structures. In present study, the optical fiber is used as a crack sensor for mortar beam. The sensing properties have been investigated by three-point bending test. Fresnel reflection of four optical fibers, which have been stuck on the surface of mortar beam with epoxy resin, is observed with OTDR (Optical Time Domain Reflectmeter). When progression of a crack occurs in the mortar, the crack or the crack propagation breaks the optical fiber. Therefore, the cracking point can be measured by Fresnel reflection. In addition, the crack propagation in the mortar beam can be also measured by the breaking sequence of four optical fibers. Hence, it has been recognized that an optical fiber by using OTDR is very useful as a crack sensor for intelligent structures.