T Ishikawa at both ends and by reducing the number of layers in composites used at the ends together with FRP U-anchors or bolt anchorage with bonded steel plates. The lower portion of Fig. 5 [4] shows typical load-deflection curves of 10 m long concrete girders with pre-stressed PBO fiber sheets at different pre-stress levels. The control girder without external reinforcement failed due to yielding of tensile steel followed by concrete crushing. The girder strengthened by three-layer PBO sheets without pre-stressing exhibited the similar bend-over point to the control, but gained significant strain hardening after the yielding of steel. With pre-stressing, he linear proportional limit was increased by 45% and the ultimate load by 65%0 a high percentage of pre-stressing could effectively change the failure mode from debonding to PBO tensile rupture, leading to a significant increase in load-carrying capacity. This technology and PBO fiber show high potential of increasing load- arrying capacity of civil structures such as the bridge girders Concrete structures a Prestress I Pretension of pfrp sheets PFRP sheets 2. Bonding and curing Release and cutting of PPRP Stress distribution in a structural section 60 PFRP sheets rupt 500 PFRP sheets debor propagation 300 ………"- 200 --·· Without reinforcement Without prestressing(3 layers PFRP sheets) 33% prestressing layers PFRPsheets) 45%prestressing( layers PFRPsheets 0 40 80100120 Displacement(mm) Figure 5. Load-deflection curve of concrete girder of 10 m span with pre-stressed PFRP
8 T. Ishikawa at both ends and by reducing the number of layers in composites used at the ends together with FRP U-anchors or bolt anchorage with bonded steel plates. The lower portion of Fig. 5 [4] shows typical load–deflection curves of 10 m long concrete girders with pre-stressed PBO fiber sheets at different pre-stress levels. The control girder without external reinforcement failed due to yielding of tensile steel followed by concrete crushing. The girder strengthened by three-layer PBO sheets without pre-stressing exhibited the similar bend-over point to the control, but gained significant strain hardening after the yielding of steel. With pre-stressing, the linear proportional limit was increased by 45% and the ultimate load by 65%. A high percentage of pre-stressing could effectively change the failure mode from debonding to PBO tensile rupture, leading to a significant increase in load-carrying capacity. This technology and PBO fiber show high potential of increasing loadcarrying capacity of civil structures such as the bridge girders. Figure 5. Load–deflection curve of concrete girder of 10 m span with pre-stressed PFRP
Trends in composite research in Japa 3. NANO-TUBES NANO-FIBERS AND NANO-FILLERS a big boom of nanocomposites research has landed also in Japan. As a virtual center of excellence' in composites technology there, ACE TeC of JAXA has led pioneering sections of nanocomposites research, particularly in mechanica properties oriented applications. An overview of research activities related nanocomposites in ACE TeC will be given first and some remarkable results will be introduced briefly The first example is a Carbon Fiber Reinforced Composite(CFRP) laminate using epoxy resin stiffened by carbon nanotubes (CNT). It is well known that CNT exhibits extremely high elastic modulus and strengths. One trend in CNT application as composite reinforcement is direct dispersion like chopped fiber in polymer with an alignment as parallel as possible, which is considered as two-phase material of CNT and polymer. Although it looks like a proper way of obtaining CNT-reinforced composites, alignment of CNT with uniform spatial dispersion in resin is not an easy task. ACE TeC pursues another route for CNT reinforced composites of three-phase material-CNT, polymer and conventional carbon fiber (CF. In this idea, CNT can be regarded as a modifier of matrix resin for increasing its mechanical properties. In this trial, the CNt used is unique, and should be referred to as carbon nanofiber(CNF): Carbere, made by GSI CREOS Corporation in Japan [5]. They refer to this product as cup-stack nanofiber(CSNF). A schematic llustration [5] of such CSNF is shown in Fig. 6 where conical graphene sheets are accumulated like a stack of paper cups and their outer diameter is in the range of 80 to 100 nm with 15 nm wall thickness, which is rather larger than usual CNT. The mechanical properties of this CSNF are compared in Fig. 7 where common carbon fiber data are plotted [5]. Two types of CSNF were employed in the difference of aspect ratios, i.e. fiber length of 500 nm to I um(AR1O)and fiber length of 2.5 to 10.0 um(AR50), respectively. These two types of CSNF are dispersed to Epikote 827 epoxy resin by the company and they can supply the dispersed \15nm\50nm CARBERE Figure 6. Schematic illustration of cup-stack type carbon nanofiber and its dim
Trends in composite research in Japan 9 3. NANO-TUBES, NANO-FIBERS AND NANO-FILLERS A big boom of nanocomposites research has landed also in Japan. As a virtual ‘center of excellence’ in composites technology there, ACE TeC of JAXA has led pioneering sections of nanocomposites research, particularly in mechanical properties oriented applications. An overview of research activities related to nanocomposites in ACE TeC will be given first and some remarkable results will be introduced briefly. The first example is a Carbon Fiber Reinforced Composite (CFRP) laminate using epoxy resin stiffened by carbon nanotubes (CNT). It is well known that CNT exhibits extremely high elastic modulus and strengths. One trend in CNT application as composite reinforcement is direct dispersion like chopped fiber in polymer with an alignment as parallel as possible, which is considered as two-phase material of CNT and polymer. Although it looks like a proper way of obtaining CNT-reinforced composites, alignment of CNT with uniform spatial dispersion in resin is not an easy task. ACE TeC pursues another route for CNT reinforced composites of three-phase material — CNT, polymer and conventional carbon fiber (CF). In this idea, CNT can be regarded as a modifier of matrix resin for increasing its mechanical properties. In this trial, the CNT used is unique, and should be referred to as carbon nanofiber (CNF): Carbere®, made by GSI CREOS Corporation in Japan [5]. They refer to this product as cup-stack nanofiber (CSNF). A schematic illustration [5] of such CSNF is shown in Fig. 6 where conical graphene sheets are accumulated like a stack of paper cups and their outer diameter is in the range of 80 to 100 nm with 15 nm wall thickness, which is rather larger than usual CNT. The mechanical properties of this CSNF are compared in Fig. 7 where common carbon fiber data are plotted [5]. Two types of CSNF were employed in the difference of aspect ratios, i.e. fiber length of 500 nm to 1 µm (AR10) and fiber length of 2.5 to 10.0 µm (AR50), respectively. These two types of CSNF are dispersed to Epikote® 827 epoxy resin by the company and they can supply the dispersed Figure 6. Schematic illustration of cup-stack type carbon nanofiber and its dimensions
T Ishikawa Usual SW-CNT Modulus: 1 TPa Strength: 600 GPa 20%15%10 0.5% 098 CARBERETM 品 T1000 ★CSNF 0 O PITCH 2000 Modulus(GPa) Figure 7. Strength and modulus plot of cup-stack type carbon nanofiber ompound of high CSNF weight content and this epoxy. At the first trial stage, a manual fabrication process of the composite plates by impregnation of the diluted compound by the same epoxy into dry carbon fiber fabrics(Torayca CO6343 T300/plain weave) was employed followed by the hot-press curing. Compression strength improvements around 15%o in the three-phase composites by a loading of CSNF in comparison with the control case of no CSNF are shown Fig. 8 as the typical example of this stage [5] where all strength data are normalized by the control data. The most efficient case, 5% loading of the weight of the resin with AR 50 CSNF, is marked by a circle in this figure. By stimulating such promising results, the supplier company of CSNf, GSI-Creos, decided to develop prepreg systems containing dispersed CsnF [6] in order to obtain more stable mechanical properties than manual fabrication processes. Another key issue in the prepreg development is an optimization of the aspect ratio of CSNF. Although the details of this process belong to the company and cannot be disclosed, one advantage for a good dispersion is many numbers of edges of graphene sheets appearing on the CSNF surface. Such edges may help to increase interaction between CSNF and polymer due to this cup-stack nature. A good dispersion of CsnF is suggested in oscopic sectional view of Fig. 9 for the three phase composites through the prepreg route 16]. Compression strength improvement in these three phase composites by the prepreg is more remarkable than manually impregnated cases as shown in Fig. 10 [6]. In this T-700 CF Ud prepreg case, the compression strength in the fiber direction is improved by some 25% in comparison with the contre (no CSNF) case. However as shown there, the elastic modulus in compression of
10 T. Ishikawa Figure 7. Strength and modulus plot of cup-stack type carbon nanofiber. compound of high CSNF weight content and this epoxy. At the first trial stage, a manual fabrication process of the composite plates by impregnation of the diluted compound by the same epoxy into dry carbon fiber fabrics (Torayca CO6343: T300/plain weave) was employed followed by the hot-press curing. Compression strength improvements around 15% in the three-phase composites by a loading of CSNF in comparison with the control case of no CSNF are shown in Fig. 8 as the typical example of this stage [5] where all strength data are normalized by the control data. The most efficient case, 5% loading of the weight of the resin with AR 50 CSNF, is marked by a circle in this figure. By stimulating such promising results, the supplier company of CSNF, GSI-Creos, decided to develop prepreg systems containing dispersed CSNF [6] in order to obtain more stable mechanical properties than manual fabrication processes. Another key issue in the prepreg development is an optimization of the aspect ratio of CSNF. Although the details of this process belong to the company and cannot be disclosed, one advantage for a good dispersion is many numbers of edges of graphene sheets appearing on the CSNF surface. Such edges may help to increase interaction between CSNF and polymer due to this cup-stack nature. A good dispersion of CSNF is suggested in the microscopic sectional view of Fig. 9 for the three phase composites through the prepreg route [6]. Compression strength improvement in these three phase composites by the prepreg is more remarkable than manually impregnated cases as shown in Fig. 10 [6]. In this T–700 CF UD prepreg case, the compression strength in the fiber direction is improved by some 25% in comparison with the control (no CSNF) case. However as shown there, the elastic modulus in compression of
Trends in composite research in Japan ■ Tension Compression .10H Fle xure 090 080 06 CoNTROL AR10-5wt% AR10-10wt AR50-5wt% AR50-10wt% Figure 8. Compression strength increase in CSNF/epoxy/CF three-phase composites. Satisfactory Dispersion Figure 9. SEM view of CSNF/epoxy/CF three phase composites by prepreg route his composite is not affected as naturally expected. These prepreg systems are considered to be very potential and they are now available in the composite raw material market. Moreover, their applications have been started to the sporting goods in Japan
Trends in composite research in Japan 11 Figure 8. Compression strength increase in CSNF/epoxy/CF three-phase composites. Figure 9. SEM view of CSNF/epoxy/CF three phase composites by prepreg route. this composite is not affected as naturally expected. These prepreg systems are considered to be very potential and they are now available in the composite raw material market. Moreover, their applications have been started to the sporting goods in Japan
T Ishikawa Compression Stength of CNT-CFRP 700 +25 400 300 B30 20 10 CNT 4 5% CNT 4.5% CF: Toray T700 180 gim? RC 38% 180 glm"RC Figure 10. Compression strength and modulus in CSNF/epoxy/CF three-phase composites by prepreg route(T700, 180 g/m- fiber aerial weight, CNT%: volume content The second research topic concerning nanotechnology composites in ACE TeC/ JAXA is an improvement of heat resistance of high temperature polymer [7 by loading of multi-wall carbon nanotube(MW-CNT). In this attempt, a baseline polymer itself is one of the ever-best high temperature polymers, Triple A Polyimide (TriA- PD), which shows much better heat resistance than NASA standard PETI-5 The detail of this resin will be explained in the next section on newly developed polymers. The final purpose here is to increase heat resistance such as glass transition temperature by adding Mw-CNT to this polymer. The Mw-CNT adopted in this attempt is fabricated through CVD technique by Carbon Nanotech Research Institute Inc(CNRD) in Japan [7]. The SeM picture of this MW-CNT is shown in Fig. 11 where its diameter and lengths show the scatter of 20 to 100 nm and several um, respectively. Chosen loading weight fractions of MW-CNT are as follows: 0. 3.3. 7.7 and 14.3%. Because the imid-oligomer is solid room temperature, MW-CNT is added to imid-oligomer powder and mechanically blended by a ball mixer(RouteD). Then the mixture is consolidated by using a hot- press for an hour at 370C. Another processing route is the liquid route in an ami acid/NMP solution with some chemical reactions including imidization(Route (m)) although a description of the details is not given here. The obtained material is a two-phase composite if we follow the previous terminology. Dynamic viscoelastic properties were measured for the obtained composites by a dynamic mechanical analyzer (DMA: Q800 of TA Instruments Corp )in the single cantilever bending method and static tensile properties were measured with a small coupon of 5 mm x 1. I mm x 80 mm(width x thickness x length). Increases in the glass transition temperature(Tg) based on the dynamic mechanical analyzer are shown in Fig 12 where Tg defined by initiation of the storage modulus reduction is 333 C for the
12 T. Ishikawa Figure 10. Compression strength and modulus in CSNF/epoxy/CF three-phase composites by prepreg route (T700, 180 g/m2 fiber aerial weight, CNT%: volume content). The second research topic concerning nanotechnology composites in ACE TeC/ JAXA is an improvement of heat resistance of high temperature polymer [7] by loading of multi-wall carbon nanotube (MW-CNT). In this attempt, a baseline polymer itself is one of the ever-best high temperature polymers, Triple A Polyimide (TriA-PI), which shows much better heat resistance than NASA standard PETI-5. The detail of this resin will be explained in the next section on newly developed polymers. The final purpose here is to increase heat resistance such as glass transition temperature by adding MW-CNT to this polymer. The MW-CNT adopted in this attempt is fabricated through CVD technique by Carbon Nanotech Research Institute Inc. (CNRI) in Japan [7]. The SEM picture of this MW-CNT is shown in Fig. 11 where its diameter and lengths show the scatter of 20 to 100 nm and several µm, respectively. Chosen loading weight fractions of MW-CNT are as follows: 0, 3.3, 7.7 and 14.3%. Because the imid-oligomer is solid at room temperature, MW-CNT is added to imid-oligomer powder and mechanically blended by a ball mixer (Route (I)). Then the mixture is consolidated by using a hotpress for an hour at 370◦C. Another processing route is the liquid route in an amid acid/NMP solution with some chemical reactions including imidization (Route (II)) although a description of the details is not given here. The obtained material is a two-phase composite if we follow the previous terminology. Dynamic viscoelastic properties were measured for the obtained composites by a dynamic mechanical analyzer (DMA: Q800 of TA Instruments Corp.) in the single cantilever bending method and static tensile properties were measured with a small coupon of 5 mm × 1.1 mm × 80 mm (width × thickness × length). Increases in the glass transition temperature (Tg) based on the dynamic mechanical analyzer are shown in Fig. 12, where Tg defined by initiation of the storage modulus reduction is 333◦C for the