Materials and Structures (2008)41: 879-890 DOI10.1617/s11527-00792914 ORIGINAL ARTICLE Determination of tensile strength of glass fiber straps Zorislav Soric Josip Galic Tatjana Rukavina Received: 19 January 2007Accepted: 19 July 2007/Published online: 20 September 2007 C RILEM 2007 Abstract Glass fibers straps have been used for E01-0.3% Modulus of elasticity of fabric strengthening of masonry and concrete structures in determined by the increase of the ing tensile strength of dry glass fiber straps as well as E. -4s8 0.3% as recommended by ASMe and the last decade. Recently, their use has become specimens strain, aF, between 0.1% greater. This paper describes the research of measur modulus of el sticity of fabric determined straps that were made of glass fibers and epoxy by the increase of the specimen's strain, coating. The effect of strap widths, the effect of F, between 0.1% and 0.5%o loading speeds and the effect of epoxy coating placed Fn Force at failure, [kNI on fiber straps, on tensile strength of straps have been fmax Tensile strength of fabrics, [kN/ analyzed. Differences of tension strengths of fiber fr Tensile strength of longitudinally straps with and without epoxy coating are shown riented FRP composite Tensile fiber strength Keywords Glass fibers. Epoxy coating tensile matrix strength Straps. Fabric. Composites Strengthening fab Tensile strength of fabrics, IMPal 1-0.5% Secant stiffness of fabric Notations L Strap area Mean value B Strap width Standard deviation Modulus of elasticity of longitudinally xsVV Volume of matrix in oriented FRP composites Volume of fibers in composit Efb Modulus of elasticity of fibers EFma Strain of strap at failure E Modulus of elasticity of matrix Coefficient of variation Ve Volume ratio of fibers in composite Z Soric(<).J Galic. T Rukavina Faculty of Civil Engineering, University of Zagreb, olume ratio of matrix in composite Kaciceva 26, 10000 Zagreb. Croatia e-mail: soric@ grad.hr J. Galic 1 Introduction T Ruk FRP products are polymer composite materials comprising fibers, and polymer matrix, thus the name
ORIGINAL ARTICLE Determination of tensile strength of glass fiber straps Zorislav Soric Æ Josip Galic Æ Tatjana Rukavina Received: 19 January 2007 / Accepted: 19 July 2007 / Published online: 20 September 2007 RILEM 2007 Abstract Glass fibers straps have been used for strengthening of masonry and concrete structures in the last decade. Recently, their use has become greater. This paper describes the research of measuring tensile strength of dry glass fiber straps as well as straps that were made of glass fibers and epoxy coating. The effect of strap widths, the effect of loading speeds and the effect of epoxy coating placed on fiber straps, on tensile strength of straps have been analyzed. Differences of tension strengths of fiber straps with and without epoxy coating are shown. Keywords Glass fibers Epoxy coating Straps Fabric Composites Strengthening Notations A Strap area B Strap width Ef Modulus of elasticity of longitudinally oriented FRP composites Efib Modulus of elasticity of fibers Em Modulus of elasticity of matrix E0.1–0.3% Modulus of elasticity of fabric determined by the increase of the specimen’s strain, eF, between 0.1% and 0.3% as recommended by ASTM [3] E0.1–0.5% modulus of elasticity of fabric determined by the increase of the specimen’s strain, eF, between 0.1% and 0.5% Fmax Force at failure, [kN] fmax Tensile strength of fabrics, [kN/m] ff Tensile strength of longitudinally oriented FRP composite ffib Tensile fiber strength fm tensile matrix strength ffab Tensile strength of fabrics, [MPa] Jmax 0.1–0.5% Secant stiffness of fabrics L Strap length x Mean value s Standard deviation Vm Volume of matrix in composite Vfib Volume of fibers in composite eFmax Strain of strap at failure m Coefficient of variation vfib Volume ratio of fibers in composite: vfib ¼ Vfib VfibþVm vm Volume ratio of matrix in composite: vm ¼ Vm VfibþVm 1 Introduction FRP products are polymer composite materials, comprising fibers, and polymer matrix, thus the name Z. Soric (&) J. Galic T. Rukavina Faculty of Civil Engineering, University of Zagreb, Kaciceva 26, 10000 Zagreb, Croatia e-mail: soric@grad.hr J. Galic e-mail: jgalic@grad.hr T. Rukavina e-mail: rukavina@grad.hr Materials and Structures (2008) 41:879–890 DOI 10.1617/s11527-007-9291-4
880 Materials and Structures(2008)41: 879-890 Fiber reinforced Polymers or abbreviated FRP. Due temperature boundary between the areas of different to their properties, small mass, and high tensile mechanical properties of the same polymer. At this ally used in military and temperature, fibers will not be damaged, but the bond aviation industries. The manufactures were seeking strength between the matrix and fibers will decrease, new applications for their FRP product and the as well as the bond between the matrix and the possibility of massive use in civil engineering surrounding material. The most commonly used industry appeared. The basic idea was to produce a coating is epoxy resin. The usual properties of the material that would satisfy the growing needs for cold curing epoxy adhesive, concrete and steel used strengthening of structural elements and replace the in civil engineering, are shown, for comparison, in steel reinforcement of concrete and/or masonry Table 1 [1]. structures in aggressive environment. It is well known that aggressive environment can accelerate the corrosion of steel that was placed as reinforce 2.2 Fibers ment in concrete or masonry elements. Polymer composites are used in civil engineering and are Polymer composites have acronyms depending on the made of Aramid Fibers(AF), Carbon Fibers(CF)and fibers they are reinforced with: GFRP(glass fibers), Glass Fibers(GF). All these fibers possess high AFRP(aramid fibers), or CFRP(carbon fibers). The tensile strength. Polymer glue bounds fibers and such fibers increase the strength and stiffness of the com posite are shaped in moulds, applying pressure, composites. In order to create composites, one of which results in: AFRP, CFRP, and GFRP products. three types of fibers can be used: (1) short fibers There are two basic components of FRP products: (50 mm), oriented in all directions;(2) longitudinally fibers and polymer matrix. In polymer composite oriented, interwoven, straps of long fibers; and (3) fiber content is between 35 and 70 vol%o long fibers in bundles or fabric woven from bundles. The fibers are 5-20 um (10 m) thick and thei physical and mechanical properties vary depending 2 Components of polymer composites on fiber's type, [1, 2]. Carbon fibers are manufactured by controlled oxidation, carbonization, and graphiti- 2.1 Polymer matrix zation comparison with the other types of The matrix is made of different polymer resins, fibers have a higher strength and modulus of elastic selection of which depends on the type of fiber that ity, as well as better corrosion, fatigue, and creep should be bounded. Polymer matrix bounds fibers resistance Glass fibers are produced by protrusion of into composites and ensures the right position and a mixture of melted quartz sand, china clay, lime direction of fibers. At the same time the matrix stone, and colemanite, through small apertures at the protects fibers from environmental damage and mixture temperature of 1, 600oC. Th partially also from the mechanical damage. The later cooled. Different types of glass fibers are hatrix is also responsible for the transfer of load manufactured. Most often used glass fibers in com- between fibers, which implies a need for a good bond posites are: A-glasses (alkaline), E-glasse between fibers and the matrix. Strength and defor- (electrical), and C-glasses (chemical) fibers. The mation of composites are determined by the content A-glass fibers contain high proportion of boric acid of fibers and polymer matrix in the composites and and aluminates, thus, they are sensitive to alkaline he quality of bond between the matrix and the fibers. corrosion. E-glasses are more resistant to alkaline An important property of a synthetic resin is water agents and are significantly stronger and stiffer resistance, especially in extremely damp environ- C-glasses are highly resistant to alkaline agents ments. Thermal properties of the matrix determine E-glass fibers are most commonly used fibers due to he resistance of the composite, because the matrix their acceptable price and good technical properties has got much smaller thermal resistance and stability Composites with the glass fibers are heavier and of than fibers. The main property of the matrix is smaller strength and modulus of elasticity, but they Tg.the glass transition temperature, i.e., the are several times less expensive than the composites
Fiber Reinforced Polymers or abbreviated FRP. Due to their properties, small mass, and high tensile strength, they were initially used in military and aviation industries. The manufactures were seeking new applications for their FRP product and the possibility of massive use in civil engineering industry appeared. The basic idea was to produce a material that would satisfy the growing needs for strengthening of structural elements and replace the steel reinforcement of concrete and/or masonry structures in aggressive environment. It is well known that aggressive environment can accelerate the corrosion of steel that was placed as reinforcement in concrete or masonry elements. Polymer composites are used in civil engineering and are made of Aramid Fibers (AF), Carbon Fibers (CF) and Glass Fibers (GF). All these fibers possess high tensile strength. Polymer glue bounds fibers and such composite are shaped in moulds, applying pressure, which results in: AFRP, CFRP, and GFRP products. There are two basic components of FRP products: fibers and polymer matrix. In polymer composites, fiber content is between 35 and 70 vol%. 2 Components of polymer composites 2.1 Polymer matrix The matrix is made of different polymer resins, selection of which depends on the type of fiber that should be bounded. Polymer matrix bounds fibers into composites and ensures the right position and direction of fibers. At the same time the matrix protects fibers from environmental damage and partially also from the mechanical damage. The matrix is also responsible for the transfer of load between fibers, which implies a need for a good bond between fibers and the matrix. Strength and deformation of composites are determined by the content of fibers and polymer matrix in the composites and the quality of bond between the matrix and the fibers. An important property of a synthetic resin is water resistance, especially in extremely damp environments. Thermal properties of the matrix determine the resistance of the composite, because the matrix has got much smaller thermal resistance and stability than fibers. The main property of the matrix is Tg, ‘‘the glass transition temperature,’’ i.e., the temperature boundary between the areas of different mechanical properties of the same polymer. At this temperature, fibers will not be damaged, but the bond strength between the matrix and fibers will decrease, as well as the bond between the matrix and the surrounding material. The most commonly used coating is epoxy resin. The usual properties of the cold curing epoxy adhesive, concrete and steel used in civil engineering, are shown, for comparison, in Table 1 [1]. 2.2 Fibers Polymer composites have acronyms depending on the fibers they are reinforced with: GFRP (glass fibers), AFRP (aramid fibers), or CFRP (carbon fibers). The fibers increase the strength and stiffness of the composites. In order to create composites, one of three types of fibers can be used: (1) short fibers (50 mm), oriented in all directions; (2) longitudinally oriented, interwoven, straps of long fibers; and (3) long fibers in bundles or fabric woven from bundles. The fibers are 5–20 lm (106 m) thick and their physical and mechanical properties vary depending on fiber’s type, [1, 2]. Carbon fibers are manufactured by controlled oxidation, carbonization, and graphitization of organic materials rich in carbon. In comparison with the other types of fibers, carbon fibers have a higher strength and modulus of elasticity, as well as better corrosion, fatigue, and creep resistance. Glass fibers are produced by protrusion of a mixture of melted quartz sand, china clay, limestone, and colemanite, through small apertures at the mixture temperature of 1,600C. Those fibers are later cooled. Different types of glass fibers are manufactured. Most often used glass fibers in composites are: A-glasses (alkaline), E-glasses (electrical), and C-glasses (chemical) fibers. The A-glass fibers contain high proportion of boric acid and aluminates, thus, they are sensitive to alkaline corrosion. E-glasses are more resistant to alkaline agents and are significantly stronger and stiffer. C-glasses are highly resistant to alkaline agents. E-glass fibers are most commonly used fibers due to their acceptable price and good technical properties. Composites with the glass fibers are heavier and of smaller strength and modulus of elasticity, but they are several times less expensive than the composites 880 Materials and Structures (2008) 41:879–890
Materials and Structures (2008)41: 879-890 Table 1 Comparison of typical properties for epoxy adhesives, concrete, and steel [ll Property(at20°C Cold-curing epoxy adhesive Concrete Density(kg/m) 1,100-1,700 2, 7.800 Modulus of elasticity (GPa) 0.520 Shear modulus(GPa) 0.2-8 80 Tensile strength(MPa) 200-600 25-150 200.600 Tensile strain at break (%o 0.015 Coefficient of thermal expansion(10/C) 25-100 ll-13 0=15 Glass transition temperature-Tg(C) 45-80 with carbon fibers. Due to the sensitivity of glass to classified under two product types:(1)Prefabricated alkaline corrosion, the matrix material must protect FRP composite elements in form of wires, bars, glass fibers in composite. Aramid is a synthetic lamellas(Fig 1a); and (2) Dry, one or two direction polymer of high specific strength. Aramid fibers are oriented interwoven fiber fabric(Fig. la, b)set in flame resistant, but not UV-resistant Composites of epoxy resin during the strengthening process aramid fibers are of high impact strength and they absorb and disperse impact energy. Typical prope ties of all three-fiber types are shown in Table 2 [1]. 3.1 Prefabricated FRP products in form of rods and lamellas 3 Polymer composite products in civil engineering Composite products that are made of fibers and Polymer composites are used in civil engineering in matrix are first produced and then incorporated into a various forms: as wires, bars, or rods for reinforce- structure(e.g, FRP bars), or glued with adhesive glue ment, cables and ropes for pre-stressing and post- on the prepared surface(as lamellas). They are ready stressing, fabrics, laminate, straps and various types for use products, similar (in shape and appearance)to of sandwich panels. Polymer composites can be steel products. Their properties for certain types of Table 2 Typical properties of fibers [1I Material Modulus of elasticity Tensile strength Ultimate tensile strain MPa) High strengt 215-235 3,500-4,800 l4-2.0 215-235 3,5006000 15-2.3 High modulus 350-500 2,500-3,100 Ultra high 500-700 2,100-2,400 Glass fibers 70 1,900-3,000 3.0-4.5 3,500-4,800 Low modulus 70-80 3,500-4,100 4.3-5.0 High modulus l15-130 3,500-4,000 5-3.5
with carbon fibers. Due to the sensitivity of glass to alkaline corrosion, the matrix material must protect glass fibers in composite. Aramid is a synthetic polymer of high specific strength. Aramid fibers are flame resistant, but not UV-resistant. Composites of aramid fibers are of high impact strength and they absorb and disperse impact energy. Typical properties of all three-fiber types are shown in Table 2 [1]. 3 Polymer composite products in civil engineering Polymer composites are used in civil engineering in various forms: as wires, bars, or rods for reinforcement, cables and ropes for pre-stressing and poststressing, fabrics, laminate, straps and various types of sandwich panels. Polymer composites can be classified under two product types: (1) Prefabricated FRP composite elements in form of wires, bars, lamellas (Fig. 1a); and (2) Dry, one or two direction oriented interwoven fiber fabric (Fig. 1a, b) set in epoxy resin during the strengthening process. 3.1 Prefabricated FRP products in form of rods and lamellas Composite products that are made of fibers and matrix are first produced and then incorporated into a structure (e.g., FRP bars), or glued with adhesive glue on the prepared surface (as lamellas). They are ready for use products, similar (in shape and appearance) to steel products. Their properties for certain types of Table 1 Comparison of typical properties for epoxy adhesives, concrete, and steel [1] Property (at 20C) Cold-curing epoxy adhesive Concrete Mild steel Density (kg/m3 ) 1,100–1,700 2,350 7,800 Modulus of elasticity (GPa) 0.5–20 20–50 205 Shear modulus (GPa) 0.2–8 8–21 80 Poisson’s ratio 0.3–0.4 0.2 0.3 Tensile strength (MPa) 9–30 1–4 200–600 Shear strength (MPa) 10–30 2–5 200–600 Compressive strength (MPa) 55–110 25–150 200–600 Tensile strain at break (%) 0.5–5 0.015 25 Coefficient of thermal expansion (106 /C) 25–100 11–13 10–15 Glass transition temperature—Tg (C) 45–80 – – Table 2 Typical properties of fibers [1] Material Modulus of elasticity (GPa) Tensile strength (MPa) Ultimate tensile strain (%) Carbon fibers High strength 215–235 3,500–4,800 1.4–2.0 Ultra high strength 215–235 3,500–6,000 1.5–2.3 High modulus 350–500 2,500–3,100 0.5–0.9 Ultra high modulus 500–700 2,100–2,400 0.2–0.4 Glass fibers A 70 1,900–3,000 3.0–4.5 E 85–90 3,500–4,800 4.5–5.5 Aramid fibers Low modulus 70–80 3,500–4,100 4.3–5.0 High modulus 115–130 3,500–4,000 2.5–3.5 Materials and Structures (2008) 41:879–890 881
Materials and Structures(2008)41: 879-890 Fig 1 (a) FRP products bars. lamellas. fabric and b)glass fiber fabric that as tested fibers depend on the volume ratio of fibers in the and dimension is based on the overall cross section of composite. The usual fiber content of fibers in the composI ite product(fiber matrix) lamellas is between 35 and 70 vol%o. The modulus of elasticity and tensile strength of final composites 3.2 Dry, one-way or two-way oriented can be calculated from the volume ratio of fibers and interwoven fibers matrix using Eqs. (1) and (2)[1] E=Eb·wib+Em·Vm (1) These products come to the market in the form of f=f·Vb+fm·v (2) without the matrix. In the process of strengthening where straps, or fabric should be applied on cleaned and even (leveled) element surface that has previously Er= modulus of elasticity of longitudinally ori ented FRP composites; been coated with a fresh synthetic glue of a specific quality. The straps or fabric should then be pressed Efib modulus of elasticity of fibers: Em=modulus of elasticity of matrix: toward treated surface, causing the glue to penetrate between fibers. Since the composite product, fibers Vtb= volume ratio of fibers in composit matrix,emerges only after the strap or fabric Vm= volume ratio of matrix in composite ously been treated with the glue, the volume proportion of the glue (matrix) significantly varies Vm volume of matrix in composite and it is very difficult to control thickness of such Vib= volume of fibers in composite composite. Therefore, the analysis of strengthenin fr= tensile strength of longitudinally oriented FRP that is oriented to the cross section area of the composite composite product is difficult to estimate. Most of the fib tensile strength of fibers; fm= tensile strength of matrix researchers recommend analysis that is based only on the properties of the fibers, i. e, their tensile strength, Equations (1)and(2) do not always match the which neglects the influence of the matrix. That kind experimental results, so it is necessary to determinate, of analysis underestimates the effectiveness of for special types of lamellas and rods, the real strengthening because the tests that were carried ou modulus of elasticity and the real tensile strength. So and the results of which will be shown later in this far,the standard for FRP product verification does not paper show that epoxy glue significantly increases the exist,but verification for plastic products is usually strength of composite product. It should be pointed made according to one of the two standards: either out that the tensile strength of fabric is smaller than ASTM D 3039/D 3039M(ASTM 1995)[3], or en the sum of strengths of all fibers In specimens, some ISO527-5(sO1997)[4 of the fibers were initially stretched more than others The real properties of the composites should be and thus they broke earler The fabric strength, fab, btained by experiment and those values will later be which could be expressed as the first part of the eq used for further analysis. The analysis of resistance (2), was tested according to the standards [1,3, 4]
fibers depend on the volume ratio of fibers in the composite. The usual fiber content of fibers in lamellas is between 35 and 70 vol%. The modulus of elasticity and tensile strength of final composites can be calculated from the volume ratio of fibers and matrix using Eqs. (1) and (2) [1]. Ef ¼ Efib mfib þ Em mm ð1Þ ff ¼ ffib mfib þ fm mm ð2Þ where: Ef = modulus of elasticity of longitudinally oriented FRP composites; Efib = modulus of elasticity of fibers; Em = modulus of elasticity of matrix; vfib = volume ratio of fibers in composite: vfib ¼ Vfib VfibþVm ; vm = volume ratio of matrix in composite: vm ¼ Vm VfibþVm ; Vm = volume of matrix in composite; Vfib = volume of fibers in composite; ff = tensile strength of longitudinally oriented FRP composite; ffib = tensile strength of fibers; fm = tensile strength of matrix; Equations (1) and (2) do not always match the experimental results, so it is necessary to determinate, for special types of lamellas and rods, the real modulus of elasticity and the real tensile strength. So far, the standard for FRP product verification does not exist, but verification for plastic products is usually made according to one of the two standards: either ASTM D 3039/D 3039M (ASTM 1995) [3], or EN ISO 527-5 (ISO 1997) [4]. The real properties of the composites should be obtained by experiment and those values will later be used for further analysis. The analysis of resistance and dimension is based on the overall cross section of the composite product (fiber + matrix). 3.2 Dry, one-way or two-way oriented interwoven fibers These products come to the market in the form of fiber straps and fabrics. They consist of fibers only, without the matrix. In the process of strengthening, straps, or fabric should be applied on cleaned and even (leveled) element surface that has previously been coated with a fresh synthetic glue of a specific quality. The straps or fabric should then be pressed toward treated surface, causing the glue to penetrate between fibers. Since the composite product, fibers + matrix, emerges only after the strap or fabric is pressed toward the element surface that has previously been treated with the glue, the volume proportion of the glue (matrix) significantly varies and it is very difficult to control thickness of such composite. Therefore, the analysis of strengthening that is oriented to the cross section area of the composite product is difficult to estimate. Most of the researchers recommend analysis that is based only on the properties of the fibers, i.e., their tensile strength, which neglects the influence of the matrix. That kind of analysis underestimates the effectiveness of strengthening because the tests that were carried out and the results of which will be shown later in this paper show that epoxy glue significantly increases the strength of composite product. It should be pointed out that the tensile strength of fabric is smaller than the sum of strengths of all fibers. In specimens, some of the fibers were initially stretched more than others and thus they broke earlier. The fabric strength, ffab, which could be expressed as the first part of the Eq. (2), was tested according to the standards [1, 3, 4]. Fig. 1 (a) FRP products: bars, lamellas, fabric and (b) glass fiber fabric that was tested 882 Materials and Structures (2008) 41:879–890
Materials and Structures (2008)41: 879-890 of the glass fiber fabric was Mapewrap G UNI-AX 900/60, provided by Italian manufacturer Mapei [7]. The glue Mape Wrap 31 that was used as matrix was produced by Mapei too. The testing of tensile strength was performed by adapting standards described in [3, 4] nd the procedure given by the standard of testing the tensile strength and secant stiffness of geotextiles [5]. All specimens were tested by the use of rubber jaws Zwick Z100 testing machine Standards [ 3, 4, prescribe testing of the FRP Fig. 2 Measuring the thickness of glass fiber fabric fabric without matrix. However. since there are no The fabric strength is expressed as a product of rigid specimens the application of these standards to tensile strength of fibers fib, and a volume ratio of fabric is very demanding. In order to determine fibers in composite vib different influences on tensile strength, and modulus of elasticity of glass fiber straps, four types of testin Geotextile is similar product to the fiber fabrics. were performed. These four types of testing are The tensile strength and secant stiffness of geotextiles (a) Influence of specimen width on the tensile could be estimated according to trength and the modulus of elasticity of glass (1996)[5]. This standard [5], recommended calcula- fiber straps tion of tensile strength and secant stiffness (b) Influence of deformation speed on the tensile geotextiles both expressed in kilonewtons per meter, strength and the modulus of elasticity of glass ( kN/m). The average thickness of fabric, that was measured in this research on 15 specimens(see (c) Influence of strengthening at strap grip areas on ig. 2), according to EN ISo 964-1 [6]was the tensile strength of glass fiber straps 0.636 mm. Measuring of thickness of fabric accord-(d) Influence of epoxy glue on the tensile strength ing to the standard for geotextiles was chosen because of glass fiber straps of the similarity of products. According to standard [6] the stress on textile surface, during measuring of 5 Influence of specimen width, on the tensile fabric's thickness. was 200 kPa. strength and the modulus of elasticity of glass fiber straps 4 Determination of tensile strength of glass fiber straps and their composites by testi The first specimen testing has measured the influence of specimen width on the tensile strength and the For the sake of testing tensile strength, narrow straps modulus of elasticity of glass fiber straps. The were cut out from the 60 cm wide fabric, which has standard requirements [3] for specimens are shown Table 3 Tensile specimen geometry recommendation [3] Fiber orientation idth b length Thickness Tab length Tab thickness Angle of /(mm) a(mm) c(mr d(mm) Tabx(° 90° unidirectional 175 2.0 Random-discontinuous 2.5
The fabric strength is expressed as a product of tensile strength of fibers ffib, and a volume ratio of fibers in composite vfib. ffab ¼ ffib mfib ð3Þ Geotextile is similar product to the fiber fabrics. The tensile strength and secant stiffness of geotextiles could be estimated according to EN ISO 10319 (1996) [5]. This standard [5], recommended calculation of tensile strength and secant stiffness of geotextiles both expressed in kilonewtons per meter, (kN/m). The average thickness of fabric, that was measured in this research on 15 specimens (see Fig. 2), according to EN ISO 964-1 [6] was 0.636 mm. Measuring of thickness of fabric according to the standard for geotextiles was chosen because of the similarity of products. According to standard [6] the stress on textile surface, during measuring of fabric’s thickness, was 200 kPa. 4 Determination of tensile strength of glass fiber straps and their composites by testing For the sake of testing tensile strength, narrow straps were cut out from the 60 cm wide fabric, which has had unidirectional fiber orientation. The type of the glass fiber fabric was Mapewrap G UNI-AX 900/60, provided by Italian manufacturer Mapei [7]. The glue MapeWrap 31 that was used as matrix was produced by Mapei too. The testing of tensile strength was performed by adapting standards described in [3, 4], but also taking into account recommendations and the procedure given by the standard of testing the tensile strength and secant stiffness of geotextiles [5]. All specimens were tested by the use of rubber jaws in Zwick Z100 testing machine. Standards [3, 4], prescribe testing of the FRP composites, lamellas, i.e., real composites, and not fabric without matrix. However, since there are no rigid specimens the application of these standards to fabric is very demanding. In order to determine different influences on tensile strength, and modulus of elasticity of glass fiber straps, four types of testing were performed. These four types of testing are: (a) Influence of specimen width on the tensile strength and the modulus of elasticity of glass fiber straps, (b) Influence of deformation speed on the tensile strength and the modulus of elasticity of glass fiber straps, (c) Influence of strengthening at strap grip areas on the tensile strength of glass fiber straps, (d) Influence of epoxy glue on the tensile strength of glass fiber straps. 5 Influence of specimen width, on the tensile strength and the modulus of elasticity of glass fiber straps The first specimen testing has measured the influence of specimen width on the tensile strength and the modulus of elasticity of glass fiber straps. The standard requirements [3] for specimens are shown Fig. 2 Measuring the thickness of glass fiber fabric Table 3 Tensile specimen geometry recommendation [3] Fiber orientation Width b (mm) Overall length l (mm) Thickness a (mm) Tab length c (mm) Tab thickness d (mm) Angle of Tab a () 0 unidirectional 15 250 1.0 56 1.5 0 or 90 90 unidirectional 25 175 2.0 25 1.5 90 Balanced and symmetric 25 250 2.5 – – Random—discontinuous 25 250 2.5 – – Materials and Structures (2008) 41:879–890 883
884 Materials and Structures(2008)41: 879-890 Fig 3 Tension test prefabricated FRP composite [31 in Table 3, and in Fig 3. The exact dimensions of the recommendation by ASTM [3] of specimens length, straps widths that were recommended by standard of 250 mm, was difficult to adopt due to the yere difficult to meet. If one would wish to cut apparatus jaws length, and that was the reason why 15.0 mm wide straps out of fabrics, as recommended the total length of l=450 mm was used (see Fig. 5) by ASTM standard [3], it should be done along the Though, the specimen length between the jaws was fibers bundle, which would cause cutting of fibers. 150 mm, i.e., the required value by standards [3]. All For that reason each specimen consisted of exact specimens in this testing were dry interwoven fibers number of fiber bundles. The each bundle was i.e., without coating of epoxy glue, not strengthened 2.63 mm wide. Thus six bundles were 15.8 mm wide at their ends or anywhere else. Extensometer, type that was close to 15.0 mm. Tensile strength values in Multisens B066608, was used for deformation mea this research were determined by testing series of five surement. The extensometer was placed in the middle different specimen types of various widths (see of the strap, at the length of 50 mm. Specimens were Fig 5a). The specimen,'s type widths were:(1) stretched at the speed of 2 mm/min. At the grip area 2.63 mm or one fibers bundle, (2) 15.8 mm(6 rubber jaws with wider clamps were used, thus there bundles),(3)50 mm(19 bundles),(4)100 mm(38 was neither tearing, nor slipping of the specimen bundles), and(5)200 mm(76 bundles). The cutting straps. The tensile strengths of the specimens were of straps out of fabric is shown in Fig 4. Each represented both in MPa fab as in [3)) and in kN/m series type consisted of five specimens. The max as in [5). The tensile strength and the modulus of elasticity are determined by adopting the measured value of the glass fabric thickness of 0.636 mm. The defining tensile strength in MPa is not always practical because it is difficult to determine the average thickness of the fabric. The fabric is inter- woven with thin threads. and even the smallest error in such determination could influence the thickness value of tensile strength. The specimens of 50 mm width show the highest tensile strength and strain at failure. In specimens some of the fibers were at beginning stretched more than others and thus they broke earlier during the testing. When the most stretched fibers in narrower straps break, they cause proportionally more cross section damage than breaking of most stretched fibers in wider straps. It Fig. 4 Straps cutting out from fabric has happened because all the fibers could not be
in Table 3, and in Fig. 3. The exact dimensions of the straps widths that were recommended by standards were difficult to meet. If one would wish to cut 15.0 mm wide straps out of fabrics, as recommended by ASTM standard [3], it should be done along the fibers bundle, which would cause cutting of fibers. For that reason each specimen consisted of exact number of fiber bundles. The each bundle was 2.63 mm wide. Thus six bundles were 15.8 mm wide that was close to 15.0 mm. Tensile strength values in this research were determined by testing series of five different specimen types of various widths (see Fig. 5a). The specimen’s type widths were: (1) 2.63 mm or one fibers bundle, (2) 15.8 mm (6 bundles), (3) 50 mm (19 bundles), (4) 100 mm (38 bundles), and (5) 200 mm (76 bundles). The cutting of straps out of fabric is shown in Fig. 4. Each series type consisted of five specimens. The recommendation by ASTM [3] of specimens length, of 250 mm, was difficult to adopt due to the apparatus jaws length, and that was the reason why the total length of l = 450 mm was used (see Fig. 5). Though, the specimen length between the jaws was 150 mm, i.e., the required value by standards [3]. All specimens in this testing were dry interwoven fibers i.e., without coating of epoxy glue, not strengthened at their ends or anywhere else. Extensometer, type Multisens B066608, was used for deformation measurement. The extensometer was placed in the middle of the strap, at the length of 50 mm. Specimens were stretched at the speed of 2 mm/min. At the grip area rubber jaws with wider clamps were used, thus there was neither tearing, nor slipping of the specimen straps. The tensile strengths of the specimens were represented both in MPa (ffab as in [3]) and in kN/m (fmax as in [5]). The tensile strength and the modulus of elasticity are determined by adopting the measured value of the glass fabric thickness of 0.636 mm. The defining tensile strength in MPa is not always practical because it is difficult to determine the average thickness of the fabric. The fabric is interwoven with thin threads, and even the smallest error in such determination could influence the thickness value of tensile strength. The specimens of 50 mm width show the highest tensile strength and strain at failure. In specimens some of the fibers were at beginning stretched more than others and thus they broke earlier during the testing. When the most stretched fibers in narrower straps break, they cause proportionally more cross section damage than breaking of most stretched fibers in wider straps. It has happened because all the fibers could not be b c a d α l a d b a Fig. 3 Tension test specimen drawings of prefabricated FRP composite [3] Fig. 4 Straps cutting out from fabric 884 Materials and Structures (2008) 41:879–890
Materials and Structures (2008)41: 879-890 Fig. 5 Dimensions of glass fiber strap specimens a)Specimens of glass fiber straps without epoxy coating b)specimen of glass fiber strap with poxy coating at grip areas prepared for tensile testin 100mm 100mm 15.8mm c)Specim d)Specime 40 mm 40 mm evenly stretched and evenly held by jaws. However, it 200 mm width is shown in Fig. 8b. At the maximum was difficult to place evenly stretched fibers of wider stress specimen fibers started to fail successively and straps in the testing machine jaws. In the beginning of therefore the stress decreased and the strain increased the stretching, only a few fibers started to stretch, and (see Fig. 6a, b). At the place of specimens failure the "force-strain" diagram had a smaller force most fibers remained stretched and broken but they growth, but a larger growth of the strain. Only after were not broken at the strait cross-section line. The I the fibers were stretched, i.e., above strain of breaking area of fibers in fabric was several centi- approximately 0.3%, the"force-strain" curve meters wide showed a larger force growth and a smaller growth of the strain. Therefore two modules of elasticity were determined, first E. -o3% for strains between 6 Influence of deformation speed on the tensile 0. 1% and 0.3%o as recommended by ASTM [3], and strength and the modulus of elasticity of glass 1-0.5% for strains between 0.1% and 0.5%0 fiber straps The secant stiffness Jo.ax.%(kN/m)according to EN Iso 10319 [5] was measured too. When most of the The second testing was done for the specimen straps fibers broke the extensometer was turned off. after width of 15. 8 mm in order to determine the influence that, the deformation was(automatically)determined of deformation speed on the tensile strength of glass from the distance between the apparatus jaws. fiber straps. The tensile strength was determined for However, the value of deformation resulted from three deformation speeds: (a)2,(b)5, and(c)10 mm the real deformation of a strap to which a small strap min. The results are shown in Table 5. The test slipping in the jaws was added. The measured values results showed that differences of tensile strengths were presented in Table 4. The"force-strain"dia- (max) obtained by different deformation speeds were AsS ams of the 15.8 mm wide specimens are shown in small. It could be concluded that different testing 6a. Figure 6b shows the"force-strain"diagrams speeds did not have a significant influence on the of the specimens with 200 mm width. a photo of the value of the tensile strength of the fabric. The speed specimen of the 15.8 mm wide strap, which was put of deformation of 5 mm/min resulted in the highest in the testing machine, is shown in Fig. 7a. Figure 7b strength of the straps, i.e., 2.6% above average shows a photo of the same specimen(15.8 mm) after strength(see Table 5). This speed also resulted in the failure. A photo of a specimen after failure in Fig. &a most equalized values of the strength, which is of 50 mm width and that of evident from the smallest value of the coefficient of
evenly stretched and evenly held by jaws. However, it was difficult to place evenly stretched fibers of wider straps in the testing machine jaws. In the beginning of the stretching, only a few fibers started to stretch, and the ‘‘force–strain’’ diagram had a smaller force growth, but a larger growth of the strain. Only after all the fibers were stretched, i.e., above strain of approximately 0.3%, the ‘‘force–strain’’ curve showed a larger force growth and a smaller growth of the strain. Therefore two modules of elasticity were determined, first E0.1–0.3% for strains between 0.1% and 0.3% as recommended by ASTM [3], and second E0.1–0.5% for strains between 0.1% and 0.5%. The secant stiffness Jmax 0.1–0.5% (kN/m) according to EN ISO 10319 [5] was measured too. When most of the fibers broke, the extensometer was turned off. After that, the deformation was (automatically) determined from the distance between the apparatus jaws. However, the value of deformation resulted from the real deformation of a strap to which a small strap slipping in the jaws was added. The measured values were presented in Table 4. The ‘‘force–strain’’ diagrams of the 15.8 mm wide specimens are shown in Fig. 6a. Figure 6b shows the ‘‘force–strain’’ diagrams of the specimens with 200 mm width. A photo of the specimen of the 15.8 mm wide strap, which was put in the testing machine, is shown in Fig. 7a. Figure 7b shows a photo of the same specimen (15.8 mm) after failure. A photo of a specimen after failure in Fig. 8a shows a specimen of 50 mm width and that of 200 mm width is shown in Fig. 8b. At the maximum stress specimen fibers started to fail successively and therefore the stress decreased and the strain increased (see Fig. 6a, b). At the place of specimen’s failure most fibers remained stretched and broken but they were not broken at the strait cross-section line. The breaking area of fibers in fabric was several centimeters wide. 6 Influence of deformation speed on the tensile strength and the modulus of elasticity of glass fiber straps The second testing was done for the specimen straps width of 15.8 mm in order to determine the influence of deformation speed on the tensile strength of glass fiber straps. The tensile strength was determined for three deformation speeds: (a) 2, (b) 5, and (c) 10 mm/ min. The results are shown in Table 5. The test results showed that differences of tensile strengths (fmax) obtained by different deformation speeds were small. It could be concluded that different testing speeds did not have a significant influence on the value of the tensile strength of the fabric. The speed of deformation of 5 mm/min resulted in the highest strength of the straps, i.e., 2.6% above average strength (see Table 5). This speed also resulted in the most equalized values of the strength, which is evident from the smallest value of the coefficient of a) Specimens of glass fiber straps without epoxy coating 2.63 mm 15.8 mm 50 mm 100 mm 450 mm 15.8 mm c) Specimen of glass fiber strap with epoxy coating at its total length 450 mm strengthened with 2 straps and epoxy coating 200 mm 15.8 mm 3 x 50 mm 40 mm ekstensometer 100 mm Direction of the stretching d) Specimen in machine jaws 40 mm 100 mm b) Specimen of glass fiber strap with epoxy coating at grip areas 100 mm 100 mm 150 mm 50 mm 50 mm Fig. 5 Dimensions of glass fiber strap specimens prepared for tensile testing Materials and Structures (2008) 41:879–890 885
886 Materials and Structures(2008)41: 879-890 Table 4 Tensile strength, modulus of elasticity, and secant stiffness of glass fiber strap specimens Width of specimens Fmax(N 1.26mm(1 bundle)x7852429857469451.3143,433.3843,56291 27,70601 759759 75920.56 19.25 9.10 9.10 58mm(6 bundles)x4,933.72312.26490.971.4935,75062 29.6527.19 42.75 0. 4,996.90 3,93621 2,50343 8.7118.17 13.98 3.50mm(19 bundles)x17,376.00347.52546411.6032,220.6536,081.5622,94787 1,549.8431.00 48.740.298,921.61 641628 8.928.92 17.78 4.100mm(38 bundles)x31,901.15319.20501.891.4531,8990335,74069 22,731.07 s3.l12.7831.15 v9.769.76 97610.32 14.20 7.95 5.200mm(76 bundles)x63,020.10315.26495.691.1737,916.0841,38743 26,322.41 2,193.071097 17.250.16 683401 50.70 3,08504 v3.483.48 3.4813.49 l172 11.72 Fig. 6 The"force-strain diagrams of glass fibe traps without epoxy coating: (a)15.8 mm wide traps, (b)200 mm wide 158mm Strain(%) Strain ( Fig. 7(a) Testing of glass fiber strap 15.8 mm wide b)glass fiber strap after afte failure variation D= 5.28. The specimens of this type of 7 Influence of strengthening at the strap grip testing were all broken in the same way, regardless of areas on the tensile strength of glass fiber straps the test speed. The failure of fabric occurred in the jaws, although certain damage was visible in the jaws treatment at the area held by jaws, on th.r strap middle 150 mm length of specimens between rubbe The third testing showed the influence area as well of the strap. Although in previous specimens the
variation t = 5.28. The specimens of this type of testing were all broken in the same way, regardless of the test speed. The failure of fabric occurred in the middle 150 mm length of specimens between rubber jaws, although certain damage was visible in the jaws area as well. 7 Influence of strengthening at the strap grip areas on the tensile strength of glass fiber straps The third testing showed the influence of strap treatment at the area held by jaws, on the strength of the strap. Although in previous specimens the a) b) Fig. 6 The ‘‘force–strain’’ diagrams of glass fiber straps without epoxy coating: (a) 15.8 mm wide straps, (b) 200 mm wide straps Table 4 Tensile strength, modulus of elasticity, and secant stiffness of glass fiber strap specimens Width of specimens Fmax (N) fmax (kN/m) ffab (MPa) eFmax (%) E0.1–0.3% (MPa) E0.1–0.5% (MPa) Jmax 0.1–0.5% (kN/m) 1. 2.6 mm (1 bundle) x 785.24 298.57 469.45 1.31 43,433.38 43,562.91 27,706.01 s 59.63 22.67 35.64 0.27 8,362.03 3,963.05 2,520.50 m 7.59 7.59 7.59 20.56 19.25 9.10 9.10 2. 15.8 mm (6 bundles) x 4,933.72 312.26 490.97 1.49 35,750.62 38,508.17 24,491.20 s 429.65 27.19 42.75 0.27 4,996.90 3,936.21 2,503.43 m 8.71 8.71 8.71 18.17 13.98 9.10 9.10 3. 50 mm (19 bundles) x 17,376.00 347.52 546.41 1.60 32,220.65 36,081.56 22,947.87 s 1,549.84 31.00 48.74 0.29 8,921.61 6,416.28 4,080.75 m 8.92 8.92 8.92 18.33 27.69 17.78 17.78 4. 100 mm (38 bundles) x 31,901.15 319.20 501.89 1.45 31,899.03 35,740.69 22,731.07 s 3,112.78 31.15 48.98 0.15 4,530.61 2,841.73 1,807.34 m 9.76 9.76 9.76 10.32 14.20 7.95 7.95 5. 200 mm (76 bundles) x 63,020.10 315.26 495.69 1.17 37,916.08 41,387.43 26,322.41 s 2,193.07 10.97 17.25 0.16 6,834.01 4,850.70 3,085.04 m 3.48 3.48 3.48 13.49 18.02 11.72 11.72 Fig. 7 (a) Testing of glass fiber strap 15.8 mm wide; (b) glass fiber strap after failure 886 Materials and Structures (2008) 41:879–890
Materials and Structures (2008)41: 879-890 Fig. 8 Glass fiber after failure 50 mm width, (b) strap of Strap 200 mm width Strap 50 mm width Table 5 Tensile strengths of glass fiber strap specimens of Test speed(mm/min) (kN/m) fab(MPa) EFmax(%o 15. 8 mm width at different x4,933.7231226 testing speeds 42965 5,446.8932385 509.20 270.38 5.28 486.51 39.17 0.34 8.05 8.05 failure of straps did not happen at the grip areas, with in the area held by the jaws due to the jaws pressure this test it was tried to avoid even the smallest However, the failure of straps occurred in the middle damage of the straps on these areas. Specimen straps part of the strap's length between the jaws From the ere strengthened with additional layers of fabric at results shown in Table 6 it is evident that the both sides of the specimen at the each of the grip strengthened specimens showed smaller tensile areas. Those additional layers were bonded with strength than unstrengthened ones. The reason for epoxy glue(see Figs. 4, 5b, 9)to the specimen. All that is in straps preparation. The area held by jaws of tests were done on specimen straps that were the strap that was coated with the epoxy glue was 15.8 mm wide with the deformation speed of area with fibers that were not all equally stretched 2 mm/min. The test results are shown in Table 6 Equal stretching of fibers was difficult to achieve, and in Fig. 1la. The damages of specimens in a form especially when working with the epoxy glue. The of cracking of dried(hardened) epoxy glue occurred uneven stretching during the testing caused that it Fig. 9 Partial strengthened glass fiber straps with epoxy ating before and(b) after failure a)specimens before testing Specimen after failure
failure of straps did not happen at the grip areas, with this test it was tried to avoid even the smallest damage of the straps on these areas. Specimen straps were strengthened with additional layers of fabric at both sides of the specimen at the each of the grip areas. Those additional layers were bonded with epoxy glue (see Figs. 4, 5b, 9) to the specimen. All tests were done on specimen straps that were 15.8 mm wide with the deformation speed of 2 mm/min. The test results are shown in Table 6 and in Fig. 11a. The damages of specimens in a form of cracking of dried (hardened) epoxy glue occurred in the area held by the jaws due to the jaws pressure. However, the failure of straps occurred in the middle part of the strap’s length between the jaws. From the results shown in Table 6 it is evident that the strengthened specimens showed smaller tensile strength than unstrengthened ones. The reason for that is in straps preparation. The area held by jaws of the strap that was coated with the epoxy glue was area with fibers that were not all equally stretched. Equal stretching of fibers was difficult to achieve, especially when working with the epoxy glue. The uneven stretching during the testing caused that it Fig. 8 Glass fiber straps after failure: (a) strap of 50 mm width, (b) strap of 200 mm width Table 5 Tensile strengths of glass fiber strap specimens of 15.8 mm width at different testing speeds Test speed (mm/min) Fmax (N) fmax (kN/m) ffab (MPa) eFmax (%) 2 x 4,933.72 312.26 490.97 1.49 s 429.65 27.19 42.75 0.27 m 8.71 8.71 8.71 18.17 5 x 5,446.89 323.85 509.20 1.74 s 270.38 17.14 26.95 0.24 m 5.28 5.28 5.28 13.70 10 x 4,888.82 309.42 486.51 1.94 s 393.63 24.91 39.17 0.34 m 8.05 8.05 8.05 17.60 Fig. 9 Partial strengthened glass fiber straps with epoxy coating at grip areas: (a) before and (b) after failure Materials and Structures (2008) 41:879–890 887
888 Materials and Structures(2008)41: 879-890 Table glass fiber 158mm H specimens of Type of specime (kN/m) with epoxy coating at grip Specimens without epoxy strengthening at grip areas x 4,933. 72 312.26 s shown in Table 4 4296527.1942.750.27 Specimens with epoxy strengthening at grip areas x 3, 245.09 205.39 322.94 1 29 s535.78339l53.320.13 v16.5l16.5116.51997 Table 7 Tensile strengths of glass fiber stra Fmax fmax(kN/ strap cMax 15.8 mm width with epoxy coati 2.1134,437.83 total length Specimens with epoxy 0.522,146.94 24.82 took time for initially less stretched fibers to become coated in their whole length with the epoxy glue in active. During testing fibers broke at different times, order to determine the influence of the coating on the first the more stretched ones and, then those that were strength. About 28 days after the coating, epoxy glue less stretched. The non-strengthened specimens expe- was finally hardened. The specimens were tested with rienced similar problems but their advantage lies in the deformation speed of 2 mm/min, without any the fact that, during testing, the stronger stretched additional treatment of the specimen's surface held fibers pulled out by small extend from the rubber by the jaws. The test results are shown in Table 7 and jaws, and therefore enabled more equal distribution in Fig. 11b. These specimens showed strengths that of stretching force on more fibers Therefore. it is were almost double and secant stiffness about 40 recommended to use specimen straps without epoxy greater than those of specimens without the epoxy coating, but to use jaws made of hard rubber. coating. The failure was brittle. The epoxy coating between fibers. a voiding the concentration of stresse 8 Influence of epoxy glue on the tensile strength of in specimens jaws areas would most probably result lass fiber straps in a higher strength of the specimen. Figure 10 shows the photos of specimens before and after the failure The fourth tests were done in order to determine the Figure Ila shows force-strain experimental dia- nfluence of epoxy glue on the tensile strength grams of specimens that were coated with the epoxy glass fiber straps. Specimens of 15.8 mm width were glue only at grip areas(see also Fig 9a). Figure 1lb Fig. 10 Strengthened glass fiber straps with epoxy b) coating at their total length (a) before, and(b)afte
took time for initially less stretched fibers to become active. During testing fibers broke at different times, first the more stretched ones and, then those that were less stretched. The non-strengthened specimens experienced similar problems but their advantage lies in the fact that, during testing, the stronger stretched fibers pulled out by small extend from the rubber jaws, and therefore enabled more equal distribution of stretching force on more fibers. Therefore, it is recommended to use specimen straps without epoxy coating, but to use jaws made of hard rubber. 8 Influence of epoxy glue on the tensile strength of glass fiber straps The fourth tests were done in order to determine the influence of epoxy glue on the tensile strength of glass fiber straps. Specimens of 15.8 mm width were coated in their whole length with the epoxy glue in order to determine the influence of the coating on the strength. About 28 days after the coating, epoxy glue was finally hardened. The specimens were tested with the deformation speed of 2 mm/min, without any additional treatment of the specimen’s surface held by the jaws. The test results are shown in Table 7 and in Fig. 11b. These specimens showed strengths that were almost double and secant stiffness about 40% greater than those of specimens without the epoxy coating. The failure was brittle. The epoxy coating ensured the more even distribution of stresses between fibers. Avoiding the concentration of stresses in specimen’s jaws areas would most probably result in a higher strength of the specimen. Figure 10 shows the photos of specimens before and after the failure. Figure 11a shows force–strain experimental diagrams of specimens that were coated with the epoxy glue only at grip areas (see also Fig. 9a). Figure 11b Table 7 Tensile strengths of glass fiber strap specimens of 15.8 mm width strengthened with epoxy coating at their total length Type of specimen Fmax (N) fmax (kN/ m) fstrap (MPa) eFmax (%) Jmax 0.1–0.5% (kN/ m) Specimens with epoxy strengthening x 8,944.85 566.13 890.14 2.11 34,437.83 s 1,114.85 70.56 110.94 0.52 2,146.94 m 12.46 12.46 12.46 24.82 6.23 Table 6 Tensile strengths of glass fiber strap specimens of 15.8 mm width without and with epoxy coating at grip areas Type of specimen Fmax (N) fmax (kN/m) ffab (MPa) eFmax (%) Specimens without epoxy strengthening at grip areas as shown in Table 4 x 4,933.72 312.26 490.97 1.49 s 429.65 27.19 42.75 0.27 m 8.71 8.71 8.71 18.17 Specimens with epoxy strengthening at grip areas x 3,245.09 205.39 322.94 1.29 s 535.78 33.91 53.32 0.13 m 16.51 16.51 16.51 9.97 Fig. 10 Strengthened glass fiber straps with epoxy coating at their total length: (a) before, and (b) after failure 888 Materials and Structures (2008) 41:879–890