Glass fibre reinforced precast concrete containing high content pozzolanic materials
* Author to whom correspondence should be addressed.
Sigma Journal of Engineering and Natural Sciences 2019, Vol. 37, Issue 2, pp. 675-686; doi.org/10.62051/ytu.sigma-journal-of-engineering-and-natural-sciences-glass-fibre-reinforced-precast-concrete-containing-high-content-pozzolanic-mater
Abstract
Keywords: Glass fiber reinforced concrete; fly ash; slag; sustainability; microstructure.
1. Introduction
Glass fiber reinforced concrete (GFRC) is a material composed of cement based matrix and glass fiber reinforcements dispersed in this matrix. Glass fiber and cement based matrix preserves their physical and chemical properties and different physical and mechanical properties are obtained by combining these materials [1,2]. The characteristics of GFRC depend on different variables such as production, mixing ratio, type, length and orientation of glass fiber and additives used. In general terms, fibers play a role in improving the bending strength, while cement based matrix provides the load transfer between the fibers [1]. The standard GFRC mixture consists of sand, cement, water, and glass fibers. Chemical additives, acrylic polymers and other mineral admixtures are used to modify the properties and increase durability of GFRC. GFRC combines the high compressive strength of the cement based matrix with the bending and tensile strength imparted by the fibers. GFRC is usually produced in thin thicknesses in the range of 10-20 mm. *
Corresponding Author: e-mail: nkabay@yildiz.edu.tr, tel: (212) 383 52 44 675
Due to its low weight compared to traditional concrete, it provides convenience in transportation, storage and inst allation. GFRC was first used by Pilkington Bros in England in 1956 as a building material, but EType glass fibers used in that period failed to yield successful results due to chemical corrosion in the cement matrix [1,2]. Glass fibers generally became fragile over time due to the alkalinity of the cement mortar. For this reason, alkali resistant (AR) glass fiber was developed in 1977 [3]. AR glass fiber is used as micro reinforcement in concrete-like composites due to its stability in high alkaline environments [4]. The main purpose of using AR glass fibers is to improve the mechanical properties of the matrix, such as tensile and flexural strength. The use of glass fiber contributes to the tensile strength of the matrix and provides a controlled development of cracks. Thus, by improving the toughness of the material, it allows the failure to occur with first fiber pull-out instead of fracture [4]. The durability of GFRC decreases over time when exposed to weather conditions, even the stability of the fibers is affected [5]. The main binding material used in concrete structures is Portland cement. It is also used intensively as binder in GFRC. Cement production causes high energy consumption and CO 2 emission, which has negative impacts on the environment. For this reason, many studies have been carried out to reduce the use of cement without increasing the cost. One of these methods is the substitution of cement with environmentally friendly pozzolanic materials in certain proportions. The proper use of pozzolanic materials improves the concrete properties especially at later ages. Calcium silicate hydrate (CSH) gel is produced by the reaction of pozzolans with calcium hydroxide (CH) which occurs due to the hydration of cement [5,6]. Most importantly, pozzolans improve the durability properties, workability and reduces the hydration temperature of the cement based materials [4]. Generally, the GFRC is manufactured by two different techniques for different uses, including spray and premix methods [3,7]. Both grades of GFRC require different physical and mechanical properties. Compared to the premix class, high-fiber sprayed GFRC provides higher tensile strength and shows a more ductile behavior. In the spraying technique, the mortar is produced in a separate place without the addition of glass fiber. The glass filament is cut within the spray gun and then sprayed into the mold with the mortar to manufacture GFRC panels with desired geometry. The aim of this study is to determine the effect of fly ash and blast furnace slag on the workability and the physical and mechanical properties of GFRC. Therefore, Portland cement was replaced with combinations of fly ash and blast furnace slag at a fixed content of 50% by weight.
2.1. Materials and Mix Design
The materials used to manufacture GFRC panels are white Portland cement (CEM I 52.5R), fly ash, blast furnace slag, sand and superplasticizer. Chemical composition and physical properties of binders are presented in Table 1. The mean particle size of cement, fly ash, slag and sand were determined as 15.45 µm, 23.19µm, 10.43µm and 561 µm respectively, and the particle size distribution of the solid materials are shown in Fig. 1. Density, tensile strength, young’s modulus and strain to failure of glass fibre is 2800 kg/m3, 1500 N/mm2, 74000 N/mm2 and 2% respectively. In the experimental study 6 mixtures were produced and the mix design is shown in Table 2. The water to binder ratio was set constant as 0.34 and the sand to binder ratio by weight is fixed as 1.0 in all mixes. Melamine based superplasticizer was used at varying proportions to achieve a desired workability. The glass fiber content was set as 3% and 4% and in some mixes the cement was replaced with varying combinations of fly ash and slag by 50% by weight. Sample panels with a nominal thickness of 20mm were manufactured for each mix by spraying method. The mixes are coded according to the composition of the GFRC (Table 2), where the
number following the “C” letter represents the percent mass of cement in the binder, the number following the “F” and “S” letters indicates the percent mass of the fly ash and slag in the binder. “G3” and “G4” represents the percent mass of glass fiber in the mix. Table 1. Chemical composition and physical properties of binders Oxide % by mass
SiO2 Fe2O3 Al2O3 CaO MgO SO3 Na2O K2O Specific Weight, g/cm3 Specific Surface Area, cm2/g
Blast Furnace Slag 40.6 1.2 12.6 35.1 5.8 0.1 0.8 0.7 2.91 4976
2.2. Casting and Curing
The manufacture of GFRC panels was performed in collaboration with 3H Precast company. Fresh cementitious matrix was sprayed with glass fiber in the molds with dimensions of 500x1000x20 mm. The panels are demolded 24h after casting and stored in laboratory conditions at 20±2 ºC and 60±10% relative humidity for 6 days. The samples are then cut out in two different orientations as shown in Fig. 2 as a standard procedure. Before mechanical tests, the samples are immersed in water for 24h so that they have been aged for seven days at the time of the tests.
2.3. Testing Procedure
2.3.2. Workability Workability of freshly mixed slurry (with no fibers) should comply with the selected processing method. Generally, high workability makes production and compaction easier. However, the strength of GFRC must not compromise or produce unacceptable bleeding to achieve high workability. Therefore use of superplasticizers appropriate for GFRC mixtures to increase workability without a reduction in strength is generally helpful. The workability is
measured before addition of the glass fiber in accordance with EN 1170-1 [8]. This test demonstrate the workability and the uniformity of the freshly produced GFRC mixtures. For this purpose a cylindrical mold with 57 mm inner diameter is placed in the center of a plate. The cylindrical mold is filled with the required amount of fresh mixture and raised vertically. The span diameter is measured after the mixture has stopped spreading as seen in the Fig. 3 to determine the slump (spread) value of mixtures.
Figure 3. Slump test 2.3.3. Dimensional variations Dimensional changes during the service of GFRC material is of primary importance. The fixing systems must be able to accommodate weight and applied loads and also all movements based on changes in humidity and temperature in service. Total dimensional changes are indicative of the combination of movements due to changes in humidity and temperature. All materials containing high amounts of ordinary Portland cement are subject to dimensional changes due to humidity changes in service environments. The dimensional changes are not only dependent on the temperature change but also because of the sunlight absorption on external GFRC surfaces. The dimensional changes are all non-linear and time-dependent. In principle, immersion in water causes expansion and drying causes shrinkage [9]. Diagrammatic representation of dimensional changes in cementitious materials is shown in Fig. 4 [10].
The dimensional variations of GFRC mixtures were determined according to EN 1170-7 [11]. The test samples after aged for 6 days are placed and kept in laboratory at a temperature of 20±3 ºC until the products have reached 10 days age. Subsequent to ageing in the laboratory, the initial length of each test piece was measured and recorded (l0, mm). Three test samples from each GFRC mix was then placed in the ventilated drying oven adjusted to 30±3 ºC. Test pieces were removed from the oven after 21 days and stabilized in the laboratory at a temperature of 20±3 ºC for 6 h. The length of the test pieces were measured and recorded (l 2, mm) to determine the drying shrinkage. Three more test samples from each GFRC mix were placed in a tank filled with water at a temperature of 20±3. After 96h immersion in the water, test pieces were removed from the tank and wiped with a damp cloth. The final length of the test pieces were measured and recorded (l1, mm) to determine the expansion of the mixtures. The residual shrinkage (Δls/l, mm/m) and expansion values (Δle/l, mm/m) of GFRC mixtures were determined by Equations (1) and (2). The value of the extreme dimensional variation (Δl c/l, mm/m) was determined by the Equation (3). ∆𝑙𝑠 𝑙 ∆𝑙𝑒 𝑙 ∆𝑙𝑐 𝑙
2.3.4. Water Absorption and Dry Density The water absorption and dry density of the samples are determined according to EN 1170-6 [12]. The test samples after aged for 6 days are placed and kept in water at 20 ±2 ºC for 24h. The samples are then removed from water, wiped with a damp cloth to remove any surface water and each test piece is weighed to determine the saturated mass (mw). The test pieces are then placed and kept in a ventilated drying oven adjusted to 105±5 ºC for 24h to determine the dry mass (md). Water absorption (γ, %) and dry density (ρ𝑑, 𝑘𝑔 ⁄𝑚3 ) was determined by using the Equations (4) and (5): γ=
2.3.1. Flexural test Flexural strength of GFRC samples was determined according to EN 1170-5 [13], at the 7th day by 4 point bending test. The test pieces were positioned in the testing machine, as shown in Fig. 5, on the two bottom supports with a span length (L) of 300 mm. For each mix 4 samples were tested with the mould face down on the two bottom supports and the other 4 samples with the mould face in contact with the top supports. Test was performed on a displacement controlled testing machine at a loading rate of 1 mm/min. Thickness and width of the samples were measured after the test in the failure zone to the nearest 0.1 mm and the modulus of rupture (σMOR. MPa) is determined by the Equation (6): 𝐹
In this equation FMOR, L, b and d represents the maximum load, span length, width and thickness of test samples, respectively.
3. Results And Discussion
This study aims to research the effect of fly ash and blast furnace slag replacement with cement on the workability, water absorption, density and early age flexural strength of GFRC. Fig. 6 presents the slump test results. Slump values varied between 73 and 107mm and substitution of cement with fly ash resulted in an enhanced workability due to the spherical shape of the fly ash particles. On the other hand slag addition did not have any significant effect on the workability. 120
Figure 6. Slump test results The extreme dimensional variations of the mixtures are presented in Fig. 7. The use of fly ash reduced the total dimensional variation by about 25% compared to the reference mixtures (C100F0S0G3 and C100F0S0G4). It is known that the shrinkage and expansion of cementitious matrix is governed by the amount of cement in the mixture. Therefore, any attempts to reduce the amount of cement provides reduction in shrinkage and expansion. The addition of slag, on the other hand, increased the dimensional variation, never the less the results obtained were still lower than the reference mixes. The increase in glass fiber content was found to slightly increase the dimensional variations of the mixtures.
Figure 7. Total dimensional variations due to shrinkage and expansion Water absorption test results of the mixes are presented in Fig. 8. The use of pozzolanic materials and the increase in fiber content had no significant effect on the water absorption of the mixes, the values varied between 10.8% and 11.5% as seen in Fig.8. Previous studies show that the replacement of cement with fly ash by up to 20% in GFRC reduces water absorption. The reason for the decrease in water absorption is attributed to the pozzolanic reaction that lead to a reduction in pore structure in cement paste [14]. However the slow reaction of pozzolanic materials, especially fly ash compared to the slag, retards this contribution. Therefore the effect of pozzolanic materials on the water absorption should be expected at later ages. Another impact of pozzolanic materials on cementitious matrix is their pore filling capability (contribution of nonreacted particles) especially at earlier ages which also is expected to reduce the water absorption. The median particle size of the binding materials shows that fly ash is coarser and slag is finer than the cement used in this study. Therefore the contribution of non-reacted fly ash particles on the early age water absorption might be hindered due to their coarser particle size compared to the cement. However, fly ash and blast furnace slag as a pozzolanic material has a significant effect on reducing the dry density of mixtures as shown in Table 3. Although densities of reference GFRC mixes were about 2000 kg/m3, lower densities were obtained by replacing cement with fly ash and slag. Table 3. Dry density of the mixtures Sample ID
Figure 8. Water absorption results of GFRC samples The flexural strength (σMOR) of the mixes was evaluated by comparison with the reference mixes C100F0SOG3 and C100F0SOG4, in which the cement was used as the sole binder and the fiber content varied as 3 and 4%. The average load-deflection curves of the mixes are shown in Fig. 9-10 and the average of flexure strength of the mixes are presented in Fig. 11. Test results showed that the flexural strength of the reference mixes are higher than the mixes containing pozzolanic materials. As seen in Fig. 11, the use of high content fly ash and blast furnace slag reduces the early age strength. The increase in the glass fiber content from 3 to 4% did not alter the flexural strength of the reference mix but improved the flexural strength of mixes containing fly ash and slag. The increase in flexural strength was noted as about 20% when higher amount of glass fiber was used in C50F50S0G4 and C50F25S25G4 mixes. C100F0S3G3
Figure 10. Load-Deflection Curves mixes with 4% glass fiber 14 12
C50f50s0g4 C50f25s25g4
Figure 11. Flexural strength of GFRC mixes at 7 days Microstructure of fly ash incorporated GFRC sample is shown in Fig. 12. The figure shows that the fibers are aligned in one direction and are stick together without any dispersion in the matrix and that they are well surrounded with the cementititous matrix.
4. Conclusions
This research investigated the effect of high content fly ash and blast furnace slag replacement with cement on the flexural strength and water absorption of GFRC. Based on the test results the following conclusions may be drawn: 1. Substitution of cement with fly ash resulted in an enhanced workability, however slag addition did not have any significant effect on the workability.
2. Replacement of cement by high amount of fly ash and blast furnace slag reduced the
flexural strength at 7 days. The increase in fiber content enhanced the flexural strength, the effect was more pronounced in mixes containing pozzolanic materials. 3. The use of pozzolanic materials and the increase in fiber content had no significant effect on the water absorption of the mixes. However, use of fly ash has a considerable effect on reducing dry density and total dimensional variation and improving the workability. 4. Further research should be performed to determine the effect of pozzolanic additions on physical and mechanical properties of GFRC at later ages.
Acknowledgements
This work is supported by research fund of Yildiz Technical University (Project number: FYL-2018-3415). The authors also thank 3H Precast company for their helps in supplying the raw materials and manufacturing the GFRC panels.
Share and Cite
AMED, B.; KABAY, N. Glass fibre reinforced precast concrete containing high content pozzolanic materials. Sigma Journal of Engineering and Natural Sciences 2019, Vol. 37, pp. 675-686. https://doi.org/10.62051/ytu.sigma-journal-of-engineering-and-natural-sciences-glass-fibre-reinforced-precast-concrete-containing-high-content-pozzolanic-mater