Mechanical properties of black sugar palm fiber-reinforced concrete

Transcription

1 Article Mechanical properties of black sugar palm fiber-reinforced concrete T. Ferdiansyah and H. Abdul Razak Journal of Reinforced Plastics and Composites 30(11) ! The Author(s) 2011 Reprints and permissions: sagepub.co.uk/journalspermissions.nav DOI: / jrp.sagepub.com Abstract This article reports the results of an investigation on the engineering properties of concrete containing black sugar palm fiber. Three fiber lengths of 15, 25, and 35 mm in four volume fractions, namely 0.2%, 0.4%, 0.6%, and 0.8%, were utilized in this investigation. The values of compressive, flexural, toughness, first crack deflection, first crack toughness, and toughness indices are reported for ages up to 90 days. It was observed that the addition of palm fibers slightly increased the flexural strength of concrete. The incorporation of the fibers had no significant effect on the compressive strength. The mix with 0.8% volume fraction and 35 mm length fiber gave higher toughness and ductility compared to other mixes. Keywords palm fiber, engineering properties, fiber length, fiber volume Introduction Since 1960 s, research work on fiber-reinforced concrete have been carried out extensively leading to a wide range of practical applications. As a result, there exists today a wide range of fiber types available for use as fibers in concrete. These include steel, asbestos, glass, ceramic, polymer, and natural fibers such as hemp, sisal, cotton, coconut, and palm. Many research works have been carried out using steel, polymer, and glass. 1 3 However, investigations on the use of natural fiber are rather limited and surely reported in the literature 4,5 compared to non-natural fibers such as steel and plastic. Principally, the sources of natural fiber are found in several plants, but not all are suitable for use as fibers in concrete. The advantage of using natural fibers is that they are readily available, environment-friendly, and cheap since the production cost is lower than nonnatural fibers. The major problems associated with the use of natural fibers are due to poor durability, low modulus of elasticity, poor bonding, and poor fire resistance. Natural fibers have been used in soil cement construction 6 but their application is mainly in non-structural components such as roofing tiles, concrete masonry blocks, slab for roofing, and construction of water tanks. 4,5 Recent studies using natural fibers such as sugarcane bagasse 4 and cellulose fiber 7 have shown that incorporation of these fibers enhanced the compressive strength, tensile strength, flexure strength, toughness, and impact resistance. This study is concerned with the engineering properties of concrete containing palm fiber as reinforcement. The effect of fibers in concrete depends on two important parameters, namely fiber volume fraction and aspect ratio. Fiber volume fraction is the amount of fibers added to a concrete mix expressed as a percentage of the total volume of the composite. Aspect ratio is calculated by dividing the fiber length by its diameter. The performance of fiber-reinforced concrete also depends on the matrix mix ingredients. 1 The matrix strength of fiber-reinforced concrete, which has a significant influence on the toughness characteristic, 2 Department of Civil Engineering, University of Malaya, Malaysia. Corresponding author: H. Abdul Razak, Department of Civil Engineering, University of Malaya, 50603, Kuala Lumpur, Malaysia hashim@um.edu.my

2 Ferdiansyah and Razak 995 is dependent on the geometry, type, volume fraction of fiber, and bond strength of fiber in the matrix. The addition of fibers to a concrete mix causes a reduction in the workability, normally indicated by low slump value. Fibers have relatively large surface area and as a result the water requirement for a given workability is much higher. This problem can be overcome by the use of water-reducing admixture or superplasticizer 1 in the concrete mix. The physical properties of the fiber influencing workability are the aspect ratio and the volume percent of fiber. In general, an aspect ratio of less than 100 and fiber content not higher than 2% represent the maximum limit to obtain a mix with good workability. 8 Higher fiber content leads to reduced workability and longer mixing time. As the fibers are added to the fresh concrete mix manually in the mixer, the dispersion is not uniform and may result in balling. Fiber dispersion in the wet mix is also an important factor in the case of properties of the concrete such as compressive strength, flexural strength, and toughness. Experimental work Materials Locally produced ordinary Portland cement (OPC) equivalent to ASTM Type 1 was used. The coarse aggregate was crushed granite ranging between 4.75 and 19 mm in size. The fine aggregate was siliceous sand and consists of the combination of particle size according to ASTM sieve sizes, namely coarse (8 16 mesh), medium (20 40 mesh), and fine (40 60 mesh). A sulfonated naphthalene formaldehyde-based superplasticizer with 40% solid content was added to achieve a slump value within the range mm for each mix at a constant water/cement ratio of 0.4. Mixing and curing water was taken directly from tap supply. The fibers used were obtained from the trunk of the black sugar palm or commonly known in Indonesia as the Ijuk tree. The Latin name of this particular species of palm is Arenga Pinnata and can be found widely in the swamp forests of South East Asia and India. The fibers were untreated, before concreting; the fibers were cleaned and cut to the required size i.e., 15, 25, and 35 mm. The physical and mechanical properties of the palm fibers are given in Table 1. The results were obtained from test on 10 fiber specimens. A total of 13 mixes, which include the plain concrete mix which acts as the control and 12 fiber-reinforced concrete mixes with fiber lengths 15, 25 and 35 mm in four volume fractions 0.2%, 0.4%, 0.6%, and 0.8% were produced. Table 2 presents the mix proportions used in the investigation. Specimen preparation and test procedures Two types of concrete specimens were used in this investigation. They were mm 3 cubes and mm 3 prisms for the determination of compressive strength and flexural toughness, respectively. Mixing of the ingredients was carried out using a pan mixer and the sequence of mixing can be summarized as follows: aggregates and cement were mixed dry for approximately 1½ minutes, and then three quarters of the mixing water was added, followed by the superplasticizer, and finally the remaining water. In the case of the fiber concrete, after the last water, the fibers were added to the wet mix and gradually in small amounts to avoid clustering. The mixing was continuing approximately 1 min after the last fiber added. The specimens were demolded after 24 h and subjected to full water curing in a tank until the age of testing. The workability of fresh concrete was measured using the slump test. The compressive strength and flexural toughness of the concrete were tested in accordance with BS 1881: Part116: 1983 and ASTM C b, respectively. The strength development was monitored at 3, 7, 28, 56, and 90 days. Results Compressive strength The compressive strength values are based on the average results of three samples. The results for 28 days are presented in Table 3. The compressive strength of palm fiber-reinforced concrete is relatively less affected by usage of fibers as compared to other properties of hardened concrete. The compressive strength at 28 days for the control was 53.6 MPa. The highest compressive strength at 28 days was 60.7 MPa for a mix having 0.2% fiber volume fractions with fiber length 35 mm (F-0.2/35) compared to the control, which is an increase of 13%. The lowest compressive strength at 28 days was 46 MPa for a mix having 0.8% fiber volume fraction with fiber length of 35 mm i.e., a reduction of 7% compared to the control. The coefficient of variation for control was 2.1%, while for palm fiber-reinforced concrete, the coefficient of variation varied between 1.2% and 4.7%. Figure 1 shows that the presence of palm fibers did not significantly increase the compressive strength. In fact, some mixes showed lower strength, especially the concrete having 0.8% fiber volume fractions with 35 and 15 mm fiber lengths. This result is in agreement with past studies, 9,4 whereby the addition of natural fiber had little effect on compressive strength. The palm fiber is only slightly hydrophilic with low water absorption, 0.5%. The effect was more significant in the

3 996 Journal of Reinforced Plastics and Composites 30(11) mix with higher fiber volume. Based on visual observation, the mix with higher fiber volume mixes basically caused more air to be entrapped and have more voids when hardened, resulting in lower compressive strength. For mixes with higher fiber volume i.e., 0.6% and 0.8%, the effect of fiber length was inconsistent. This inconsistency was also found by Ozyildirim et al. 10 In these mixes, the effect of fiber volume was more significant than fiber length. It was proved by the increase in strength which only occurred for the mix with 25 mm fiber length, but the mixes with 15 and 35 mm fiber lengths showed a decrease in compressive strength. This is due to the clustering of the fibers, making it difficult to have a random distribution of fibers in the wet mix. Hence, the concrete becomes more difficult to compact. This phenomenon is referred to as the balling effect. Table 1. Properties of palm fiber Properties a Diameter mm Yield stress MPa Modulus of elasticity MPa Elongation at break 2 5% Specific gravity 1.05 Absorption capacity 0.5% a Values based on average of 10 specimens. Flexural strength The results for flexural strength at 28 days are presented in Table 3. Each value is the average of three specimens tested. The coefficient of variation for the control was 1.4%, while for palm fiber-reinforced concrete, the coefficient of variation ranged between 1.3% and 7.2%. At 28 days, the highest flexural strength was 6.05 MPa for F-0.8/35, while the control achieved a value of 5.55 MPa. It was believed that the large number of fibers improves the bonding in the matrix. The increase of the flexural tensile strength also may be attributed to the fact that the fibers suppressed the development of microcracks into macrocracks and consequently the apparent tensile strength of the matrix increases. 11 This increase was relatively small whereby the improvement obtained using fiber was less than 10% compared to concrete without fiber. This slight increase in flexural strength with addition of natural fiber was also reported by other researchers. 4,12 The development of flexural strength with age is shown in Figure 2. The fiber content had a positive effect on the flexural strength. Generally, the flexural strength increased with age from 7 days onwards, but some mixes also exhibited small reduction in strength after 56 days. The reasons behind the inconsistency may have been by the structural deterioration of lignin. Lignins comprise phenolic polymers, which when present in the cell walls of plants and with cellulose are responsible for the stiffness and rigidity of the plant stems. Lignins are especially associated with wood plants, since up to 30% of the organic matter Table 2. Mix proportions Aggregate Fiber Mixtures Cement (kg/m 3 ) Water (kg/m 3 ) Fine (kg/m 3 ) Coarse (kg/m 3 ) Volume (%) Length (mm) F-0.2/15 a F-0.2/ F-0.2/ F-0.4/ F-0.4/ F-0.4/ F-0.6/ F-0.6/ F-0.6/ F-0.8/ F-0.8/ F-0.8/ a Mix with volume fraction 0.2% and length 15 mm.

4 Ferdiansyah and Razak 997 Table 3. Mechanical properties of fiber-reinforced concrete at 28 days Fiber Mixtures Volume (%) Length (mm) Compressive strength (MPa) Flexural strength (MPa) F-0.2/ F-0.2/ F-0.2/ F-0.4/ F-0.4/ F-0.4/ F-0.6/ F-0.6/ F-0.6/ F-0.8/ F-0.8/ F-0.8/ consists of lignin. Since the structure of lignin was deteriorated, the stiffness and rigidity of the fiber also decreased. Toughness According to ASTM C1018, toughness is the energy equivalent to the area under the load deflection curve up to a specified deflection. For the purpose of this study, the limit on the final deflection is fixed at 2 mm. The addition of palm fiber resulted in concrete with increased toughness compared with the plain concrete, as given in Table 4. However, this was only apparent for the mixes with high fiber volume. The mixes with 0.2% and 0.4% fiber volume showed little or no improvement in toughness properties compared with the plain concrete. It was observed that the load deflection behavior of these mixes was similar to that of the plain concrete, whereby the failure mode was brittle and the test specimens fractured into two halves, therein unable to sustain any further loading. The results at 28 days for the fiber mixes with 0.6% and 0.8% volume fraction indicated that the toughness increased with volume fraction and fiber length. The effect was more pronounced for the mix with maximum fiber length and volume fraction. The relative increase in toughness due to an increase in volume fraction from 0.6% to 0.8% was about 8 10%, while the relative increase due to increase in fiber length was much smaller, which was approximately 2 5%. The complete results of toughness for all testing ages are shown in Figure 3. According to a study conducted by Gao et al., 13 the fiber content had a more significant influence on toughness compared to fiber length. However, in this study, the trend was not apparent at 90 days. There was a marked increase in toughness from 7 to 28 days, which was due to the effect of concrete maturity, but at later ages, the increase became insignificant. As a matter of fact, for the mix with 0.8% volume fraction, there was a slight depreciation in toughness with age. There are two probable reasons for this phenomenon. First, the alkali pore water attacks the fibers reacting with the lignin and hemicelluloses present in the middle lamellae of the wood fiber, thus weakening the link between the individual fiber cells that constitute the natural fiber. 9,12 This affects the tensile strength of the individual fibers. Second, the addition of a chemical admixture may have a weakening effect on the chemical bonding between the fiber and the matrix causing a decrease in residual load and the corresponding toughness value computed. 14 These effects seem to be more pronounced in the mix with high fiber volume and longer fiber lengths, wherein the distribution of the fibers in the mix is likely to be uniform. First crack deflection According to Trottier and Banthia, 2 the first crack deflection is defined as the value corresponding to the point on the load deflection plot when the matrix has ruptured and whatever load remaining is solely taken up by the fiber. This is interpreted as the point of maximum load on the plot. The average values of the first

5 998 Journal of Reinforced Plastics and Composites 30(11) %, 15mm 0.2%, 25mm 0.2%, 35mm Compressive strength (MPa) %, 15mm 0.4%, 25mm 0.4%, 35mm 0.6%, 15mm 0.6%, 25mm 0.6%, 35mm %, 15mm 0.8%, 25mm 0.8%, 35mm Age (days) Figure 1. The development of compressive strength with age for all mixes. crack deflection for the control and fiber concrete mixes are given in Table 4. It was apparent that the first crack deflection, which also relates to ductility, increased with the addition of the fibers. Overall, it can be observed that the increase was between 30% and 50% compared to that of the control, and that longer fiber length and higher volume fraction contributed to higher values of first crack

6 Ferdiansyah and Razak %, 15mm 0.2%, 25mm 0.2%, 35mm Flexural strength (MPa) %, 15mm 0.4%, 25mm 0.4%, 35mm 0.6%, 15mm 0.6%, 25mm 0.6%, 35mm %, 15mm 0.8%, 25mm 0.8%, 35mm Age (days) Figure 2. The development of flexural strength with age for all mixes.

7 1000 Journal of Reinforced Plastics and Composites 30(11) Table 4. Toughness characteristics of fiber-reinforced concrete at 28 days First crack Toughness indices Mixtures Deflection (mm) Toughness (kn.mm) Toughness (kn.mm) I 5 I 10 I F-0.2/ F-0.2/ F-0.2/ F-0.4/ F-0.4/ F-0.4/ F-0.6/ F-0.6/ F-0.6/ F-0.8/ F-0.8/ F-0.8/ Toughness (kn.mm) %, 15mm 0.6%, 25mm 0.6%, 35mm %, 15mm 0.8%, 25mm 0.8%, 35mm Age (days) Figure 3. The development of toughness with age. deflection. Results from past investigations conducted by Marikunte et al. 1 and Trottier and Banthia 2 confirmed that the addition of fibers increased the first crack deflection in varying magnitudes ranging between 200% and 400%. First crack toughness The development of first crack toughness with age for all the fiber mixes is shown in Figure 4. The results indicated that there was a significant increase in first crack

8 Ferdiansyah and Razak %, 15mm 0.2%, 25mm 0.2%, 35mm First crack toughness (kn.mm) %, 15mm 0.4%, 25mm 0.4%, 35mm %, 15mm 0.6%, 25mm 0.6%, 35mm %, 15mm 0.8%, 25mm 0.8%, 35mm Age (days) Figure 4. The development of first crack toughness with age for all mixes.

9 1002 Journal of Reinforced Plastics and Composites 30(11) I I %, 15mm 0.6%, 25mm 0.6%, 35mm 8 6 I 5 Toughness indices I %, 15mm 0.8%, 25mm 0.8%, 35mm 10 I I Age (days) Figure 5. The development of toughness indices with age. toughness due to fiber addition with substantial reduction in maximum crack width and delay of the occurrence of the first crack deflection as tabulated in Table 4. However, the depreciation in values with age was less significant. For the computation of the first crack toughness, the region prior to peak load was utilized. This presumes that the load-carrying capacity is provided solely by the strength of the matrix and the contribution of the fibers is merely in bridging across the microcracks which delays the initiation of the first crack. At 28 days, the increase amounted to about 3 50% compared to the control, as indicated in Table 4. The highest first crack toughness at 28 days was 60.7 MPa for a mix having 0.8% fiber volume fraction with a fiber length of 35 mm. The results at 28 days also indicated that the first crack toughness increased with volume fraction and fiber length. An investigation conducted by Nanni 15 deduced that the addition of steel fiber only slightly increased the first crack stress. Shah 11 also inferred that the type and

10 Ferdiansyah and Razak 1003 amount of fibers did not significantly enhance the first crack stress of fiber-reinforced concrete. Toughness indices The toughness index characterizes the toughness up to a specified end-point deflection according to ASTM C1018. In this study, three toughness indices were computed, which are I 5, I 10, and I 20. For the control, 0.2% and 0.4% fiber volume mixes, the toughness index values were unity, since it did not possess any postcracking strength, which was evident from the load deflection plots. The results in Figure 5 showed that there was a slight or gradual increase in toughness indices after 28 days for all the fiber mixes, which reflected a different trend when compared with the toughness values. Based on the load deflection plots, there was an increase in peak load with age as a result of concrete maturity, coupled with an increase in the unstable region. The values of the indices were computed from the plots based on the specific end-point deflection which lies in the unstable zone, subsequently giving an apparent increase of the values computed as the concrete ages. In general, fiber volume and fiber length did not have any significant effect on the toughness index values. This reinforces the earlier argument which suggests that the role of the fibers in the post-cracking stage is confined to acting as ties across the major crack, which only affects the load-carrying capacity or the residual load value after the matrix cracks. On the load deflection plot, the residual load determines the stable zone, which follows the unstable zone. The ability to sustain loads after cracking is considerably dependent on the tensile strength of individual fibers and the bond between the fiber and matrix. Based on the values of the indices at 28 days, as given in Table 4, the performance of the F-0.6/35 mix was superior compared to the other mixes, which recorded values 4.7 and 8.6 for I 5 and I 10, respectively. However, for I 20, the F-0.8/35 mix gave the highest value. The values of the indices obtained were similar to those obtained by Ezeldin and Lowe 7 in their investigation on the mechanical properties of steel fiber-reinforced concrete. Based on the results of this study, it appears that palm fiber can be as effective as steel fibers in terms of enhancing a specific toughness property. Conclusions The findings of an experimental investigation on the hardened properties of black sugar palm fiber-reinforced concrete are reported. The following conclusions can be derived: 1. The compressive strength did not show any significant improvement and a consistent trend with addition the fibers. In contrast, some mixes exhibited a drop in strength, especially mixes with fiber volume fraction of 0.8% due to difficulty in compaction. 2. In general, the addition of fibers increased the flexural strength, but was not significant, with increase less than 10% compared the control. Fiber length had a more influencing role on flexural strength compared to fiber volume. 3. The incorporation of palm fiber into the mix had a positive effect on the ductility of the concrete and increased toughness characteristics. 4. The ability to sustain the applied load after cracking was observed only in the 0.6% and 0.8% fiber volume mixes. In contrast, the post-cracking behavior of the 0.2% and 0.4% fiber volume mixes resembled that of the plain concrete which failed in a brittle manner. 5. The toughness at early age showed significant increases up to 28 days due to the effect of concrete maturity. However, at later ages, the increase becomes minimal and even experienced a reduction as a result of the reaction between the fibers and the alkaline pore water. The effect of the reaction is more pronounced in the mix with high fiber volume fraction. 6. From this study, the effect of fiber length and volume fraction on the toughness characteristics showed no definite trend and becomes less predominant with age. Funding This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. References 1. Marikunte S, Aldea C and Shah SP. Durability of glass fiber reinforced cement composites: effect of silica fume and metakaolin. Adv Cem Based Mater 1997; 5: Trottier JF and Banthia N. Toughness characteristic of steel-fiber reinforced concrete. J Mater Civil Eng 1994; 6(2): Zheng Z and Feldman D. Synthetic fibre-reinforced concrete. Prog Polym Sci 1995; 20: Stephens D. Natural fibre reinforced concrete blocks. In: 20th WEDC Conference, Colombo, Sri Lanka, 1994, pp Yao W and Li Z. Flexural behavior of bamboo-fiber-reinforced mortar laminates. Cem Concr Res 2003; 33:

11 1004 Journal of Reinforced Plastics and Composites 30(11) 6. Ghavami K, Filho RDT and Barbosa NP. Behaviour of composite soil reinforced with natural fibres. Cem Concr Compos 1999; 21: Ezeldin AS and Lowe SR. Mechanical properties of steel fiber reinforced rapid-set materials. ACI Mater J 1991; 88(4): Berg TW. Fiber Reinforced Concrete. The Technical Advisor, < fiber_rc/fiber_rc.html> (1993, accessed 2003). 9. Sarigaphuti M, Shah SP and Vinson KD. Shrinkage cracking and durability characteristics of cellulose fiber reinforced concrete. ACI Mater J 1993; 90(4): Ozyildirim C, Moen C and Hladky S. Investigation of fiber-reinforced concrete for use in transportation structures. Virginia: Virginia Transportation Research Council, VTRC 97-R15, Shah SP. Do the fibers increase the tensile strength of cement-based matrixes? ACI Mater J 1991; 88(6): Marikunte S and Soroushian P. Statistical evaluation of long-term durability characteristic of cellulose fiber reinforced cement composites. ACI Mater J 1994; 91(6): Gao J, Sun W and Morino K. Mechanical properties of steel fiber-reinforced, high-strength, lightweight concrete. Cem Concr Compos 1997; 19: Zia P, Ahmad S and Leming M. High-performance concretes a state-of-art report ( ), FHWA-RD , Georgetown Pike Mclean, Virginia: US Department of Transportation, Federal Highway Administration, Nanni A. Splitting-tension test for fiber reinforced concrete. ACI Mater J 1988; 81(4):