Assessment of mechanical properties of fibrous mortar and interlocking soil stabilised block (ISSB) for low-cost masonry housing

F. Qamar^a*, T. Thomas^b, M. Ali^c

^a. University of Warwick (Warwick, UK)

^b. Engineering Dept. University of Warwick (Warwick, UK)

^c. Department of Civil Engineering, Capital University of Science and Technology (Islamabad, Pakistan)

*f.qamar@warwick.ac.uk

1. INTRODUCTIONTOP

Due to increase of population in developing countries the need forlow-cost residential housing has increased considerably around the world (1, 2). At the same time need for a faster construction system has led to the change in conventional approach of masonry construction to interlocking construction technique. Interlocking masonry units or blocks can be laid without layers of mortar and require less labour (3-5). Interlocking masonry units differ from conventional blocks in that these units can be assembled with geometrical features built into blocks without the need of a mortar layer (6). Most interlocking blocks available in the industry differ in geometry, material and dimensional features. As wellas advantages of removal of bedding mortar and reduction in cost, interlocking blocks have some limitations. These include the requirement of strict dimensional tolerance and reduced lateral resistance. The structural behaviour of an interlocking block wall of a masonry house may differ from that of a conventional masonry wall when subjected to in-plane and out-of-plane loading due to geometrical imperfection (7-10). Masonry walls are mostly used in resisting compressive and horizontal (out-of-plane) loading like wind and earthquake. Therefore, the structural performance of the interlocking wall under these loadings is a primary concern(11,12). In previous studies different types of interlocking block, mainly depending on material, geometry and interlocking mechanism, were used. Different types of mechanical property were explored, as detailed in Table 1. In research undertaken by Fundi et al. 2018 cement-stabilised soil blocks were used to test the behaviour of small wall (900 mm length and 1200 mm height) for vertical and horizontal loading. The optimisation of compressive strength of blocks by the addition of cement stabiliser was also determined and a minimum of 2.5 MPa compressive strength found which could be increased to 4.5 MPa at 28 days’ curing by the addition of lime in soil and cement. In another study by Anand (6)the compressive strength of interlocking concrete blocks was reported to be 5.42 MPa for single block (400 mm x 150 mm x 100 mm) and 3.77 MPa for small wall (400 mm x 600 mm x 100 mm). Shear and tensile strengths were also reported to be 0.48 MPa and 0.21 MPa respectively. In a study by Lee (1), where interlocking block columns were built and interlocking holes were infilled with mortar and the reinforcement, the material properties and failure behaviour were studied and the compressive strength of 14.28 MPa was reported for interlocking blocks. In research conducted by Jaafar (3) hollow interlocking blocks were used and a correlation developed between the compressive strength of a single block, prism and wall panel. It was found that the compressive strength of a single block was 15.2 MPa. The prism and wall panel compressive strengths were 0.47 and 0.39 respectively, times the strength of interlocking block. In all research studies the compressive strength of different types of interlocking block was measured and found to be in the range of 2.5 -16.48 MPa. Other mechanical properties like shear strength and compressive toughness were also measured and found to be in the range of 0.48-2.65 MPa and 0.56, respectively. It became evident from the outcome of different studies that properties of interlocking block vary depending on geometry, material and interlocking mechanism and could affect the overall strength of masonry wall.

As reported in the literature, interlocking block wall showed reduced lateral resistance to wind and earthquake. Therefore, to enhance lateral resistance, different techniques, like plain or reinforced grouting, surface bonding and plastering, were used (6). Use of natural fibres in mortar was considered in a variety of studies and the mechanical properties of specimens with and without fibres were compared, as detailed in Table 2. In a study by Pereira (13) mechanical properties like compressive and bending strength and toughness of plain mortar and mortar with sisal fibres were compared. It was observed that there was a 21% decrease in compressive strength with the addition of fibres, but a 60% and 600% increase in bending strength and toughness respectively were reported. In research by Savastano (14,15) natural fibres like sisal and banana with cementitious composite were tested under three-point bending. Mechanical properties, including flexural strength, modulus of elasticity and toughness index were obtained and compared with those of non-fibrous composite. It was found that the addition of fibres resulted in 100% improvement in flexural strength and significant enhancement of toughness index. This could be of interest where cement-based components are exposed to lateral loading like wind and earthquake. Basalt-fibre reinforced mortar was tested by Zych (16) and Asadi (17). A 20% increase in bending strength and 15% reduction in compressive strength were found to be due to the addition of fibres. This finding/ result isconsistentwith that of Pereira (13). In a study by Ali (18) coconut-fibre reinforced concrete cylinders and beamlets were tested for dynamic and mechanical properties and it was found that fibre content of 5% gave best properties. Lertwattanaruk (19) used coconut and oil palm fibre in cementitious composite which was tested for mechanical properties prior to its recommendation for residential housing application. Differing fibre content of 5%, 10% and 15% by mass of cement was used and it was found that increasing the fibre content resulted in lower compressive and flexural strengths. In a study by Toihidul Islam(20) impact resistance of masonry units bounded with fibrous mortar was investigated. Mechanical properties including compressive and flexural strength were also evaluated for fibrous cementitious composites. It was found that 0.5% fibre content in the mortar improved impact resistance of masonry unit, whereas bond strength was reduced with an increase in fibre content from 0 - 0.5%. The outcome of these various studies has indicated that type and content of fibre influences the mechanical properties of cementitious material.

In otherstudies, the performance of interlocking block masonry wall was tested and it was found that the compressive properties of the wall were directly related to the strength of masonry units (8). Therefore, in this study material properties of interlocking soil-stabilised block (ISSB) are explored for their possible use in a typical masonry house, as shown in the Figure 1.

Figure 1 shows the plan, elevation and section of a proposed masonry house. Walls consist of single ISSB width having block projecting outside wallat regular intervals for its stability. A timber roof truss is anchored to walls with steel ties. The structure is mortar-free and lateral resistance is to be ensured by interlocking mechanism and plastering. The mechanical properties of ISSB and fibrous mortar reported by other researchers are detailed in Tables 1 and 2. It was found that for ISSB only compressive strength had been measured: other properties like modulus of elasticity, pre/post crack energy absorbed, total energy absorbed and toughness index had not been evaluated. As these latter properties are required for modelling wall performance they need to be measured. To the author’s knowledge there is no study where mechanical properties of interlocking blocks had been evaluated relatingto the overall performance of a masonry structure. Therefore it is important to assess the mechanical properties for ISSB. On the other hand, for fibrous cementitious matrix as a plaster (reinforced coating), some evidence of mechanical properties like compressive strength, flexural strength and toughness index is available, but only for natural and artificial fibres possessing high tensile strengths. Therefore, the mechanical properties (as detailed earlier) of sisal, rice straw and treated rice straw cementitious matrix were experimentally evaluated. 2% and 5% of fibre content, by mass of cement and 50 mm in length, were used. Mortar cube samples were tested as per standard BS EN 1015-11(21). Surface morphology of sisal, rice straw and treated rice straw were compared using Bruker 3D optical microscope to evaluate the surface roughness of fibres. Microscopic observation of tested mortar samples was carried out to study the bond between the fibres and mortar. Empirical relations were also developed between compressive strength and modulus of elasticity with the help of experimental data, for the interlocking block, plain and fibrous mortar.

2. EXPERIMENTAL PROCEDURE TOP

2.1. Materials and composition TOP

In the preparation of interlocking soil-stabilised blocks soil and ordinary Portland cement were used. Soil was outsourced from Boko quarries Dar es Salam and it was obtained at a depth of 1 m below the earth surface in order to eliminate the addition of humus materials. The cement was sourced from Twiga cement factory Dar es Salam. Chemical and physical properties as adopted from the manufacturer are shown in Table 3. Similar chemical composition of portland cement was reported by Jaurez (4) and Zych (16). The physical properties of cement reported in Table 3 were also described by Lertwattanaruk (19). The properties of soil used for ISSB are detailed in Table 4, as reported by (7). In this study it was reported that the addition of fibres resulted in an increase in porosity and water absorption and a decrease in density as compared with non-fibrous samples, which was consistent with findings of other studies by (22- 25).

The same cement was used for the preparation of plain and fibrous mortar cubes. Three types/conditions of natural fibre were used, including sisal, rice straw and treated rice straw as shown in Figure 2. Sisal and rice straw are classified as naturalfibres, which are further categorised as plant fibres based on their origin (25). The length of the fibres was about 50 mm. The width of rice straw was about 7mm and the diameter of sisal fibres was reported to be 100±300 µmm (26, 27). These were obtained locally from Kibaha district Dar es Salam. In the case of treated rice straw fibres were washed using portable water and then boiled for half an hour. The boiled water was then drained and fibres were washed again and then left to dry in the sun before use in mortar cubes. In mortar cube preparation river sand obtained from Kibaha district Dar es Salam was also used. The properties of sand as obtained from supplier are detailed in Table 5. Similar properties of sand were reported by Khonsari (28). Sand of relative density of 2.51, absorption of 1.77% and fineness modulus of 2.14, classified as fine aggregates, was used.

2.2 Mix ratio and specimen preparation TOP

2.2.1 Plain and fibrous mortar cubes TOP

The mix proportions for plain and fibrous mortar cubes are listed in Table 6. Cement and sand were dry mixed and water was added to make a workable mix. For fibrous mortar cubes fibres were added to a dry mix of cement, sand and water was added later to make it a workable mix. 2% and 5% (by mass of cement) fibre contents were used. Mortar mixes for fibrous and non-fibrous samples were cast into cubes of dimensions 100 mm x 100 mm x 100 mm. In order to ensure uniformity in test conditions the cubes were poured from the same batch of mortar. The specimens were kept in their moulds and demoulded after 7 days. They were stored at ambient temperature and humidity for at least 28 days before testing as per BS EN 1015-11(21).

2.2.2 ISSBs TOP

ISSBs were available in the National Housing and Building Research Agency Tanzania (NHBRA) which were prepared using manually pressed machine with dimensions of 300 mm (length) x 150 mm (width) x 100 mm (height). The soil-cement ratio for ISSBs was 12: 1.

2.3 Testing TOP

The cubes and blocks were tested as per standards (21, 29) in a universal testing machine and were instrumented as shown in Figures 3 and 4, respectively, to derive the stress-strain response. The loading surface was kept plane and parallel by the use of steel plates for cubes and a wooden frame for blocks. Generally, three samples of each case were tested to obtain an average. In total 18 samples of mortar cubes with different combinations as detailed in Table 6 were tested. Nine samples of blocks were tested with three arrangements, as shown in the Figure 4. The mechanical properties, which include compressive strength, pre and post crack energy absorbed, total energy absorbed, modulus of elasticity and toughness index were observed. Surface morphology of sisal, rice straw and treated rice straw was compared using Bruker 3D optical microscope to evaluate the surface roughness of fibres. Microscopic observation of tested mortar samples was carried out to study the bond between the fibres and mortar. Empirical relations were also developed between compressive strength and modulus of elasticity with the help of experimental data, for the interlocking block, plain and fibrous mortar.

3. RESULTS AND ANALYSIS TOP

3.1. Compressive properties of mortar cubes TOP

3.1.1. Compressive behaviour of mortar cubesTOP

The compressive stress-strain curves are shown in Figure 5 (a-f) for each sample of fibrous and non-fibrous mortar cubes. In all cases samples with fibres showed higher stress and strain as compared with plain mortar sample resulting in enhanced ductility due to the presence of fibre. In Figures 5c and 5e, which represent 2% and 5% sisal fibres samples, much higher stiffness and ductility are observed than in the case of all other samples. This is due to the high tensile strength of sisal fibre and its presence in the mix bridges the cracks and resists deformation. Figure 6 (a-f) shows the fractured surfaces of tested samples. Figure 6a shows the tested plain mortar sample and crushing failure of the sample is observed leading to the spalling. Figures 6b and 6d (samples with 2% and 5% rice straw) show buldging effect due to the presence of fibre and provide the ability of taking greater load. Figures 6c and 6e (samples with 2% and 5% sisal fibre) show the bridging effect due to their high tensile strength. This bridging effectalso shows the ability to delay the progression of cracks causing increased failure load. In the case of fibrous samples, the intentionally broken sample revealed that the debonding of fibres was also observed, instead of breaking of fibres as shown in Figure 6d. This greater debonding and reduced fibre fracture may be due to the higher tensile strength of fibres and lower bond strength.

3.1.2. Compressive strength, compressive pre-crack/post-crack, absorbed energies and compressive toughness index of mortar cubesTOP

Compressive strength was taken as the peak value of stress from the stress-strain curve. The area below the stress-strain curve up to the stress of first crack was defined as pre-crack absorbed energy PE_p. The area below the stress-strain curve from the first crack stress to maximum load was consideredto be post crack energy absorbed CE_p. The total area below the stress-strain curve from initial to maximum load was categorised as total energy absorbed TE_p. Compressive toughness index (CTI)p was determined by the ratio of total energy absorbed (TEp) and pre-crack energy absorbed PEp. Table 7 shows the average values of first crack load, maximum load, compressive strength, PEp, CEp, TEp and CTIp of samples A to F. It can be observed 17% and 5% improvement in the first crack load by addition of 2% and 5% sisal fibres, respectively, as compared to plain sample. The maximum load was noted to be 8% higher for the 2% sisal fibre sample only as compared with the plain sample. This could be due to the high tensile strength of sisal fibre and the better bond with the cementitious matrix. In comparison withthe compressive strength of the plain mortar sample to 2% sisal fibre sample, a3% increment was observed, whereas in all other cases reduction in compressive strength for fibrous samples was noted as compared with that of the plain mortar sample. This was consistent with the finding of many studies (13, 30) where addition of fibre resulted in reduction of compressive strength, mainly due to decrease in the density of sample and because ofthe creation of voids due to the presence of fibre. The diameter/width of rice straw was greater than that of sisal fibre. Therefore greater reduction in compressive strength was observed in rice straw samples.The CEp of all samples A to F was calculated as 0.017 MJ/m³, 0.014 MJ/m³, 0.03 MJ/m³, 0.016 MJ/m³, 0.03 MJ/m³ and 0.005 MJ/m³, respectively, which showed significant improvement of 76% for both 2% and 5% sisal fibre samples as compared with that of the plain mortar sample. The PEp of plain and fibrous samples A to F wasrecorded as 0.0035 MJ/m³, 0.0043 MJ/m³, 0.012 MJ/m³, 0.004 MJ/m³, 0.009 MJ/m³ and 0.005 MJ/m³ respectively. This also indicated an increment of 2.21 and 1.57 times for 2% and 5% sisal fibre samples respectively, as compared with the plain mortar sample. The TEp of samples A to F was observed as 0.021 MJ/m³, 0.018 MJ/m³, 0.041 MJ/m³, 0.02 MJ/m³, 0.036 MJ/m³ and 0.011 MJ/m³respectively. This showed 95% and 71% improvement for 2% and 5% sisal fibre samples respectively, as compared with that of the plain sample. The compressive toughness index showed values of 1.41 and 1.31 for 2% and 5% sisal fibres respectively, and 1.30 and 1.23 for 2% and 5% rice straw samples respectively, as compared with 1.21 for the plain mortar sample, which was the indication of fibre bridging which helped in limiting the size of cracks and provided enhanced resistance to stresses after the initiation of cracks.

Comparison of compressive strength of plain and fibrous mortar samples is shown in Figure 7. The reduction of 2.1 times, 3.16 times and 2.71 times was noted for 2% rice straw, 5% rice straw and 2% treated rice straw, respectively, as compared with plain samples. However, a 6% increment was found with the addition of 2% sisal fibre than the plain mortar sample, whereas a 5% reduction in compressive strength was observed in the 5% sisal fibre mortar sample. This showed that the content and type of fibre influenced the compressive strength of properties and lowering the content of fibre could increase compressive strength.

3.1.3. Modulus of elasticity of mortar cubes TOP

Modulus of elasticity E_p was calculated as the ratio of stress change to strain change within elastic limits. The averaged E_p of samples A to F is given in Table 7. The E_p for plain mortar was 2.99 GPa and for 2% and 5% rice straw it was 2.05 GPa and 1.13 GPa respectively. This showed a 31% and 62% reduction thanthe plain mortar sample. For the treated 2% rice straw sample the value of E was 1.79 GPa. This showed a 40% reduction than the plain sample. The value of E for the 2% and 5% sisal fibre samples was calculated to be 7.17 GPa and 2.23 GPa. This showed 1.39 times increment in the 2% sisal fibre sample, a 25% reduction in the 5% sisal fibre sample as compared with the plain sample. It can be observed that, as with other properties, the 2% sisal fibre sample showed best values for modulus of elasticity as compared with plain and rice-straw mortar samples due to its high tensile properties and lower fibre content.

3.2. Compressive properties of ISSB TOP

3.2.1. Compressive behaviour of ISSBTOP

The compressive stress-strain curves are shown in Figure 8(a-c) for each sample of single block, 1*2 blocks (1 block wide and 2 blocks high) and 2*2 blocks (2 blocks wide and 2 blocks high). Single-block samples showed higher stress and strain as compared with the other two sets of blocks. The reduction in stress-strain of the 1*2 blocks and 2*2 blocks compared with the single block might have been due to interlocking interaction between the different block units. Figure 9(a-c) shows the fractured mechanism of tested samples. Figure 9a shows the fracture surface of the single block sample and shear/vertical cracks can be observed in the region of interlocking. This is due to stress peak in the interlocking region where the resistance area is found to be smaller than in the rest of the block. Figures 9b and 9c show the failure of 1*2 blocks and 2*2 blocks, which indicates crack development at an inclined angle and not on the joints. This led to blocks’ failing through crushing by opening of small gaps on individual blocks and finally initiation of cracks without opening up of joints. This could be due to the efficiency of interlocking mechanism of the blocks.

3.2.2. Compressive strength, compressive pre/post crack, absorbed energies and compressive toughness index of ISSBTOP

The compressive strength was taken as the peak value of stress from stress strain curve of ISSB. The area below the stress-strain curve up to the stress of first crack was defined as the pre-crack absorbed energy PE_b. The area below the stress-strain curve from the first crack stress to maximum load was regarded as the post-crack energy absorbed (CE)_b. The total area below the stress-strain curve from initial to maximum load was categorised as the total energy absorbed (TE)_b. The ratio of total energy absorbed and first-crack energy absorbed was taken as the compressive toughness index (CTI)_b. Table 8 shows the values of first-crack load, maximum load, compressive strength, PE_b, CE_b, TE_b and CTI_b of samples for single block, 1*2 blocks and 2*2 blocks. The first crack load of 11.7 kN, 34.8 kN, and 51.1 kN were found for single block, 1*2 block and 2*2 block samples respectively. Maximum loads of 20 kN, 72.5 kN and 105 kN were recorded for single block, 1*2 blocks and 2*2 blocks samples respectively. The values for first-crack and maximum loads were increased by 3-4 times as the number of blocks increased with better load distribution area. The compressive strengths of single block, 1*2 blocks and 2*2 blocks samples were noted as 1.53 MPa, 0.83 MPa, and 0.92 MPa respectively. Comparing compressive strength of a single block with 1*2 blocks and 2*2 blocks a reduction of 45% and 39% respectively was noted. This decrease could be due to a slenderness effect of the samples with more than one block. This also compared well with the fact that as the sample height increased the compressive strength reduced, which was also reported by other researchers (3). Comparison of average compressive strengths of each sample is shown in Figure 10.

The CE_b of three specimen configurations (single block, 1*2 blocks and 2*2 blocks) was calculated as 0.006 MJ/m³, 0.002 MJ/m³ and 0.006 MJ/m³ respectively, which showed significant improvement of 66% for 1*2 blocks specimen over the single block and no difference was observed in 2*2 blocks specimen. This increment for 1*2 blocks specimen was due to efficiency of interlocking interface between two blocks. The PE_b of three specimen configurations was recorded as 0.0011 MJ/m³, 0.0005 MJ/m³ and 0.0009 MJ/m³, respectively. This had indicated 54% and 18% reduction in post-crack energy absorbed for 1*2 and 2*2 blocks specimens respectively, as compared with single block. This showed that the interlocking mechanism did not perform well after crack propagation. A similar outcome was observed for the values of TE_b and CTI_b, which showed a reduction of 64% for TE_b as compared with single block and a reduction of 13% for CTI_b as compared with single block. This proved that the interlocking mechanism did not perform efficiently after the propagation of cracks and required inclusion of another component like plaster to improve post-crack behaviour.

3.2.3. Modulus of elasticity of ISSBTOP

Modulus of elasticity (E_b) was calculated as the ratio of stress change to strain change. The E_b is detailed for each specimen in Table 8. It was found that the value of E_b for the single-block sample was 0.201 GPa and for 1*2 blocks and 2*2 blocks specimens it was calculated as 0.209 GPa and 0.234 GPa respectively. This showed a 4% and 16% increase from that of the single- block specimen, which was an indication of the interlocking mechanism capacity’s taking more deformation than was the case with the single-block specimen.

3.3. Microscopic analysis TOP

3.3.1. Surface images of fibres TOP

Natural fibres have different chemical composition depending on variation in cultivation techniques, soil and environment. Pretreatment of fibres like washing with tap water and retention in boiling water affects their properties. Morphological changes in the fibres were observed using Bruker 3D optical microscope. Figure 11 (a-c) shows the surface images of sisal, rice straw and treated rice straw. It was observed that treated rice straw had a smoother surface than sisal and untreated rice straw, which resulted in a poor bond between mortar and treated fibre. This was evident from the lowest strength of the treated rice-straw sample as compared with all other samples. It was also noted from Figures 11a and 11b that surface profile for sisal was higher than that of rice straw, giving a high value of surface roughness, which could be an important factor in developing good bond between sisal and cementitious material.

3.3.2. Surface images of fibrous and plain-mortar cube samples TOP

Figure 12 (a-d) shows the microscopic images of plain and fibrous mortar cube samples. These samples were produced from the cube samples after carrying out compression testing. Figure 12a shows micro cracks in the cement paste. In Figure 12b a sample of sisal-fibre mortar cube shows the embedment of sisal fibre in cement paste without any gap or void, indicating a proper bond of fibre in the cement matrix. This resulted in better mechanical properties in sisal fibrous cubes than in other samples. Figures 12c and 12d show the untreated and treated rice-straw samples respectively, which indicated the presence of micro cracks and voids in the cement paste and a loose bond with rice straw. This poor bond and void resulted in reduced post-crack performance and toughness.

4. DISCUSSION TOP

4.1. Relationship between material properties and low-cost house performance TOP

In low-cost masonry housing main-component walling is mostly employed to resist compressive and lateral load (31, 32, 33). Therefore, the structural performance of the wall under these loadings is of prime importance. This depends on the characteristics and mechanical properties of the materials used, which include compressive strength, modulus of elasticity, pre/post/total energy absorption capacity (PE and CE, respectively) and toughness (CTI). Therefore, it is important to explore the best mechanical properties of material to enhance the performance of walling and in essence overall performance of masonry house. In this study mechanical properties of plain and fibrous mortar cube samples and ISSBs were examined. The addition of 2% sisal fibre cube samples showed the most promising results, indicating a 3% increment in compressive strength, 76% increment in PE_p, 2.21 times increase in CE_p, 95% increase in TE_p and 17% increase in CTI_p as compared with the plain mortar cube samples. For ISSB there was a 45% and 39% reduction in compressive strength for 1*2 and 2*2 blocks respectively, as compared with single block. In other mechanical properties like CE_b a 66% improvement was observed. The values of PE_b, TE_b and CTI_b for 1*2 and 2*2 blocks were reduced by 54%,64% and 13% respectively, as compared with those of single block. This shows that the use of mortar with 2% sisal fibre for interlocking structure is likely to give optimised mechanical properties and hence improve the overall performance of masonry structure.

4.2. Empirical equations TOP

4.2.1. Empirical relation between modulus of elasticity and compressive strength of mortar cubes TOP

The following empirical equations [1-2] were developed (Figure 13) with the help of averaged experimental data. These equations are based on limited data (only three values) in order to check the trend of variation due to addition of different content of fibre. These are established by means of best fit curve (R² ranging from 0.90 to 1.0) with the simplification of coefficients and exponents of input variables numerically to predict the modulus of elasticity (E) in GPa:

where C is compressive strength in MPa and K = 1, 1.1 and 0.95 for plain, 2% rice straw and 5% rice straw specimen respectively in equation [1]. For equation [2], K (GPa) = 0.293, 0.294 and 0.291 for plain, 2% sisal and 5% sisal fibre reinforced mortar, respectively, Y (1/MPa²) =1 and Z (GPa/MPa) = 1. It may be noted that for each value an average of three readings is taken. In the case of sisal-reinforced mortar a convex quadratic increase is observed, whereas in rice- reinforced mortar a powered linear increase is found.

Table 9 details the experimental and empirical values of modulus of elasticity (GPa). It can be seen that a correlation holds between compressive strength and modulus of elasticity and these being directly proportional. This direct relationship is attributed to the capacity to take greater load, which is increased due to the presence of fibre of high tensile strength and the bridging phenomenon. It can therefore be stated that an increase in compressive strength results in higher value of modulus of elasticity. Figure 14 compares the experimental and numerical values of modulus of elasticity in plain and fibrous samples. It can be seen that the values of E obtained from equations [1] and [2] are very close to the experimental values. There is good agreement between empirical and experimental values. The percentage errors are 3.7%, 1.2-3.6% and 2.4%-3.5% for plain, rice-straw and sisal mortar samples respectively.

4.2.2. Empirical relation between modulus of elasticity and compressive strength of ISSB TOP

The following empirical equation [3] is developed (Figure 15) with the help of averaged experimental data. This equation is based on nine samples of blocks representing three no. of samples for single block, 1*2 and 2*2 blocks respectively. This is established by means of best fit curve (R²= 1.0) with the simplification of coefficients and exponents of input variables numerically to predict the modulus of elasticity (E) in GPa:

where C is compressive strength in MPa and K (1/MPa) = 0.98, 0.65, 0.60 and Z (GPa/MPa) = 1, 0.95, 0.9 for single block, 1*2 and 2*2 blocks respectively. As far as the behaviour of blocks is concerned, empirical modelling reveals that there is concave variation among single blocks, 1*2 blocks and 2*2 blocks.

Table 10 details the experimental and empirical values of modulus of elasticity (GPa) of ISSB. It can be seen that compressive strength and modulus of elasticity of ISSB are correlative and are inversely proportional. This inverse proportionality is due to the different configuration of blocks. The reduction in compressive strength and increase in modulus of elasticity of other configurations as compared with single block is due to block interface joint which leads to large displacements. Therefore it can be stated that a decrease in compressive strength results in a higher value of modulus of elasticity. Figure 16 compares the experimental and numerical values of modulus of elasticity for single block, 1*2 and 2*2 blocks. It can be seen that the values of E obtained from equation [3] are very close to the experimental values. The percentage error is 0.7%-6.6% for different block configurations.

5. CONCLUSIONS TOP

The mechanical properties (compressive strength, modulus of elasticity, pre and post crack energy absorbed and toughness index) of interlocking soil-stabilised blocks and fibrous/non-fibrous mortar were experimentally investigated to determine how best to enhance the performance of proposed interlocked fibrous plastered (i.e. reinforced coating) low-cost housing. The following conclusions are drawn:

In light of the findings and observed behaviour the addition of 2% sisal fibre in plaster (i.e. reinforced coating) of interlocked masonry walling is likely to be effective in improving the performance including the lateral resistance of low-cost masonry house. Further investigation is required to evaluate the performance of ISSB column or wall with fibre-reinforced plaster when subjected to lateral load like wind and earthquake.

ACKNOWLEDGEMENTSTOP

The authors would like to thank B Chilla & H Hatibu (NHBRA) and organizations who helped them throughout this research work which was funded by Energy and Low Income Tropical Housing Project (ELITH). The careful review and valuable recommendations by the anonymous reviewers are appreciatively acknowledged.

REFERENCESTOP