- Email: [email protected]
Influence of sand grading on the characteristics of mortars and soil–cement block masonry
Available online at www.sciencedirect.com
Construction and Building
MATERIALS
Construction and Building Materials 22 (2008) 1614–1623
www.elsevier.com/locate/conbuildmat
Influence of sand grading on the characteristics of mortars and soil–cement block masonry B.V. Venkatarama Reddy a
a,*
, Ajay Gupta
b
Department of Civil Engineering, Indian Institute of Science, Bangalore 560 012, India b Vintech Consultants, K 1/128, Chitranjan Park, New Delhi 110 019, India
Received 8 August 2006; received in revised form 20 June 2007; accepted 25 June 2007 Available online 13 August 2007
Abstract Sand constitutes bulk of the mortar volume. Sand grading can influence the characteristics of mortar and masonry. Influence of sand grading on the characteristics of two types of mortars and soil–cement block masonry are examined in this paper. Three different sand gradings were used to examine the workability, strength, water retentivity, drying shrinkage and stress–strain characteristics of cement mortar and cement–lime mortar. Bond strength, compressive strength and stress–strain characteristics of soil–cement block masonry were also examined using these mortars. Major findings of the study are: (a) for a given consistency mortar with fine sand requires 25–30% more water, (b) as the sand becomes fine mortar compressive strength and modulus decreases while drying shrinkage increases, (c) fine sand reduces the tensile bond strength of masonry, whereas masonry compressive strength is not sensitive to sand grading variations and (d) masonry modulus reduces as the sand used in the mortar becomes finer. 2007 Elsevier Ltd. All rights reserved. Keywords: Mortar; Sand grading; Soil–cement block; Masonry; Stress–strain relation; Masonry strength
1. Introduction Sand is the common ingredient for masonry mortars even though varieties of cementitious materials are used for mortars. Sand constitutes bulk of the mortar volume. Composition of sand and its grading can influence the characteristics of mortars in fresh as well as in hardened state. Also, it could influence brick–mortar adhesion and other masonry characteristics. There are limited studies on the influence of sand grading on the characteristics of mortars and masonry. Drew and Braj [1] studied the effect of sand characteristics (fineness modulus, void ratio, specific surface, etc.) and water content of the mix on the mortar strength. The tests were performed on Scottish sand samples collected from 30 different places. They observed that water–cement ratio is the largest single factor affecting the compressive strength of mortar irrespec*
Corresponding author. Tel.: +91 80 2293 3126; fax: +91 80 2360 0404. E-mail address: [email protected] (B.V. Venkatarama Reddy).
0950-0618/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.conbuildmat.2007.06.014
tive of different types of sand and sand grading. With the increase in void ratio (the percentage of voids in loose conditions) of sand, the water requirement for mortar increases for standard consistency. As the fineness modulus of sand decreases the requirement of water for a particular mix proportion increases. Specific surface of sand has no relation with the water requirement of mortar mix and it does not influence the mortar strength. Balen and Gemert [2] examined some properties of fresh mortar and observed that for a given consistency, mortars with very fine sand required up to 50% more water than similar mortars having normal sand grading. They also observed that the fine sand mortars have better water retentivity as compared to normal sand grading for a given mortar proportion. I.S. 2116 [3] gives grading limits for sand for use in masonry mortars as shown in Fig. 1. Similar sand grading limits can be found in ASTM C 144 [4], BS: 4551 [5] and many other codes of practice. Anderson and Held [6] investigated the influence of sand grading on the bond strength of cement–lime mortar with three types of bricks. They found that the sand grading
B.V. Venkatarama Reddy, A. Gupta / Construction and Building Materials 22 (2008) 1614–1623 100
% Finer
80 60 40
Natural sand Medium sand Fine sand Upper limit Lower limit
20 0 0.1
1
10
1615
through an experimental programme. Grading curves of natural river sand was varied by removing a portion of the coarse fraction thus obtaining three different gradations for the sand. Characteristics of two types of mortars (cement mortar and cement–lime mortar) and soil–cement block masonry using these mortars were examined. Cement–lime mortar is the most commonly used mortar for masonry throughout the world. In India, cement mortar (1 cement:6 sand) is commonly used for load bearing masonry structures. Hence, in this investigation both cement mortar and cement–lime mortar have been selected. Table 1 gives details of the experimental programme.
Particle Size (mm)
3. Materials used in the investigation
Fig. 1. Particle size distribution of sand.
significantly affects the tensile bond strength of mortar to brick. Mortars using fine sand results in lower bond strength. Groot [7] examined the tensile brick–mortar bond strength by means of cross couplet tests following ASTM C 952 [8] guidelines. Tests were performed on extruded clay bricks (dry), machine moulded bricks (prewetted 15% by mass) and calcium silicate bricks (prewetted 7% by mass). Portland cement mortar, lime–cement mortar and masonry cement mortar with different sand grading ranging from finer to coarser were used to cast the specimens. He observed that the mortars with coarser sand give better bond strength than with finer sand and the type of the masonry unit influences the bond strength more significantly rather than the grading of the sand. There are limited numbers of focused studies pertaining to influence of sand grading on mortar and masonry characteristics. Hardly any literature exists on the effect of sand grading on the characteristics of mortar related to soil– cement block masonry. Hence, the present investigation is focused on the influence of sand grading on the characteristics of mortars in the fresh and hardened state, and its influence on the characteristics of soil–cement block masonry.
3.1. Cement and lime Ordinary Portland cement conforming to I.S. 8112 [9] was used for the manufacture of soil–cement blocks as well as for the mortars. Locally available commercial grade calcium hydroxide (lime) was used for cement–lime mortar. This is a non-hydraulic type lime. 3.2. Sand Natural river sand was used in the experiments. Influence of sand grading and its fineness modulus on various properties was examined by reconstituting the natural sand having particles finer than 4.75, 1.18 and 0.5 mm. The grain size distribution curves of natural sand and reconstituted sands are shown in Fig. 1. Upper and lower bound limiting gradations for sand as specified in I.S. 2116 [3] code are also shown in the figure. The natural sand gradation curve falls outside the limiting gradation curves. Details of fineness modulus (FM) and designations of three types of sands used in the experiments are given in Table 1. FM of natural sand, medium sand and fine sand is 3.21, 2.45 and 1.72, respectively.
2. Scope of the work and the experimental programme 3.3. Soil–cement blocks Influence of sand grading on characteristics of mortars in the fresh as well as in the hardened state and the characteristics of soil–cement block masonry were examined
Soil–cement blocks are manufactured by compacting a mixture of soil, sand and cement at optimum moisture into
Table 1 Test programme for mortar and masonry characteristics Mortar
Sand
Proportion (by volume)
Designation
C
L
S
1 1 1 1 1 1
– – – 1 1 1
6 6 6 6 6 6
CMN CMM CMF CLMN CLMM CLMF
Finer than (mm)
4.75 1.18 0.50 4.75 1.18 0.50
Properties investigated Designation
Natural Medium Fine Natural Medium Fine
FM
3.21 2.45 1.72 3.21 2.45 1.72
Mortar A p p p p p p
B p p p p p p
Soil–cement block masonry CS p p p p p p
D p p p p p p
E p p p p p p
T p p p p p p
CS p p p p p p
E p p p p p p
C: cement; L: lime; S: sand; A: flow characteristics; B: water retentivity; CS: compressive strength; D: drying shrinkage; E: stress–strain characteristics; T: tensile bond strength; FM: fineness modulus.
B.V. Venkatarama Reddy, A. Gupta / Construction and Building Materials 22 (2008) 1614–1623
a dense block using a machine. Such blocks are used for load bearing masonry of 2–3 stories height buildings in India and elsewhere [10–15]. Fig. 2 shows a load bearing soil–cement block masonry residential building in India. More information on the technology of soil–cement blocks can be found in the investigations of Lunt [16], Olivier and Mesbah [17], Heathcote [18], Venkatarama Reddy and Jagadish [19], Walker and Stace [20], Walker [21], Venkatarama Reddy and Walker [22], Venkatarama Reddy and Gupta [23,24] and many other publications. Soil–cement blocks of size: 305 · 143 · 100 mm were used in this investigation. The blocks were prepared using 12% cement by weight. Commonly used range of cement contents for the manufacture of soil–cement blocks lie in between 5% and 12% depending upon the block strength desired, the type of structure, etc. Hence, in the present study 12% cement has been used for the soil–cement block manufacture, which generally yields blocks of strength sufficient for three storey load bearing residential buildings. Locally available red loamy soil, sand and ordinary Portland cement were used for the block manufacture. After 28 days of curing the blocks were dried inside the laboratory for 30 days and then used for the experiments. The blocks were soaked in water for 48 h prior to testing. I.S. 3495 [25] code guidelines were followed for determining the wet compressive strength. Water absorption (saturated water content) for the blocks was determined using 24-h immersion cold water test as per the guidelines of I.S. 3495 [25] code. Wet compressive strength and water absorption values for the block are 7.19 MPa (mean of 20 specimens) and 11.4% (mean of 6 specimens), respectively. The stress–strain relationship for the soil–cement block (after soaking in water for 48 h prior to test) was obtained by measuring the longitudinal strains using a 200 mm demec gauge. Fig. 3 shows the stress–strain relationship for the soil–cement block representing the mean of six specimens. Initial tangent modulus and strain at peak stress for the block are 6000 and 0.0033, respectively.
3.5
Compressive stress (MPa)
1616
3 2.5 2 1.5 1 0.5 0 0
0.001
0.002
0.003
0.004
0.005
Longitudinal strain Fig. 3. Stress–strain curve for the soil–cement block.
4. Testing procedure 4.1. Determination of workability of mortars through flow table tests It is essential to have sufficient workability for the mortar in order to facilitate spreading of the mortar on the horizontal bed joints and filling the vertical joints. Apart from the composition of the mix, generally water–cement ratio controls the workability of the mortar. Workability of the mortar is generally characterized by conducting tests like dropping ball test, cone impression test and slump test. A few trial tests on the consistency of cement mortar using cone impression test revealed that the depth of cone penetration is not sensitive when water–cement ratio of the cement mortar (1:6 proportion) is higher than 1.1. Thus, this test did not provide actual cone penetration depth at higher water–cement ratio due to segregation in fresh mortar samples. Similar problems were anticipated for dropping ball test. Hence in the present study consistency/ workability characteristics of mortar were measured using
Fig. 2. Load bearing soil–cement block residential building.
B.V. Venkatarama Reddy, A. Gupta / Construction and Building Materials 22 (2008) 1614–1623
a flow table test. BS: 4551 [5] code guidelines were followed to carry out experiments to determine the flow of mortars. 4.2. Compressive strength of mortar Compressive strength of the mortar was obtained by testing 70 mm size cube specimens. Thoroughly mixed mortar sample is filled into a metal mould in three layers, each layer is tamped 25 times using a standard tamping rod (specified in I.S. 2250 [26]). After 24 h of casting the cubes were removed from the metal moulds and then soaked in water for curing. After 28 days curing the cubes were tested for compressive strength in saturated condition. The mean of six cubes tested is reported as compressive strength of the mortar.
were cast using soil–cement blocks and the two types of mortars. The initial moisture content of the block during casting of the prism specimens can affect the bond strength. Partially saturated blocks (75% saturation) lead to maximum bond strength [29,24]. Thus, to avoid the interference of the moisture content of the block on bond strength the initial moisture content of the blocks at the time of casting of couplets was kept constant by soaking them in water for a period of 4 min prior to casting (experiments conducted on these blocks showed that the blocks attain about 75% saturation when soaked in water for 4 min). The mortar bed joint thickness of 12 mm was maintained in all the cases. Fig. 4 shows the soil–cement block cross couplet specimen. Mortar flow was kept constant at 100% for casting the specimens. After 28 days of curing under wet burlap the couplets were soaked in water for 48 h prior to test.
4.3. Water retentivity Water retentivity is defined as the ability of the fresh mortar to hold/retain water when placed in contact with absorbent masonry units. Water retentivity of the mortar depends on various factors like mix proportion, water– cement ratio, type of cemenititous binders, etc. Standard codes of practice like I.S. 2250 [26], ASTM C 91 [27] and BS: 4551 [5] give procedures to determine water retentivity of the mortar. In the present investigation, water retentivity was examined for various mortar mixes by adopting the BS: 4551 [5] code guidelines. 4.4. Drying shrinkage value of mortar ASTM C 1148 [28] code procedure was followed to determine drying shrinkage of the mortar. The drying shrinkage of the mortar, as determined by this method, is the measure of decrease in length of test specimen in unrestrained condition, under drying condition, after an initial period of curing. The average of five mortar specimens is reported as drying shrinkage value of mortar as specified by the code.
Fig. 4. Soil–cement block couplet specimen.
4.5. Stress–strain characteristics of mortars Stress–strain relationships for the mortars were obtained by testing mortar prisms of size 150 · 150 · 300 mm. After 28 days of moist burlap curing the prisms were soaked in water for a period of 48 h prior to testing. Prisms were tested in a compression testing machine having a constant piston displacement rate of 1.25 mm/min. The longitudinal strains were measured using 200 mm demec gauge. Three specimens were tested for each mortar proportion and the mean values are reported. 4.6. Tensile bond strength The tensile bond strength of soil–cement block and mortar interface was determined by adopting the guidelines outlined in ASTM C 952 [8] code. Cross couplet specimen
1617
Fig. 5. Soil–cement block masonry prism.
1618
B.V. Venkatarama Reddy, A. Gupta / Construction and Building Materials 22 (2008) 1614–1623
4.7. Compressive strength and stress–strain relationships for soil–cement block masonry prisms Compressive strength of soil–cement block masonry was determined by testing the masonry prisms. Four blocks high stack bonded masonry prisms (size: 305 · 143 · 436 mm) were used. A mortar joint thickness of 12 mm was maintained for all the prisms. The blocks were partially saturated by soaking them in water for a period of 4 min prior to casting of the masonry prisms. The prisms were capped with 12 mm thick 1:3 cement mortar and then cured for 28 days under a wet burlap. The prisms were tested (after soaking them in water for 48 h prior to testing) in a universal testing machine and the longitudinal compressive strains were measured by using a 200 mm demec gauge. Fig. 5 shows the details of soil–cement block masonry prism. 5. Results and discussion 5.1. Flow characteristics of mortars The flow value versus water–cement ratio for cement mortar and cement–lime mortar are shown in Figs. 6 and 7, respectively. The following observations can be made from the results shown in these figures:
140 120
Flow (%)
100
Natural sand (CMN) Medium sand (CMM) Fine sand (CMF)
80 60 40 20 0 1.1
1.3
1.5
1.7
1.9
2.1
2.3
Water-cement ratio Fig. 6. Flow versus water–cement ratio for cement mortar.
160 140
Flow (%)
(1) The flow value increases with the increase in water– cement ratio of the mortar for both cement mortar and cement–lime mortar irrespective of sand grading. The relationship between flow value and water– cement ratio is linear for both the mortar types and for all three grades of sand, except in case of cement mortar with fine sand, which has bilinear variation. (2) The mortar flow is very sensitive to water–cement ratio of the mix. For example in case of cement mortar, there is a three-fold increase in flow for about 25– 30% increase in water–cement ratio of the mix for all the three grades of sand. (3) Mortar with fine sand requires more water to attain similar flow values when compared to mortars with natural and medium sand for both cement mortar and cement–lime mortar. For example in case of cement mortar to achieve a flow of 90%, mortar with natural sand and medium sand requires water– cement ratio of 1.60, whereas mortar with fine sand needs water–cement ratio of 2.00. Similarly for cement–lime mortar to attain a flow of 90%, a water–cement ratio of 1.75 is required for mortars with natural and medium sand, whereas mortar with fine sand needs water–cement ratio of 2.35. This may be attributed to the fact that mortar having fine sand has large surface area. The results are comparable with the study conducted by Balen and Gemert [2]. (4) Cement–lime mortars require higher water–cement ratio when compared to cement mortars, to achieve similar flow values irrespective of type of sand used. The cement–lime mortars with natural and medium sand require 8% more water when compared to cement mortar to achieve flow values beyond 80%, whereas in case of fine sand it is about 17% more water.
120
Natural sand (CLMN) Medium sand (CLMM) Fine sand (CLMF)
Gupta [30] conducted a field study to asses the consistency of mortars by conducting flow table tests on fresh mortars used in the filed. He concludes that a flow value of 100% is commonly used in the field for the construction of load bearing masonry in India. The water–cement ratio required for both these mortars using different grades of sand, to achieve 100% flow can be obtained from the relationships shown in Figs. 6 and 7 and the values are given in Table 2 for both the mortars. These results clearly show that to achieve 100% flow water–cement ratio demand increases with increase in fineness of sand for both the mortars.
100
5.2. Compressive strength and water retentivity of mortars
80 60 40 20 0 1.3
1.5
1.7
1.9
2.1
2.3
2.5
2.7
Water-cement ratio Fig. 7. Flow versus water–cement ratio for cement–lime mortar.
Compressive strength of the mortar was determined by testing 70 mm cubes, whereas water retentivity of the mortar was examined by following the guidelines of BS: 4551 [5] code. The compressive strength and water retentivity values for both the mortars are given in Table 2. Three types of sand grading have been attempted in each category of mor-
B.V. Venkatarama Reddy, A. Gupta / Construction and Building Materials 22 (2008) 1614–1623
1619
Table 2 Compressive strength, water retentivity and elastic properties of mortars (mortar flow = 100%) Mortar type
Water–cement ratio
Dry density (g/cm3)
Compressive strength (MPa)
Water retentivity (%)
Initial tangent modulus (MPa)
Strain at peak stress
CMN CMM CMF CLMN CLMM CLMF
1.65 1.69 2.04 1.79 1.83 2.40
1.90 1.88 1.79 1.90 1.87 1.63
5.40 5.22 3.30 5.94 4.45 2.04
74 82 85 82 85 83
4534 7764 4245 4820 7737 2704
0.0040 0.0012 0.0017 0.0026 0.0016 0.0017
Compressive stress (MPa)
1. Water–cement ratio increases as the fineness of sand increases in order to maintain 100% flow, for both the mortars. Increase in water–cement ratio leads to decrease in dry density of the mortar and hence reduction in mortar strength with the increase in fineness of sand. The compressive strength of cement mortar and cement–lime mortar using natural sand at 100% flow is in the range of 5–6 MPa. For a given flow value of 100% the compressive strength of cement–lime mortar is more sensitive to fineness of sand as compared to cement mortar. The study conducted by Drew and Braj [1] on cement mortars also shows that as the water content of the mix is increased the dry density reduces and results in lower compressive strength of mortar. 2. Water retentivity increases with increase in fineness of sand for cement mortars. There is a 15% increase in water retentivity when the natural sand is replaced by fine sand. Whereas in case of cement–lime mortar, there is only a marginal variation in water retentivity values as the sand grading changes. Generally, cement–lime mortars possess better water retentivity than cement mortars due to the presence of fine lime particles in cement–lime mortar.
4 3.5 3 2.5 2 1.5 Natural sand (CMN) Medium sand (CMM) Fine sand (CMF)
1 0.5
0 0
0.001
0.002
0.003
0.004
0.005
Longitudinal strain Fig. 8. Stress–strain characteristics for cement mortar at a flow value of 100%. 3.5
Compressive stress (MPa)
tar to understand the influence of sand fineness on compressive strength and water retentivity of mortars. The following points emerge from the results given in Table 2:
3 2.5 2 1.5 1 Natural sand (CLMN) Medium sand (CLMM) Fine sand (CLMF)
0.5 0 0
0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035 0.004
Longitudinal strain
5.3. Stress–strain characteristics of mortars The stress–strain characteristics like initial tangent modulus and the stain at peak stress values for both the mortars are given in Table 2. Figs. 8 and 9 show the stress–strain relationships for cement mortar and cement–lime mortar using three different grades of sand. The stress–strain curves have been generated by keeping the mortar flow at 100% for all the cases. The following observations can be made from these results: (1) Mortars using medium sand (fineness modulus of 2.45) give highest modulus for both the mortars. In case of cement mortar with medium sand the modulus is 50% more as compared to other two types of sands. Cement–lime mortar with fine sand and natural sand have modulus values of 60% and 35% of that for the medium sand, respectively. Mortars with fine sand give lowest modules.
Fig. 9. Stress–strain characteristics for cement–lime mortar at a flow value of 100%.
(2) The strain at peak stress for cement mortar with natural sand is significantly high of the order of 0.004, when compared to mortar with medium and fine sand. The strain value decreases as the sand becomes finer. In case of cement–lime mortar with natural sand strain at peak stress is 0.0026. This strain reduces to about 0.0016 for mortar with medium and fine sand. 5.4. Drying shrinkage Shrinkage that takes place during hardening of the mortar can be called as drying shrinkage and most of it takes place in the first few months. A part of drying shrinkage is recovered on immersion of mortar in water. The drying
B.V. Venkatarama Reddy, A. Gupta / Construction and Building Materials 22 (2008) 1614–1623
shrinkage of mortar could depend on various factors like water–cement ratio, cement content, sand grading, curing period, etc. In masonry construction, drying shrinkage of mortar causes shrinkage cracks observed at the masonry unit–mortar interface and it can also result in impaired bond [31,32]. The drying shrinkage of mortar was tested in the laboratory through a mortar prism of size 250 · 25 · 25 mm in unrestrained condition. It differs from that experienced in a masonry wall where the drying shrinkage of the mortar is influenced by restraint offered by masonry units, masonry unit absorption characteristics, thickness of mortar bed joint, etc. Thus drying shrinkage value of the mortar examined in the laboratory is more useful for comparative purposes. The drying shrinkage of cement mortar and cement– lime mortar was examined. All the tests were performed at a mortar flow of 100%. The drying shrinkage of mortars at 25 days of drying duration is taken as the ultimate drying shrinkage. The results given in the following sections represent the mean of five specimens. The variation in drying shrinkage of the cement mortar and cement–lime mortar specimens with the duration of drying is shown in Figs. 10 and 11, respectively. A comparison in the variation of
Drying shrinkage (%)
0.12 0.1 0.08 0.06 Natural sand (CMN)
0.04
Medium sand
0.02
0 0
5
10
15
20
25
30
Drying duration (days) Fig. 10. Variation in drying shrinkage with drying duration for cement mortar.
0.16
Drying shrinakge (%)
0.14
0.16 0.14
Drying shrinkage (%)
1620
Cement mortar Cement lime mortar
0.12 0.1 0.08 0.06 0.04 0.02 0 Natural
Medium
Fine
Sandtype Fig. 12. Variation in ultimate drying shrinkage with sand grading for cement mortar and cement–lime mortar.
ultimate drying shrinkage of cement mortars and cement– lime mortars with the type of sand used is shown in Fig. 12. The following points emerge from the results on drying shrinkage: (a) The ultimate drying shrinkage values lie in the range of 0.057–0.151% for cement mortar and cement–lime mortars for various grades of sand. The drying shrinkage of these mortars reaches the ultimate drying shrinkage in initial 3–4 days of drying as illustrated in Figs. 10 and 11. The ultimate drying shrinkage is generally more in case of cement mortar as compared to cement–lime mortar, except for mortars using fine sand. (b) The fineness of sand in mortars greatly affects the ultimate drying shrinkage values as shown in Fig. 12. There is an increase in drying shrinkage value when the sand becomes finer. For example, in case of cement mortar, there is more than one third increase in the ultimate drying shrinkage when the sand grading is changed from coarser to finer (fineness modulus changing from 3.21 to 1.72). On the other hand, 2.7 times increase in the ultimate drying shrinkage is observed in the case of cement–lime mortar for similar change in sand grading.
012. 0.1
5.5. Influence of sand grading on tensile bond strength
0.08 0.06 0.04
Natural sand (CLMN) Medium sand (CLMM) Fine sand (CLMF)
0.02 0 0
5
10
15
20
25
30
Drying duration (days) Fig. 11. Variation in drying shrinkage with drying duration for cement– lime mortar.
The influence of sand grading on tensile bond strength of soil–cement block couplets was examined using both the mortars. These results are tabulated in Table 3. The variation in tensile bond strength with the grading of sand is shown in Fig. 13. The tensile bond strength values lie in the range of 0.08–0.252 MPa for the couplets using both the types of mortars with three different grades of sand. In case of cement mortar, the tensile bond strength steadily decreases as the sand grading changes from coarser to fine
B.V. Venkatarama Reddy, A. Gupta / Construction and Building Materials 22 (2008) 1614–1623
Mortar type
CMN CMM CMF CLMN CLMM CLMF
Tensile bond strength (MPa)
0.181 0.100 0.080 0.233 0.252 0.093
(0.15–0.21) (0.041–0.213) (0.05–0.111) (0.198–0.259) (0.201–0.297) (0.076–0.119)
Masonry prism strength (MPa)
Masonry Initial tangent modulus (MPa)
Strain at peak stress
4.55 4.84 4.55 5.27 4.67 4.57
4163 4087 3084 4500 4770 2864
0.0044 0.0046 0.0060 0.0052 0.0040 0.0073
4 3 2
Natural sand(CMN) Medium snad(CMM)
1
Fine sand (CMF)
0 0
0.001
0.002
0.003
0.004
0.005
0.006
0.007
Longitudinal strain
6
Compressive stress (MPa)
Tensile bond strength (MPa)
5
Fig. 14. Influence of sand grading on stress–strain relationships for soil– cement block masonry with cement mortar.
Values in parenthesis indicate the range of values.
0.3
6
Compressive stress (MPa)
Table 3 Bond strength, compressive strength and elastic properties of soil–cement block masonry
1621
Cement mortar Cement-lime mortar
0.25 0.2 0.15 0.1 0.05
5 4 3 Natural sand (CLMN)
2
Medium sand (CLMM)
1
Fine sand (CLMF)
0 0
0 Natural
Medium
Fine
0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 0.009
Longitudinal strain
Type of sand Fig. 13. Influence of sand grading on tensile bond strength of soil–cement block couplets.
(fineness modulus changing from 3.21 to 1.72). There is 56% reduction in the tensile bond strength as the sand grading is changed from coarser to fine. On the other hand, in case of cement–lime mortar there is hardly any variation in bond strength between coarse and medium sand. But tensile bond strength decreases by 60% when sand grading is changed from medium to fine for cement–lime mortar. Thus, for both the mortars use of very fine sand reduces the tensile bond strength significantly. 5.6. Compressive strength and stress–strain characteristics of masonry Compressive strength and stress–strain relationships for the soil–cement block masonry were obtained by using both the types of mortars with three types of sand gradings. Stress–strain relationships for the masonry using cement mortar and cement–lime mortar are shown in Figs. 14 and 15, respectively. Details of compressive strength, initial tangent modulus and strain at peak stress of soil–cement block masonry using both the mortars are given in Table 3. The following observations can be made from the results given in the table and figures:
Fig. 15. Influence of sand grading on stress–strain relationship for soil– cement block masonry using cement–lime mortar.
(1) In case of cement mortar the masonry compressive strength shows a minor variation, when the sand grading is changed from coarser to fine (fineness modulus changing from 3.21 to 1.72). The masonry strength lies in the range of 4.55–4.84 MPa for the three grades of sand. The masonry strength using cement–lime mortar is in the range of 4.57– 5.27 MPa. The masonry prisms made of cement–lime mortar show a decrease in masonry compressive strength, as the sand used in the mortar becomes fine. There is a 13% decrease in compressive strength when the sand grading is changed from coarser to fine. (2) Initial tangent modulus for masonry with cement mortar lies in the range of 3000–4100 MPa, whereas for cement–lime mortar it is in between 2800 and 4800 MPa. There is a significant reduction in masonry modulus when the grading of sand is changed from coarse to fine. For example in case of masonry using cement mortar there is about 25% reduction in initial tangent modulus when the sand grading is changed from coarse to fine, whereas for cement–lime mortar a decrease of about 40% is observed. The masonry prisms with fine sand for both cement mortar and cement–lime mortar have
1622
B.V. Venkatarama Reddy, A. Gupta / Construction and Building Materials 22 (2008) 1614–1623
higher values of strain at peak stress, of the order of 0.006 and 0.0073, respectively; as compared to masonry prisms with other two grades of sand (strain at peak stress less than 0.0052). Thus, mortar containing fine sand (low fineness modulus) leads to more ductile masonry as indicated by modulus and strain at peak stress values. 6. Concluding remarks Influence of sand grading on certain properties of cement mortar, cement–lime mortar and the soil–cement block masonry were examined. The following conclusions can be drawn from this investigation: (a) To achieve any given consistency, the mortars using fine sand require 25–30% more water. (b) Compressive strength of cement–lime mortar is more sensitive to the fineness of sand as compared to cement mortar. For a given consistency the compressive strength of mortar decreases with increase in fineness of sand. (c) Water retentivity of cement mortar increases with increase in fineness of sand. Water retentivity of cement–lime mortar is not sensitive to the variation in sand gradation. (d) Mortar modulus and strain at peak stress decrease as the sand fineness increases. There is 40% reduction in modulus and strain values as the sand grading is changed from coarse to fine. (e) Sand grading greatly affects the drying shrinkage values of both the mortars. Drying shrinkage increased by 1/3rd and 2.7 times for cement mortar and cement–lime mortar, respectively, as the sand grading changed from coarse to fine. (f) Tensile bond strength of soil–cement block couplets is sensitive to the fineness of sand used for the mortars. There is 55–60% reduction in tensile bond strength of soil–cement block couplets as fineness modulus of sand changes from 3.21 to 1.72. Thus, for both the mortars use of fine sand reduces the tensile bond strength of soil–cement block couplets significantly. (g) Soil–cement block masonry prisms show marginal variation (10%) in compressive strength as the sand grading changes from coarse to fine. But there is 25– 40% reduction in modulus of soil–cement block masonry as the sand grading is changed from coarse to fine. Reduction in masonry modulus is accompanied by increase in values of strains at peak stress for both the mortars. It is interesting to note here, that even though the grading curves for medium and fine sand used in the investigations fall within the grading limits specified by I.S. 8112 code, there is considerable variation in the properties of mortar as well as soil–cement block masonry using these two grades of sand.
References [1] Drew Currie, Braj Sinha. Survey of Scottish sands and their characteristics which affect mortar strength. Chem Ind 1981;19:639–46. [2] Balen Van K, Gemert Van D. Lime–cement mortars compared to the commonly used cement mortars in Belgium. Belgium: K.U. Leuven (University); 1990. [3] I.S. 2116-80 (reaffirmed 1998). Indian standard specification for sand for masonry mortars. New Delhi, India: Bureau of Indian Standards; 1998. [4] ASTM C 144-04. Standard specification for aggregate for masonry mortar. USA: American Society for Testing and Materials; 2004. [5] BS: 4551. British standard methods of testing mortars, screeds and plasters. UK: British Standards Institution; 1980. [6] Anderson C, Held LC. The effect of sand grading on mortar properties and tensile bond strength of brickwork specimens. Proc Br Mason Soc 1986;1:1–6. [7] Groot Caspar. Effects of water on mortar–brick bond. PhD thesis, Faculty of Civil Engineering, Delft University of Technology, The Netherlands; 1993. [8] ASTM C 952-91. Standard test method for bond strength of mortar to masonry units. American Society for Testing and Materials; 1991. [9] I.S. 8112. Indian standard specification for 43 grade ordinary Portland cement. New Delhi, India: Bureau of Indian Standards; 1989. [10] Fitzmaurice R. Manual on stabilized soil construction for housing. New York: United Nations; 1958. [11] U.N. Report. Soil–cement – its use in building. New York, USA: Department of Economic and Social Affairs, United Nations; 1964. [12] Theunissen Ph. Building with earth Dimension 3. Bimonthly Rev Inform Serv Belgian Admin Dev Cooper 1985;4:10–2. [13] Jagdish KS. The progress of stabilised soil construction in India. In: Proceedings of the national seminar on stabilised mud blocks for housing and building, vol. 1; 1988. p. 17–43. [14] Hugo Houben, Hubert Guillaud. Earth construction – a comprehensive guide. UK/France: ITDG Publishing/CRATerre-EAG; 2003. [15] Walker P, Venkatarama Reddy BV, Mesbah Ali, Morel JeanCluaude. The case for compressed earth block construction. In: Proceedings of the sixth international seminar on structural masonry for developing countries; 2000. p. 27–35. [16] Lunt MG. Stabilized soil blocks for buildings. Overseas building notes, vol. 184; 1980. [17] Olivier M, Mesbah Ali. Influence of different parameters on the resistance of earth, used as a building materials. In: Proceedings of the international conference on mud architecture; 1987. [18] Heathcote K. Compressive strength of cement stabilized pressed earth blocks. Build Res Inform 1991;19(2):101–5. [19] Venkatarama Reddy BV, Jagdish KS. Influence of soil composition on the strength and durability of soil–cement blocks. Indian Concr J 1995:517–24. [20] Walker P, Stace T. Properties of some cement stabilized compressed earth blocks and mortars. Mater Struct 1997;30:545–51. [21] Walker P. Strength and erosion characteristics of earth blocks and earth block masonry. J Mater Civil Eng (ASCE) 2004;16(5):497–506. [22] Venkatarama Reddy BV, Walker Peter. Stabilised mud blocks: problems, prospects. In: Proceedings of the international earth building conference EarthBuild2005; 2005. p. 63–5. [23] Venkatarama Reddy BV, Gupta Ajay. Characteristics of soil–cement blocks using highly sandy soils. Mater Struct (RILEM) 2005;38(280):651–8. [24] Venkatarama Reddy BV, Gupta Ajay. Strength and elastic properties of stabilised mud block masonry using cement–soil mortars. J Mater Civil Eng (ASCE) 2006;18(3):472–6. [25] I.S. 3495. Methods of tests for burnt building bricks – part 2. Determination of water absorption. New Delhi, India: Bureau of Indian Standards; 1992.
B.V. Venkatarama Reddy, A. Gupta / Construction and Building Materials 22 (2008) 1614–1623 [26] I.S. 2250. Indian standard code of practice for preparation and use of masonry mortars. New Delhi, India: Bureau of Indian Standards; 1981. [27] ASTM C 91-95. Standard specification for masonry cement. USA: American Society for Testing and Materials; 1995. [28] ASTM C 1148-92a. Standard test methods for measuring the drying shrinkage of masonry mortar. USA: American Society for Testing and Materials; 1992. [29] Venu Madhava Rao K, Venkatarama Reddy BV, Jagadish KS. Flexural bond strength of masonry using various blocks and mortars. Mater Struct (RILEM) 1996;29:119–24.
1623
[30] Gupta Ajay. Studies on characteristics of cement–soil mortars and soil–cement block masonry. MSc(Eng) thesis, Department of Civil Engineering, Indian Institute of Science, Bangalore, India; 2003. [31] Baker LR. Some factors affecting the bond strength of brickwork. In: Proceedings of the fifth international brick masonry conference, vol. II-9; 1979. p. 84–9. [32] Lawrence SJ, Cao HT. Microstructure of the interface between brick and mortar. In: Proceedings of the eighth international brick/block masonry conference; 1988. p. 194–204.
Construction and Building
MATERIALS
Construction and Building Materials 22 (2008) 1614–1623
www.elsevier.com/locate/conbuildmat
Influence of sand grading on the characteristics of mortars and soil–cement block masonry B.V. Venkatarama Reddy a
a,*
, Ajay Gupta
b
Department of Civil Engineering, Indian Institute of Science, Bangalore 560 012, India b Vintech Consultants, K 1/128, Chitranjan Park, New Delhi 110 019, India
Received 8 August 2006; received in revised form 20 June 2007; accepted 25 June 2007 Available online 13 August 2007
Abstract Sand constitutes bulk of the mortar volume. Sand grading can influence the characteristics of mortar and masonry. Influence of sand grading on the characteristics of two types of mortars and soil–cement block masonry are examined in this paper. Three different sand gradings were used to examine the workability, strength, water retentivity, drying shrinkage and stress–strain characteristics of cement mortar and cement–lime mortar. Bond strength, compressive strength and stress–strain characteristics of soil–cement block masonry were also examined using these mortars. Major findings of the study are: (a) for a given consistency mortar with fine sand requires 25–30% more water, (b) as the sand becomes fine mortar compressive strength and modulus decreases while drying shrinkage increases, (c) fine sand reduces the tensile bond strength of masonry, whereas masonry compressive strength is not sensitive to sand grading variations and (d) masonry modulus reduces as the sand used in the mortar becomes finer. 2007 Elsevier Ltd. All rights reserved. Keywords: Mortar; Sand grading; Soil–cement block; Masonry; Stress–strain relation; Masonry strength
1. Introduction Sand is the common ingredient for masonry mortars even though varieties of cementitious materials are used for mortars. Sand constitutes bulk of the mortar volume. Composition of sand and its grading can influence the characteristics of mortars in fresh as well as in hardened state. Also, it could influence brick–mortar adhesion and other masonry characteristics. There are limited studies on the influence of sand grading on the characteristics of mortars and masonry. Drew and Braj [1] studied the effect of sand characteristics (fineness modulus, void ratio, specific surface, etc.) and water content of the mix on the mortar strength. The tests were performed on Scottish sand samples collected from 30 different places. They observed that water–cement ratio is the largest single factor affecting the compressive strength of mortar irrespec*
Corresponding author. Tel.: +91 80 2293 3126; fax: +91 80 2360 0404. E-mail address: [email protected] (B.V. Venkatarama Reddy).
0950-0618/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.conbuildmat.2007.06.014
tive of different types of sand and sand grading. With the increase in void ratio (the percentage of voids in loose conditions) of sand, the water requirement for mortar increases for standard consistency. As the fineness modulus of sand decreases the requirement of water for a particular mix proportion increases. Specific surface of sand has no relation with the water requirement of mortar mix and it does not influence the mortar strength. Balen and Gemert [2] examined some properties of fresh mortar and observed that for a given consistency, mortars with very fine sand required up to 50% more water than similar mortars having normal sand grading. They also observed that the fine sand mortars have better water retentivity as compared to normal sand grading for a given mortar proportion. I.S. 2116 [3] gives grading limits for sand for use in masonry mortars as shown in Fig. 1. Similar sand grading limits can be found in ASTM C 144 [4], BS: 4551 [5] and many other codes of practice. Anderson and Held [6] investigated the influence of sand grading on the bond strength of cement–lime mortar with three types of bricks. They found that the sand grading
B.V. Venkatarama Reddy, A. Gupta / Construction and Building Materials 22 (2008) 1614–1623 100
% Finer
80 60 40
Natural sand Medium sand Fine sand Upper limit Lower limit
20 0 0.1
1
10
1615
through an experimental programme. Grading curves of natural river sand was varied by removing a portion of the coarse fraction thus obtaining three different gradations for the sand. Characteristics of two types of mortars (cement mortar and cement–lime mortar) and soil–cement block masonry using these mortars were examined. Cement–lime mortar is the most commonly used mortar for masonry throughout the world. In India, cement mortar (1 cement:6 sand) is commonly used for load bearing masonry structures. Hence, in this investigation both cement mortar and cement–lime mortar have been selected. Table 1 gives details of the experimental programme.
Particle Size (mm)
3. Materials used in the investigation
Fig. 1. Particle size distribution of sand.
significantly affects the tensile bond strength of mortar to brick. Mortars using fine sand results in lower bond strength. Groot [7] examined the tensile brick–mortar bond strength by means of cross couplet tests following ASTM C 952 [8] guidelines. Tests were performed on extruded clay bricks (dry), machine moulded bricks (prewetted 15% by mass) and calcium silicate bricks (prewetted 7% by mass). Portland cement mortar, lime–cement mortar and masonry cement mortar with different sand grading ranging from finer to coarser were used to cast the specimens. He observed that the mortars with coarser sand give better bond strength than with finer sand and the type of the masonry unit influences the bond strength more significantly rather than the grading of the sand. There are limited numbers of focused studies pertaining to influence of sand grading on mortar and masonry characteristics. Hardly any literature exists on the effect of sand grading on the characteristics of mortar related to soil– cement block masonry. Hence, the present investigation is focused on the influence of sand grading on the characteristics of mortars in the fresh and hardened state, and its influence on the characteristics of soil–cement block masonry.
3.1. Cement and lime Ordinary Portland cement conforming to I.S. 8112 [9] was used for the manufacture of soil–cement blocks as well as for the mortars. Locally available commercial grade calcium hydroxide (lime) was used for cement–lime mortar. This is a non-hydraulic type lime. 3.2. Sand Natural river sand was used in the experiments. Influence of sand grading and its fineness modulus on various properties was examined by reconstituting the natural sand having particles finer than 4.75, 1.18 and 0.5 mm. The grain size distribution curves of natural sand and reconstituted sands are shown in Fig. 1. Upper and lower bound limiting gradations for sand as specified in I.S. 2116 [3] code are also shown in the figure. The natural sand gradation curve falls outside the limiting gradation curves. Details of fineness modulus (FM) and designations of three types of sands used in the experiments are given in Table 1. FM of natural sand, medium sand and fine sand is 3.21, 2.45 and 1.72, respectively.
2. Scope of the work and the experimental programme 3.3. Soil–cement blocks Influence of sand grading on characteristics of mortars in the fresh as well as in the hardened state and the characteristics of soil–cement block masonry were examined
Soil–cement blocks are manufactured by compacting a mixture of soil, sand and cement at optimum moisture into
Table 1 Test programme for mortar and masonry characteristics Mortar
Sand
Proportion (by volume)
Designation
C
L
S
1 1 1 1 1 1
– – – 1 1 1
6 6 6 6 6 6
CMN CMM CMF CLMN CLMM CLMF
Finer than (mm)
4.75 1.18 0.50 4.75 1.18 0.50
Properties investigated Designation
Natural Medium Fine Natural Medium Fine
FM
3.21 2.45 1.72 3.21 2.45 1.72
Mortar A p p p p p p
B p p p p p p
Soil–cement block masonry CS p p p p p p
D p p p p p p
E p p p p p p
T p p p p p p
CS p p p p p p
E p p p p p p
C: cement; L: lime; S: sand; A: flow characteristics; B: water retentivity; CS: compressive strength; D: drying shrinkage; E: stress–strain characteristics; T: tensile bond strength; FM: fineness modulus.
B.V. Venkatarama Reddy, A. Gupta / Construction and Building Materials 22 (2008) 1614–1623
a dense block using a machine. Such blocks are used for load bearing masonry of 2–3 stories height buildings in India and elsewhere [10–15]. Fig. 2 shows a load bearing soil–cement block masonry residential building in India. More information on the technology of soil–cement blocks can be found in the investigations of Lunt [16], Olivier and Mesbah [17], Heathcote [18], Venkatarama Reddy and Jagadish [19], Walker and Stace [20], Walker [21], Venkatarama Reddy and Walker [22], Venkatarama Reddy and Gupta [23,24] and many other publications. Soil–cement blocks of size: 305 · 143 · 100 mm were used in this investigation. The blocks were prepared using 12% cement by weight. Commonly used range of cement contents for the manufacture of soil–cement blocks lie in between 5% and 12% depending upon the block strength desired, the type of structure, etc. Hence, in the present study 12% cement has been used for the soil–cement block manufacture, which generally yields blocks of strength sufficient for three storey load bearing residential buildings. Locally available red loamy soil, sand and ordinary Portland cement were used for the block manufacture. After 28 days of curing the blocks were dried inside the laboratory for 30 days and then used for the experiments. The blocks were soaked in water for 48 h prior to testing. I.S. 3495 [25] code guidelines were followed for determining the wet compressive strength. Water absorption (saturated water content) for the blocks was determined using 24-h immersion cold water test as per the guidelines of I.S. 3495 [25] code. Wet compressive strength and water absorption values for the block are 7.19 MPa (mean of 20 specimens) and 11.4% (mean of 6 specimens), respectively. The stress–strain relationship for the soil–cement block (after soaking in water for 48 h prior to test) was obtained by measuring the longitudinal strains using a 200 mm demec gauge. Fig. 3 shows the stress–strain relationship for the soil–cement block representing the mean of six specimens. Initial tangent modulus and strain at peak stress for the block are 6000 and 0.0033, respectively.
3.5
Compressive stress (MPa)
1616
3 2.5 2 1.5 1 0.5 0 0
0.001
0.002
0.003
0.004
0.005
Longitudinal strain Fig. 3. Stress–strain curve for the soil–cement block.
4. Testing procedure 4.1. Determination of workability of mortars through flow table tests It is essential to have sufficient workability for the mortar in order to facilitate spreading of the mortar on the horizontal bed joints and filling the vertical joints. Apart from the composition of the mix, generally water–cement ratio controls the workability of the mortar. Workability of the mortar is generally characterized by conducting tests like dropping ball test, cone impression test and slump test. A few trial tests on the consistency of cement mortar using cone impression test revealed that the depth of cone penetration is not sensitive when water–cement ratio of the cement mortar (1:6 proportion) is higher than 1.1. Thus, this test did not provide actual cone penetration depth at higher water–cement ratio due to segregation in fresh mortar samples. Similar problems were anticipated for dropping ball test. Hence in the present study consistency/ workability characteristics of mortar were measured using
Fig. 2. Load bearing soil–cement block residential building.
B.V. Venkatarama Reddy, A. Gupta / Construction and Building Materials 22 (2008) 1614–1623
a flow table test. BS: 4551 [5] code guidelines were followed to carry out experiments to determine the flow of mortars. 4.2. Compressive strength of mortar Compressive strength of the mortar was obtained by testing 70 mm size cube specimens. Thoroughly mixed mortar sample is filled into a metal mould in three layers, each layer is tamped 25 times using a standard tamping rod (specified in I.S. 2250 [26]). After 24 h of casting the cubes were removed from the metal moulds and then soaked in water for curing. After 28 days curing the cubes were tested for compressive strength in saturated condition. The mean of six cubes tested is reported as compressive strength of the mortar.
were cast using soil–cement blocks and the two types of mortars. The initial moisture content of the block during casting of the prism specimens can affect the bond strength. Partially saturated blocks (75% saturation) lead to maximum bond strength [29,24]. Thus, to avoid the interference of the moisture content of the block on bond strength the initial moisture content of the blocks at the time of casting of couplets was kept constant by soaking them in water for a period of 4 min prior to casting (experiments conducted on these blocks showed that the blocks attain about 75% saturation when soaked in water for 4 min). The mortar bed joint thickness of 12 mm was maintained in all the cases. Fig. 4 shows the soil–cement block cross couplet specimen. Mortar flow was kept constant at 100% for casting the specimens. After 28 days of curing under wet burlap the couplets were soaked in water for 48 h prior to test.
4.3. Water retentivity Water retentivity is defined as the ability of the fresh mortar to hold/retain water when placed in contact with absorbent masonry units. Water retentivity of the mortar depends on various factors like mix proportion, water– cement ratio, type of cemenititous binders, etc. Standard codes of practice like I.S. 2250 [26], ASTM C 91 [27] and BS: 4551 [5] give procedures to determine water retentivity of the mortar. In the present investigation, water retentivity was examined for various mortar mixes by adopting the BS: 4551 [5] code guidelines. 4.4. Drying shrinkage value of mortar ASTM C 1148 [28] code procedure was followed to determine drying shrinkage of the mortar. The drying shrinkage of the mortar, as determined by this method, is the measure of decrease in length of test specimen in unrestrained condition, under drying condition, after an initial period of curing. The average of five mortar specimens is reported as drying shrinkage value of mortar as specified by the code.
Fig. 4. Soil–cement block couplet specimen.
4.5. Stress–strain characteristics of mortars Stress–strain relationships for the mortars were obtained by testing mortar prisms of size 150 · 150 · 300 mm. After 28 days of moist burlap curing the prisms were soaked in water for a period of 48 h prior to testing. Prisms were tested in a compression testing machine having a constant piston displacement rate of 1.25 mm/min. The longitudinal strains were measured using 200 mm demec gauge. Three specimens were tested for each mortar proportion and the mean values are reported. 4.6. Tensile bond strength The tensile bond strength of soil–cement block and mortar interface was determined by adopting the guidelines outlined in ASTM C 952 [8] code. Cross couplet specimen
1617
Fig. 5. Soil–cement block masonry prism.
1618
B.V. Venkatarama Reddy, A. Gupta / Construction and Building Materials 22 (2008) 1614–1623
4.7. Compressive strength and stress–strain relationships for soil–cement block masonry prisms Compressive strength of soil–cement block masonry was determined by testing the masonry prisms. Four blocks high stack bonded masonry prisms (size: 305 · 143 · 436 mm) were used. A mortar joint thickness of 12 mm was maintained for all the prisms. The blocks were partially saturated by soaking them in water for a period of 4 min prior to casting of the masonry prisms. The prisms were capped with 12 mm thick 1:3 cement mortar and then cured for 28 days under a wet burlap. The prisms were tested (after soaking them in water for 48 h prior to testing) in a universal testing machine and the longitudinal compressive strains were measured by using a 200 mm demec gauge. Fig. 5 shows the details of soil–cement block masonry prism. 5. Results and discussion 5.1. Flow characteristics of mortars The flow value versus water–cement ratio for cement mortar and cement–lime mortar are shown in Figs. 6 and 7, respectively. The following observations can be made from the results shown in these figures:
140 120
Flow (%)
100
Natural sand (CMN) Medium sand (CMM) Fine sand (CMF)
80 60 40 20 0 1.1
1.3
1.5
1.7
1.9
2.1
2.3
Water-cement ratio Fig. 6. Flow versus water–cement ratio for cement mortar.
160 140
Flow (%)
(1) The flow value increases with the increase in water– cement ratio of the mortar for both cement mortar and cement–lime mortar irrespective of sand grading. The relationship between flow value and water– cement ratio is linear for both the mortar types and for all three grades of sand, except in case of cement mortar with fine sand, which has bilinear variation. (2) The mortar flow is very sensitive to water–cement ratio of the mix. For example in case of cement mortar, there is a three-fold increase in flow for about 25– 30% increase in water–cement ratio of the mix for all the three grades of sand. (3) Mortar with fine sand requires more water to attain similar flow values when compared to mortars with natural and medium sand for both cement mortar and cement–lime mortar. For example in case of cement mortar to achieve a flow of 90%, mortar with natural sand and medium sand requires water– cement ratio of 1.60, whereas mortar with fine sand needs water–cement ratio of 2.00. Similarly for cement–lime mortar to attain a flow of 90%, a water–cement ratio of 1.75 is required for mortars with natural and medium sand, whereas mortar with fine sand needs water–cement ratio of 2.35. This may be attributed to the fact that mortar having fine sand has large surface area. The results are comparable with the study conducted by Balen and Gemert [2]. (4) Cement–lime mortars require higher water–cement ratio when compared to cement mortars, to achieve similar flow values irrespective of type of sand used. The cement–lime mortars with natural and medium sand require 8% more water when compared to cement mortar to achieve flow values beyond 80%, whereas in case of fine sand it is about 17% more water.
120
Natural sand (CLMN) Medium sand (CLMM) Fine sand (CLMF)
Gupta [30] conducted a field study to asses the consistency of mortars by conducting flow table tests on fresh mortars used in the filed. He concludes that a flow value of 100% is commonly used in the field for the construction of load bearing masonry in India. The water–cement ratio required for both these mortars using different grades of sand, to achieve 100% flow can be obtained from the relationships shown in Figs. 6 and 7 and the values are given in Table 2 for both the mortars. These results clearly show that to achieve 100% flow water–cement ratio demand increases with increase in fineness of sand for both the mortars.
100
5.2. Compressive strength and water retentivity of mortars
80 60 40 20 0 1.3
1.5
1.7
1.9
2.1
2.3
2.5
2.7
Water-cement ratio Fig. 7. Flow versus water–cement ratio for cement–lime mortar.
Compressive strength of the mortar was determined by testing 70 mm cubes, whereas water retentivity of the mortar was examined by following the guidelines of BS: 4551 [5] code. The compressive strength and water retentivity values for both the mortars are given in Table 2. Three types of sand grading have been attempted in each category of mor-
B.V. Venkatarama Reddy, A. Gupta / Construction and Building Materials 22 (2008) 1614–1623
1619
Table 2 Compressive strength, water retentivity and elastic properties of mortars (mortar flow = 100%) Mortar type
Water–cement ratio
Dry density (g/cm3)
Compressive strength (MPa)
Water retentivity (%)
Initial tangent modulus (MPa)
Strain at peak stress
CMN CMM CMF CLMN CLMM CLMF
1.65 1.69 2.04 1.79 1.83 2.40
1.90 1.88 1.79 1.90 1.87 1.63
5.40 5.22 3.30 5.94 4.45 2.04
74 82 85 82 85 83
4534 7764 4245 4820 7737 2704
0.0040 0.0012 0.0017 0.0026 0.0016 0.0017
Compressive stress (MPa)
1. Water–cement ratio increases as the fineness of sand increases in order to maintain 100% flow, for both the mortars. Increase in water–cement ratio leads to decrease in dry density of the mortar and hence reduction in mortar strength with the increase in fineness of sand. The compressive strength of cement mortar and cement–lime mortar using natural sand at 100% flow is in the range of 5–6 MPa. For a given flow value of 100% the compressive strength of cement–lime mortar is more sensitive to fineness of sand as compared to cement mortar. The study conducted by Drew and Braj [1] on cement mortars also shows that as the water content of the mix is increased the dry density reduces and results in lower compressive strength of mortar. 2. Water retentivity increases with increase in fineness of sand for cement mortars. There is a 15% increase in water retentivity when the natural sand is replaced by fine sand. Whereas in case of cement–lime mortar, there is only a marginal variation in water retentivity values as the sand grading changes. Generally, cement–lime mortars possess better water retentivity than cement mortars due to the presence of fine lime particles in cement–lime mortar.
4 3.5 3 2.5 2 1.5 Natural sand (CMN) Medium sand (CMM) Fine sand (CMF)
1 0.5
0 0
0.001
0.002
0.003
0.004
0.005
Longitudinal strain Fig. 8. Stress–strain characteristics for cement mortar at a flow value of 100%. 3.5
Compressive stress (MPa)
tar to understand the influence of sand fineness on compressive strength and water retentivity of mortars. The following points emerge from the results given in Table 2:
3 2.5 2 1.5 1 Natural sand (CLMN) Medium sand (CLMM) Fine sand (CLMF)
0.5 0 0
0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035 0.004
Longitudinal strain
5.3. Stress–strain characteristics of mortars The stress–strain characteristics like initial tangent modulus and the stain at peak stress values for both the mortars are given in Table 2. Figs. 8 and 9 show the stress–strain relationships for cement mortar and cement–lime mortar using three different grades of sand. The stress–strain curves have been generated by keeping the mortar flow at 100% for all the cases. The following observations can be made from these results: (1) Mortars using medium sand (fineness modulus of 2.45) give highest modulus for both the mortars. In case of cement mortar with medium sand the modulus is 50% more as compared to other two types of sands. Cement–lime mortar with fine sand and natural sand have modulus values of 60% and 35% of that for the medium sand, respectively. Mortars with fine sand give lowest modules.
Fig. 9. Stress–strain characteristics for cement–lime mortar at a flow value of 100%.
(2) The strain at peak stress for cement mortar with natural sand is significantly high of the order of 0.004, when compared to mortar with medium and fine sand. The strain value decreases as the sand becomes finer. In case of cement–lime mortar with natural sand strain at peak stress is 0.0026. This strain reduces to about 0.0016 for mortar with medium and fine sand. 5.4. Drying shrinkage Shrinkage that takes place during hardening of the mortar can be called as drying shrinkage and most of it takes place in the first few months. A part of drying shrinkage is recovered on immersion of mortar in water. The drying
B.V. Venkatarama Reddy, A. Gupta / Construction and Building Materials 22 (2008) 1614–1623
shrinkage of mortar could depend on various factors like water–cement ratio, cement content, sand grading, curing period, etc. In masonry construction, drying shrinkage of mortar causes shrinkage cracks observed at the masonry unit–mortar interface and it can also result in impaired bond [31,32]. The drying shrinkage of mortar was tested in the laboratory through a mortar prism of size 250 · 25 · 25 mm in unrestrained condition. It differs from that experienced in a masonry wall where the drying shrinkage of the mortar is influenced by restraint offered by masonry units, masonry unit absorption characteristics, thickness of mortar bed joint, etc. Thus drying shrinkage value of the mortar examined in the laboratory is more useful for comparative purposes. The drying shrinkage of cement mortar and cement– lime mortar was examined. All the tests were performed at a mortar flow of 100%. The drying shrinkage of mortars at 25 days of drying duration is taken as the ultimate drying shrinkage. The results given in the following sections represent the mean of five specimens. The variation in drying shrinkage of the cement mortar and cement–lime mortar specimens with the duration of drying is shown in Figs. 10 and 11, respectively. A comparison in the variation of
Drying shrinkage (%)
0.12 0.1 0.08 0.06 Natural sand (CMN)
0.04
Medium sand
0.02
0 0
5
10
15
20
25
30
Drying duration (days) Fig. 10. Variation in drying shrinkage with drying duration for cement mortar.
0.16
Drying shrinakge (%)
0.14
0.16 0.14
Drying shrinkage (%)
1620
Cement mortar Cement lime mortar
0.12 0.1 0.08 0.06 0.04 0.02 0 Natural
Medium
Fine
Sandtype Fig. 12. Variation in ultimate drying shrinkage with sand grading for cement mortar and cement–lime mortar.
ultimate drying shrinkage of cement mortars and cement– lime mortars with the type of sand used is shown in Fig. 12. The following points emerge from the results on drying shrinkage: (a) The ultimate drying shrinkage values lie in the range of 0.057–0.151% for cement mortar and cement–lime mortars for various grades of sand. The drying shrinkage of these mortars reaches the ultimate drying shrinkage in initial 3–4 days of drying as illustrated in Figs. 10 and 11. The ultimate drying shrinkage is generally more in case of cement mortar as compared to cement–lime mortar, except for mortars using fine sand. (b) The fineness of sand in mortars greatly affects the ultimate drying shrinkage values as shown in Fig. 12. There is an increase in drying shrinkage value when the sand becomes finer. For example, in case of cement mortar, there is more than one third increase in the ultimate drying shrinkage when the sand grading is changed from coarser to finer (fineness modulus changing from 3.21 to 1.72). On the other hand, 2.7 times increase in the ultimate drying shrinkage is observed in the case of cement–lime mortar for similar change in sand grading.
012. 0.1
5.5. Influence of sand grading on tensile bond strength
0.08 0.06 0.04
Natural sand (CLMN) Medium sand (CLMM) Fine sand (CLMF)
0.02 0 0
5
10
15
20
25
30
Drying duration (days) Fig. 11. Variation in drying shrinkage with drying duration for cement– lime mortar.
The influence of sand grading on tensile bond strength of soil–cement block couplets was examined using both the mortars. These results are tabulated in Table 3. The variation in tensile bond strength with the grading of sand is shown in Fig. 13. The tensile bond strength values lie in the range of 0.08–0.252 MPa for the couplets using both the types of mortars with three different grades of sand. In case of cement mortar, the tensile bond strength steadily decreases as the sand grading changes from coarser to fine
B.V. Venkatarama Reddy, A. Gupta / Construction and Building Materials 22 (2008) 1614–1623
Mortar type
CMN CMM CMF CLMN CLMM CLMF
Tensile bond strength (MPa)
0.181 0.100 0.080 0.233 0.252 0.093
(0.15–0.21) (0.041–0.213) (0.05–0.111) (0.198–0.259) (0.201–0.297) (0.076–0.119)
Masonry prism strength (MPa)
Masonry Initial tangent modulus (MPa)
Strain at peak stress
4.55 4.84 4.55 5.27 4.67 4.57
4163 4087 3084 4500 4770 2864
0.0044 0.0046 0.0060 0.0052 0.0040 0.0073
4 3 2
Natural sand(CMN) Medium snad(CMM)
1
Fine sand (CMF)
0 0
0.001
0.002
0.003
0.004
0.005
0.006
0.007
Longitudinal strain
6
Compressive stress (MPa)
Tensile bond strength (MPa)
5
Fig. 14. Influence of sand grading on stress–strain relationships for soil– cement block masonry with cement mortar.
Values in parenthesis indicate the range of values.
0.3
6
Compressive stress (MPa)
Table 3 Bond strength, compressive strength and elastic properties of soil–cement block masonry
1621
Cement mortar Cement-lime mortar
0.25 0.2 0.15 0.1 0.05
5 4 3 Natural sand (CLMN)
2
Medium sand (CLMM)
1
Fine sand (CLMF)
0 0
0 Natural
Medium
Fine
0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 0.009
Longitudinal strain
Type of sand Fig. 13. Influence of sand grading on tensile bond strength of soil–cement block couplets.
(fineness modulus changing from 3.21 to 1.72). There is 56% reduction in the tensile bond strength as the sand grading is changed from coarser to fine. On the other hand, in case of cement–lime mortar there is hardly any variation in bond strength between coarse and medium sand. But tensile bond strength decreases by 60% when sand grading is changed from medium to fine for cement–lime mortar. Thus, for both the mortars use of very fine sand reduces the tensile bond strength significantly. 5.6. Compressive strength and stress–strain characteristics of masonry Compressive strength and stress–strain relationships for the soil–cement block masonry were obtained by using both the types of mortars with three types of sand gradings. Stress–strain relationships for the masonry using cement mortar and cement–lime mortar are shown in Figs. 14 and 15, respectively. Details of compressive strength, initial tangent modulus and strain at peak stress of soil–cement block masonry using both the mortars are given in Table 3. The following observations can be made from the results given in the table and figures:
Fig. 15. Influence of sand grading on stress–strain relationship for soil– cement block masonry using cement–lime mortar.
(1) In case of cement mortar the masonry compressive strength shows a minor variation, when the sand grading is changed from coarser to fine (fineness modulus changing from 3.21 to 1.72). The masonry strength lies in the range of 4.55–4.84 MPa for the three grades of sand. The masonry strength using cement–lime mortar is in the range of 4.57– 5.27 MPa. The masonry prisms made of cement–lime mortar show a decrease in masonry compressive strength, as the sand used in the mortar becomes fine. There is a 13% decrease in compressive strength when the sand grading is changed from coarser to fine. (2) Initial tangent modulus for masonry with cement mortar lies in the range of 3000–4100 MPa, whereas for cement–lime mortar it is in between 2800 and 4800 MPa. There is a significant reduction in masonry modulus when the grading of sand is changed from coarse to fine. For example in case of masonry using cement mortar there is about 25% reduction in initial tangent modulus when the sand grading is changed from coarse to fine, whereas for cement–lime mortar a decrease of about 40% is observed. The masonry prisms with fine sand for both cement mortar and cement–lime mortar have
1622
B.V. Venkatarama Reddy, A. Gupta / Construction and Building Materials 22 (2008) 1614–1623
higher values of strain at peak stress, of the order of 0.006 and 0.0073, respectively; as compared to masonry prisms with other two grades of sand (strain at peak stress less than 0.0052). Thus, mortar containing fine sand (low fineness modulus) leads to more ductile masonry as indicated by modulus and strain at peak stress values. 6. Concluding remarks Influence of sand grading on certain properties of cement mortar, cement–lime mortar and the soil–cement block masonry were examined. The following conclusions can be drawn from this investigation: (a) To achieve any given consistency, the mortars using fine sand require 25–30% more water. (b) Compressive strength of cement–lime mortar is more sensitive to the fineness of sand as compared to cement mortar. For a given consistency the compressive strength of mortar decreases with increase in fineness of sand. (c) Water retentivity of cement mortar increases with increase in fineness of sand. Water retentivity of cement–lime mortar is not sensitive to the variation in sand gradation. (d) Mortar modulus and strain at peak stress decrease as the sand fineness increases. There is 40% reduction in modulus and strain values as the sand grading is changed from coarse to fine. (e) Sand grading greatly affects the drying shrinkage values of both the mortars. Drying shrinkage increased by 1/3rd and 2.7 times for cement mortar and cement–lime mortar, respectively, as the sand grading changed from coarse to fine. (f) Tensile bond strength of soil–cement block couplets is sensitive to the fineness of sand used for the mortars. There is 55–60% reduction in tensile bond strength of soil–cement block couplets as fineness modulus of sand changes from 3.21 to 1.72. Thus, for both the mortars use of fine sand reduces the tensile bond strength of soil–cement block couplets significantly. (g) Soil–cement block masonry prisms show marginal variation (10%) in compressive strength as the sand grading changes from coarse to fine. But there is 25– 40% reduction in modulus of soil–cement block masonry as the sand grading is changed from coarse to fine. Reduction in masonry modulus is accompanied by increase in values of strains at peak stress for both the mortars. It is interesting to note here, that even though the grading curves for medium and fine sand used in the investigations fall within the grading limits specified by I.S. 8112 code, there is considerable variation in the properties of mortar as well as soil–cement block masonry using these two grades of sand.
References [1] Drew Currie, Braj Sinha. Survey of Scottish sands and their characteristics which affect mortar strength. Chem Ind 1981;19:639–46. [2] Balen Van K, Gemert Van D. Lime–cement mortars compared to the commonly used cement mortars in Belgium. Belgium: K.U. Leuven (University); 1990. [3] I.S. 2116-80 (reaffirmed 1998). Indian standard specification for sand for masonry mortars. New Delhi, India: Bureau of Indian Standards; 1998. [4] ASTM C 144-04. Standard specification for aggregate for masonry mortar. USA: American Society for Testing and Materials; 2004. [5] BS: 4551. British standard methods of testing mortars, screeds and plasters. UK: British Standards Institution; 1980. [6] Anderson C, Held LC. The effect of sand grading on mortar properties and tensile bond strength of brickwork specimens. Proc Br Mason Soc 1986;1:1–6. [7] Groot Caspar. Effects of water on mortar–brick bond. PhD thesis, Faculty of Civil Engineering, Delft University of Technology, The Netherlands; 1993. [8] ASTM C 952-91. Standard test method for bond strength of mortar to masonry units. American Society for Testing and Materials; 1991. [9] I.S. 8112. Indian standard specification for 43 grade ordinary Portland cement. New Delhi, India: Bureau of Indian Standards; 1989. [10] Fitzmaurice R. Manual on stabilized soil construction for housing. New York: United Nations; 1958. [11] U.N. Report. Soil–cement – its use in building. New York, USA: Department of Economic and Social Affairs, United Nations; 1964. [12] Theunissen Ph. Building with earth Dimension 3. Bimonthly Rev Inform Serv Belgian Admin Dev Cooper 1985;4:10–2. [13] Jagdish KS. The progress of stabilised soil construction in India. In: Proceedings of the national seminar on stabilised mud blocks for housing and building, vol. 1; 1988. p. 17–43. [14] Hugo Houben, Hubert Guillaud. Earth construction – a comprehensive guide. UK/France: ITDG Publishing/CRATerre-EAG; 2003. [15] Walker P, Venkatarama Reddy BV, Mesbah Ali, Morel JeanCluaude. The case for compressed earth block construction. In: Proceedings of the sixth international seminar on structural masonry for developing countries; 2000. p. 27–35. [16] Lunt MG. Stabilized soil blocks for buildings. Overseas building notes, vol. 184; 1980. [17] Olivier M, Mesbah Ali. Influence of different parameters on the resistance of earth, used as a building materials. In: Proceedings of the international conference on mud architecture; 1987. [18] Heathcote K. Compressive strength of cement stabilized pressed earth blocks. Build Res Inform 1991;19(2):101–5. [19] Venkatarama Reddy BV, Jagdish KS. Influence of soil composition on the strength and durability of soil–cement blocks. Indian Concr J 1995:517–24. [20] Walker P, Stace T. Properties of some cement stabilized compressed earth blocks and mortars. Mater Struct 1997;30:545–51. [21] Walker P. Strength and erosion characteristics of earth blocks and earth block masonry. J Mater Civil Eng (ASCE) 2004;16(5):497–506. [22] Venkatarama Reddy BV, Walker Peter. Stabilised mud blocks: problems, prospects. In: Proceedings of the international earth building conference EarthBuild2005; 2005. p. 63–5. [23] Venkatarama Reddy BV, Gupta Ajay. Characteristics of soil–cement blocks using highly sandy soils. Mater Struct (RILEM) 2005;38(280):651–8. [24] Venkatarama Reddy BV, Gupta Ajay. Strength and elastic properties of stabilised mud block masonry using cement–soil mortars. J Mater Civil Eng (ASCE) 2006;18(3):472–6. [25] I.S. 3495. Methods of tests for burnt building bricks – part 2. Determination of water absorption. New Delhi, India: Bureau of Indian Standards; 1992.
B.V. Venkatarama Reddy, A. Gupta / Construction and Building Materials 22 (2008) 1614–1623 [26] I.S. 2250. Indian standard code of practice for preparation and use of masonry mortars. New Delhi, India: Bureau of Indian Standards; 1981. [27] ASTM C 91-95. Standard specification for masonry cement. USA: American Society for Testing and Materials; 1995. [28] ASTM C 1148-92a. Standard test methods for measuring the drying shrinkage of masonry mortar. USA: American Society for Testing and Materials; 1992. [29] Venu Madhava Rao K, Venkatarama Reddy BV, Jagadish KS. Flexural bond strength of masonry using various blocks and mortars. Mater Struct (RILEM) 1996;29:119–24.
1623
[30] Gupta Ajay. Studies on characteristics of cement–soil mortars and soil–cement block masonry. MSc(Eng) thesis, Department of Civil Engineering, Indian Institute of Science, Bangalore, India; 2003. [31] Baker LR. Some factors affecting the bond strength of brickwork. In: Proceedings of the fifth international brick masonry conference, vol. II-9; 1979. p. 84–9. [32] Lawrence SJ, Cao HT. Microstructure of the interface between brick and mortar. In: Proceedings of the eighth international brick/block masonry conference; 1988. p. 194–204.