The objective of this study is to introduce a type of concrete that is more cost effective when compared to other concretes, yet is still compliant with the BS8110 standards of mechanical characteristics for design and construction of structural concrete. This paper presents part of the test results of an ongoing study in an attempt to use the low cost solid waste material 'almond shell' (AS), as a coarse aggregate in the production of structural lightweight concrete. Test results in this paper determine various structural and physical properties, which are elastic modulus, compressive strength, and the bond strength between the concrete and aggregate. There is also an investigation into the flexural behaviour of the AS concrete. Based on these experimental results, it was determined that although AS concrete has a lower modulus of elasticity than other test samples, middle-scale beam tests revealed that deflections under expected design service loads are within acceptable limits. The BS8110 standards specify acceptable limits for the span-deflection ratios of lightweight structural concrete, and this study shows that AS concrete complies with this, with values of deflection between 252 and 263. Laboratory investigations showed satisfactory performance of the AS concrete within the remit of the BS8110 code, and it can be concluded that for lightweight concrete as a structural material, almond shell shows great as a coarse aggregate. This result especially has significance for low-cost applications such as in housing construction and for use in earthquake prone areas.
The issue of resource reduction has become more prominent in recent years, and global pollution has resulted in a challenge for engineers to seek and develop new materials based on resources that are renewable. To achieve this, researchers are examining the use of waste materials, recycled items, and by-products from the construction industry. These waste materials are already commonly included as aggregate materials in the production of lightweight structural concrete. There has already been significant amounts of study into the physical properties and structural application of lightweight concrete produced with the inclusion of aggregate materials, which has been focused on investigating aggregates that are manufactured, naturally occurring, and also waste materials from industry.
Stone fruit belongs to the same family of fruit as the peach, which is known as the Rosaceae family. It is native to the Middle East, and it thrives only in a dry, hot environment. Almonds are the single seed from a type of stone fruit, and they are grown throughout California in the USA, in the entire Mediterranean Region, and also in Australia, Africa, Turkey and Iran. The Almond, as a stone fruit, is not easily grown in a wet environment, meaning these are prime locations for its cultivation. Currently, research efforts have been directed towards the potential for using almond shell (AS) as a coarse aggregate in the production of structural lightweight concrete. In Iran, there is over 70 tonnes of Almond shell waste produced at each cropping, meaning it is a prime location for making use of this waste material. Recycling AS leads to many benefits, such as maximisation of the use of almond shell, preservation of natural resources and maintenance of ecological balance. In addition, the current economic and social climate means that in many countries across the world there is a higher recent deficit in low-cost, affordable housing, meaning that AS has potential as a cheaper alternative to the conventional aggregates in fulfilling this demand.
In structural concrete, aggregates that have a density of 1200Kg/m3 (dry unit weight) are commonly used, and are classified as lightweight aggregates (Owens, 1993). AS aggregate has a unit weight of 600-700 kg/m3, which is approximately 55% lighter in comparison with traditional aggregate materials, such as crushed stone. As a result of using lighter aggregate material, the structural concrete produced will have a lower weight. This AS aggregate lightweight concrete has been very recently developed and employed as a structural material in the construction industry, meaning that its structural performance still requires investigation. This study attempts to determine some important characteristics of AS aggregate concrete, to result in wider acceptance of AS as a lightweight aggregate material alternative in producing concrete that can be used as a structural, lightweight concrete for use in the construction industry, in particular for the application of low-cost residential dwellings. To achieve this, this investigation will attempt to determine the elastic modulus, compressive strength, and the bond strength between the concrete and aggregate. There is also a study into the flexural behaviour of the AS concrete samples
MIX DESIGN OF AS CONCRETE
The components of the AS concrete in this study included ordinary Portland cement (ASTM Type 1), with crushed sand as a fine aggregate, almond shell (AS) as a coarse aggregate, and a superplasticiser to reduce the water content. In this study, a Sodium Naphthalene Sulphonate Formaldehyde Condensate was used, of Type F (Collepardi et al. 1993). The chemical structure of the SNF based superplasticiser is shown in Figure 1.
Figure 1. Chemical structure of naphthalene sulphonate formaldehyde based superplasticiser.
The almond shell aggregates for this study were acquired from local almond shell mills. AS is usually available in particles of various shapes, which as shown in Figure 2 are generally highly irregular. To prepare the AS particles as aggregates, they were sieved with the following regime: the particles were first passed through the No.1/2 sieve, with the particles that were retained being discarded. The remaining AS was then passed through the No.4 sieve, with the particles retained from this second sieving being used for the experiment. The particle size distribution of the AS aggregate is shown in Figure 3, and the properties of the crushed sand and AS are shown in Table 1.
Figure 2. Various shapes of AS aggregate.
Figure 3. Particle size distribution of AS aggregate.
Table 1. Properties of crushed river sand and AS
Almond shell (AS)
Maximum grain size, mm
AS particle thickness, mm
Bulk unit weight, kg/m3
Los Angeles abrasion value, %
Aggregate impact value, %
Aggregate crushing value, %
24-h water absorption, %
Based on the properties of the fine and coarse aggregate available for this study, the necessary proportions of materials in the concrete mix were estimated, and subsequently these trial mixes were altered and modified to achieve a practical and satisfactory result. The acceptable mix comprised 450 kg/m3 cement, 810 kg/m3 sand, and 440 kg/m3 AS, with a free water to cement ratio of 0.35. The cement content used in production of concrete in this research was acceptable for lightweight structural content, based on limits determined by Mindess et al. (2003). Superplasticiser amount used was based on values suggested by the manufacturer, and were set at 1.5% by cement weight (6.75 kg/m3. This value was fixed for the duration of the experiment. The AS aggregate used was mixed when it satisfied the condition of 'saturated surface dry', or 'SSD'. This is where the particle surfaces are saturated, but the insides are dry. This was achieved by submerging the aggregate in potable water for 24 hours.
The objective of this investigation was to determine if the structural and physical properties of AS concrete that were specified in the previous section are within acceptable limits for structural application. To this end, various laboratory tests were carried out. Compression strength tests were carried out to conform to the standard BS 1881: Part 116, on concrete samples that were cubes of dimension 100mm. The initial modulus of elasticity was determined based on the ASTM C 469-87a standard, on concrete cylinder samples of 150Ã-300 mm. The strength of the aggregate-cement bond in the AS concrete was calculated on concrete cylinder samples of 100x200mm, by way of pullout tests. The flexural properties of the AS concrete were studied by way of tests on full-scale models of concrete beams. For each test, three samples were studied, and the results obtained were an average for these three values. For the flexural tests, there was one beam studied, but with three tension reinforcements.
The concrete samples were produced by mixing the constituent materials in a rotating drum, which was compliant with the relevant standard for aggregates that are SSD, which is section 6.3 of BS 1881: Part 125. In order to limit the evaporation of moisture from the concrete, which would result in subsequent shrinkage, once the moulds were cast, a plastic sheet was used to cover the samples. This was left in-situ for 24 Â± 3 hours in the laboratory at ambient conditions of 24-28 Â°C and relative humidity of 85% - 95%. After this, the cube and cylindrical specimens were transferred into a 25-30 Â°C water tank until testing begun. The samples for flexural testing, the full-scale prototype beams, were moist-cured over an additional six days. They were then stored until testing in the laboratory in ambient conditions. The bond between the reinforcement and AS concrete was investigated by application of the bond-slip calculation. To this end, the pullout test was used. The samples were deformed concrete bars with diameters of 10mm, 23mm and 13mm. The tests were performed at various stages of aging, starting at 3 days, then at 7 days, and subsequently at 28 days, 56 days, 90 days and finally at 180 days. The formula (1) was then applied to determine the strength of the bond.
Ï„ = Stress in the bond, MPa
F = Load applied to the bond, N
d = Diameter of the bar, nominal, mm
l = embedment length (mm)
Three separate reinforced sample beams were produced and studied. These test beams were all of the same dimensions, which were 150mmÃ-230mmx3200mm, and they clearly have a rectangular cross section. Once in the test apparatus, the effective length under study was reduced to 3000 mm. These dimensions for the samples were chosen to be sufficient in magnitude in order to accurately represent an actual structural component. The tension reinforcement for the three beam samples were 2Ø10, 2Ø12, and 2Ø13. To mount the experiment, two steel hanger bars were used, of 8mm in diameter. In order to prevent shear mode failure, and in order to only investigate the properties specified, shear links were employed. The samples were subjected to the loading profile known as the four point bending test, and the resulting deflections in the pure bending zone were measured by way of linear voltage displacement transducers, or LVDTs. These were selected to read up to a displacement of 100mm, and three were used. The experiment and beam specifications are shown in Figure 4.
Figure 4. Beam testing setup and details.
RESULTS AND DISCUSSION
The fresh concrete density during this study ranged from 1863 to 1897 kg/m3, and the air content in the specimens was in the range of 4.2% to 4.9%, which is large when compared to other lightweight concrete types. It is suggested that this is a results of the diverse and non-uniform topography of the individual AS particles, which prevented their full compaction with each other; however, this air content value for the AS concrete samples is still within the acceptable values of 4% to 8% specified in the ACI 213R-87 samples. The slump test results for the AS fresh concrete workability evaluation was performed, which determined that AS concrete was in the range of 6 to 8 cm. Analysis of these test results revealed that the AS concrete had a medium degree of workability, which was within the acceptable range of a workable concrete. The properties after 28 days of the hardened AS aggregate concrete are shown in Table 2.
Table 2. Physical properties of AS concrete.
Air-dry density, kg/m3
Compressive strength, MPa
Modulus of elasticity, GPa
Pullout bond strength, MPa
Concretes that have a density of less than 1900 kg/m3 are classified as lightweight concretes, and from Table 2 it can be seen the density of AS concrete is within this range, meaning it can be termed as a lightweight concrete. Compared to normal concretes, which have a density of around 2400 kg/m3, AS concrete is approximately 25.5% lighter. This shows that use of AS concrete would eliminate 25.5% of dead load when used in construction, whilst still maintaining satisfactory physical properties. Another benefit of this reduced weight is that the damaging effect on concrete structures exposed to disastrous, structural weight-dependent earthquake forces and inertia forces can be lessened.
In order to evaluate compressive bearing capacity of AS concrete, compressive strength tests were performed on 10 cm cubes at an age of 28 days. The results determine an average compressive strength of 32.5 MPa, which is approximately 90% higher than the minimum required compressive strength provided by the ASTM C330 standard, which for lightweight structural concrete is set at 17 MPa. Although AS is of organic origin, it was shown that there was negative impact on the results from biological decay, as the cubes retained acceptable strengths even after 180 days. This has been showed in figure 5.
Figure 5. Compressive strength development of AS concrete.
Different damage and failure mechanisms were observed at different sample ages. At the earlier stages of testing between 3 days and 28 days, the failure in the concrete samples under compression loading was primarily due to failure in the interface between the cement paste and the AS aggregate, where the crack follows a path around the aggregate (Figure 6a). At later stages, from 42 to 180 days, this mortar-aggregate interface bond is stronger, and hence the crack grows through the aggregate as illustrated in Figure 6b.
Crack path around AS aggregates
Crack path through
Figure 6. Paths of the cracks at (a) preliminary and (b) late stages.
The bond strength development of AS concrete is illustrated in Figure 7. Based on the pullout test performed, the bond strength of AS concrete was found to be about 2.7 to 3.5 times higher than is required by the BS 8110 standard. The failure in the test specimens was all by the same mechanism, which resulted in the concrete sample cover being split. This failure mode was catastrophic and very sudden. Longitudinal cracking in the region was observed to accompany the failure, which spread over the entire beam length prior to failure. The reason for this is that radial cracks were formed from the loading that was transferred from the steel hanger bars to the concrete. The splitting failure then occurs as the cracks that form result in bond forces being projected outwards from this point, and the confining concrete cover is subsequently cracked. The bond strength of the AS concrete was approximately 26% to 29% of the compressive strength. This is within similar values for Aerocrete (Chitharanjan et al., 1988), sintered pulverized fuel ash concrete (Orangun, 1967), and other similar lightweight structural concretes currently commonly employed in construction.
Figure 7. Bond strength development of AS concrete.
The determination of elastic modulus is crucial, and it is a vital parameter when designing concrete structures due to the fact that it is necessary for evaluating deflections and determining potential cracking and failure. Figure 8 shows a stress-strain curve for the AS concrete. The strain value corresponding to the maximum stress is approximately 0.005 micro-strains. One particular concern for the AS concrete is that it has a low modulus of elasticity compared to other similar concretes. This was studied by prototype beam testing.
Figure 8. Stress-strain curve for AS concrete.
All of the tested beam samples exhibited flexural failure as expected typically for lightweight structural concrete. The failure was slow and gradual, and due to the fact that all the beams were under-reinforced, it resulted in the first failure mode being in the tensile reinforcement. This was before the concrete cover was crushed due to loading in the bending zone. To determine a predicted value for the beams' ultimate moments, a rectangular stress block analysis was used as recommended by the BS 8110 standard. Once the experiment was complete, the actual test beams' ultimate moments were about 22% to 31% greater when compared to the predicted moments for normal weight concrete. These tests demonstrate that the BS 8110 standard is useful for providing a conservative prediction for determining ultimate moments in singly reinforced AS concrete beams. The deflection resulting from loading is a vital parameter in design of a structural member. When applying the expected service load to the samples, which includes the dead load and live load, the mid span deflection obtained was 11.43mm, 11.62mm, and 11.87mm for the three test beams. Despite the lower elastic modulus for AS concrete compared with other lightweight concretes, the design service load deflections observed were compliant with the BS 8110 standard. The experimental deflections were in the region of 252 - 263 mm. The moment/deflection graphs for the three beams are shown in figure 9.
Figure 9. Moment-deflection curves for the test beams.
From the results of this study, it has been determined that almond shell is compliant with required standards for use as an aggregate for inclusion in production of lightweight structural concrete. It especially has potential for use in applications where a low to medium strength is required, and low cost is necessary, such as in affordable housing. By integrating this waste material into concrete mixtures, not only can a reduction in concrete weight and production costs be achieved, but there is the added benefit that the ecological cyclic system can be preserved. Based on this research study, the following conclusions can be drawn:
The evaluation of the workability of fresh AS concrete by the slump test was 6 to 8 cm, which showed that the AS concrete had a medium degree of workability. It could therefore be defined as workable concrete.
The air content percentage in the concrete, which was 4.2 - 4.9%, could be reduced by optimising the gradation curve. However, the air content in the AS concrete is within the allowable range of 4-8% recommended by ACI 213R-87.
The compressive strength of the AS concrete after 28 days was 32.5 MPa, which satisfies the requirement for structural lightweight concrete.
Despite the varying shapes of AS aggregate particles, the bonding properties of AS concrete are up to the standard as shown by other commonly used lightweight structural concretes. It should be noted that this bond and its mechanical properties were obtained by using a cement content of 450 kg/m3.
Although AS concrete has a low modulus of elasticity of 6.73 GPa, full scale beam tests showed that the deflection under the expected design service loads was acceptable. The ratios of the effective span length to maximum mid span deflection ranged from 252 to 263. This is compliant with the acceptable ranges specified in BS 8110.
From the experiment, it has been shown that the actual experimental values of ultimate moments for the test beams were 22% to 31% higher than those predicted by the BS 8110 standard. The standard can therefore be used to give a conservative estimate.
ACI 213R-87, Guide for Structural Lightweight Aggregate Concrete, American Concrete Institute.
ASTM C330, Standard Specification for Lightweight Aggregates for Structural Concrete, Annual Book of ASTM Standards ASTM C 469-87a, Standard Test Method for Static Modulus of Elasticity and Poisson's Ratio of Concrete in Compression, Annual Book of ASTM Standards.
BS 1881, Part 116, Method for Determination of Compressive Strength of Concrete Cubes, British Standards Institution, London.
BS 1881, Part 125, Methods for Mixing and Sampling Fresh Concrete Samples in the Laboratory, British Standards Institution, London.
BS 8110, Structural use of Concrete Part 1, Code of Practice for Design and Construction, British Standards Institution, London, 1985.
Chitharanjan N., Sundararajan R. and Manoharan P.D., \Development of Aerocrete: A New Lightweight High Strength Material", The International Journal of Cement Composites and Lightweight Concrete, 10, 27-38, 1988.
Collepardi M., Coppola L., Cerulli T., Ferrari G., Pistolesi C., Zaffaroni P., and Quek F., Â«Zero Slump Loss Superplasticized ConcreteÂ», Proceedings of the Congress "Our World in Concrete and Structures", Singapore, pp 73-80, 1993.
Mannan M.A. and Ganapathy C., \Behavior of Lightweight Concrete in Marine Environments", Proceedings of the International Conference on Ocean Engineering, Chennai, India, 409-413, 2001.
Mindess S., Young J.F. and Darwin D., Concrete, 2nd Edition, Prentice Hall, USA, 2003. Orangun C.O., \The Bond Resistance between Steel and Lightweight-Aggregate (Lytag) Concrete", Building Science, 2, 21-28, 1967.
Owens P.L., Lightweight Aggregates for Structural Concrete, Structural Lightweight Aggregate Concrete, edited by J.L. Clarke, Blackie Academic & Professional, London, 1993.