- Email: [email protected]
Fracture behaviour of a fibre reinforced earthen material under static and impact flexural loading
Construction and Building Materials 109 (2016) 109–119
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Fracture behaviour of a fibre reinforced earthen material under static and impact flexural loading F. Aymerich a, L. Fenu b, L. Francesconi a,⇑, P. Meloni a a b
Department of Mechanical, Chemical and Materials Engineering, University of Cagliari, Italy Department of Civil and Environmental Engineering and Architecture, University of Cagliari, Italy
h i g h l i g h t s We studied the effect of hemp fibres on the mechanical performance of an earthen material. The samples were subjected to static and impact bending. Hemp fibres greatly increase the fracture resistance and the energy absorption capacity. Both unreinforced and reinforced materials show a strong sensitivity to the rate of loading. The post-cracking performance is improved by increasing the fibre fraction and length.
a r t i c l e
i n f o
Article history: Received 28 June 2015 Received in revised form 27 December 2015 Accepted 27 January 2016 Available online 6 February 2016 Keywords: Earthen materials Hemp fibres Impact Fracture Toughness
a b s t r a c t The study investigates the enhancements in the load carrying capacity, crack resistance and energy absorption properties provided by the addition of hemp fibres in an earthen material. Notched earthen samples reinforced with two fibre contents (2% and 3% in weight) and three fibre lengths (10, 20, and 30 mm) were manufactured and tested under static and impact bending to investigate and compare the influence of the reinforcement on the fracture resistance of the soil material at low and high strain rates. The results of the experimental analyses show that the incorporation of fibres greatly improves the peak load, the post-crack strength, the ductility and the energy dissipation of soil under both static and impact bending. The mechanical response of both unreinforced and reinforced samples is significantly affected by the rate of loading, with samples exhibiting higher values of strength and absorbed energy under impact than under static bending. For both static and impact loading, the post-crack response of the material at large deformations is clearly improved by increasing the fibre content and, at the same fibre content, by increasing the fibre length. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Earth is the most widely utilised natural construction material in the world and unfired earth is by large the prevalent material used for dwelling structures in many hot, arid or temperate regions, with estimates indicating that 30–50% of the world’s population live in houses made of unbaked earth-based materials [1,2]. Earth construction methods are varied, including techniques such as rammed earth, cob, wattle and daub, brick masonry, with soil or mud bricks (adobes) being probably the most common methods of earth construction worldwide [3,4].
⇑ Corresponding author. E-mail addresses: [email protected] (F. Aymerich), [email protected] (L. Francesconi). http://dx.doi.org/10.1016/j.conbuildmat.2016.01.046 0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
The use of earth as a construction material is a natural solution in developing countries, since the raw materials, typically soil and sand, are abundant and locally available, and can be processed with simple technologies, without requiring expensive tools or specialised manufacturing skills. Even though the majority of earthen architecture is located in less developed countries, the past decade has witnessed a growing interest in the use of earthen building materials in industrialised countries; this renewed interest has been prompted not only by specific requirements for conservation and rehabilitation of architectural heritage sites [5], but also by increasing demands for more sustainable and energyefficient materials and techniques for the building and construction industry [6]. In this respect, as compared to conventional building materials, earth-based materials provide significant reductions in environmental impact, because of the low toxicity and high recyclability properties, the minimal transportation costs
110
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119
(as a consequence of the local availability of raw materials), and the high energy efficiency and ecological compatibility of the manufacturing processes. [6,7]. As an example, it has been estimated that building a square metre of masonry with compressed earth blocks consumes about 15 times less energy than a square metre of fired bricks [8]. Similarly, a recent case study [9] revealed that the embodied energy of a three-storey rammed earth building was about one third than that of an analogous building made of conventional clay fired bricks and between one ninth and one fourth of that of a reinforced concrete framed structure building. Further benefits of earthen building materials are related to their excellent performances in terms of heat storage capacity, thermal inertia and hygrothermal behaviour, which ensure low fluctuations in indoor air temperature and relative humidity [10]. Good acoustic attenuation and fire resistance capabilities are additional advantages of this class of materials [11]. However, there are still many challenges that have to be addressed before the potential of earth materials may be fully exploited for modern buildings in industrialised or economically developed countries. In addition to the high labour costs and the lack of standards and codes for characterisation and design, the use of earth in construction has been restricted by limitations inherent to the material, which suffers from drying shrinkage and high susceptibility to water erosion, as well as from low tensile and flexural strength, poor ductility and limited fracture resistance [2–4]. The low tensile strength and the brittle behaviour are probably the major issues for structural use of earth-based materials, particularly for buildings and constructions exposed to seismic hazard [12,13]. Vegetal or animal fibres, such as straw, flax, jute, sisal and wool [1,2,14–16], have been often used in the past to improve the mechanical and deformation properties of the soil material for specific applications. While a number of investigations have been conducted to study the effect of reinforcing fibres on the compression behaviour of earthen materials [14,17–20], fewer studies are available on the influence of fibres on the response of the materials under flexural or tensile loadings [20–33]. Bouhicha and coworkers [21] examined the flexural response of prismatic earth samples reinforced with chopped barley straw, using fibre/soil ratios up to 3.5% in weight and fibre lengths ranging between 10 and 60 mm. It was found that the flexural strength increased with fibre fraction and with fibre length, and that higher deflections were achieved at failure in reinforced specimens. The flexural response of different types of clay soils stabilised with polypropylene or wool fibres was investigated in [22–24]. The analyses indicate that the fibrous reinforcement reduces the shrinkage and improves the flexural strength of the soil samples, even though the potential benefits achievable by the addition of fibres depend on both the class of soil and the type of fibres. Similarly, significant increases in strength and deformation at failure were observed in unbaked soil samples subjected to bending loads when reinforced with sisal [25], kenaf [26,27], or banana fibres [28]. Danso and co-workers conducted splitting tensile tests to examine the tensile strength of soil blocks stabilised with different fibres and observed that the samples failed gradually, with multiple fine cracks and substantial deformation at final failure [29]. Improvements in ductility were also reported for soil–cement samples enhanced with flax [30] or sisal fibres [31]. Little attention has been up to now specifically devoted to the characterisation of the toughness and energy absorption properties of fibre reinforced earthen materials. Unnotched and notched earth blocks samples reinforced with straw fibres and subjected to three point bending were investigated in [32] to examine the resistance to crack propagation of the material with increasing crack openings. The improvements in flexural strength and post-fracture
performance achievable by wool fibres in notched earthen samples were examined by the authors in a previous study [33]. It was found that the addition of fibres greatly improves the flexural strength and the energy absorption of the samples after first cracking, and that longer fibres provide better toughness and residual strength properties at large deflections. To date, to the authors’ knowledge, no studies have been conducted to investigate the effect of fibrous reinforcement on the fracture resistance and energy absorption capabilities of earthbased materials under dynamic loading conditions. The poor strength and the brittleness of soil materials make earth constructions particularly vulnerable to the high levels of seismic forces that develop during earthquakes. For this reason, accurate analyses and reliable characterisations of the contribution of the fibres to the enhancement of the fracture resistance and deformation capacity of earth-based materials under high strain rates are strongly required for safe utilisation of these materials in earthquakeprone regions. As a note of caution, it should be however pointed out that while the deformation capacity of the earthen material is critical in controlling the earthquake response of layered or rammed earth constructions, other constructive aspects, such as the quality of joints between earthen blocks and the lack of horizontal reinforcement, play a primary role in the seismic behaviour of adobe or block constructions [34]. This study investigates the improvements in strength, crack resistance, post-cracking performance and energy absorbing capability achievable under static and dynamic (impact) loading conditions by the addition of hemp fibres in an unbaked soil material. Notched samples reinforced with different fibre contents and fibre lenghts were tested in a three point bending configuration under both quasi-static and low-velocity impact loading conditions. The results obtained during the experimental analyses are illustrated and discussed in the paper to compare the mechanical response and the damage resistance of the various materials at low and high strain rates. 2. Materials and experimental methods 2.1. Materials The soil used in this study was extracted from quaternary sediments in the area of the Tirso River, near Oristano (Sardinia, Italy). The particle-size distribution of the soil, illustrated in the graph of Fig. 1, was characterised by sieve and hydrometer analyses carried out in accordance with ASTM standard D422 [35]. The liquid and plastic limits of the soil are respectively 29% and 17%, and the soil may be classified as CL (Lean Clay), i.e. inorganic clay with low plasticity and low liquid limit, in the Unified Soil Classification System (USCS) [36]. X-ray diffraction (XRD) revealed the quartz as the dominant component, in association with other minerals such as plagioclase and illite. Natural bast fibres collected from the phloem (or bast) of the hemp plant were selected to reinforce the soil material. Hemp fibres are cellulosic in nature and consist of bundles of elementary, single-cell, fibres bonded together by pectin [37]. Technical hemp fibres with an average diameter of 0.2 mm and an average tensile strength of 320 MPa were used in the study.
Fig. 1. Particle-size distribution of the soil.
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119 Table 1 Mixture compositions.
a
111
3. Experimental results and discussion Fractiona (%)
Fibre length (mm)
Water/soil weight ratio
3.1. Static tests
Unreinforced soil
–
–
0.2
Reinforced 2%
2 2 2
10 20 30
0.3 0.3 0.3
Reinforced 3%
3 3 3
10 20 30
0.3 0.3 0.3
Fig. 4 presents typical load–deflection curves obtained during static bending of unreinforced and hemp-reinforced samples. The plots of Fig. 4 clearly show that the addition of fibres greatly enhances the mechanical performance of the soil, leading to large increases in the maximum load carried by the samples and in the residual strength of the material at large deflections. It is also seen that the presence of fibres does not substantially change the initial stiffness of the material, as indicated by the similar slopes of the force deflection curves of unreinforced and reinforced samples in the linear elastic region. The marginal effect of the fibrous reinforcement on the initial behaviour of the material may be explained by the low amount of reinforcement and was already observed in other investigations for different soil and fibre materials [31,33]. Unreinforced and reinforced samples exhibit however totally different mechanical responses after the onset of first cracking phenomena at the notch tip. As soon as a macro crack develops at the notch of unreinforced samples, the fracture propagates quickly towards the upper surface of the beam, leading to the separation of the sample into two halves, with a total loss of loadcarrying capacity (Fig. 5); the abrupt drop to zero of the load, observed at the end of the linear portion of the load–deflection curve, is clearly associated to the unstable growth of the crack in the mid-section of the sample. Samples made of reinforced soil, in contrast, can carry increasing loads after the first cracking occurs at the edge of the notch. As visible in Fig. 6, after the initiation of first cracking (which is typically associated to a slight slope reduction of the force–deflection curve after the linear stage), the fibres continue to transmit a significant amount of load across the cracked surface, thus promoting the development of additional cracking in the soil matrix and giving rise to a damage region characterised by the accumulation of multiple crack paths, as opposed to the single crack scenario occurring in unreinforced soil samples. The effectiveness of the bridging action of fibres is enhanced by the good bond between fibre and soil, which was observed by optical microscopy of the fracture regions of the reinforced samples. The good interfacial adhesion is clearly visible in the micrographs of Fig. 7, which show large amounts of soil matrix around the surface of fibres across the cracked area of a sample tested under static bending. Similar features were observed at the fibre surfaces of samples subjected to impact loading. As shown in the graphs of Fig. 4, the load sustained by the cracked fibre-reinforced samples grows with increasing deflection until reaching a maximum value and then starts decreasing at a slow rate; during the post-peak load stage the samples are still capable of retaining a significant degree of residual strength even at large bending deflections. As an example, from the force–deflection plots reported in Fig. 4 it is seen that the load carried at 15 mm deflection by a sample reinforced with 2% of 20 mm long fibres is more than 20% of the peak load reached during the test. The results of the static bending tests show that the incorporation of fibres in the soil drastically enhances the post-cracking behaviour of the material, with great improvements in ductility, energy dissipation and residual load-carrying capacities at large deformation levels. The superior fracture resistance and load bearing properties of the reinforced soil may be attributed to the introduction of toughening or energy absorbing mechanisms such as fibre bridging across cracked surfaces (as seen in Fig. 6), fracture of the fibre/soil interface and frictional pullout of fibres from the surrounding soil matrix. The major role played by these mechanisms, which can be
Fibre weight fraction is evaluated on the total weight of the mixture.
Soil mixtures with two fibre weight fractions (2% and 3%) and three fibre lengths (10 mm, 20 mm and 30 mm) were prepared to characterise the influence of fibre weight fraction and fibre length on the mechanical properties of the reinforced materials. 2.2. Manufacturing of samples Soil, hemp fibres and distiled water were used to manufacture the samples. The soil was first manually screened to remove inclusions and debris such as rock fragments, leaves, roots and other unwanted organic materials. The resultant material was then crushed using a mortar and a pestle, sieved using a 4 mm mesh, and finally desiccated in a drying oven at 40 Celsius degree for 2 h to eliminate moisture. The hemp fibres were hand combed, cut in three different lengths (10 mm, 20 mm, and 30 mm), and used to prepare six mixture compositions with different fibre length and content. Screened soil, water and fibres were thoroughly mixed by hand in a mortar until a mixture of uniform consistency was obtained. Table 1 summarises the different mix compositions examined in the study. The water content was chosen so as to achieve similar workability for the different mixture compositions. The mixture was finally poured into plastic prismatic moulds and manually compacted with a tamper to form bricks 70 70 160 mm3 in size. All specimens were cured at room temperature for 30 days (for 48 h inside the moulds and for 28 days after demoulding) and subsequently post-cured in an oven at 40 °C for 15 days. A 35 mm deep notch, as visible in Fig. 2, was finally cut at the centre of the samples by a diamond coated circular saw blade. 2.3. Testing procedures The potential of hemp fibres for the enhancement of the fracture performance of the soil material was investigated by subjecting the samples to static and impact flexural loadings. The experimental analyses were specifically aimed at characterising the influence of fibre content and fibre length on the crack control capabilities and energy absorption properties imparted by the reinforcement at low and high strain rates. The same three-point bending configuration, with a span length (distance between the supports) of 140 mm, was used to load the specimens during both static and impact tests. The centre load was imposed through a steel solid cylinder with a diameter of 10 mm. Static tests were carried out using a closed-loop servo-electric 5 kN universal testing machine operating in displacement control at a rate of 1 mm/min. The mid-span deflection of the sample was determined by acquiring simultaneously the signal of the built-in displacement transducer of the testing machine (which measures the deflection at the loaded side of the beam) and that of a linear variable differential transducer (LVDT) positioned at the bottom side of the beam in the vicinity of the notch. Since no significant differences were observed between the two measured values, only the deflections recorded on the upper surface of the samples were considered in the analyses. Static tests were terminated when the mid-span deflection exceeded 16 mm. Impact tests were conducted using an instrumented drop-weight impact testing machine (Fig. 3), equipped with a pneumatically assisted braking system to avoid multiple strikes in the event of a rebound of the impactor. During tests, a 4.83 kg impactor, provided with a loading cylinder analogous to that used for static tests, was placed at the required height and dropped onto the sample. All samples were impacted with an energy of 50 J, which corresponds to a drop height of approximately 1.05 m and to an impact velocity slightly higher than 4.5 m/s. The contact force between the loading cylinder and the sample was measured by a piezoelectric force transducer mounted in line with the impactor axis. An infra-red sensor was used to obtain the velocity of the impactor just before contact with the sample. The histories of the displacement and of the kinetic energy of the impactor were obtained as a function of time by integration of the signal acquired by the force transducer. A minimum number of five samples were tested for any (static and impact) loading condition and for any of the soil mixtures examined in the study.
112
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119
Fig. 2. Geometry of samples (a) and view of a reinforced sample before testing (b).
Fig. 3. Instrumented drop weight testing machine (a) and schematic layout of the testing setup (b).
Fig. 4. Typical load–deflection curves of unreinforced and reinforced samples subjected to static bending. The curve of the unreinforced sample is shifted for clarity.
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119
113
Fig. 5. Initiation and propagation of cracks from the notch of an unreinforced sample.
Fig. 6. Initiation and propagation of a crack from the notch of a reinforced sample (fiber content = 3%; fiber length = 10 mm).
Fig. 7. Optical micrographs of surfaces of hemp fibers at the fracture region of reinforced samples subjected to static bending.
only activated when macro-cracks are sufficiently open, is also suggested by the fact that better post-cracking properties are recorded in samples reinforced with higher fibre content and longer fibres, as evident from the force–deflection plots of Fig. 4.
It is immediately seen, for example, that the load carried by the samples at specific deflection levels increases with increasing both fibre fraction and fibre length. To better characterise this dependence over the whole deformation range, two deflection-based
114
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119
indicators of the post-cracking response of the material (namely, the energy absorption and the load carrying capacity of the samples [33,38]) were evaluated and compared at predetermined mid-span displacements. The energy absorption of the samples was calculated as the area under the force–displacement curve up to the selected deflection, while the load carrying capacity was evaluated as the load sustained by the sample at the chosen deflection level. The diagrams of Figs. 8 and 9 illustrate the effect of fibre content and fibre length on the energy absorption and residual strength properties of the samples at 5, 10 and 15 mm deflections. The data reported in the graphs indicate that for small deflections (5 mm), which correspond to a situation where cracks are not sufficiently wide to mobilise the fibre-based mechanisms responsible for crack control, the post-cracking performance of the samples is not influenced by the amount of fibres and by their length. In contrast, the indicators evaluated at 10 mm and 15 mm deflections clearly show that higher fibre contents and longer fibres correspond, for large deflections, to better fracture resistance and energy absorbing properties of the enhanced material. 3.2. Impact tests The fracture resistance and the influence of fibre content and fibre length on the post-cracking behaviour of the material was further investigated at high strain rates by three point bending tests carried out with the instrumented drop weight machine shown in Fig. 3. The samples were placed on a support anvil with the same span length used during static tests (140 mm) and then impacted with an energy of 50 J by a mass provided with a cylindrical loading head. Because of the high levels of accelerations experienced by the sample during the impact (the typical duration of an impact ranged
from about 2 ms for unreinforced samples to a maximum of 15 ms for reinforced samples), the signal recorded by the force transducer mounted on the impactor does not accurately represent the actual bending load acting on the sample, since part of the measured force is balanced by the inertial forces required to accelerate the material from rest [39,40]. An effective approach to account for the presence of inertial effects and derive the true stressing load sustained by the sample was proposed by Banthia and coworkers [41]. Following this approach, the true impact force for bending was obtained in this study by subtracting the generalised inertial force from the force measured at the impactor, where the generalised inertial force was calculated as a function of the acceleration at the mid-span, assuming a linear distribution of the acceleration along the sample. The acceleration at mid-span was measured by a piezoelectric accelerometer that was attached to the bottom surface of the sample in the vicinity of the notch (see Fig. 3). Fig. 10 shows typical true force versus deflection curves obtained during impact tests for samples with the different mixture compositions under investigation. As seen in the graphs, the response to impact of unreinforced soil samples is characterised by a steep load increase up to a peak value of about 4 kN, after which the load drops abruptly to zero, as a result of the brittle fragmentation of the sample following the unstable propagation of a crack from the notch tip. Compared to plain soil samples, reinforced samples exhibit higher peak loads – even though the gain is not as large as that measured during static tests- and show a more gradual load decrease after reaching the maximum load, with remarkable post-cracking strength and energy absorption properties at large deflections. Similar to the static case, the residual strength and the fracture energy of reinforced samples increase with increasing fibre content
Fig. 8. Average value of energy absorbed during static bending at 5, 10 and 15 mm deflections for samples reinforced with 2% (a) and 3% (b) fiber content. Error bars indicate plus and minus one standard deviation.
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119
115
Fig. 9. Average value of load-carrying capacity measured during static bending at 5, 10 and 15 mm deflections for samples reinforced with 2% (a) and 3% (b) fiber content. Error bars indicate plus and minus one standard deviation.
Fig. 10. Typical load–deflection curves of unreinforced and reinforced samples subjected to impact bending (impact energy = 50 J).
and fibre length, as evident from the comparison of the curves of Fig. 10. The strong dependence of the fracture resistance of the material on the amount and length of fibres is also suggested by the simple inspection of the fractured appearance of impacted samples (Fig. 11), which are always characterised by smaller
residual deformations and crack openings for higher fibre contents and larger fibre lengths. With specific reference to the energy absorption capacity, the average fracture energy required to achieve complete separation of the unreinforced samples (as assessed by evaluating the area
116
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119
Fig. 11. Appearance of reinforced samples after impact testing.
Fig. 12. Average value of energy absorbed during impact bending at 5, 10 and 15 mm deflections for samples reinforced with 2% (a) and 3% (b) fiber content. Error bars indicate plus and minus one standard deviation.
under the force–displacement curves) was found to be about 15 J, and thus much smaller than the total 50 J energy imparted to the samples. By contrast, the impact energy of 50 J was not sufficiently high to induce the complete failure of the reinforced samples, as evident from the typical appearance of the impacted samples shown in Fig. 11. In this regard, it should be pointed out that the load decrease visible in the final region of the force–deflection curves of reinforced samples corresponds to the unloading occurring during the rebound of the impactor, and not to the brittle failure of the sample, as in the case of the unreinforced material. The graphs of Fig. 12 report the average values of the energy absorbed by reinforced samples under impact at selected deflections. The experimental results show that, for small deflections, increases of the fibre content or of the fibre length do not lead to
improvements in the energy absorption properties of the materials (Fig. 12; 5 mm deflection); at large deformations, in contrast, the energy absorbed during impact increases with increasing fibre weight fraction and, for a given fibre weight fraction, with increasing fibre length (Fig. 12; 10 and 15 mm deflections). This trend is analogous to that observed under static bending (see Fig. 8), and indicates that the fracture response of the fibre-reinforced materials at high strain rates is again governed by key toughening mechanisms, such as fibre bridging and fibre pullout, which can only be mobilised for sufficiently large crack opening displacements. The influence of fibre content and fibre length at large deformations is also evident from the analysis of the plots of Fig. 13, which report the loads carried by samples with the different mixture compositions at 10 and 15 mm deflections.
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119
117
Fig. 13. Average value of load-carrying capacity measured during impact bending at 10 and 15 mm deflections for samples reinforced with 2% (a) and 3% (b) fiber content. Error bars indicate plus and minus one standard deviation.
Fig. 14. Typical load–deflection curves of reinforced samples subjected to static bending after being subjected to a 50 J impact.
If we compare the impact and the static performances of the samples, we notice that the peak load reached by unreinforced samples during impact tests is much higher than that recorded during quasistatic loading, thus indicating a strong sensitivity of the plain soil
material to the load rate. Similarly, a significant effect of the strain rate may be observed in the structural response of the reinforced materials, as evident by a comparison of the force deflection traces reported in Figs. 4 and 10. The analysis of the static and dynamic
118
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119
curves shows that the reinforced samples, irrespectively of fibre fraction and fibre length, exhibit higher peak loads, superior residual strengths at large deflections, and larger energy dissipations under impact loading than under static conditions. As an example, a comparison of the static and impact responses of samples reinforced with 3% of 30 mm fibres indicates that the peak load and the residual strength at 15 mm deflection measured during impact bending are, respectively, almost two times and three times higher than those attained during static bending. To further explore the efficiency of the reinforcement, samples subjected to 50 J impact were finally tested under static bending to compare the residual properties of the different materials in terms of load-carrying and energy dissipation capacities. Typical static force–deflection curves obtained on impacted samples with different mixture compositions are shown in Fig. 14. The plots show that the samples were generally capable of providing some residual load carrying capacity after the impact and that, as expected, higher fibre contents and longer fibres correspond to much better post–impact structural performances. These results indicate that remarkable levels of residual properties, such as strength, fracture resistance, energy dissipation, capacity to sustain large deflections, can be achieved after dynamic loading events in soil materials with the appropriate selection of the fibrous reinforcement.
4. Conclusions Notched earth-based samples made of a clayey soil reinforced with different amounts of hemp fibres (2% and 3% in weight; 10, 20 and 30 mm in length) were manufactured and tested under both static and impact bending to investigate and compare the influence of the reinforcement on the fracture resistance and energy absorption properties of the material at low and high strain rates. The addition of hemp fibres improves significantly the fracture resistance of the plain soil under both static and impact loadings, with large increases in energy absorption and load carrying capabilities of the samples for large deformation regimes. In particular, it was found that while fibre content and fibre length do not significantly affect the mechanical response of the samples at small deflections, the post-cracking performance at large deflections is enhanced by increasing the fibre fraction and, for a constant fibre content, by increasing the length of fibres. The key mechanisms responsible for the enhancement of the strength and toughness of the soil are the bridging of fibres across cracked surfaces and the debonding with subsequent pullout of fibres from the soil matrix. The beneficial action of these mechanisms may be effectively exploited only for sufficiently open cracks, thus explaining the marginal influence of the values of fibre content and fibre length on the post-cracking properties of the samples at small deflections. Both unreinforced and reinforced earthen materials show a strong sensitivity to the rate of loading, with significantly higher peak loads and energy absorptions under dynamic bending as compared to static bending. Improvements by two times in the peak load and three times in the residual strength at 15 mm deflections were observed for impacted reinforced samples over analogous samples subjected to static bending. Similarly to the static case, increases of the fibre content or of the fibre length lead to improvements in the strength and energy absorption properties of the materials only when sufficiently large deflections are attained during impact. The results of the experimental study presented in the paper show that the addition of hemp fibres is an effective and efficient way to greatly enhance the fracture resistance, the ductility and
the load bearing properties of earth-based building materials under both static and impact loading conditions. Further analyses are however required to better understand the role and the relative importance of the toughening and strengthening mechanisms introduced by the incorporation of the fibrous reinforcement in the base soil material.
Acknowledgement Research funding was provided by the Sardinian Regional Government (LR N.7 7-8-2007, Tender 3, 2011).
References [1] G. Minke, Building with Earth. Design and Technology of a Sustainable Architecture, Birkhäuser, Basel, 2006. [2] E. Avrami, H. Guillaud, M. Hardy (Eds.), Terra Literature Review. An Overview of Research in Earthen Architecture Conservation, The Getty Conservation Institute, Los Angeles, 2008. [3] M. Calkins, Materials for Sustainable Sites, Wiley, Hoboken, 2009. [4] F. Pacheco-Torgal, S. Jalali, Earth construction: lessons from the past for future eco-efficient construction, Constr. Build. Mater. 29 (2012) 512–519. [5] C. Mileto, F. Vegas, V. Cristini (Eds.), Rammed Earth Conservation, CRC Press, Boca Raton, 2012. [6] Torgal F. Pacheco, S. Jalali, Eco-efficient Construction and Building Materials, Springer, London, 2011. [7] B. Berge, The Ecology of Building Materials, Architectural Press, Oxford, 2000. [8] M.S. Zami, A. Lee, Economic benefits of contemporary earth construction in low-cost urban housing – state-of-the-art review, J. Build. Appraisal 5 (2010) 259–271. [9] B.V. Venkatarama Reddy, G. Leuzinger, V.S. Sreeram, Low embodied energy cement stabilised rammed earth building—a case study, Energy Build. 68 (2014) 541–546. [10] M.R. Hall, S. Casey, Hygrothermal behaviour and occupant comfort in modern earth buildings, in: M.R Hall, R. Lindsay, M. Krayenhoff (Eds.), Modern Earth Buildings. Materials, Engineering, Construction and Applications, Woodhead, Oxford, 2012, pp. 17–40. [11] C.J. Hopfe, M.R. Hall, Fabric insulation, thermal bridging and acoustics in modern earth buildings, in: M.R. Hall, R. Lindsay, M. Krayenhoff (Eds.), Modern Earth Buildings. Materials, Engineering, Construction and Applications, Woodhead, Oxford, 2012, pp. 41–71. [12] Tolles E. Leroy, E.E. Kimbro, F.A. Webster, W.S. Ginell, Seismic Stabilization of Historic Adobe Structures, The Getty Conservation Institute, Los Angeles, 2000. [13] M. Blondet, Mitigation of seismic risk on earthen buildings, in: H. Gökçekusß, U. Türker, J.W. LaMoreaux (Eds.), Survival and Sustainability, Environmental Earth Sciences, Springer-Verlag, Berlin, 2011, pp. 391–400. [14] E. Quagliarini, S. Lenci, The influence of natural stabilizers and natural fibres on the mechanical properties of ancient Roman adobe bricks, J. Cult. Heritage 11 (2010) 309–314. [15] D. Silveira, H. Varum, A. Costa, T. Martins, H. Pereira, J. Almeida, Mechanical properties of adobe bricks in ancient constructions, Constr. Build. Mater. 28 (1) (2012) 36–44. [16] F. Parisi, D. Asprone, L. Fenu, A. Prota, Experimental characterization of Italian composite adobe bricks reinforced with straw fibers, Compos. Struct. 122 (2015) 300–307. [17] K. Ghavami, R.D. Toledo Filho, N.P. Barbosa, Behaviour of composite soil reinforced with natural fibres, Cem. Concr. Compos. 21 (1999) 39–48. [18] H. Binici, O. Aksogan, T. Shah, Investigation of fibre reinforced mud brick as a building material, Constr. Build. Mater. 19 (4) (2005) 313–318. [19] Q. Piattoni, E. Quagliarini, S. Lenci, Experimental analysis and modelling of the mechanical behaviour of earthen bricks, Constr. Build. Mater. 25 (4) (2011) 2067–2075. [20] H. Danso, D.B. Martinson, M. Ali, C. Mant, Performance characteristics of enhanced soil blocks: a quantitative review, Build. Res. Inform. 43 (2) (2015) 253–262. [21] M. Bouhicha, F. Aouissi, S. Kenai, Performance of composite soil reinforced with barley straw, Cem. Concr. Compos. 27 (2005) 617–621. [22] C. Galán-Marín, C. Rivera-Gómez, J. Petric, Clay-based composite stabilized with natural polymer and fibre, Constr. Build. Mater. 24 (8) (2010) 1462–1468. [23] C. Galán-Marín, C. Rivera-Gómez, F. Bradley, The mechanical properties and molecular bonding characteristics of clay-based natural composites reinforced with animal fibres, J. Biobased Mater. Bioenergy 7 (1) (2013) 143–151. [24] C. Rivera-Gómez, C. Galán-Marín, F. Bradley, Analysis of the influence of the fiber type in polymer matrix/fiber bond using natural organic polymer stabilizer, Polymers 6 (2014) 977–994. [25] R. Mattone, Sisal fibre reinforced soil with cement or cactus pulp in bahareque technique, Cem. Concr. Compos. 27 (2005) 611–616. [26] Y. Millogo, J.C. Morel, J.E. Aubert, K. Ghavami, Experimental analysis of pressed adobe blocks reinforced with Hibiscus cannabinus fibers, Constr. Build. Mater. 52 (2014) 71–78.
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119 [27] Y. Millogo, J.E. Aubert, E. Hamard, J.C. Morel, How properties of kenaf fibers from burkina faso contribute to the reinforcement of earth blocks, Materials 8 (2015) 2332–2345. [28] M. Mostafa, N. Uddin, Effect of banana fibers on the compressive and flexural strength of compressed earth blocks, Buildings 5 (2015) 282–296. [29] H. Danso, D.B. Martinson, M. Ali, J. Williams, Effect of fibre aspect ratio on mechanical properties of soil building blocks, Constr. Build. Mater. 83 (2015) 314–319. [30] M. Segetin, K. Jayaraman, X. Xu, Harakeke reinforcement of soil-cement building materials: manufacturability and properties, Build. Environ. 42 (2007) 3066–3079. [31] A.P.S. Martins, F.A. Andrade Silva, R.D. Toledo Filho, Mechanical behavior of self-compacting soil-cement-sisal fiber composites, Key Eng. Mater. 634 (2015) 421–432. [32] F. Clementi, S. Lenci, T. Sadowski, Fracture characteristics of unfired earth, Int. J. Fract. 149 (2008) 193–198. [33] F. Aymerich, L. Fenu, P. Meloni, Effect of reinforcing wool fibres on fracture and energy absorption properties of an earthen material, Constr. Build. Mater. 27 (1) (2012) 66–72.
119
[34] A.S. Arya, T. Boen, Y. Ishiyama, Guidelines for Earthquake Resistant NonEngineered Construction, UNESCO, Paris, 2014. [35] ASTM Standard D422-63, Standard Test Method for Particle-size Analysis of Soils, ASTM International, West Conshohocken (PA), 2007, http://dx.doi.org/ 10.1520/D0422-63R07.. [36] ASTM Standard D2487-11, Standard Practice for Classification of Soils for Engineering Purposes (Unified Soil Classification System), ASTM International, West Conshohocken (PA), 2011, http://dx.doi.org/10.1520/D2487-11. . [37] A.K. Mohanty, M. Misra, L.T. Drzal, Natural Fibers, Biopolymers, and Biocomposites, CRC Press, Boca Raton, 2005. [38] V.S. Gopalaratnam, R. Gettu, On the characterization of flexural toughness in fiber reinforced concretes, Cem. Concr. Compos. 17 (1995) 239–254. [39] N. Wang, S. Mindess, K. Ko, Fibre reinforced concrete beams under impact loading, Cem. Concr. Res. 26 (3) (1996) 363–376. [40] S.N. Soleimani, N. Banthia, A novel drop weight impact setup for testing reinforced concrete beams, Exp. Tech. 38 (3) (2014) 72–79. [41] N. Banthia, S. Mindess, A. Bentur, M. Pigeon, Impact testing of concrete using a drop-weight impact machine, Exp. Mech. 29 (1) (1989) 63–69.
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Fracture behaviour of a fibre reinforced earthen material under static and impact flexural loading F. Aymerich a, L. Fenu b, L. Francesconi a,⇑, P. Meloni a a b
Department of Mechanical, Chemical and Materials Engineering, University of Cagliari, Italy Department of Civil and Environmental Engineering and Architecture, University of Cagliari, Italy
h i g h l i g h t s We studied the effect of hemp fibres on the mechanical performance of an earthen material. The samples were subjected to static and impact bending. Hemp fibres greatly increase the fracture resistance and the energy absorption capacity. Both unreinforced and reinforced materials show a strong sensitivity to the rate of loading. The post-cracking performance is improved by increasing the fibre fraction and length.
a r t i c l e
i n f o
Article history: Received 28 June 2015 Received in revised form 27 December 2015 Accepted 27 January 2016 Available online 6 February 2016 Keywords: Earthen materials Hemp fibres Impact Fracture Toughness
a b s t r a c t The study investigates the enhancements in the load carrying capacity, crack resistance and energy absorption properties provided by the addition of hemp fibres in an earthen material. Notched earthen samples reinforced with two fibre contents (2% and 3% in weight) and three fibre lengths (10, 20, and 30 mm) were manufactured and tested under static and impact bending to investigate and compare the influence of the reinforcement on the fracture resistance of the soil material at low and high strain rates. The results of the experimental analyses show that the incorporation of fibres greatly improves the peak load, the post-crack strength, the ductility and the energy dissipation of soil under both static and impact bending. The mechanical response of both unreinforced and reinforced samples is significantly affected by the rate of loading, with samples exhibiting higher values of strength and absorbed energy under impact than under static bending. For both static and impact loading, the post-crack response of the material at large deformations is clearly improved by increasing the fibre content and, at the same fibre content, by increasing the fibre length. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Earth is the most widely utilised natural construction material in the world and unfired earth is by large the prevalent material used for dwelling structures in many hot, arid or temperate regions, with estimates indicating that 30–50% of the world’s population live in houses made of unbaked earth-based materials [1,2]. Earth construction methods are varied, including techniques such as rammed earth, cob, wattle and daub, brick masonry, with soil or mud bricks (adobes) being probably the most common methods of earth construction worldwide [3,4].
⇑ Corresponding author. E-mail addresses: [email protected] (F. Aymerich), [email protected] (L. Francesconi). http://dx.doi.org/10.1016/j.conbuildmat.2016.01.046 0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
The use of earth as a construction material is a natural solution in developing countries, since the raw materials, typically soil and sand, are abundant and locally available, and can be processed with simple technologies, without requiring expensive tools or specialised manufacturing skills. Even though the majority of earthen architecture is located in less developed countries, the past decade has witnessed a growing interest in the use of earthen building materials in industrialised countries; this renewed interest has been prompted not only by specific requirements for conservation and rehabilitation of architectural heritage sites [5], but also by increasing demands for more sustainable and energyefficient materials and techniques for the building and construction industry [6]. In this respect, as compared to conventional building materials, earth-based materials provide significant reductions in environmental impact, because of the low toxicity and high recyclability properties, the minimal transportation costs
110
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119
(as a consequence of the local availability of raw materials), and the high energy efficiency and ecological compatibility of the manufacturing processes. [6,7]. As an example, it has been estimated that building a square metre of masonry with compressed earth blocks consumes about 15 times less energy than a square metre of fired bricks [8]. Similarly, a recent case study [9] revealed that the embodied energy of a three-storey rammed earth building was about one third than that of an analogous building made of conventional clay fired bricks and between one ninth and one fourth of that of a reinforced concrete framed structure building. Further benefits of earthen building materials are related to their excellent performances in terms of heat storage capacity, thermal inertia and hygrothermal behaviour, which ensure low fluctuations in indoor air temperature and relative humidity [10]. Good acoustic attenuation and fire resistance capabilities are additional advantages of this class of materials [11]. However, there are still many challenges that have to be addressed before the potential of earth materials may be fully exploited for modern buildings in industrialised or economically developed countries. In addition to the high labour costs and the lack of standards and codes for characterisation and design, the use of earth in construction has been restricted by limitations inherent to the material, which suffers from drying shrinkage and high susceptibility to water erosion, as well as from low tensile and flexural strength, poor ductility and limited fracture resistance [2–4]. The low tensile strength and the brittle behaviour are probably the major issues for structural use of earth-based materials, particularly for buildings and constructions exposed to seismic hazard [12,13]. Vegetal or animal fibres, such as straw, flax, jute, sisal and wool [1,2,14–16], have been often used in the past to improve the mechanical and deformation properties of the soil material for specific applications. While a number of investigations have been conducted to study the effect of reinforcing fibres on the compression behaviour of earthen materials [14,17–20], fewer studies are available on the influence of fibres on the response of the materials under flexural or tensile loadings [20–33]. Bouhicha and coworkers [21] examined the flexural response of prismatic earth samples reinforced with chopped barley straw, using fibre/soil ratios up to 3.5% in weight and fibre lengths ranging between 10 and 60 mm. It was found that the flexural strength increased with fibre fraction and with fibre length, and that higher deflections were achieved at failure in reinforced specimens. The flexural response of different types of clay soils stabilised with polypropylene or wool fibres was investigated in [22–24]. The analyses indicate that the fibrous reinforcement reduces the shrinkage and improves the flexural strength of the soil samples, even though the potential benefits achievable by the addition of fibres depend on both the class of soil and the type of fibres. Similarly, significant increases in strength and deformation at failure were observed in unbaked soil samples subjected to bending loads when reinforced with sisal [25], kenaf [26,27], or banana fibres [28]. Danso and co-workers conducted splitting tensile tests to examine the tensile strength of soil blocks stabilised with different fibres and observed that the samples failed gradually, with multiple fine cracks and substantial deformation at final failure [29]. Improvements in ductility were also reported for soil–cement samples enhanced with flax [30] or sisal fibres [31]. Little attention has been up to now specifically devoted to the characterisation of the toughness and energy absorption properties of fibre reinforced earthen materials. Unnotched and notched earth blocks samples reinforced with straw fibres and subjected to three point bending were investigated in [32] to examine the resistance to crack propagation of the material with increasing crack openings. The improvements in flexural strength and post-fracture
performance achievable by wool fibres in notched earthen samples were examined by the authors in a previous study [33]. It was found that the addition of fibres greatly improves the flexural strength and the energy absorption of the samples after first cracking, and that longer fibres provide better toughness and residual strength properties at large deflections. To date, to the authors’ knowledge, no studies have been conducted to investigate the effect of fibrous reinforcement on the fracture resistance and energy absorption capabilities of earthbased materials under dynamic loading conditions. The poor strength and the brittleness of soil materials make earth constructions particularly vulnerable to the high levels of seismic forces that develop during earthquakes. For this reason, accurate analyses and reliable characterisations of the contribution of the fibres to the enhancement of the fracture resistance and deformation capacity of earth-based materials under high strain rates are strongly required for safe utilisation of these materials in earthquakeprone regions. As a note of caution, it should be however pointed out that while the deformation capacity of the earthen material is critical in controlling the earthquake response of layered or rammed earth constructions, other constructive aspects, such as the quality of joints between earthen blocks and the lack of horizontal reinforcement, play a primary role in the seismic behaviour of adobe or block constructions [34]. This study investigates the improvements in strength, crack resistance, post-cracking performance and energy absorbing capability achievable under static and dynamic (impact) loading conditions by the addition of hemp fibres in an unbaked soil material. Notched samples reinforced with different fibre contents and fibre lenghts were tested in a three point bending configuration under both quasi-static and low-velocity impact loading conditions. The results obtained during the experimental analyses are illustrated and discussed in the paper to compare the mechanical response and the damage resistance of the various materials at low and high strain rates. 2. Materials and experimental methods 2.1. Materials The soil used in this study was extracted from quaternary sediments in the area of the Tirso River, near Oristano (Sardinia, Italy). The particle-size distribution of the soil, illustrated in the graph of Fig. 1, was characterised by sieve and hydrometer analyses carried out in accordance with ASTM standard D422 [35]. The liquid and plastic limits of the soil are respectively 29% and 17%, and the soil may be classified as CL (Lean Clay), i.e. inorganic clay with low plasticity and low liquid limit, in the Unified Soil Classification System (USCS) [36]. X-ray diffraction (XRD) revealed the quartz as the dominant component, in association with other minerals such as plagioclase and illite. Natural bast fibres collected from the phloem (or bast) of the hemp plant were selected to reinforce the soil material. Hemp fibres are cellulosic in nature and consist of bundles of elementary, single-cell, fibres bonded together by pectin [37]. Technical hemp fibres with an average diameter of 0.2 mm and an average tensile strength of 320 MPa were used in the study.
Fig. 1. Particle-size distribution of the soil.
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119 Table 1 Mixture compositions.
a
111
3. Experimental results and discussion Fractiona (%)
Fibre length (mm)
Water/soil weight ratio
3.1. Static tests
Unreinforced soil
–
–
0.2
Reinforced 2%
2 2 2
10 20 30
0.3 0.3 0.3
Reinforced 3%
3 3 3
10 20 30
0.3 0.3 0.3
Fig. 4 presents typical load–deflection curves obtained during static bending of unreinforced and hemp-reinforced samples. The plots of Fig. 4 clearly show that the addition of fibres greatly enhances the mechanical performance of the soil, leading to large increases in the maximum load carried by the samples and in the residual strength of the material at large deflections. It is also seen that the presence of fibres does not substantially change the initial stiffness of the material, as indicated by the similar slopes of the force deflection curves of unreinforced and reinforced samples in the linear elastic region. The marginal effect of the fibrous reinforcement on the initial behaviour of the material may be explained by the low amount of reinforcement and was already observed in other investigations for different soil and fibre materials [31,33]. Unreinforced and reinforced samples exhibit however totally different mechanical responses after the onset of first cracking phenomena at the notch tip. As soon as a macro crack develops at the notch of unreinforced samples, the fracture propagates quickly towards the upper surface of the beam, leading to the separation of the sample into two halves, with a total loss of loadcarrying capacity (Fig. 5); the abrupt drop to zero of the load, observed at the end of the linear portion of the load–deflection curve, is clearly associated to the unstable growth of the crack in the mid-section of the sample. Samples made of reinforced soil, in contrast, can carry increasing loads after the first cracking occurs at the edge of the notch. As visible in Fig. 6, after the initiation of first cracking (which is typically associated to a slight slope reduction of the force–deflection curve after the linear stage), the fibres continue to transmit a significant amount of load across the cracked surface, thus promoting the development of additional cracking in the soil matrix and giving rise to a damage region characterised by the accumulation of multiple crack paths, as opposed to the single crack scenario occurring in unreinforced soil samples. The effectiveness of the bridging action of fibres is enhanced by the good bond between fibre and soil, which was observed by optical microscopy of the fracture regions of the reinforced samples. The good interfacial adhesion is clearly visible in the micrographs of Fig. 7, which show large amounts of soil matrix around the surface of fibres across the cracked area of a sample tested under static bending. Similar features were observed at the fibre surfaces of samples subjected to impact loading. As shown in the graphs of Fig. 4, the load sustained by the cracked fibre-reinforced samples grows with increasing deflection until reaching a maximum value and then starts decreasing at a slow rate; during the post-peak load stage the samples are still capable of retaining a significant degree of residual strength even at large bending deflections. As an example, from the force–deflection plots reported in Fig. 4 it is seen that the load carried at 15 mm deflection by a sample reinforced with 2% of 20 mm long fibres is more than 20% of the peak load reached during the test. The results of the static bending tests show that the incorporation of fibres in the soil drastically enhances the post-cracking behaviour of the material, with great improvements in ductility, energy dissipation and residual load-carrying capacities at large deformation levels. The superior fracture resistance and load bearing properties of the reinforced soil may be attributed to the introduction of toughening or energy absorbing mechanisms such as fibre bridging across cracked surfaces (as seen in Fig. 6), fracture of the fibre/soil interface and frictional pullout of fibres from the surrounding soil matrix. The major role played by these mechanisms, which can be
Fibre weight fraction is evaluated on the total weight of the mixture.
Soil mixtures with two fibre weight fractions (2% and 3%) and three fibre lengths (10 mm, 20 mm and 30 mm) were prepared to characterise the influence of fibre weight fraction and fibre length on the mechanical properties of the reinforced materials. 2.2. Manufacturing of samples Soil, hemp fibres and distiled water were used to manufacture the samples. The soil was first manually screened to remove inclusions and debris such as rock fragments, leaves, roots and other unwanted organic materials. The resultant material was then crushed using a mortar and a pestle, sieved using a 4 mm mesh, and finally desiccated in a drying oven at 40 Celsius degree for 2 h to eliminate moisture. The hemp fibres were hand combed, cut in three different lengths (10 mm, 20 mm, and 30 mm), and used to prepare six mixture compositions with different fibre length and content. Screened soil, water and fibres were thoroughly mixed by hand in a mortar until a mixture of uniform consistency was obtained. Table 1 summarises the different mix compositions examined in the study. The water content was chosen so as to achieve similar workability for the different mixture compositions. The mixture was finally poured into plastic prismatic moulds and manually compacted with a tamper to form bricks 70 70 160 mm3 in size. All specimens were cured at room temperature for 30 days (for 48 h inside the moulds and for 28 days after demoulding) and subsequently post-cured in an oven at 40 °C for 15 days. A 35 mm deep notch, as visible in Fig. 2, was finally cut at the centre of the samples by a diamond coated circular saw blade. 2.3. Testing procedures The potential of hemp fibres for the enhancement of the fracture performance of the soil material was investigated by subjecting the samples to static and impact flexural loadings. The experimental analyses were specifically aimed at characterising the influence of fibre content and fibre length on the crack control capabilities and energy absorption properties imparted by the reinforcement at low and high strain rates. The same three-point bending configuration, with a span length (distance between the supports) of 140 mm, was used to load the specimens during both static and impact tests. The centre load was imposed through a steel solid cylinder with a diameter of 10 mm. Static tests were carried out using a closed-loop servo-electric 5 kN universal testing machine operating in displacement control at a rate of 1 mm/min. The mid-span deflection of the sample was determined by acquiring simultaneously the signal of the built-in displacement transducer of the testing machine (which measures the deflection at the loaded side of the beam) and that of a linear variable differential transducer (LVDT) positioned at the bottom side of the beam in the vicinity of the notch. Since no significant differences were observed between the two measured values, only the deflections recorded on the upper surface of the samples were considered in the analyses. Static tests were terminated when the mid-span deflection exceeded 16 mm. Impact tests were conducted using an instrumented drop-weight impact testing machine (Fig. 3), equipped with a pneumatically assisted braking system to avoid multiple strikes in the event of a rebound of the impactor. During tests, a 4.83 kg impactor, provided with a loading cylinder analogous to that used for static tests, was placed at the required height and dropped onto the sample. All samples were impacted with an energy of 50 J, which corresponds to a drop height of approximately 1.05 m and to an impact velocity slightly higher than 4.5 m/s. The contact force between the loading cylinder and the sample was measured by a piezoelectric force transducer mounted in line with the impactor axis. An infra-red sensor was used to obtain the velocity of the impactor just before contact with the sample. The histories of the displacement and of the kinetic energy of the impactor were obtained as a function of time by integration of the signal acquired by the force transducer. A minimum number of five samples were tested for any (static and impact) loading condition and for any of the soil mixtures examined in the study.
112
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119
Fig. 2. Geometry of samples (a) and view of a reinforced sample before testing (b).
Fig. 3. Instrumented drop weight testing machine (a) and schematic layout of the testing setup (b).
Fig. 4. Typical load–deflection curves of unreinforced and reinforced samples subjected to static bending. The curve of the unreinforced sample is shifted for clarity.
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119
113
Fig. 5. Initiation and propagation of cracks from the notch of an unreinforced sample.
Fig. 6. Initiation and propagation of a crack from the notch of a reinforced sample (fiber content = 3%; fiber length = 10 mm).
Fig. 7. Optical micrographs of surfaces of hemp fibers at the fracture region of reinforced samples subjected to static bending.
only activated when macro-cracks are sufficiently open, is also suggested by the fact that better post-cracking properties are recorded in samples reinforced with higher fibre content and longer fibres, as evident from the force–deflection plots of Fig. 4.
It is immediately seen, for example, that the load carried by the samples at specific deflection levels increases with increasing both fibre fraction and fibre length. To better characterise this dependence over the whole deformation range, two deflection-based
114
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119
indicators of the post-cracking response of the material (namely, the energy absorption and the load carrying capacity of the samples [33,38]) were evaluated and compared at predetermined mid-span displacements. The energy absorption of the samples was calculated as the area under the force–displacement curve up to the selected deflection, while the load carrying capacity was evaluated as the load sustained by the sample at the chosen deflection level. The diagrams of Figs. 8 and 9 illustrate the effect of fibre content and fibre length on the energy absorption and residual strength properties of the samples at 5, 10 and 15 mm deflections. The data reported in the graphs indicate that for small deflections (5 mm), which correspond to a situation where cracks are not sufficiently wide to mobilise the fibre-based mechanisms responsible for crack control, the post-cracking performance of the samples is not influenced by the amount of fibres and by their length. In contrast, the indicators evaluated at 10 mm and 15 mm deflections clearly show that higher fibre contents and longer fibres correspond, for large deflections, to better fracture resistance and energy absorbing properties of the enhanced material. 3.2. Impact tests The fracture resistance and the influence of fibre content and fibre length on the post-cracking behaviour of the material was further investigated at high strain rates by three point bending tests carried out with the instrumented drop weight machine shown in Fig. 3. The samples were placed on a support anvil with the same span length used during static tests (140 mm) and then impacted with an energy of 50 J by a mass provided with a cylindrical loading head. Because of the high levels of accelerations experienced by the sample during the impact (the typical duration of an impact ranged
from about 2 ms for unreinforced samples to a maximum of 15 ms for reinforced samples), the signal recorded by the force transducer mounted on the impactor does not accurately represent the actual bending load acting on the sample, since part of the measured force is balanced by the inertial forces required to accelerate the material from rest [39,40]. An effective approach to account for the presence of inertial effects and derive the true stressing load sustained by the sample was proposed by Banthia and coworkers [41]. Following this approach, the true impact force for bending was obtained in this study by subtracting the generalised inertial force from the force measured at the impactor, where the generalised inertial force was calculated as a function of the acceleration at the mid-span, assuming a linear distribution of the acceleration along the sample. The acceleration at mid-span was measured by a piezoelectric accelerometer that was attached to the bottom surface of the sample in the vicinity of the notch (see Fig. 3). Fig. 10 shows typical true force versus deflection curves obtained during impact tests for samples with the different mixture compositions under investigation. As seen in the graphs, the response to impact of unreinforced soil samples is characterised by a steep load increase up to a peak value of about 4 kN, after which the load drops abruptly to zero, as a result of the brittle fragmentation of the sample following the unstable propagation of a crack from the notch tip. Compared to plain soil samples, reinforced samples exhibit higher peak loads – even though the gain is not as large as that measured during static tests- and show a more gradual load decrease after reaching the maximum load, with remarkable post-cracking strength and energy absorption properties at large deflections. Similar to the static case, the residual strength and the fracture energy of reinforced samples increase with increasing fibre content
Fig. 8. Average value of energy absorbed during static bending at 5, 10 and 15 mm deflections for samples reinforced with 2% (a) and 3% (b) fiber content. Error bars indicate plus and minus one standard deviation.
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119
115
Fig. 9. Average value of load-carrying capacity measured during static bending at 5, 10 and 15 mm deflections for samples reinforced with 2% (a) and 3% (b) fiber content. Error bars indicate plus and minus one standard deviation.
Fig. 10. Typical load–deflection curves of unreinforced and reinforced samples subjected to impact bending (impact energy = 50 J).
and fibre length, as evident from the comparison of the curves of Fig. 10. The strong dependence of the fracture resistance of the material on the amount and length of fibres is also suggested by the simple inspection of the fractured appearance of impacted samples (Fig. 11), which are always characterised by smaller
residual deformations and crack openings for higher fibre contents and larger fibre lengths. With specific reference to the energy absorption capacity, the average fracture energy required to achieve complete separation of the unreinforced samples (as assessed by evaluating the area
116
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119
Fig. 11. Appearance of reinforced samples after impact testing.
Fig. 12. Average value of energy absorbed during impact bending at 5, 10 and 15 mm deflections for samples reinforced with 2% (a) and 3% (b) fiber content. Error bars indicate plus and minus one standard deviation.
under the force–displacement curves) was found to be about 15 J, and thus much smaller than the total 50 J energy imparted to the samples. By contrast, the impact energy of 50 J was not sufficiently high to induce the complete failure of the reinforced samples, as evident from the typical appearance of the impacted samples shown in Fig. 11. In this regard, it should be pointed out that the load decrease visible in the final region of the force–deflection curves of reinforced samples corresponds to the unloading occurring during the rebound of the impactor, and not to the brittle failure of the sample, as in the case of the unreinforced material. The graphs of Fig. 12 report the average values of the energy absorbed by reinforced samples under impact at selected deflections. The experimental results show that, for small deflections, increases of the fibre content or of the fibre length do not lead to
improvements in the energy absorption properties of the materials (Fig. 12; 5 mm deflection); at large deformations, in contrast, the energy absorbed during impact increases with increasing fibre weight fraction and, for a given fibre weight fraction, with increasing fibre length (Fig. 12; 10 and 15 mm deflections). This trend is analogous to that observed under static bending (see Fig. 8), and indicates that the fracture response of the fibre-reinforced materials at high strain rates is again governed by key toughening mechanisms, such as fibre bridging and fibre pullout, which can only be mobilised for sufficiently large crack opening displacements. The influence of fibre content and fibre length at large deformations is also evident from the analysis of the plots of Fig. 13, which report the loads carried by samples with the different mixture compositions at 10 and 15 mm deflections.
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119
117
Fig. 13. Average value of load-carrying capacity measured during impact bending at 10 and 15 mm deflections for samples reinforced with 2% (a) and 3% (b) fiber content. Error bars indicate plus and minus one standard deviation.
Fig. 14. Typical load–deflection curves of reinforced samples subjected to static bending after being subjected to a 50 J impact.
If we compare the impact and the static performances of the samples, we notice that the peak load reached by unreinforced samples during impact tests is much higher than that recorded during quasistatic loading, thus indicating a strong sensitivity of the plain soil
material to the load rate. Similarly, a significant effect of the strain rate may be observed in the structural response of the reinforced materials, as evident by a comparison of the force deflection traces reported in Figs. 4 and 10. The analysis of the static and dynamic
118
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119
curves shows that the reinforced samples, irrespectively of fibre fraction and fibre length, exhibit higher peak loads, superior residual strengths at large deflections, and larger energy dissipations under impact loading than under static conditions. As an example, a comparison of the static and impact responses of samples reinforced with 3% of 30 mm fibres indicates that the peak load and the residual strength at 15 mm deflection measured during impact bending are, respectively, almost two times and three times higher than those attained during static bending. To further explore the efficiency of the reinforcement, samples subjected to 50 J impact were finally tested under static bending to compare the residual properties of the different materials in terms of load-carrying and energy dissipation capacities. Typical static force–deflection curves obtained on impacted samples with different mixture compositions are shown in Fig. 14. The plots show that the samples were generally capable of providing some residual load carrying capacity after the impact and that, as expected, higher fibre contents and longer fibres correspond to much better post–impact structural performances. These results indicate that remarkable levels of residual properties, such as strength, fracture resistance, energy dissipation, capacity to sustain large deflections, can be achieved after dynamic loading events in soil materials with the appropriate selection of the fibrous reinforcement.
4. Conclusions Notched earth-based samples made of a clayey soil reinforced with different amounts of hemp fibres (2% and 3% in weight; 10, 20 and 30 mm in length) were manufactured and tested under both static and impact bending to investigate and compare the influence of the reinforcement on the fracture resistance and energy absorption properties of the material at low and high strain rates. The addition of hemp fibres improves significantly the fracture resistance of the plain soil under both static and impact loadings, with large increases in energy absorption and load carrying capabilities of the samples for large deformation regimes. In particular, it was found that while fibre content and fibre length do not significantly affect the mechanical response of the samples at small deflections, the post-cracking performance at large deflections is enhanced by increasing the fibre fraction and, for a constant fibre content, by increasing the length of fibres. The key mechanisms responsible for the enhancement of the strength and toughness of the soil are the bridging of fibres across cracked surfaces and the debonding with subsequent pullout of fibres from the soil matrix. The beneficial action of these mechanisms may be effectively exploited only for sufficiently open cracks, thus explaining the marginal influence of the values of fibre content and fibre length on the post-cracking properties of the samples at small deflections. Both unreinforced and reinforced earthen materials show a strong sensitivity to the rate of loading, with significantly higher peak loads and energy absorptions under dynamic bending as compared to static bending. Improvements by two times in the peak load and three times in the residual strength at 15 mm deflections were observed for impacted reinforced samples over analogous samples subjected to static bending. Similarly to the static case, increases of the fibre content or of the fibre length lead to improvements in the strength and energy absorption properties of the materials only when sufficiently large deflections are attained during impact. The results of the experimental study presented in the paper show that the addition of hemp fibres is an effective and efficient way to greatly enhance the fracture resistance, the ductility and
the load bearing properties of earth-based building materials under both static and impact loading conditions. Further analyses are however required to better understand the role and the relative importance of the toughening and strengthening mechanisms introduced by the incorporation of the fibrous reinforcement in the base soil material.
Acknowledgement Research funding was provided by the Sardinian Regional Government (LR N.7 7-8-2007, Tender 3, 2011).
References [1] G. Minke, Building with Earth. Design and Technology of a Sustainable Architecture, Birkhäuser, Basel, 2006. [2] E. Avrami, H. Guillaud, M. Hardy (Eds.), Terra Literature Review. An Overview of Research in Earthen Architecture Conservation, The Getty Conservation Institute, Los Angeles, 2008. [3] M. Calkins, Materials for Sustainable Sites, Wiley, Hoboken, 2009. [4] F. Pacheco-Torgal, S. Jalali, Earth construction: lessons from the past for future eco-efficient construction, Constr. Build. Mater. 29 (2012) 512–519. [5] C. Mileto, F. Vegas, V. Cristini (Eds.), Rammed Earth Conservation, CRC Press, Boca Raton, 2012. [6] Torgal F. Pacheco, S. Jalali, Eco-efficient Construction and Building Materials, Springer, London, 2011. [7] B. Berge, The Ecology of Building Materials, Architectural Press, Oxford, 2000. [8] M.S. Zami, A. Lee, Economic benefits of contemporary earth construction in low-cost urban housing – state-of-the-art review, J. Build. Appraisal 5 (2010) 259–271. [9] B.V. Venkatarama Reddy, G. Leuzinger, V.S. Sreeram, Low embodied energy cement stabilised rammed earth building—a case study, Energy Build. 68 (2014) 541–546. [10] M.R. Hall, S. Casey, Hygrothermal behaviour and occupant comfort in modern earth buildings, in: M.R Hall, R. Lindsay, M. Krayenhoff (Eds.), Modern Earth Buildings. Materials, Engineering, Construction and Applications, Woodhead, Oxford, 2012, pp. 17–40. [11] C.J. Hopfe, M.R. Hall, Fabric insulation, thermal bridging and acoustics in modern earth buildings, in: M.R. Hall, R. Lindsay, M. Krayenhoff (Eds.), Modern Earth Buildings. Materials, Engineering, Construction and Applications, Woodhead, Oxford, 2012, pp. 41–71. [12] Tolles E. Leroy, E.E. Kimbro, F.A. Webster, W.S. Ginell, Seismic Stabilization of Historic Adobe Structures, The Getty Conservation Institute, Los Angeles, 2000. [13] M. Blondet, Mitigation of seismic risk on earthen buildings, in: H. Gökçekusß, U. Türker, J.W. LaMoreaux (Eds.), Survival and Sustainability, Environmental Earth Sciences, Springer-Verlag, Berlin, 2011, pp. 391–400. [14] E. Quagliarini, S. Lenci, The influence of natural stabilizers and natural fibres on the mechanical properties of ancient Roman adobe bricks, J. Cult. Heritage 11 (2010) 309–314. [15] D. Silveira, H. Varum, A. Costa, T. Martins, H. Pereira, J. Almeida, Mechanical properties of adobe bricks in ancient constructions, Constr. Build. Mater. 28 (1) (2012) 36–44. [16] F. Parisi, D. Asprone, L. Fenu, A. Prota, Experimental characterization of Italian composite adobe bricks reinforced with straw fibers, Compos. Struct. 122 (2015) 300–307. [17] K. Ghavami, R.D. Toledo Filho, N.P. Barbosa, Behaviour of composite soil reinforced with natural fibres, Cem. Concr. Compos. 21 (1999) 39–48. [18] H. Binici, O. Aksogan, T. Shah, Investigation of fibre reinforced mud brick as a building material, Constr. Build. Mater. 19 (4) (2005) 313–318. [19] Q. Piattoni, E. Quagliarini, S. Lenci, Experimental analysis and modelling of the mechanical behaviour of earthen bricks, Constr. Build. Mater. 25 (4) (2011) 2067–2075. [20] H. Danso, D.B. Martinson, M. Ali, C. Mant, Performance characteristics of enhanced soil blocks: a quantitative review, Build. Res. Inform. 43 (2) (2015) 253–262. [21] M. Bouhicha, F. Aouissi, S. Kenai, Performance of composite soil reinforced with barley straw, Cem. Concr. Compos. 27 (2005) 617–621. [22] C. Galán-Marín, C. Rivera-Gómez, J. Petric, Clay-based composite stabilized with natural polymer and fibre, Constr. Build. Mater. 24 (8) (2010) 1462–1468. [23] C. Galán-Marín, C. Rivera-Gómez, F. Bradley, The mechanical properties and molecular bonding characteristics of clay-based natural composites reinforced with animal fibres, J. Biobased Mater. Bioenergy 7 (1) (2013) 143–151. [24] C. Rivera-Gómez, C. Galán-Marín, F. Bradley, Analysis of the influence of the fiber type in polymer matrix/fiber bond using natural organic polymer stabilizer, Polymers 6 (2014) 977–994. [25] R. Mattone, Sisal fibre reinforced soil with cement or cactus pulp in bahareque technique, Cem. Concr. Compos. 27 (2005) 611–616. [26] Y. Millogo, J.C. Morel, J.E. Aubert, K. Ghavami, Experimental analysis of pressed adobe blocks reinforced with Hibiscus cannabinus fibers, Constr. Build. Mater. 52 (2014) 71–78.
F. Aymerich et al. / Construction and Building Materials 109 (2016) 109–119 [27] Y. Millogo, J.E. Aubert, E. Hamard, J.C. Morel, How properties of kenaf fibers from burkina faso contribute to the reinforcement of earth blocks, Materials 8 (2015) 2332–2345. [28] M. Mostafa, N. Uddin, Effect of banana fibers on the compressive and flexural strength of compressed earth blocks, Buildings 5 (2015) 282–296. [29] H. Danso, D.B. Martinson, M. Ali, J. Williams, Effect of fibre aspect ratio on mechanical properties of soil building blocks, Constr. Build. Mater. 83 (2015) 314–319. [30] M. Segetin, K. Jayaraman, X. Xu, Harakeke reinforcement of soil-cement building materials: manufacturability and properties, Build. Environ. 42 (2007) 3066–3079. [31] A.P.S. Martins, F.A. Andrade Silva, R.D. Toledo Filho, Mechanical behavior of self-compacting soil-cement-sisal fiber composites, Key Eng. Mater. 634 (2015) 421–432. [32] F. Clementi, S. Lenci, T. Sadowski, Fracture characteristics of unfired earth, Int. J. Fract. 149 (2008) 193–198. [33] F. Aymerich, L. Fenu, P. Meloni, Effect of reinforcing wool fibres on fracture and energy absorption properties of an earthen material, Constr. Build. Mater. 27 (1) (2012) 66–72.
119
[34] A.S. Arya, T. Boen, Y. Ishiyama, Guidelines for Earthquake Resistant NonEngineered Construction, UNESCO, Paris, 2014. [35] ASTM Standard D422-63, Standard Test Method for Particle-size Analysis of Soils, ASTM International, West Conshohocken (PA), 2007, http://dx.doi.org/ 10.1520/D0422-63R07.