Enhancing Bio-Based Concrete Mechanical Properties: A Novel Approach with Composite Sandwiches and Confined Cylinders

doi:10.21203/rs.3.rs-3932572/v1

Download PDF

Research Article

Enhancing Bio-Based Concrete Mechanical Properties: A Novel Approach with Composite Sandwiches and Confined Cylinders

https://doi.org/10.21203/rs.3.rs-3932572/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 01 Jul, 2024

Read the published version in Materials and Structures →

You are reading this latest preprint version

This paper addresses the growing use of bio-based materials in Europe, thanks to their low embodied energy and carbon sequestration. Despite favorable hygrothermal and acoustic properties, the inherent challenge lies in the low mechanical properties of biobased concrete. This study presents an innovative approach to strengthen hemp concrete through natural FRCM (Fabric Reinforced Cementitious Matrix), using two distinct reinforcement techniques. Firstly, a bending reinforcement consists of applying natural FRCM as the outer skins of a composite sandwich, with hemp concrete as the core. The effect of textile layers and pre-impregnation on the FRCM mechanical properties within the composite sandwich is evaluated. Secondly, compressive reinforcement entails confining hemp concrete specimens with FRCM. The results show a significant improvement in the mechanical properties of hemp concrete, with bending and compressive reinforcement leading to increases in the mechanical strength up to 17530% and 258%, respectively. Configurations involving mineral-impregnated fabric (PM FRCM) demonstrate superior mechanical reinforcement since it allows a better interphase bond between fabric and cementitious matrix. Different failure modes are observed between reference (non-reinforced) specimens and reinforced specimens, with bending reinforced specimens exhibiting shear failure and debonding at the interface of the composite sandwich, while reference specimens fail in bending. Moreover, compressive reinforced specimens undergo crushing of hemp concrete after tensioning rupture of the fabric, while reference specimens present angular shear path in the middle of the specimens. The results underscore the promise of FRCM in mechanically reinforcing biobased concrete, opening new opportunities for their expanded use in the construction industry.

FRCM

composite sandwich

strengthening

confined cylinder

pre-impregnation

mechanical properties

In France, 43% of energy consumption and 25% of greenhouse gas emissions are attributable to the building sector, half of which are related to building materials and equipment [1][2]. To address energy efficiency and climate change, there’s growing interest in bio-based materials in construction, due to their low embodied energy and carbon sequestration [3][4]. Bio-based concrete exhibits interesting hygrothermal, and acoustic properties, due to its low density, porous structure, moisture buffer capability, low thermal conductivity, and high permeability [5], [6], [7], [8], [9], [10]. Researchers have explored various plant aggregates, with hemp shives being particularly promising due to their low fertilization and irrigation requirements, good humidity control, wide availability in the industry, ecological balance, and growing efficiency [11], [12]. Hempcrete, often referred to as lime-hemp concrete is the most common type of bio-based concrete addressed in the literature. It has a low environmental impact due to carbon sequestration during hemp growth and lime carbonation during the setting of concrete (Ca(OH)₂ + CO₂ ◊ CaCO₃ + H₂O) [13], [14], [15].

However, despite its many hygrothermal benefits [16], [17], [18], hempcrete exhibits low mechanical strength compared to conventional building materials [7], [9], limiting its use as a load-bearing material [19]. In contrast, the use of hempcrete to fill structural timber studwork has been shown to improve the racking performance of timber stud walls and structural frames by preventing buckling and resisting in-plane forces [20], [21], [22]. The low mechanical strength of hempcrete can be explained by several reasons such as the weak interfacial bonding between the bio-aggregates and the cementitious paste, volumetric variations between constituents as humidity levels change [23], [24], [25], hemicellulose effects on the setting mechanism of cement paste [26], [27], and the capillary force generated by the porosity of plant aggregates [9], [28], [29], [30].

To overcome these challenges, several treatments have been addressed, including lignocellulosic aggregate modifications such as chemical treatments with alkaline solutions, [31], [32], [33], coating with linseed oil [30], or even hot water washing treatments[34], [35]. Other treatments involve binder improvement, whether by using additives [36], [37], [38] or by choosing a bio-compatible binder [12], [39]. In the same context, Garikapati and Sadeghian have reinforced a flax concrete by using a jute fabric placed either at middle or at the quarter of the sandwich [40].

However, despite all these treatments and methods, hempcrete’s mechanical properties have never hitherto been improved sufficiently to allow it to be used in structural applications. Hitherto, hempcrete has been used mainly for its insulating properties (e.g. wood-frame filling).

This work aims to enhance the mechanical performance of a hempcrete by reinforcing it with a Fabric-Reinforced Cementitious Matrix (FRCM) composite, consisting of a flax fabric embedded in a mineral matrix. The combination of natural FRCM and hempcrete can produce an innovative hybrid composite, combining mechanical and hygrothermal performance. As far of authors’ knowledge, there has been no previous study in the literature mentioning the reinforcement of vegetal concrete with natural FRCM composites.

FRCM consists of composite materials composed of continuous yarns embedded in a mineral matrix [41]. These composites present a range of advantages, including excellent mechanical properties, good fire resistance, lightweight, ease of implementation, cost-efficiency, permeability, and compatibility with masonry and concrete structures[42], [43]. Natural FRCM composites have proven to be an efficient solution in reinforcing element structures – especially against seismic risks [44].

In this work, two reinforcement methods were elaborated: (i) flexural reinforcement by fabrication of a composite sandwich with hempcrete as a core and FRCM as skins, and (ii) compressive reinforcement by confining cylindrical hempcrete specimens with FRCM. The mechanical behavior of the reinforced specimens is assessed by considering the impact of both the reinforcement ratio and pre-impregnation of the textile. The promising results pave the way for new opportunities in enhancing bio-based concretes

This section describes the raw materials, hempcrete design, specimen preparation procedure, and the test setup for both flexural and compressive tests.

2.1 Raw Materials

2.1.1 Hemp shives

In this study, hemp shives were used as plant aggregates, and their physical and thermal properties were examined in accordance with the RILEM recommendations[45], as listed in Table 1.

The size of hemp shives was determined through image analysis, where a 5 g sample of particles was studied to cover at least 100 pixels. The images of the particles were analyzed using ImageJ software to measure both height and width.

Bulk density ${\rho }_{hemp}$ was determined using a specific protocol whereby a 1 kg sample portion was separated until approximately 350 g of hemp material was obtained [45]. Two cylindrical shells with diameters of 11 cm and 16 cm, whose height was twice their diameter, were used. The hemp material was placed into the dry cylindrical mold and the quantity was adjusted to half the volume of the cylinder. The cylinder was tapped to obtain a horizontal surface. Then, the cylinder with the hemp particles was weighed (m_hemp). After removing the material, the cylinder was filled with water, weighed again (m_water). Since the volume of both water and hemp remained the same for each experiment and the water density was known (ρ_water = 997 kg/m³), the calculation of hemp particles bulk density of hemp particles was performed using the following Eq. (1),

$${\rho }_{hemp}=\frac{{m}_{hemp}\times {\rho }_{water}}{{m}_{water}}$$

The determination of water content involved drying the aggregates and employing a specific protocol to calculate their water content at various immersion times, as outlined in a specified protocol [45] (Fig. 1). After sampling, a resulting of 200 g of hemp shives were dried in an oven at 60°C until the mass change of the sample in 24 hours was less than 0.1%, corresponding to about 30 g. These particles were placed in the water-permeable plastic bag, and the bag was immersed in water for 1 minute. The bag was placed in a “salad spinner” that was spun 100 times at an approximately 2 revolutions per second. The weight of the spun bag of hemp was noted. The steps of immersion of the hemp shives and weighing measurement were repeated with the same sample at different immersion times. The weight measurements were used to calculate the Initial Water Content (IWC: water absorption after 1 minute of immersion), the Final Water Content (FWC: water absorption after 48 h of immersion) and the diffusion rate of water in the particles (K₁: slope of the curve of Eq. 2).

$$W\left(t\right)=IWC+{K}_{1}.\text{log}\left(t\right)$$

Where W(t) is the water content at time t.

The hemp shive’s thermal conductivity λ was determined using an experimental protocol employing a hot wire under transient condition. The tests were performed on a dry-state hemp shives-filled cylinder with a 0.1 W for 120 s, repeated across five samples. The instruments used to collect data were Neotim Alimentation FP2C, Neotim conductivity wire probe, and the computer program FLIR.

There is a proportional relationship between the temperature rise ∆T and logarithmic heating time (t) as shown in Eq. 3, where q is the heat flux per meter. The thermal conductivity λ is obtained from the slope of the curve plotted by the software.

$$\varDelta T=\frac{q}{4\pi \lambda }[\text{ln}\left(t\right)+const.]$$

Table 1

Physical and hygrothermal properties of hemp shives
	Mean	CV (%)
Width (mm)	0.7-9	3
Length (mm)	0.56–25.45	2
${\rho }_{hemp}$ (kg/m³)	112	3
IWC (%)	225	6
FWC (%)	440	4
λ (W/m.k)	0.0475	4

2.1.2 Mineral binders

Two different types of mineral binders were used in this work: calcium sulfoaluminate cement CSA, which was used for the bending-reinforced specimens, and natural prompt cement PNC, which was used for the compression-reinforced specimens.

2.1.2.1 Calcium sulfoaluminate cement (CSA)

The mineral binder utilized in the case of the bending test for the reference (hemp concrete) and reinforced specimens (hemp concrete + FRCM) was CSA cement [46]. In general, CSA cement consists of a mixture of 82% CSA clinker (ye’elimite) and 18% anhydride (calcium sulfate). The densities of CSA clinker and anhydride are 3.05 g/cm³ and 2.94 g/cm³, respectively [46].

The choice to use sulfoaluminate cement instead of Portland cement was directly related to the ability of CSA cement to reduce energy consumption and CO₂ emissions during its production, as well as its contribution to a lower alkaline medium [47]. The hydration of CSA cement results in the generation of ettringite, as shown in Eq. 4, replacing the conventional formation of portlandite and hydrates of Portland cement [48]. Moreover, one of the main advantages of using CSA cement over Portland cement is its rapid strength gain, since CSA can reach a strength of 40 MPa early on, after only 6 hours of aging [48].

C₄A₃ Š + 2CŠ+38H ◊ C₆AŠ₃ H₃₂ + 2AH₃ (4)

For this study, the W/C ratios of the mixtures of that for FRCM and hemp concrete were 0.35 and 0.93, respectively. The ratio of 0.35 used for FRCM resulted in a rheology favorable for impregnation of the textile in the mineral matrix and a good adhesive interactions with hemp shives.

The tensile properties of the matrix were characterized by three-point bending tests on three prismatic specimens of 160 × 40 × 40 mm³, in accordance with French standard NF EN 196-1 [49]. The average flexural stress determined after 28-days curing period was 5.6 MPa (CV 12%).

2.1.2.2 Prompt natural cement (PNC)

Prompt natural cement denoted PNC was used as the mineral binder for the reference and reinforced specimens tested under compression. PNC is a natural hydraulic binder produced by an industrial process based on the firing of a unique raw material in the physical form of clayey limestone, followed by very fine grinding. The use of PNC lowers the environmental impact, as CO₂ emissions are reduced by about 20% compared to Portland cement [50]. PNC has the special property of quick setting and hardening time [50].

In this study, the W/C ratios of the mixture used to produce FRCM and hemp concrete were 0.5 and 0.93, respectively. In the FRCM mixture, a setting retarder called Tempo is added to the PNC due to the quick setting time of PNC, with a retarder/binder ratio of 0.0032 in mass. Setting is delayed by 10 to 30 minutes with the addition of the retarder [51].

According to the data sheet issued by Vicat, the compressive strength and stiffness modulus of PNC at a W/C ratio of 0.5 after 28 days are 12 MPa and 19.2 GPa, respectively [50].

2.1.3 Flax Textile

In this study, a unidirectional flax fabric was used to reinforce the hemp concrete. The physical properties of the textile are presented in Table 2 [52].

Flax fabric underwent tensile tests on their bare surface without any cement included, in accordance with the standard ISO 13934-1[53], in order to determine the mechanical properties in the dry state. The average tensile strength and the elastic modulus obtained are 123 MPa and 7 GPa respectively. The elastic modulus was determined between a stress range of 0.58σ_text,_max to 0.83σ_text,_max, where σ_text,_max represent the maximum stress of the tested flax textile.

Table 2

Physical properties of unidirectional flax textile used (supplier’s data)
Fabric geometry	Unidirectional
Ply thickness (mm)	0.43
Linear Tex	105
Fabric weight (g/m²)	280

2.2 Specimen preparation

The effect of reinforcing hemp concrete with FRCM materials was evaluated using two different reinforcement methods: bending and compressive reinforcement.

For bending reinforcement, parallelepiped specimens with dimensions of length: 40 cm, width: 10 cm, and height: 10 cm, were prepared. On the other hand, for compressive reinforcement, cylindrical specimens of 12 cm diameter and 22 cm height were manufactured. All specimens were produced by hand layup.

This section describes the manufacturing process of the reference specimens (hemp concrete without reinforcement) and the reinforced specimens for both bending and compression tests.

2.2.1 Hemp concrete design

Figures 2 and 3 describe the preparation steps of bending and compression reference specimens, respectively. Both were formulated under binder-to-aggregate-to-water mass ratio of 1:0.5:0.93. Initially, particles of hemp were dropped into a plastic container (Fig. 2.a). The uniform mixing process involved gradually adding water to hemp shives, wetting them before introducing cement powder (Fig. 2.b). The powder was then hydrated by adding the remaining water (Fig. 2.c). The mixture was mixed continuously by hand until all hemp shives were well covered with and adhered to the cement paste (Figs. 2.d & 2.e). Finally, the resulting mixture was cast in parallelepiped and cylindrical molds and manually compacted with a wooden stake in three layers (Fig. 2.f and Fig. 3.b). After 48 hours, specimens were demolded (Fig. 2.g and Fig. 3.c) and placed in a conditioned room (20°C and 50% RH) for 28 days, to facilitate proper setting and curing mechanisms.

The densities of the bending and compression hemp concrete specimens at the end of this curing period were 190 kg/cm³ and 210 kg/cm³, respectively.

2.2.2 Reinforced hemp concrete fabrication

2.2.2.1 Bending reinforcement

Figure 4 illustrates the procedure of producing bending-reinforced specimens through a composite sandwich. A demolding agent was applied in the metal mold, followed by the placement of a first matrix layer at the mold’s bottom (Fig. 4.a). A pre-cut flax textile (40 cm long and 10 cm wide) was then placed (Fig. 4.b), and manual pressure was applied on the fabric to ensure good penetration of the cementitious matrix into the fabric yarns. The lower 10 mm-thick FRCM skin was formed with a second matrix layer (Fig. 4.c). The previously prepared hemp concrete was added in three layers above the lower FRCM (Fig. 4.d), compacted using a wooden stake. A fresh CSA matrix layer was deposited over the core (Fig. 4.e), initiating the upper FRCM skin. The CSA matrix layer was stirred well to ensure a good interphase bond between the hemp core and the matrix layer. Then, a pre-cut flax fabric was positioned (Fig. 4.f), and manual pressure was applied to ensure better cement penetration into fabric yarns. Another matrix layer completed the formation of the upper 10 mm-thick FRCM skin of the composite sandwich (Fig. 4.g). In cases of double fabric layers, steps in (Figs. 4.a, 4.b, 4.c, 4.e, 4.f, 4.g) were repeated twice while maintaining the overall thickness of the FRCM skins at 10 mm. After 48 hours, the bending-reinforced specimens were demolded (Fig. 4.h) and placed in a conditioned room (20°C and 50% RH) for 28 days before testing.

2.2.2.2 Compressive reinforcement

Figure 5 shows the procedure of making compressive reinforced specimens by confining a cylinder with FRCM. After demolding reference hemp concrete specimens prepared as detailed in section 2.2.1, a fresh PNC matrix layer was placed around the entire cylinder (Fig. 5.a), then a precut length of flax fabric (44 cm long and 22 cm wide) was manually wrapped around the matrix layer with an overlap of 10 cm (Fig. 5.b). The flax fabric was then covered with a second PNC matrix layer to form the confinement layer (Fig. 5.c). The sample surface was smoothed with a spatula (Fig. 5.d). One hour later, the compression-reinforced specimens were stored in a conditioned room at 20°C and 50% RH, for 28 days before testing.

2.3 Experimental configurations

For bending-reinforced specimens, the experimental configurations varied based on two main parameters: the reinforcement ratio, and the pre-impregnation of the fabric.

The reinforcement ratio was related to the number of flax fabric layers incorporated in FRCM skins. In this work, B.R.1L and B.R.2L were studied – i.e., the FRCM skins contain one and two fabric layers, respectively.

However, the effect of impregnation was investigated by varying the condition of the incorporated fabric between two different configurations:

Non-impregnated fabric, denoted NP
Pre-impregnated fabric with CSA matrix, denoted PM[54].

In the case of the compression-reinforced specimens, only one experimental configuration with the designation C.R was examined which refers to composites of 1 layer of non-impregnated fabric.

Additionally, B.REF and C.REF designated the bending and compression reference specimens, respectively.

For each configuration of C.REF, C.R, B.REF, B.R.1L and B.R.2L with the different fabric configurations of NP and PM, at least 3 specimens were prepared.

2.4 Test setups

The bending reference and bending-reinforced specimens were mechanically characterized by 4-point bending in accordance to BS EN 12390-5 standards[55], using a Zwick/Roell universal machine with a 50 kN load capacity, at a testing rate of 1.5 mm/min. The 4-point bending test setup is shown in Fig. 6 (a), where the distance between the roller supports on which the specimens were placed was 30 cm, and the distance between the load points was 10 cm.

On the other hand, the mechanical properties of the compression reference and compression-reinforced specimens were determined by compression tests using the same Zwick/Roell universal machine, at a testing rate of 3 mm/min. Since, hemp concrete shows an “atypical behavior”, it was necessary to adapt standards tests[18]. The compression test setup is shown in Fig. 6 (b).

3.1 Mechanical Behavior

3.1.1 Flexural Behavior

Figure 7 represents the flexural behavior of the reference hemp concrete specimens compared to the bending-reinforced specimens with 1L.FRCM and 2L.FRCM, respectively. A representative curve was plotted for each configuration.

The mechanical behavior of the bending-reinforced specimens was generally characterized by high ductility and superior mechanical properties compared to the reference hemp concrete specimens. In this regard, Le Troëdec et al, have investigated that the addition of natural fibers in a cement matrix led to a typical composite behavior for which ductility is enhanced [56]. However, the mechanical behavior of reference hemp concrete showed notably lower mechanical strength due to several factors, including high porosity, disordered arrangement of hemp particles and weak bonding between binder and shives. Similar findings were reported in the study of [57] where flexural behavior of hemp concrete was characterized by 2 phases: a linear phase described by the ability of matrix to absorb the mechanical load up to the peak stress, followed by a progressive failure of the specimen at the interface matrix-aggregates. The same authors have established an exponential relationship between the flexural strength and the density of the hemp concrete. Generally, it was remarked that as density increases, the flexural strength increases.

A common trend can be seen in the cases of B.R.1L.NP and B.R.1L.PM, where the flexural behavior was generally divided into 3 main zones. In the first linear zone, both the FRCM skins and the hemp concrete core contributed to the mechanical response, due to the perfect bonding at the interfaces of the composite sandwich. This zone ends after the appearance of the first significant crack at the level of FRCM skins, which occurred at a very low deflection. In a parallel study, Garikapati and Sadeghian have remarked a very short initial linear section until the occurrence of a first crack at the bottom of a reinforced flax concrete[40]. A second non-linear zone then occurred, characterized by the development and propagation of multiple flexural cracks in the lower FRCM skins, resulting in successive hardening of the composite sandwich until failure. During this zone, the bending-reinforced specimens behaved in a composite manner, where forces were transferred between the matrix and the fabric. Each crack in the FRCM skins was reflected by a force drop in the flexural behavior of the composite sandwiches. At the end of this zone, and after the peak load was reached, a sharp drop in load was observed. This drop corresponds to the start of the third zone, due to the failure of the flax fabric tension, leading to complete load transfer to the hemp core and so to the interfacial debonding between the FRCM skins and hemp concrete core of the composite sandwich. Specimens of B.R.1L.PM exhibited more pronounced deflection hardening, while soft flexural hardening can be observed in the case of B.R.1L.NP. These findings were mainly related to the ability of PM fabrics to sustain higher load and to further strengthen the bending-reinforced specimens due to the bond promoted between the inner yarns and the matrix.

The same generality in trending zones was applied for the configurations of B.R.2L.NP and B.R.2L.PM. A little gap interference in the behavior of the 2L composite sandwiches can be seen, due to the uniform stress redistribution and stability that the additional fabric layer brought to the configurations of the bending-reinforced specimens. However, it can be seen that the addition of a second flax textile has indulged in a better mechanical strength and deformation of the composite sandwich. According to Garikapati and Sadeghian, this can be explained by the location of the flax textile which implied a big difference in the contribution of the fabric to the mechanical response of the composite sandwich[40]. As a matter of fact, the authors[40] have found that reinforced flax concrete with jute textiles positioned at ½ height (approximately at the middle of the specimen, which is the case of 1L composites in this study) were not able to resist to the same flexural load and displacement of a flax concrete reinforced with 2 jute fabrics positioned at ¼ height (approximately at the quarter of the specimen, which is the case of 2L composites in this study). Hence, flax textiles positioned at quarter height of the FRCM skins exhibited an early activation, thereby a more effective contribution to the mechanical response.

In addition, it can be observed that more cracks occurred in the case of B.R.2L.PM than that of B.R.2L.NP. As a matter of fact, the mineral impregnation of the fabric led to better anchorage between the CSA matrix grain and flax yarns, which then improved the bond interphase at the level of FRCM skins, meaning they contribute better to the mechanical response of the composite sandwich.

3.1.2 Compression Behavior

Figure 8 represents the compressive behavior of the reference hemp concrete specimen C.REF in comparison to compression-reinforced specimen C.R. The curve of C.REF showed typical elastoplastic behavior that was divided in three different zones: linear, plateau, and failure. In the first linear zone, the curve exhibited quasi-elastic behavior. At the time, a cracking noise was heard, indicating that PNC binder was failing, and the hemp particles collapsed, marking the start of a second zone, in which the curve reached a plateau. In this second zone, significant deformation of the hemp concrete was observed with only small increases in stress levels, until the ultimate stress value was reached. Beyond this value, failure occurred, characterized by a continuous decrease in stress over the strain that, with respect to the increased noise, corresponding to the collapse of the hemp particles and PNC binder failure. This failure feature was also reported by Glouannec et al. [58], who observed that tall specimens (height > width) showed a clear fracture path, while the change in mechanical strength was followed by a decrease in stress. The compressive behavior is similar to that observed in previous studies of uniaxial compression tests on hemp-lime concrete [59], [60]. However, the plateau region obtained in this study was long, contrary to the finding in the literature that cylindrical hemp concrete specimens exhibited a short plateau [60].

The stress–strain curve for the C.R specimens was characterized by a first quasi-linear zone that ends with the initiation of the first crack in the FRCM. In the first zone, the mechanical response was governed by the PNC matrix of the FRCM layer. A second non-linear zone then appeared where several cracks occurred in the FRCM, representing strain hardening behavior up to failure of the specimen. This was reflected on the curve of the C.R configuration by the drop and recovery of compressive strength until the ultimate stress was reached. Beyond the ultimate strength, the stress was noted to decrease sharply by about 17%, due to significant detachment of the FRCM layer from the hemp concrete. This was immediately recovered by pseudo compressive hardening until the FRCM layer was totally destroyed. According to Piero Colajanni et al. [61], this behavior is attributable to a delay in the activation of the confinement mechanism due to slippage between the fiber and the cementitious matrix, and/or to a difficulty in applying the fiber perfectly bonded to the support. Once the flax fabric was ruptured, the compression load was completely transferred to the hemp concrete which results in a continuous drop of compressive stress until the specimen ultimately failed.

From a general overview of the trends of axial stress-strain for both configurations, it can be seen that the confinement of hemp concrete by flax fabric enhanced the mechanical performance of these specimens with a substantial gain in strength and ductility. The low compressive stress for the reference hemp concrete C.REF was mainly due to the weak interfacial zone between the aggregates and mineral binder, preventing the aggregates from contributing efficiently to the mechanical response, resulting in early failure of the specimen [62]. According to Bouloc [63], the reason behind the low compressive strength of hemp concrete is due to the ductile behavior of hemp shives and their irregular arrangements within the specimen. Nguyen linked the low mechanical strength obtained in hemp concrete to the porosity of hemp shives[64]. However, the compression-reinforced specimens C.R exhibit higher compressive strength at an early stage (low strain) due to the mechanical properties of FRCM. The compression-reinforced specimens provided a mechanical stress 5 times higher than the reference specimens at a strain of 5%. This was attributable to the strength and rigidity of flax fabric, which bore the applied load for a sufficient time while increasing the strain to failure, before the load transferred to the hemp concrete due to fabric rupture. The more pronounced softening branch of the C.R indicates a larger strain to rupture compared to that of reference specimens. A gain of 34% in maximum strain was obtained between the reference and the compression reinforced specimens. This improvement can be attributed to the ability of the FRCM to absorb and maintain the mechanical load by the aims of cracks formation until the point of flax fabric rupture. This finding also highlights the high deformability of the flax fabric, as it enabled the specimens to carry loads even after reaching the peak value, thereby enhancing overall ductility.

3.2 Failure and cracking mode

3.2.1 Bending-reinforced specimens

The crack generation at the lower FRCM skin of the composite sandwich is illustrated in Fig. 9. These cracks varied according to the reinforcement ratio and the impregnation method. For all the bending-reinforced specimens, the first crack occurred right at the center of the lower FRCM skin. The B.R.1L.NP specimens showed an average of 4 cracks in the lower FRCM skin with an average crack spacing of 64 mm and crack opening of 2 mm (Fig. 9.a). On the other hand, the B.R.1L.PM specimens showed 20 cracks on average, with crack spacing of 11.5 mm and crack opening of 0.54 mm (Fig. 9.b). The rise in crack occurrence was linked to the effect of mineral impregnation, improving the adhesion bond between the internal flax yarns and the cementitious matrix. The improved anchorage ability of the fabric meant it was able to contribute more effectively to the mechanical response. The reason behind the finer crack openings for B.R.1L.PM (Fig. 9.b) compared to B.R.1L.NP (Fig. 9.a) was the fabric’s high anchorage ability. This is evidenced by the more significant force reduction in the flexural behavior of B.R.1L.PM compared to that of B.R.1L.NP (see Fig. 7). Likewise, the addition of a second layer of flax fabric to the configurations of 2L composite sandwiches caused more cracks to initiate, as a result of the excessive contribution of the additional fabric reinforcement to the mechanical response. In the same context, the cracks occurrence increased from 4 for B.R.1L.NP to 21 for B.R.2L.NP (Fig. 9.c), and from 20 for B.R.1L.PM to 25 for B.R.2L.PM (Fig. 9.d).

Figure 10 illustrates the failure mode for both the bending reference and bending-reinforced specimens. The failure mode for the bending reference specimens was characterized by bending failure at the central point, equidistant between the two applied load points (Fig. 10.a). However, the failure mode for the bending-reinforced specimens was characterized by progressive shear failure and debonding of the FRCM skins from the hemp concrete core in the composite sandwich (Fig. 10.b). This failure mode was primarily attributable to the low shear resistance of hemp concrete. The debonding between the upper skin and the core of the sandwich was due to the poor adhesive bond between the cementitious matrix layer and the hemp concrete. The low bonding force was due to the lack of connectors between the FRCM skins and the core of the sandwich and to the weak adherence between the hydrated hemp shives and the CSA matrix layer of the FRCM skins. However, in a similar study conducted by Frazão et al. [65] in which flexural tests was applied to sandwich panels combining sisal fiber-cement composites and fiber-reinforced lightweight concrete, no signs of shear cracks were observed in the sandwich core. The authors attributed this outcome to the high tensile resistance of the cement composite containing long sisal fibers, which formed the skin of the sandwich, as well as its strong bond with the fiber-reinforced lightweight concrete core. In contrast, da Gloria and Toledo Filho[66] found diagonal shear or flexural cracks in the sandwich core after 3-point bending tests on sandwich panels made of wood bio-concrete as a core and sisal-fiber-reinforced cement composites as skins. Shear and flexural failure occurred on a core made of a cement/wood ratio of 0.5 and 2.5, respectively. Notably, in both studies, no delamination was observed between the upper skin and the sandwich core during the bending tests.

3.2.2 Compression-reinforced specimens

Figure 11 shows the failure mode for the compressive reference C.REF and reinforced specimens C.R.

In the case of compression reference specimens, shear failure can be seen in the form of an angular path in the middle of the specimens as highlighted in Fig. 11.a. This angular failure that split the C.REF specimens into two separate parts was typically related to the failure of bond between the hemp shives and the mineral binder, causing local shear rupture. However, the results of a previous study on the mechanical characterization of hemp lime concrete showed that cylindrical specimens experienced failure in the upper half of the specimen [22]. According to the authors, this was due to slumping of the material during molding and the early stages of curing, which means that the bottom half of the specimen is slightly denser. Additionally, the orientation of the hemp shives has a direct impact on the failure mode, whether the load was parallel or perpendicular to the direction of casting force [67].

Figure 11.b represents the cracking and failure mode for the C.R specimens. A major crack occurred at the flax overlap location at a strain rate of 14% and propagated from the upper part of the confined cylinder to the bottom part, as highlighted in Fig. 11.b. This crack was announced by subsequent noises highlighting the detachment of the flax fabric from the PNC matrix, which led to a 21% decrease in stiffness, as can be seen from Fig. 8. The crack opening was wider in the vicinity of the metal plates compared to other areas along the crack curve, due to the stress concentration near the testing plates. After that crack, failure of the compression-reinforced specimens occurred abruptly by the crushing of the hemp concrete in the vicinity of the longer cracks, due to tensioning rupture of the flax fabric in the overlapping region of the cylinders. Though the hemp concrete core was crushed when the flax fabric was ruptured under tension, the compression-reinforced specimens, overall, remained in good condition compared to the ordinary hemp concrete (reference). A recent study has gathered the data from 76 experimental tests in the literature where the failure mode of FRCM-confined concrete was crushing failure of the reinforced elements due to telescopic jacket textile failure after the formation of wide vertical cracks in the textile overlapping region [61], [68], [69]. Although the use of fibers with FRCM composites or wraps externally bonded to the concrete surface has been introduced as a popular technique to delay damage and avoid early failure [70], the mechanism of interaction between FRCM composites (wraps) and the internal concrete surface is as yet unknown due to lack of experimental and analytical models [71].

3.3 Mechanical properties

In this section, several mechanical parameters were evaluated to investigate the effect of the bending- and compression-reinforcement techniques on the mechanical properties of hemp concrete.

3.3.1 Bending-reinforced specimens

On the basis of studying the improvement of the mechanical properties by bending reinforcement, several mechanical properties were determined in this context: the stiffness coefficient in the first zone K₁ and the second zone K_2, the maximum bending load F_max, the load corresponding to when the first crack occurred F₁, and the stored mechanical energy in the specimens W. The mechanical energy was calculated as the area under the force-deflection curve up to 40 mm of mid span deflection. The effect of the reinforcement ratio and pre-impregnation technique of the FRCM skins on the mechanical properties of composite sandwiches were analyzed as well.

The values of the mechanical properties for B.REF and B.R specimens are listed in Table 3. For each parameter, the mean value and the coefficient of variation (CV) were calculated. It must be noted that in the case of B.REF specimens, no values of F₁ were reported, since the reference specimens do not follow the criteria on the basis of which this parameter was determined.

Table 3

Mechanical properties for different configurations of bending-reinforced and bending reference specimens.
	K₁ (kN/mm)		K₂ (kN/mm)		F₁ (kN)		F_max (kN)		W (mJ)
	Mean	CV (%)	Mean	CV (%)	Mean	CV (%)	Mean	CV (%)	Mean	CV (%)
B.R.1L.NP	0.53	16	0.03	16	1.34	15	2.36	17	55	6
B.R.1L.PM	1.14	9	0.05	18	1.95	10	4.33	11	116	8
B.R.2L.NP	1.30	14	0.09	15	2.05	7	6.33	10	153	7
B.R.2L.PM	1.37	19	0.13	2	2.82	3	8.11	5	188	5
B.REF	0.025	13	-	-	-	-	0.05	6	0.18e^− 6	2

Figure 12 illustrates the maximum bending load F_max and the first crack load F₁ for the B.REF specimens and all the B.R sandwiches (B.R.1L.NP, B.R.1L.PM, B.R.2L.NP, and B.R.2L.PM).

From an overview of Fig. 12, it can be observed that enormous increase in the maximum bending load carried F_max occurred between the reference specimens and the bending-reinforced specimens. For instance, F_max increased up to 5030% and 9313% when the reinforced configurations applied were B.R.1L.NP and B.R.1L.PM, respectively. This improvement of F_max went even higher to reach 13660%, and 17530% when B.REF was compared to B.R.2L.NP and B.R.2L.PM, respectively. This tremendous increase was due to two main reasons: firstly the very low mechanical endurance of the reference hemp concrete, and secondly the effective performance of FRCM composite structures in strengthening the hemp concrete core.

An increase of 53% and 45% in terms of F₁ was observed when the fabric layer was doubled from 1L to 2L for the B.R.NP and B.R.PM configurations, respectively. This gain in F₁ load was related to the additional strength the second flax fabric imparts to the configuration, which enhanced the mechanical resistance of the FRCM. Similarly, the addition of a second fabric layer for the bending-reinforced specimens led to a general gain in terms of F_max of 168% and 87% for the B.R.NP and B.R.PM configurations, respectively. The slighter increase in the case of B.R.PM compared to B.R.NP when the flax fabric was doubled from 1L to 2L can be related to the early failure of the sandwich composite caused by core shearing. This failure is caused by the lack of connectors in the composite, which limits the effective transfer of loads between the skins and the core. As a result, the addition of a second pre-impregnated textile to the FRCM composite is less effective in improving sandwich performance.

The mineral impregnation led to a general improvement in terms of F₁ and F_max. However, these improvements were not as significant as the effect of reinforcement ratio on the FRCM composites. For instance, the mineral impregnation made a gain in F₁ of 46% and 38% for the composite sandwiches of 1L and 2L, respectively. This improvement was due to the better anchorage ability of internal flax yarns which made a strong adhesion bond between the flax fabric and cementitious matrix in the FRCM skins. As a matter of fact, the matrix-impregnated flax fabric contributes more efficiently to the mechanical response of the sandwich.

Figure 13 shows the stiffness coefficient K₁ and K₂ for the reference specimens and the bending-reinforced composites. Comparing the B.REF and B.R configurations, a huge gain in stiffness at the linear zone can be noticed, due to the FRCM skins applied to the reference hemp concrete. The stiffness coefficient K₁ increased by 2020%, 4460%, 5100%, and 5380% when the reinforcement configurations applied were B.R.1L.NP, B.R.1L.PM, B.R.2L.NP, and B.R.2L.PM, respectively. This can be attributed to the ability of the FRCM composite skins to maintain a stiffer configuration all along the linear zone with only slight flexural deflection, due to its high mechanical resistance/weight ratio.

Doubling the flax layer in the B.R.NP composites resulted in a 145% increase in the stiffness coefficient K₁, while impregnating the textile resulted in a 115% increase. However, in the case of the B.R.PM composites, the stiffness only increased by 20% when the flax layer was doubled and by 5% when the textile was impregnated. In the second zone of the flexural behavior of the reinforced specimens, doubling the flax layer resulted in an increase in K₂ of 200% and 160% for the B.R.NP and B.R.PM configurations, respectively. The significant rise in K₂ can be attributed to the substantial contribution of the flax textile to the mechanical behavior of the reinforced specimens during this phase.

It should be noted that, as with F_max, improving the mechanical properties of the FRCM (by textile impregnation or the addition of flax layers) does not have a significant effect on the mechanical behavior of the sandwich beyond a certain threshold. This limitation is due to the limited shear resistance of the core and the absence of connectors, which restrict further improvements in the mechanical properties of the sandwich.

Figure 14 shows the mechanical energy W for the reference and the different bending-reinforced configurations. This energy represents the area under the force-deflection curve for each configuration, indicating the amount of work absorbed by each specimen. The absorbed work is directly related to the flexural hardening performance of each configuration. The results showed the same general trend as the maximum effort F_max. In fact, when the fabric layer is doubled, W increases by approximately 180% and 63% for the B.R.NP and B.R.PM configurations, respectively. Furthermore, the mineral pre-impregnation of the textile resulted in an increase in the energy W. This was due to the improved adherence, which increased the strength and promoted the formation of multiple cracks, thus increasing the ductility of the sandwich. However, this improvement was more pronounced for the 1L specimens (112%) compared to the 2L specimens (23%).

3.3.2 Compression-reinforced specimens

When analyzing the improvement in mechanical properties in the case of compression reinforcement, several mechanical parameters were determined: the stress at the first crack σ_1, the maximum mechanical strength σ_max, the Young’s modulus E, and the maximum strain ɛ_max. This latter was calculated at a drop of 80% of the peak load corresponding to the testing parameters at rupture threshold. The values of the mechanical properties for the C.REF and C.R specimens are listed in Table 4. For each parameter, the mean value and the CV were calculated.

Table 4

Mechanical properties of compressive reinforced and compressive reference specimens.
	E (MPa)		σ ₁ (MPa)		σ _max (MPa)		ɛ_max (%)
	Mean	CV (%)	Mean	CV (%)	Mean	CV (%)	Mean	CV(%)
C.REF	4.7	7	-	-	0.26	12	18	2
C.R	30	10	0.77	4	0.93	12	25	8

Figure 15 illustrates the mechanical strength (σ_max) and the rigidity (E) for the two configurations, C.REF and C.R specimens. The flax fabric wrapped around the hemp concrete cylinder resulted in a 258% increase in mechanical strength compared to the reference specimens. This remarkable improvement can be attributed to two factors. Firstly, the hemp concrete had low compressive strength due to the poor bonding between the hemp shives and the binder and the imperfect arrangement of the hemp particles. Secondly, the use of FRCM confinement proved to be highly effective for reinforcing materials with low mechanical strength. Previous studies have shown that the degree of improvement in compressive strength depends more on the axial capacity of the reference specimen than on the specific confinement technique used [72].

The application of FRCM confinement had a significant effect on the rigidity of the reference specimens, resulting in a remarkable increase of 531%. This effect is evident when comparing the ascending linear zones of the reference and reinforced specimens, as shown in Fig. 8. This increase is due to the high initial stiffness of the FRCM (in the pre-crack zone), which is related to the mechanical properties of both the cementitious matrix and the flax fabric. In contrast, other studies have mentioned that the incorporation of natural fibers into cementitious matrix can led to a decrease in its stiffness[73], [74].

Bending and compressive reinforcement techniques for hemp concrete were investigated in this study. The effect of reinforcement ratio and textile pre-impregnation on the mechanical properties for the bending-reinforced specimens were analyzed. The results are promising and pave the way for exploring new possibilities in the reinforcement of vegetal concrete. On the basis of the experimental results, the following conclusions can be drawn:

Reference hemp concrete specimens exhibit very poor mechanical properties compared to reinforced specimens, especially in the case of bending reinforcement.
A tremendous enhancement in the mechanical properties of reference hemp concrete was obtained by the applied bending or compressive reinforcement technique, due to the effectiveness of FRCM in reinforcing lightweight structures.
In the case of bending-reinforced specimens with two impregnated flax layers, the increase in maximum bending load and stiffness coefficient in the first zone reached up to 17530%, and 5380%, respectively, compared to the reference specimens.
The doubling of the fabric layer brought about a general improvement in the mechanical response of the bending-reinforced specimens due to the additional contribution of the fabric to the performance.
Mineral pre-impregnation led to better mechanical properties of the bending-reinforced specimens in terms of load and stiffness due to the promoted adhesion between the inner flax yarns and the CSA matrix.
The crack openings in the lower FRCM skin were finest in the case of bending reinforced specimens with impregnated flax textiles due to the better bonding interphase between the fabric and the CSA matrix.
The mineral impregnation of the fabric contributed to more crack initiation in the FRCM due to the enhanced anchorage ability of flax fabric embedded in the CSA matrix.
The failure mode for the reference bending specimens was characterized by critical bending failure at the center of the specimens, whereas the failure for the bending-reinforced specimens was progressive shear failure and the debonding of the FRCM skins from the hemp core.
The failure mode for the reference compressive specimens was sharp shear failure, whereas the failure mode for the compressive reinforced specimens was crushing of the hemp concrete due to tensile rupture of the flax fabric at the overlap point.
The mechanical energy absorbed by the bending reinforced specimens was increased up to 180% by doubling the fabric layer, and similarly by up to 112% by mineral impregnation of the flax fabric. The increase was directly correlated to the ability of the reinforcement ratio and mineral impregnation to maintain better flexural hardening.
FRCM confinement provides a substantial gain in compressive strength and ductility. This is related to the mechanical properties and to the ease of flexibility of flax fabric.
Compression reinforced specimens exhibited a gain of 258% and 531% in terms of mechanical strength and stiffness, respectively, compared to the reference specimens.

Future perspectives of the current study include the incorporation of connectors between the skins and the core of the sandwich composite to avoid debonding at the composite interface and to improve the mechanical behavior of the sandwich. In addition, different configurations of compressive reinforcement will be investigated, as the results of the current study are preliminary and serve as a basis for future research in this area.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability statement

The data that support the findings of this study are available from the corresponding author, upon reasonable request.

B. à É, Carbone P « Les bâtiments exemplaires ». Consulté le: 30 juin 2023. [En ligne]. Disponible sur: http://www-maj.batiment-energiecarbone.e2.rie.gouv.fr/les-batiments-exemplaires-r29.html
«, Données et études statistiques pour le changement climatique, l’énergie, l’environnement, le logement, et les transports », Données et études statistiques pour le changement climatique, l’énergie, l’environnement, le logement, et les transports. Consulté le: 6 avril 2023. [En ligne]. Disponible sur: https://www.statistiques.developpement-durable.gouv.fr
« Habitat III (2015) : Third United Nations Conference on Housing and Sustainable Urban Development », J. Educ. Sustain. Dev., vol. 9, n^o 2, p. 125–125, sept. 10.1177/0169796X15600604c
Jami T, Karade SR, Singh etLP (2019) « A review of the properties of hemp concrete for green building applications », J. Clean. Prod., vol. 239, p. 117852, déc. 10.1016/j.jclepro.2019.117852
Arizzi A, Cultrone G, Brümmer M, et, Viles H (2015) « A chemical, morphological and mineralogical study on the interaction between hemp hurds and aerial and natural hydraulic lime particles: Implications for mortar manufacturing », Constr. Build. Mater., vol. 75, p. 375–384, janv. 10.1016/j.conbuildmat.2014.11.026
Arnaud L (2000) « Mechanical and thermal properties of hemp mortars and wools: experimental and theoretical approaches ». Bioresour Hemp
Arnaud L et, Gourlay E (2012) « Experimental study of parameters influencing mechanical properties of hemp concretes », Constr. Build. Mater., vol. 28, n^o 1, p. 50–56, mars 10.1016/j.conbuildmat.2011.07.052
Evrard A (2006) « Sorption behaviour of Lime-Hemp Concrete and its relation to indoor comfort and energy demand &#187
Khazma M, Hajj NE, Goullieux A, Dheilly RM, et, Queneudec M (2008) « Influence of sucrose addition on the performance of a lignocellulosic composite with a cementious matrix », Compos. Part Appl. Sci. Manuf., vol. 39, n^o 12, p. 1901–1908, déc. 10.1016/j.compositesa.2008.09.014
Solomon S, Plattner G-K, Knutti R, et, Friedlingstein P (2009) « Irreversible climate change due to carbon dioxide emissions », Proc. Natl. Acad. Sci., vol. 106, n^o 6, p. 1704–1709, févr. 10.1073/pnas.0812721106
Bourdot A et al (oct. 2017) « Characterization of a hemp-based agro-material: Influence of starch ratio and hemp shive size on physical, mechanical, and hygrothermal properties ». Energy Build 153:501–512. 10.1016/j.enbuild.2017.08.022
del Valle-Zermeño R, Aubert JE, Laborel-Préneron A, Formosa J (2016) et J. M. Chimenos, « Preliminary study of the mechanical and hygrothermal properties of hemp-magnesium phosphate cements », Constr. Build. Mater., vol. 105, p. 62–68, févr. 10.1016/j.conbuildmat.2015.12.081
Boutin MP, Flamin C, Quinton S, et, Gosse G (2005) « Analyse du cycle de vie: Compounds thermoplastiques chargés fibres de chanvre et Mur en béton de chanvre banché sur ossature bois », Rapp. D’Étude INRA Lille Réf MAP, vol. 4, p. B1
de Bruijn PB, Jeppsson K-H, Sandin K, et, Nilsson C (2009) « Mechanical properties of lime–hemp concrete containing shives and fibres », Biosyst. Eng., vol. 103, n^o 4, p. 474–479, août 10.1016/j.biosystemseng.2009.02.005
Pretot S, Collet F, et, Garnier C (2014) « Life cycle assessment of a hemp concrete wall: Impact of thickness and coating », Build. Environ., vol. 72, p. 223–231, févr. 10.1016/j.buildenv.2013.11.010
Collet F et, Pretot S (2014) « Thermal conductivity of hemp concretes: Variation with formulation, density and water content », Constr. Build. Mater., vol. 65, p. 612–619, août 10.1016/j.conbuildmat.2014.05.039
Tran Le AD, Maalouf C, Mai TH, Wurtz E, et, Collet F (2010) « Transient hygrothermal behaviour of a hemp concrete building envelope », Energy Build., vol. 42, n^o 10, p. 1797–1806, oct. 10.1016/j.enbuild.2010.05.016
Véronique C « Propriétés mécaniques, thermiques et acoustiques d’un matériau à base de particules végétales: approche expérimentale et modélisation théorique &#187
Nozahic V et, Amziane S (nov. 2012) « Influence of sunflower aggregates surface treatments on physical properties and adhesion with a mineral binder ». Compos Part Appl Sci Manuf 43:1837–1849. n^o 1110.1016/j.compositesa.2012.07.011
Gross C et, Walker P (sept. 2014) « Racking performance of timber studwork and hemp-lime walling ». Constr Build Mater 66:429–435. 10.1016/j.conbuildmat.2014.05.054
Gross CD « Structural enhancement of timber framing using hemp-lime &#187
Hirst EA, Walker P, Paine KA, et, Yates T « Characterisation of Low Density Hemp-Lime Composite Building Materials under Compression Loading »
Mohr BJ, Biernacki JJ, et, Kurtis KE (2006) « Microstructural and chemical effects of wet/dry cycling on pulp fiber–cement composites », Cem. Concr. Res., vol. 36, n^o 7, p. 1240–1251, juill. 10.1016/j.cemconres.2006.03.020
Monreal P, Mboumba-Mamboundou LB, Dheilly RM, et, Quéneudec M (2011) « Effects of aggregate coating on the hygral properties of lignocellulosic composites », Cem. Concr. Compos., vol. 33, n^o 2, p. 301–308, févr. 10.1016/j.cemconcomp.2010.10.017
Savastano H et, Agopyan V (1999) « Transition zone studies of vegetable fibre-cement paste composites », Cem. Concr. Compos., vol. 21, n^o 1, p. 49–57, janv. 10.1016/S0958-9465(98)00038-9
Laskowski JS, Liu Q, et, O’Connor CT (oct. 2007) « Current understanding of the mechanism of polysaccharide adsorption at the mineral/aqueous solution interface ». Int J Min Process 84:59–68. n^o 110.1016/j.minpro.2007.03.006
Vázquez E, Hendriks CF, et, Janssen GMT (2004) PRO 40: International RILEM Conference on the Use of Recycled Materials in Buildings and Structures (Volume 2). RILEM Publications
Nozahic V, Amziane S, Torrent G, Saïdi K, et, De Baynast H (2012) « Design of green concrete made of plant-derived aggregates and a pumice–lime binder », Cem. Concr. Compos., vol. 34, n^o 2, p. 231–241, févr. 10.1016/j.cemconcomp.2011.09.002
Rowell RM et, Rowell RM (2005) Éd., Handbook of Wood Chemistry and Wood Composites. Boca Raton: CRC Press, 10.1201/9780203492437
Sellami A, Merzoud M, et, Amziane S (oct. 2013) « Improvement of mechanical properties of green concrete by treatment of the vegetals fibers ». Constr Build Mater 47:1117–1124. 10.1016/j.conbuildmat.2013.05.073
Banks WB (1973) « Water uptake by scots pine sapwood, and its restriction by the use of water repellents », Wood Sci. Technol., vol. 7, n^o 4, p. 271–284, déc. 10.1007/BF00351073
Karade SR (2010) « Cement-bonded composites from lignocellulosic wastes », Constr. Build. Mater., vol. 24, n^o 8, p. 1323–1330, août 10.1016/j.conbuildmat.2010.02.003
Tonoli GHD, Rodrigues Filho UP, Savastano H, Bras J, Belgacem MN (2009) et F. A. Rocco Lahr, « Cellulose modified fibres in cement based composites », Compos. Part Appl. Sci. Manuf., vol. 40, n^o 12, p. 2046–2053, déc. 10.1016/j.compositesa.2009.09.016
Govin A, Peschard A, et, Guyonnet R (2006) « Modification of cement hydration at early ages by natural and heated wood », Cem. Concr. Compos., vol. 28, n^o 1, p. 12–20, janv. 10.1016/j.cemconcomp.2005.09.002
Olorunnisola AO (2008) « Effects of pre-treatment of rattan (Laccosperma secundiflorum) on the hydration of Portland cement and the development of a new compatibility index », Cem. Concr. Compos., vol. 30, n^o 1, p. 37–43, janv. 10.1016/j.cemconcomp.2007.08.002
Walker R « A study of the properties of lime-hemp concrete with pozzolans », thesis, Trinity College (Dublin, Ireland). Department of Civil, Structural and Environmental Engineering, 2013. Consulté le: 6 avril 2023. [En ligne]. Disponible sur: http://www.tara.tcd.ie/handle/2262/79389
Wei YM, Zhou YG, et, Tomita B (2000) « Study of hydration behavior of wood cement-based composite II: effect of chemical additives on the hydration characteristics and strengths of wood-cement composites », J. Wood Sci., vol. 46, n^o 6, p. 444–451, déc. 10.1007/BF00765802
Zhengtian L et, Moslemi AA (1985) « Influence of chemical additives on the hydration characteristics of western larch wood-cement-water mixtures », For. Prod. J. USA, Consulté le: 6 avril 2023. [En ligne]. Disponible sur: https://scholar.google.com/scholar_lookup?title=Influence+of+chemical+additives+on+the+hydration+characteristics+of+western+larch+wood-cement-water+mixtures&author=Zhengtian%2C+L.&publication_year=1985
Ahmad MR, Pan Y, et, Chen B (2021) « Physical and mechanical properties of sustainable vegetal concrete exposed to extreme weather conditions », Constr. Build. Mater., vol. 287, p. 123024, juin 10.1016/j.conbuildmat.2021.123024
Garikapati KP et, Sadeghian P (2020) « Mechanical behavior of flax-lime concrete blocks made of waste flax shives and lime binder reinforced with jute fabric », J. Build. Eng., vol. 29, p. 101187, mai 10.1016/j.jobe.2020.101187
Dalalbashi A, Ghiassi B, Oliveira etDV (nov. 2021) « A multi-level investigation on the mechanical response of TRM-strengthened masonry ». Mater Struct 54:224. n^o 610.1617/s11527-021-01817-4
Hallonet A, Michel L, et, Ferrier E (sept. 2016) « Investigation of the bond behavior of flax FRP strengthened RC structures through double lap shear testing ». Compos Part B Eng 100:247–256. 10.1016/j.compositesb.2016.06.049
Homoro O, Michel M, et, Baranger TN (2022) « Improvement of the mechanical properties of a glass multifilament yarn reinforced ettringitic matrix using an innovative pre-impregnation process », Eur. J. Environ. Civ. Eng., vol. 26, n^o 5, p. 1977–1992, avr. 10.1080/19648189.2020.1744038
Romanazzi A, Oliveira DV, et, Silva RA (oct. 2019) « Experimental Investigation on the Bond Behavior of a Compatible TRM-based Solution for Rammed Earth Heritage ». Int J Archit Herit 13:1042–1060. n^o 710.1080/15583058.2019.1619881
Amziane S, Collet F, Lawrence M, Magniont C, Picandet V, et, Sonebi M (2017) « Recommendation of the RILEM TC 236-BBM: characterisation testing of hemp shiv to determine the initial water content, water absorption, dry density, particle size distribution and thermal conductivity ». Mater Struct 50:167. n^o 310.1617/s11527-017-1029-3
« ALPENAT ». Consulté le: 6 avril 2023. [En ligne]. Disponible sur: https://www.ciment-vicat.fr/Nos-Ciments/Les-ultra-performants/ALPENAT-UP/ALPENAT
Afroughsabet V, Biolzi L, et, Cattaneo S (2019) « Evaluation of Engineering Properties of Calcium Sulfoaluminate Cement-based Concretes Reinforced with Different Types of Fibers », Materials, vol. 12, n^o 13, Art. n^o 13, janv. 10.3390/ma12132151
Péra J et, Ambroise J (2004) « New applications of calcium sulfoaluminate cement », Cem. Concr. Res., vol. 34, n^o 4, p. 671–676, avr. 10.1016/j.cemconres.2003.10.019
« NF EN 196-1 », Afnor EDITIONS Consulté le: 6 avril 2023. [En ligne]. Disponible sur: https://www.boutique.afnor.org/fr-fr/norme/nf-en-1961/methodes-dessais-des-ciments-partie-1-determination-des-resistances/fa184622/57803
« cahier_technique_cr_cnp_eng. pdf ». Consulté le: 6 avril 2023. [En ligne]. Disponible sur: http://www.romanportland.net/files/doc/cahier_technique_cr_cnp_eng.pdf
« ciment naturel PROMPT ». Consulté le: 27 avril 2023. [En ligne]. Disponible sur: https://www.ciment-vicat.fr/Nos-Ciments/Les-ultra-performants/Ciment-naturel-PROMPT
« ampliTex® 5025 - Bcomp Ltd - PDF Catalogs | Documentation | Boating Brochures ». Consulté le: 6 avril 2023. [En ligne]. Disponible sur: https://pdf.nauticexpo.com/pdf/bcomp-ltd/amplitex-5025/51016-100536.html
14:00–17:00, « ISO 13934-1:1999 », ISO. Consulté le: 6 avril 2023. [En ligne]. Disponible sur: https://www.iso.org/standard/23363.html
Bardouh R, Homoro O, Amziane etS (2024) « Effect of reinforcement ratio and textile pre-impregnation on the mechanical properties of flax textile reinforced sulfoaluminate matrix », J. Compos. Mater., vol. 58, n^o 2, p. 265–284, janv. 10.1177/00219983231221146
« BS EN 12390- 5:2019 - TC | 31 Jul 2019 | BSI Knowledge ». Consulté le: 5 février 2024. [En ligne]. Disponible sur: https://knowledge.bsigroup.com/products/testing-hardened-concrete-flexural-strength-of-test-specimens-1?version=tracked
Le Troëdec M, Peyratout C, Chotard T, BONNET J-P, Smith A (2007) et R. Guinebretière, « Physico-chemical modifications of the interactions between hemp fibres and a lime mineral matrix: impacts on mechanical properties of mortars », juin
Elfordy S, Lucas F, Tancret F, Scudeller Y, et, Goudet L (2008) « Mechanical and thermal properties of lime and hemp concrete (hempcrete) manufactured by a projection process », Constr. Build. Mater., vol. 22, n^o 10, p. 2116–2123, oct. 10.1016/j.conbuildmat.2007.07.016
Glouannec P et al (2011) « Propriétés physiques de bétons de chanvre », Matér. Tech., vol. 99, n^o 6, Art. n^o 6, 10.1051/mattech/2011047
Pavia S, Walker R, et, McGinn J (2015) « Effect of testing variables on the hydration and compressive strength of lime hemp concrete », Acad. J. Civ. Eng., vol. 33, n^o 2, Art. n^o 2, juin 10.26168/icbbm2015.98
Walker R, Pavia S, et, Mitchell R (2014) « Mechanical properties and durability of hemp-lime concretes », Constr. Build. Mater., vol. 61, p. 340–348, juin 10.1016/j.conbuildmat.2014.02.065
Colajanni P, De Domenico F, Recupero A, et, Spinella N (2014) « Concrete columns confined with fibre reinforced cementitious mortars: Experimentation and modelling », Constr. Build. Mater., vol. 52, p. 375–384, févr. 10.1016/j.conbuildmat.2013.11.048
Amziane S, Nozahic V, et, Sonebi M (2015) « Design of mechanically enhanced concrete using hemp shiv », Acad. J. Civ. Eng., vol. 33, n^o 2, Art. n^o 2, juin 10.26168/icbbm2015.65
Bouloc P (2006) Le chanvre industriel: production et utilisations. France Agricole Editions
Nguyen TT (2010) « Contribution à l’étude de la formulation et du procédé de fabrication d’éléments de construction en béton de chanvre », phdthesis, Université de Bretagne Sud, doi: 10/document
Frazão C, Barros J, Toledo Filho R, Ferreira S, et, Gonçalves D (2018) « Development of sandwich panels combining Sisal Fiber-Cement Composites and Fiber-Reinforced Lightweight Concrete », Cem. Concr. Compos., vol. 86, p. 206–223, févr. 10.1016/j.cemconcomp.2017.11.008
da Gloria MYR (2021) et R. D. Toledo Filho, « Innovative sandwich panels made of wood bio-concrete and sisal fiber reinforced cement composites », Constr. Build. Mater., vol. 272, p. 121636, févr. 10.1016/j.conbuildmat.2020.121636
Williams J, Lawrence M, Walker etP (2018) « The influence of constituents on the properties of the bio-aggregate composite hemp-lime », Constr. Build. Mater., vol. 159, p. 9–17, janv. 10.1016/j.conbuildmat.2017.10.109
Irandegani MA, Zhang D, et, Shadabfar M (2022) « Compressive strength of concrete cylindrical columns confined with fabric-reinforced cementitious matrix composites under monotonic loading: Application of machine learning techniques », Structures, vol. 42, p. 205–220, août 10.1016/j.istruc.2022.05.111
Ombres L (2014) « Concrete confinement with a cement based high strength composite material », Compos. Struct., vol. 109, p. 294–304, mars 10.1016/j.compstruct.2013.10.037
Koutas LN, Tetta Z, Bournas DA, et, Triantafillou TC (2019) « Strengthening of Concrete Structures with Textile Reinforced Mortars: State-of-the-Art Review ». J Compos Constr 23:03118001. n^o 110.1061/(ASCE)CC.1943-5614.0000882
Ben Seghier MEA, Kechtegar B, Nait Amar M, Correia JAFO, et, Trung N-T (oct. 2021) « Simulation of the ultimate conditions of fibre-reinforced polymer confined concrete using hybrid intelligence models ». Eng Fail Anal 128:105605. 10.1016/j.engfailanal.2021.105605
Fossetti M et, Minafò G (2017) « Comparative experimental analysis on the compressive behaviour of masonry columns strengthened by FRP, BFRCM or steel wires », Compos. Part B Eng., vol. 112, p. 112–124, mars 10.1016/j.compositesb.2016.12.048
Dalmay P, Smith A, Chotard T, Sahay-Turner P, Gloaguen V, et, Krausz P (2010) « Properties of cellulosic fibre reinforced plaster: influence of hemp or flax fibres on the properties of set gypsum », J. Mater. Sci., vol. 45, n^o 3, p. 793–803, févr. 10.1007/s10853-009-4002-x
Iucolano F, Boccarusso L, et, Langella A (oct. 2019) « Hemp as eco-friendly substitute of glass fibres for gypsum reinforcement: Impact and flexural behaviour ». Compos Part B Eng 175:107073. 10.1016/j.compositesb.2019.107073

Download PDF

Journal Publication

published 01 Jul, 2024

Read the published version in Materials and Structures →

Reviewers agreed at journal
27 Feb, 2024
Reviewers invited by journal
27 Feb, 2024
Editor invited by journal
09 Feb, 2024
Editor assigned by journal
06 Feb, 2024
First submitted to journal
05 Feb, 2024

You are reading this latest preprint version

Enhancing Bio-Based Concrete Mechanical Properties: A Novel Approach with Composite Sandwiches and Confined Cylinders

Status:

Journal Publication

Version 1

Abstract

Figures

1 Introduction

2 Materials and Methods

2.1 Raw Materials

2.1.1 Hemp shives

2.1.2 Mineral binders

2.1.2.1 Calcium sulfoaluminate cement (CSA)

2.1.2.2 Prompt natural cement (PNC)

2.1.3 Flax Textile

2.2 Specimen preparation

2.2.1 Hemp concrete design

2.2.2 Reinforced hemp concrete fabrication

2.2.2.1 Bending reinforcement

2.2.2.2 Compressive reinforcement

2.3 Experimental configurations

2.4 Test setups

3 Results and discussion

3.1 Mechanical Behavior

3.1.1 Flexural Behavior

3.1.2 Compression Behavior

3.2 Failure and cracking mode

3.2.1 Bending-reinforced specimens

3.2.2 Compression-reinforced specimens

3.3 Mechanical properties

3.3.1 Bending-reinforced specimens

3.3.2 Compression-reinforced specimens

4 Conclusion

Declarations

Declaration of Competing Interest

Data availability statement

References

Status:

Journal Publication

Version 1