New Lime-Based Hybrid Composite of Sugarcane Bagasse and Hemp as Aggregates

: Bio-based materials help reduce the consumption of non-renewable resources, contributing to the development of sustainable construction. Industrial Hemp Concrete (IHC), which uses hemp stalk (HS) as an aggregate and a lime-based binder, is a bio-based material with various applications. This research developed a new hybrid composite in order to improve the mechanical strength and durability of hemp concrete, with the incorporation of sugarcane bagasse (SCB) as an aggregate, a resource of a renewable origin that is abundant in several countries. Different formulations were used, which were molded and pressed manually, evaluating their cohesion and compactness. The performance of the developed hybrid composite was measured considering mechanical, thermal, and durability properties. The compression test results showed an increase of 19–24% for composites with 75% hemp and 25% SCB. Thermal conductivity and thermal resistance coefﬁcients were also improved, reaching 0.098 (W/m ◦ C) and 0.489 (m 2 ◦ C/W), respectively. This aggregate combination also showed the lowest water absorption coefﬁcient (reducing by 35%) and the best performance in durability tests compared to IHC. The resistance to freeze–thaw is highlighted, increasing 400%. The main reason is the inﬂuence of the SCB addition because the short and thin ﬁber form helps to maintain the physical integrity of the composite by ﬁlling the spaces between the hemp aggregates.


Introduction
The construction sector plays a major role in the CO 2 emissions produced by all activities related to the construction process (around 40%) [1]. Therefore, it is necessary to look for alternative materials, technologies, and construction methods that significantly reduce the consumption of resources and energy by providing better energy efficiency without causing pollution and harming health and ecosystems. In this sense, plant components could be used as a possible input to lighten concrete mixes [2].
Since the early nineties, a new building material was obtained by mixing hemp particles (the fibrous non-fraction of the hemp stalk called "hurds"), and a lime-based binder was developed. This allowed for the construction of sustainable buildings (for the construction of new buildings or the renovation of existing buildings) as a filler material for a load-bearing structure [3]. Since then, Industrial Hemp Concrete (HC) has been widely studied in construction as a versatile bio-based material option; it is "carbon-negative" because the hemp plant absorbs more carbon from the atmosphere than it emits during its production and application on site [4] and captures carbon through the carbonation process of lime. According to Jami et al. [5], it is a cellulosic aggregate concrete using hemp as aggregate and a lime-based binder, which could be referred to as cellulosic aggregate concrete.
However, small changes in its composition and manufacturing process can cause many variations in the product result, which can be a barrier to widespread its utilization [4]. Therefore, considering the chemical and physical properties of agricultural waste ash (AWA) and its worldwide availability, the effective use of alternative cementitious materials in large quantities is recommended. In this way, the environmental impact can be further reduced by reducing clinker production and the open burning of agricultural waste [36]. The availability of bagasse ash is very high, as can be seen from the statistics, for use as alkali activated binder precursors or as a blending material in cement production [21].
For this reason, we also take Souza's study [37]. SCB, in combination with lime as a binder, showed the best results in terms of mechanical strength; 0.495 MPa of maximum compressive strength and 0.76 MPa of flexural strength were obtained with the lime addition, with a composition of 30 to 35% SCB and 70 to 65% lime, experimenting also with other materials such as soil and binders such as metakaolin in combination with lime. Additionally, Kumar et al. [13] added SCB as a fiber for soil blocks, and the authors concluded that stabilized soil blocks containing 0.5% or 1.0% by weight of SCB and 12% by weight of cement provided the best compromise between strength and durability, meeting the recommendations corresponding to the design standards, representing a promising solution for the construction of affordable, ecological, and low-rise housing.
With respect to SCBA studies, such as from Charitha et al. [34], we characterize it with high pozzolanic capacity due to its high content of SiO 2 , Al2O 3 , and Fe 2 O 3 , totaling more than 70%, classified according to ASTM C618, proving its effectiveness in studies from Dineshkumar [16], Vasudha [17], and Bahurudeen [38] where SCBA was partially added in proportions between 20 and 30% to replace the role cement with cementitious material, evaluating the durability properties, such as water absorption, carbonation, and corrosion resistance. The results obtained showed adequate support to the replacement.
In the present study, based on the master's thesis of Zúniga [39], in the search for improvement of mechanical properties and durability of the hempcrete (IHC), the combination of industrial hemp (IH) and sugar cane bagasse (SCB) as an aggregate was studied. Different formulations were made with the addition of binders such as hydrated lime, brick dust, and sugarcane bagasse ash (SCBA) to evaluate their behavior in terms of mechanical properties, thermal behavior, and durability.

Materials
The hemp used was a commercial crushed and pre-corticated hemp stem (also known as shiv), cultivated and processed in France by the Chanvrière de I'Aube (LCDA) company. Its physical and chemical properties, according to the technical data sheet provided by the manufacturer, can be seen in Table 1. No pretreatment was carried out for its application. The length of hemp used varied mainly between 10 and 20 mm, with diameters ranging mainly from 1 to 4 mm.
The sugar cane bagasse used, of the Yuba variety, from Madeira Island, at the sugar mill, was subjected to a milling process in a 3-cylinder mill; in the laboratory, it was only air-dried (20 • C and 50% RH), and no treatment was applied. See Figure 1.
The length of SCB varied between 10 and 30 mm, but most were close to 15 mm, with diameters ranging from 0.2 to 0.5 mm. SCB was typically constituted by cellulose (50%), hemicellulose (25%), and lignin (20%) [37]. However, its chemical composition could also be observed by analyzing its ash (SCBA), as shown in Table 2. It was mainly composed of silica, iron, aluminum, and calcium.
The lime used was calcium hydroxide or calcic lime, CL (commonly known as slaked or hydrated lime), named Lusical H100 (CL 90-S), with more than 80% free lime and 5% magnesium oxide, which is for construction and civil engineering applications or materials. As a complement, SCBA was used as pozzolan. It is a residue from the production of sugar and honey by the Montelimar Corporation through its cogeneration plant Green Power S.A. for the generation of clean electricity from biomass. Brick dust (BD) from construction waste was also used, which was processed at the Construction Materials Laboratory of the University of Minho. It is important to highlight that we sought to use materials with a Resources 2023, 12, 55 4 of 20 low environmental impact, which was why cement and mineral aggregates have not been included in the studied compositions. In Figure 2, there is the presentation of the lime used (CL) and the two additional binders added, sugar cane bagasse ash (SCBA and brick powder (BD) and Table 2 shows the chemical composition of these. The sugar cane bagasse used, of the Yuba variety, from Madeira Island, at the mill, was subjected to a milling process in a 3-cylinder mill; in the laboratory, it was air-dried (20 °C and 50% RH), and no treatment was applied. See Figure 1. The length of SCB varied between 10 and 30 mm, but most were close to 15 mm diameters ranging from 0.2 to 0.5 mm. SCB was typically constituted by cellulose ( hemicellulose (25%), and lignin (20%) [37]. However, its chemical composition coul be observed by analyzing its ash (SCBA), as shown in Table 2. It was mainly compo silica, iron, aluminum, and calcium.
The lime used was calcium hydroxide or calcic lime, CL (commonly known as s or hydrated lime), named Lusical H100 (CL 90-S), with more than 80% free lime an magnesium oxide, which is for construction and civil engineering applications or m als. As a complement, SCBA was used as pozzolan. It is a residue from the product Resources 2023, 12, x FOR PEER REVIEW 5 sugar and honey by the Montelimar Corporation through its cogeneration plant G Power S.A. for the generation of clean electricity from biomass. Brick dust (BD) from struction waste was also used, which was processed at the Construction Materials L atory of the University of Minho. It is important to highlight that we sought to use rials with a low environmental impact, which was why cement and mineral aggre have not been included in the studied compositions. In Figure 2, there is the presen of the lime used (CL) and the two additional binders added, sugar cane bagasse ash ( and brick powder (BD) and Table 2 shows the chemical composition of these.

Compositions
To select the reference mixtures of hemp concrete, different studies were considered, indicating better mechanical results with the use of lime [23,40] in a hemp/lime ratio of 25-75% to 30-70% by mass, respectively.
Based on a previous experimental process carried out in the laboratory and the references studied in the state of the art, 8 mixtures with the best compressive strength results, considering the lower percentage of lime. It began to be used a 25/75% aggregates/binders ratio, after a 30/70% ratio in order to try a binder reduction. However, for better results, a 25/75% ratio was used in the last mixture. The lime amount was also reduced with a partial substitution of other binders. The mixes that were combined with the two additional binders (BD-SCBA) and the mechanical strength performance improved with the different % addition of SCB to the hemp concrete. The mixtures only with brick dust (BD) or with SCB ash (SCBA) were tested to evaluate the adhesion of these materials with the aggregates and their isolated performance in terms of resistance and durability. The last mixture use lime and SCBA without BD since no advantages were found in the addition of this material. Table 3 identifies and shows the mixtures studied. The mixes identification is according to the materials abbreviations with the initial letter of the binder.

Methods Applied
The procedures of the mixture preparation were based on Murphy et al.'s [22] research, where no pre-treatment of hemp wetting was carried out since they did not contribute significant benefits to the properties. The binders were mixed in their powder presentation, and then a part of water was added until a pasty consistency was obtained. Then, the aggregates were added gradually, until the paste covered the aggregates during an average mixing time of 6 min, evaluating the consistency of the mixture with the manual elaboration of a ball. If the ball maintained the shape and presented good cohesion, it meant that the workability of the material was adequate to produce the specimens, as has been used by Sousa to aid in the verification of the adhesion of materials [37], see Figure 3.

Methods Applied
The procedures of the mixture preparation were based on Murphy et al.'s [22] research, where no pre-treatment of hemp wetting was carried out since they did not contribute significant benefits to the properties. The binders were mixed in their powder presentation, and then a part of water was added until a pasty consistency was obtained. Then, the aggregates were added gradually, until the paste covered the aggregates during an average mixing time of 6 min, evaluating the consistency of the mixture with the manual elaboration of a ball. If the ball maintained the shape and presented good cohesion, it meant that the workability of the material was adequate to produce the specimens, as has been used by Sousa to aid in the verification of the adhesion of materials [37], see Figure 3.  It was then placed in oiled-prior metal molds, lightly pressed, and vibrated for 40 sec. After 24 h of curing (20 °C and 50% RH), it was removed from the molds and placed in a laboratory environment (20 °C and 50% RH) for 28 days. The molds were prepared in It was then placed in oiled-prior metal molds, lightly pressed, and vibrated for 40 sec. After 24 h of curing (20 • C and 50% RH), it was removed from the molds and placed in a laboratory environment (20 • C and 50% RH) for 28 days. The molds were prepared in measurements of 40 × 40 × 160 mm, 50 × 50 × 50 mm, and 150 × 150 × 50 mm, according to the requirements of the different tests that were carried out.

Compressive Strength Performance
In the case of hemp concrete, there are no standards to evaluate its mechanical resistance. This study takes the BS EN 1015-11 2019 [41] (Compression) standard. Adjusting the conditions of the material with an application of 20N/s by displacement of the force applied on the specimen until the rupture of the same. For the application of this test, specimens of 50 × 50 × 50 mm were tested with a curing time of 28 days (3 specimens for each mixture). The Figure 4 shows the specimens before the test and one during the test.

Compressive Strength Performance
In the case of hemp concrete, there are no standards to evaluate its mechanical resistance. This study takes the BS EN 1015-11 2019 [41] (Compression) standard. Adjusting the conditions of the material with an application of 20N/s by displacement of the force applied on the specimen until the rupture of the same. For the application of this test, specimens of 50 × 50 × 50 mm were tested with a curing time of 28 days (3 specimens for each mixture). The Figure 4 shows the specimens before the test and one during the test.

Thermal Conductivity Test
The Prototherm prototype [42] was used to evaluate the thermal properties. The methodology followed for this test was the internal procedure of the LMC (Laboratory of Construction Materials of the Department of Civil Engineering of the University of Minho), adapted from ISO-9869-1994 [43]. Figure 5 shows the prototherm with the heat source and a specimen to be evaluated. The test consists of placing a heat source and heat sink at the trapezoidal end towards a 60 cm wide rectangular section. Temperature sensors

Thermal Conductivity Test
The Prototherm prototype [42] was used to evaluate the thermal properties. The methodology followed for this test was the internal procedure of the LMC (Laboratory of Construction Materials of the Department of Civil Engineering of the University of Minho), adapted from ISO-9869-1994 [43]. Figure 5 shows the prototherm with the heat source and a specimen to be evaluated. The test consists of placing a heat source and heat sink at the trapezoidal end towards a 60 cm wide rectangular section. Temperature sensors were arranged to measure the temperatures of the environment, the heat source, the heat sink, and the two internal environments: internal channel 1 and 2.

Thermal Conductivity Test
The Prototherm prototype [42] was used to evaluate the thermal properties. The methodology followed for this test was the internal procedure of the LMC (Laboratory of Construction Materials of the Department of Civil Engineering of the University of Minho), adapted from ISO-9869-1994 [43]. Figure 5 shows the prototherm with the heat source and a specimen to be evaluated. The test consists of placing a heat source and heat sink at the trapezoidal end towards a 60 cm wide rectangular section. Temperature sensors were arranged to measure the temperatures of the environment, the heat source, the heat sink, and the two internal environments: internal channel 1 and 2. In this test, the heat source was set to 200 °C to obtain an average internal inlet channel temperature of 39-40 °C constant for 24 h. For the application of this test, one sample of 150 × 150 × 50 mm per mix was tested with a curing time of 60 days because a higher percentage of carbonation had already occurred. In this test, the heat source was set to 200 • C to obtain an average internal inlet channel temperature of 39-40 • C constant for 24 h. For the application of this test, one sample of 150 × 150 × 50 mm per mix was tested with a curing time of 60 days because a higher percentage of carbonation had already occurred.

Coefficient of Water Absorption by Capillarity
The capillarity test was performed in accordance with BS EN 1015-18 [44] and C 1585-04 [45] for the determination of the capillary water absorption coefficient of hardened concrete and for the measurement of the water absorption rate of hydraulic concrete, taking into account the materials characteristics of in this study.
As a preliminary procedure, the samples were subjected to a period of drying in an oven at 60 • C, until a constant mass, after which a silicone coating was applied on the lateral faces, leaving only the upper and lower faces free, the latter being exposed to a humid surface. Weight control was carried out on a time scale until constant values were reached in the defined period. For the application of this test, the dimensions of the samples were 50 × 50 × 50 mm, with a curing time of 60 days (3 specimens for each mixture), see Figure 6.
Resources 2023, 12, x FOR PEER REVIEW 8 of 22

Coefficient of Water Absorption by Capillarity
The capillarity test was performed in accordance with BS EN 1015-18 [44] and C 1585-04 [45] for the determination of the capillary water absorption coefficient of hardened concrete and for the measurement of the water absorption rate of hydraulic concrete, taking into account the materials characteristics of in this study.
As a preliminary procedure, the samples were subjected to a period of drying in an oven at 60 °C, until a constant mass, after which a silicone coating was applied on the lateral faces, leaving only the upper and lower faces free, the latter being exposed to a humid surface. Weight control was carried out on a time scale until constant values were reached in the defined period. For the application of this test, the dimensions of the samples were 50 × 50 × 50 mm, with a curing time of 60 days (3 specimens for each mixture), see Figure 6.

Freezing and Thawing Resistance
For hemp concrete, there is no standard for the application of this type of test; thus, ASTM C 666/C 666M-03 [46] was followed, adapting it to the material conditions. Previously, the mixtures were subjected to a mass control to characterize the initial conditions of the samples. In the freeze-thaw chamber, the samples were placed on a surface that allowed contact with water on only one side of the sample, with a water level of 15 mm, to produce the absorption effect of the fibers, considering that it is an outdoor side that is generally exposed to climatic variations.

Freezing and Thawing Resistance
For hemp concrete, there is no standard for the application of this type of test; thus, ASTM C 666/C 666M-03 [46] was followed, adapting it to the material conditions. Previously, the mixtures were subjected to a mass control to characterize the initial conditions of the samples. In the freeze-thaw chamber, the samples were placed on a surface that allowed contact with water on only one side of the sample, with a water level of 15 mm, to produce the absorption effect of the fibers, considering that it is an outdoor side that is generally exposed to climatic variations.
The samples were subjected to freezing and thawing cycles of 21,600 s for each cycle, and, in Figure 7, you can see the chamber used in the process until the specimen was visibly destroyed, for which 24 h visual controls were carried out. For the application of this test, the dimensions of the samples were 50 × 50 × 50 mm, being 3 specimens per mixture, with a curing time of 60 days.

Freezing and Thawing Resistance
For hemp concrete, there is no standard for the application of this type of test; thus, ASTM C 666/C 666M-03 [46] was followed, adapting it to the material conditions. Previously, the mixtures were subjected to a mass control to characterize the initial conditions of the samples. In the freeze-thaw chamber, the samples were placed on a surface that allowed contact with water on only one side of the sample, with a water level of 15 mm, to produce the absorption effect of the fibers, considering that it is an outdoor side that is generally exposed to climatic variations.
The samples were subjected to freezing and thawing cycles of 21,600 s for each cycle, and, in Figure 7, you can see the chamber used in the process until the specimen was visibly destroyed, for which 24 h visual controls were carried out. For the application of this test, the dimensions of the samples were 50 × 50 × 50 mm, being 3 specimens per mixture, with a curing time of 60 days.

Degradation by Exposure to Simulated Rain
Considering the characteristics of the materials in this study, the accelerated test proposal of Rezende et al. [47] was used to evaluate the degradation by rain simulation. The exposure time (5 h) was the equivalent of 100 years, with mass recording before and after the test to calculate the percentage of rain absorption and erosion wear.
The effects were assessed by a mechanical flexural strength test, according to BS EN 1015-11 2019 (Flexure) [41], compared to samples that had not been subjected to this simulation, in addition to a weight control before and after the test. For the application of this test, the dimensions of the specimens were 150 × 150 × 50 mm, with 3 specimens for each mixture, with a curing time of 60 days. A coating of a dry mortar formulated from natural hydraulic lime was applied. Figure 8 shows the test process.
Resources 2023, 12, x FOR PEER REVIEW 9 of 22

Degradation by Exposure to Simulated Rain
Considering the characteristics of the materials in this study, the accelerated test proposal of Rezende et al. [47] was used to evaluate the degradation by rain simulation. The exposure time (5 h) was the equivalent of 100 years, with mass recording before and after the test to calculate the percentage of rain absorption and erosion wear.
The effects were assessed by a mechanical flexural strength test, according to BS EN 1015-11 2019 (Flexure) [41], compared to samples that had not been subjected to this simulation, in addition to a weight control before and after the test. For the application of this test, the dimensions of the specimens were 150 × 150 × 50 mm, with 3 specimens for each mixture, with a curing time of 60 days. A coating of a dry mortar formulated from natural hydraulic lime was applied. Figure 8 shows the test process.

Saline Exposure
The salt exposure test was performed in accordance with BSI 3900-F2 1973 [48] for the determination of resistance to moisture, for the neutral salt test, adjusting the procedures to the type of material.
The cure time of the samples was 28 days, and the exposure time was 60 days. The temperature range of the experiment was between 24 and 48 °C, with cycles of 24 h to  The salt exposure test was performed in accordance with BSI 3900-F2 1973 [48] for the determination of resistance to moisture, for the neutral salt test, adjusting the procedures to the type of material.
The cure time of the samples was 28 days, and the exposure time was 60 days. The temperature range of the experiment was between 24 and 48 • C, with cycles of 24 h to subject the samples to the vapor of the liquid medium previously prepared, in 1.5 L of drinking water and 200 g of coarse salt.
The method to evaluate the effect of the salt vapor consisted of testing the mechanical resistance to compression, taking up the BS EN 1015-11 2019 [41] standards (compression), and comparing the results with those of the samples that were not subjected to this test with the same curing time. Weight control was performed at the beginning and at the end of the test. For the application of this test, the dimensions of the specimens were 50 × 50 × 50 mm, with 3 specimens per mixture, see Figure 9.

Compositions (Fresh State)
The fresh state of the mixtures, shown in Table 3, was visually evaluated using an optical microscope (indicate magnification). Figure 10 shows the images of the eight mixes, in their fresh state, which allowed for the observation of spaces between the fibers, the integration, and the coating of the binders on the aggregates.
In the HL-ref concrete composition (Figure 10a), the hydrated lime adhered to the hemp particles and filled the spaces between them, making the mixture more cohesive. It was found that it was visible when there was greater coverage of the aggregates because there was better adhesion of the binder to these, and they provided better cohesion in the mixtures and in filling the voids. In the formation of the new composite, SCB pieces filled these spaces and bonded the hemp shavings together, creating a network between the spaces. It was also noted that the BD did not mix homogeneously with the rest of the binders, presenting small concentrations of BD (Figure 10c, f-h).

Compositions (Fresh State)
The fresh state of the mixtures, shown in Table 3, was visually evaluated using an optical microscope (indicate magnification). Figure 10 shows the images of the eight mixes, in their fresh state, which allowed for the observation of spaces between the fibers, the integration, and the coating of the binders on the aggregates.

Compositions (Fresh State)
The fresh state of the mixtures, shown in Table 3, was visually evaluated using an optical microscope (indicate magnification). Figure 10 shows the images of the eight mixes, in their fresh state, which allowed for the observation of spaces between the fibers, the integration, and the coating of the binders on the aggregates.
In the HL-ref concrete composition (Figure 10a), the hydrated lime adhered to the hemp particles and filled the spaces between them, making the mixture more cohesive. It was found that it was visible when there was greater coverage of the aggregates because there was better adhesion of the binder to these, and they provided better cohesion in the mixtures and in filling the voids. In the formation of the new composite, SCB pieces filled these spaces and bonded the hemp shavings together, creating a network between the spaces. It was also noted that the BD did not mix homogeneously with the rest of the binders, presenting small concentrations of BD (Figure 10c, f-h).  Figure 11 shows the compressive strength results for the mixtures. Considering the reference situation (HL-ref), one can observe that there is no loss in the mechanical re- In the HL-ref concrete composition (Figure 10a), the hydrated lime adhered to the hemp particles and filled the spaces between them, making the mixture more cohesive. It was found that it was visible when there was greater coverage of the aggregates because there was better adhesion of the binder to these, and they provided better cohesion in the mixtures and in filling the voids. In the formation of the new composite, SCB pieces filled these spaces and bonded the hemp shavings together, creating a network between the spaces. It was also noted that the BD did not mix homogeneously with the rest of the binders, presenting small concentrations of BD (Figure 10c,f-h). The satisfactory performance in composites based on lime was already shown in other research [23,40,49,50], with an average compressive strength of 0.48 MPa, which meant that that there was a better compressive strength at a proportion of more than 70% of the binder. In the mixtures where there was no hydrated lime as binder, HS-B and HS-A, there was a sharp drop in the compressive strength, which, in a certain way, was already expected since pozzolanic materials were effectively efficient when combined with calcium hydroxide [51]. Considering the mixtures that used three different binders (hydrated lime, brick dust, and ash) and aggregates (hemp and SCB) divided in different ratios (with 25, 50, and 75% of SCB), it could be seen that the higher the addition of SCB the lower the resistance. In addition, the compressive strength was unsatisfactory, varying between 0.05 and 0.11 MPa. Figure 12 shows the state of the test specimens after the application of the compression test. The satisfactory performance in composites based on lime was already shown in other research [23,40,49,50], with an average compressive strength of 0.48 MPa, which meant that that there was a better compressive strength at a proportion of more than 70% of the binder.

Compressive Strength Performance
In the mixtures where there was no hydrated lime as binder, HS-B and HS-A, there was a sharp drop in the compressive strength, which, in a certain way, was already expected since pozzolanic materials were effectively efficient when combined with calcium hydroxide [51]. Considering the mixtures that used three different binders (hydrated lime, brick dust, and ash) and aggregates (hemp and SCB) divided in different ratios (with 25, 50, and 75% of SCB), it could be seen that the higher the addition of SCB the lower the resistance. In addition, the compressive strength was unsatisfactory, varying between 0.05 and 0.11 MPa. Figure 12 shows the state of the test specimens after the application of the compression test.
In other study, it was also shown that brick dust did not improve the mechanical properties of hemp concrete, as concluded by Sinka and Sahmenko (2013) [52] in their study using hemp composites with hydraulic lime and ground brick as pozzolanic mix. It showed little hydraulic activity and could only be used as a substitute for lime for microfilling, reducing costs, but not to increase compressive strength. The reductions in strength with pozzolans can be justified as the aggregate pozzolanic performance reducing when its fineness was smaller than the size of the aggregate. In addition, at early stages, the pozzolan had an almost single filler effect, and the greatest increase in the strength by pozzolanic reaction only was clearly observed at 180 and 850 days [53]. In the mixtures where there was no hydrated lime as binder, HS-B and HS-A, there was a sharp drop in the compressive strength, which, in a certain way, was already expected since pozzolanic materials were effectively efficient when combined with calcium hydroxide [51]. Considering the mixtures that used three different binders (hydrated lime, brick dust, and ash) and aggregates (hemp and SCB) divided in different ratios (with 25, 50, and 75% of SCB), it could be seen that the higher the addition of SCB the lower the resistance. In addition, the compressive strength was unsatisfactory, varying between 0.05 and 0.11 MPa. Figure 12 shows the state of the test specimens after the application of the compression test.  In this case, not even the combination of BD with the ash of SCB showed efficiency. In addition to that, as already mentioned, the mixtures in the fresh state showed little cohesion and particle aggregation.
Another important issue is the aggregates/binders' ratio, which was once in the mixtures with brick and ash at a 30/70% ratio, which was tested to assess the possibility of reducing the amount of binder material. However, due the little cohesion observed, a final composition was made with lime and ash only, again with the ratio of the initial compositions of reference 25/75% ratio, and, as shown, the results were better and higher than the reference mixtures only with lime, as mentioned before.

Thermal Conductivity Test
The comparison of the results between the mixtures shows that the mixtures, based on pozzolans, have the lowest thermal conductivities, confirming the results of Abdellatef's study [54] that added 10% crushed brick to the hempcrete and had the lowest thermal conductivity. As can be seen in Figure 5, the HS25-L-B-A, HS50-L-B-A, and HS75-L-B-A mixtures presented values very close to each other, 0.089 W/m • C on average, while the mixture based only on lime and SCBA was closer to the values obtained for the H-L-ref and HS-L mixtures, around 0.11 W/m • C, whose results were compatible with those found in the literature, between 0.06 and 0.14 W/m • C [54].
These were also comparable with the results of P. Leão and E. Araújo, with a conductivity coefficient of 0.11 and 0.12 W/m • C, with a similar ratio of aggregates/binders (25/75%).
The results obtained from the tests showed that none of the mixtures were within the parameters to be considered conventional thermal insulating materials, according to Portuguese regulations. According to ITE50 [55], an insulation material must have a thermal conductivity lower than 0.065 W/m • C (blue line in Figure 6) and a thermal resistance higher than 0.030 m 2 • C/W. Regarding thermal resistance, Figure 13 shows that all the mixtures had good thermal resistance, with results above 0.030 m 2 • C/W (red line in Figure 5), with an average of 0.533 m 2 • C/W. However, these rules were made for materials only with thermal functions, but the objective of this material was to be used to substitute the set masonry/insulation or other non-structural wall materials that do not need additional insulation. As such, these thermal conductivity values are closer to these materials: aerated concrete (0.16); calcium silicate board (0.17); prefabricated timber wall panels (0.12); and timber blocks (650 kg/m 3 ) (0.14) [56]. The smaller this value is, the better its behavior is, and it could be considered that the developed composite presented good thermal performance.  Figure 14 shows the capillary water absorption coefficients of the mixtures studied. It was observed that the lime-based mixes without pozzolan materials presented the closest coefficients, with an average of 4.78 kg/(m² h 0.5 ). In the mixes that were based on the three and two binders, the absorption coefficient decreased by 45.73% when the three binders studied were mixed and only around 30% with the SCBA mixes (HS-A and HS-L-A), with respect to the lime-based mixes.

Coefficient of Water Absorption by Capillarity
In the HS25-L-B-A, HS50-L-B-A, HS75-L-B-A, and HS-L-A mixes, the effect of SCBA in combination with lime could be observed, the percentage of water absorption was directly related to the permeability and the porosity of the sample, due to the porosity of the sample, due to the pozzolanic performance of the SCBA, and it had the same effect as in the high-performance concrete, it decreased the water absorption and porosity of the mixes [10]. The added 25% percent of ash could be considered as being effective, as it improved performance by decreasing water absorption. Ganesan et al. used 20% from SCBA also improved the water absorption behavior, due to the formation of a dense microstructure in the mixed concrete from 20% [57].    Figure 14 shows the capillary water absorption coefficients of the mixtures studied. It was observed that the lime-based mixes without pozzolan materials presented the closest coefficients, with an average of 4.78 kg/(m 2 h 0.5 ). In the mixes that were based on the three and two binders, the absorption coefficient decreased by 45.73% when the three binders studied were mixed and only around 30% with the SCBA mixes (HS-A and HS-L-A), with respect to the lime-based mixes.  Figure 14 shows the capillary water absorption coefficients of the mixtures studied. It was observed that the lime-based mixes without pozzolan materials presented the closest coefficients, with an average of 4.78 kg/(m² h 0.5 ). In the mixes that were based on the three and two binders, the absorption coefficient decreased by 45.73% when the three binders studied were mixed and only around 30% with the SCBA mixes (HS-A and HS-L-A), with respect to the lime-based mixes.

Coefficient of Water Absorption by Capillarity
In the HS25-L-B-A, HS50-L-B-A, HS75-L-B-A, and HS-L-A mixes, the effect of SCBA in combination with lime could be observed, the percentage of water absorption was directly related to the permeability and the porosity of the sample, due to the porosity of the sample, due to the pozzolanic performance of the SCBA, and it had the same effect as in the high-performance concrete, it decreased the water absorption and porosity of the mixes [10]. The added 25% percent of ash could be considered as being effective, as it improved performance by decreasing water absorption. Ganesan et al. used 20% from SCBA also improved the water absorption behavior, due to the formation of a dense microstructure in the mixed concrete from 20% [57].   In the HS25-L-B-A, HS50-L-B-A, HS75-L-B-A, and HS-L-A mixes, the effect of SCBA in combination with lime could be observed, the percentage of water absorption was directly related to the permeability and the porosity of the sample, due to the porosity of the sample, due to the pozzolanic performance of the SCBA, and it had the same effect as in the high-performance concrete, it decreased the water absorption and porosity of the mixes [10]. The added 25% percent of ash could be considered as being effective, as it improved performance by decreasing water absorption. Ganesan et al. used 20% from SCBA also improved the water absorption behavior, due to the formation of a dense microstructure in the mixed concrete from 20% [57]. Figure 15 shows the behavior of the samples over time, in contact with the water, until stability is reached. HS25-L-B-A, HS50-L-B-A, HS75-L-B-A, and HS-L-A samples present a linear trend, with an imperceptible variation, achieving stability after 5 √ min, while the other mixtures only achieve a linear stability until after 22 √ min. This indicates that these mixtures will have a larger open porosity that will facilitate water passage in the early stages of the test.  Concerning the results with brick dust and ash mixtures, although other authors mentioned that the mortar matrix structure only became denser with the formation of the hydration components of pozzolanic activity, decreasing the porosity [53], in this case, the reduced absorption at an early age, also showed a filler effect on the voids. These materials mixes could act as water retainers once with the introduction of small pores that could cut the capillary network [58]. However, for a better understanding of the direct relationship between the absorption and porosity of these composite materials, further tests would have to be carried out, such as absorption by immersion.

Freezing and Thawing Resistance
For the application of the test, the samples were grouped according to binders. The H-L-ref and HS-L mixtures were subjected to seven freeze-thaw cycles and had a density loss of 8.93% and 6.09%, respectively. The samples showed cracks at the level of visible water absorption, between 25 and 35 mm. The H-L-ref mix was the most damaged, losing its shape and part of the material, while HS-L maintained its dimensions and presented narrower cracks. Both mixtures presented cracks on the upper face and lost the binder at their bases, leaving the aggregates exposed. Figure 16 shows the level of degradation after freeze/thaw cycles of five blends (the beginning-top figure; and the end-bottom figure), completed 30.95 freeze/thaw cycles, and had an average density loss of 2.37%. The HS-A (Figure 17e) blend had the highest density loss (20.47%), reaching the end of the test full of water with completely disintegrated areas. The HS50-L-B-A (Figure 17c) mix presented the lowest loss (1.35%), with a composition of 50% SCB as an aggregate. It can be observed that the sample of the HS-L-A (Figure 17d) mix maintained its dimensions but was completely wetted with a density loss of 1.59%. In general, the mixes maintained their shapes, resisted the cycles, and did not present cracks, but some samples ended up without binder coverage, with exposed fibers or lost part of the material. Concerning the results with brick dust and ash mixtures, although other authors mentioned that the mortar matrix structure only became denser with the formation of the hydration components of pozzolanic activity, decreasing the porosity [53], in this case, the reduced absorption at an early age, also showed a filler effect on the voids. These materials mixes could act as water retainers once with the introduction of small pores that could cut the capillary network [58]. However, for a better understanding of the direct relationship between the absorption and porosity of these composite materials, further tests would have to be carried out, such as absorption by immersion.

Freezing and Thawing Resistance
For the application of the test, the samples were grouped according to binders. The H-L-ref and HS-L mixtures were subjected to seven freeze-thaw cycles and had a density loss of 8.93% and 6.09%, respectively. The samples showed cracks at the level of visible water absorption, between 25 and 35 mm. The H-L-ref mix was the most damaged, losing its shape and part of the material, while HS-L maintained its dimensions and presented narrower cracks. Both mixtures presented cracks on the upper face and lost the binder at their bases, leaving the aggregates exposed. Figure 16 shows the level of degradation after freeze/thaw cycles of five blends (the beginning-top figure; and the end-bottom figure), completed 30.95 freeze/thaw cycles, and had an average density loss of 2.37%. The HS-A (Figure 17e) blend had the highest density loss (20.47%), reaching the end of the test full of water with completely disintegrated areas. The HS50-L-B-A (Figure 17c) mix presented the lowest loss (1.35%), with a composition of 50% SCB as an aggregate. It can be observed that the sample of the HS-L-A (Figure 17d) mix maintained its dimensions but was completely wetted with a density loss of 1.59%. In general, the mixes maintained their shapes, resisted the cycles, and did not present cracks, but some samples ended up without binder coverage, with exposed fibers or lost part of the material.
The HS-B mix performed separately due to its composition. The samples were completely filled with water, and, although they maintained their shape, they had some loss of material.
The addition of SCBA in combination with lime increased the cementitious properties to form hydrated calcium silicate (the pozzolan effects), confirming the study of Botas et al. (2010) [59] on the freeze-thaw resistance of lime mortars. They concluded that the increase in mechanical strength of air lime and hydraulic lime mortars was generally associated with a lower resistance to freeze-thaw cycles due to an increase in porosity, most likely accompanied by a reduction in pore size. Thus, when there was a greater number of pores with smaller dimensions, the performance of mortars against freeze-thaw cycles worsened. The HS-B mix performed separately due to its composition. The samples were completely filled with water, and, although they maintained their shape, they had some loss of material.
The addition of SCBA in combination with lime increased the cementitious properties to form hydrated calcium silicate (the pozzolan effects), confirming the study of Botas et al. (2010) [59] on the freeze-thaw resistance of lime mortars. They concluded that the increase in mechanical strength of air lime and hydraulic lime mortars was generally associated with a lower resistance to freeze-thaw cycles due to an increase in porosity, most likely accompanied by a reduction in pore size. Thus, when there was a greater number of pores with smaller dimensions, the performance of mortars against freeze-thaw cycles worsened. Table 4 shows the results obtained after the rain-erosion degradation test. In general, the mixes had an average mass loss (MLr) of 5.35%, except for the HS-D that showed the smallest mass loss (3.78%) and the highest absorption (Abr) percentage 42.80%. Regarding flexural behavior, the results (F) are for the mixes that were not subjected to the rain simulation test, with a curing time of 28 days plus the period of 14 days for the application of the finish, in total, 42 days before the flexural test, without being subjected to the rain simulation. The results (Fr) are for the mixes that were subjected to the rain erosion degradation test, with the same curing and finishing time, plus 3 days for the test (time of exposure to water and the drying period until reaching a constant mass), totaling 45 days for all samples before performing the flexural test. Comparing these results, in general, there was no significant difference in the flexural strength, mainly for the mixtures The HS-B mix performed separately due to its composition. The samples were completely filled with water, and, although they maintained their shape, they had some loss of material.

Degradation by Exposure to Simulated Rain
The addition of SCBA in combination with lime increased the cementitious properties to form hydrated calcium silicate (the pozzolan effects), confirming the study of Botas et al. (2010) [59] on the freeze-thaw resistance of lime mortars. They concluded that the increase in mechanical strength of air lime and hydraulic lime mortars was generally associated with a lower resistance to freeze-thaw cycles due to an increase in porosity, most likely accompanied by a reduction in pore size. Thus, when there was a greater number of pores with smaller dimensions, the performance of mortars against freeze-thaw cycles worsened. Table 4 shows the results obtained after the rain-erosion degradation test. In general, the mixes had an average mass loss (MLr) of 5.35%, except for the HS-D that showed the smallest mass loss (3.78%) and the highest absorption (Abr) percentage 42.80%. Regarding flexural behavior, the results (F) are for the mixes that were not subjected to the rain simulation test, with a curing time of 28 days plus the period of 14 days for the application of the finish, in total, 42 days before the flexural test, without being subjected to the rain simulation. The results (Fr) are for the mixes that were subjected to the rain erosion degradation test, with the same curing and finishing time, plus 3 days for the test (time of exposure to water and the drying period until reaching a constant mass), totaling 45 days for all samples before performing the flexural test. Comparing these results, in general, there was no significant difference in the flexural strength, mainly for the mixtures  Table 4 shows the results obtained after the rain-erosion degradation test. In general, the mixes had an average mass loss (ML r ) of 5.35%, except for the HS-D that showed the smallest mass loss (3.78%) and the highest absorption (Ab r ) percentage 42.80%. Regarding flexural behavior, the results (F) are for the mixes that were not subjected to the rain simulation test, with a curing time of 28 days plus the period of 14 days for the application of the finish, in total, 42 days before the flexural test, without being subjected to the rain simulation. The results (Fr) are for the mixes that were subjected to the rain erosion degradation test, with the same curing and finishing time, plus 3 days for the test (time of exposure to water and the drying period until reaching a constant mass), totaling 45 days for all samples before performing the flexural test. Comparing these results, in general, there was no significant difference in the flexural strength, mainly for the mixtures without ash, but they show a reduction of around 12% in HS75-L-B-A and HS-L-A, as can be seen in Table 4. However, it was again verified that ashes decreased the water absorption capacity of the mixes. In general, the new composite, with SCB incorporation, obtained good results when exposed to the rain simulation for a period of 100 years, compatible with the study of Miller's hemp concrete in his life cycle analysis [60], with an average life of more than 100 years. Figure 17 shows an example of the difference in weight gained after the rain simulation test of the different test specimens.

Degradation by Exposure to Simulated Rain
Comparing the HS-L with the hempcrete mixture (HL-ref), there were no significant differences. Nevertheless, the last mixture HS-L-A showed the lowest water absorption, and the loss of mass did not compromise its effectiveness because the value obtained was higher than that of the other mixtures.

Saline Exposure
The density and compressive strength of the mixtures were evaluated during salt exposure. For density, during the test, the samples did not show large variations, although the trend was that the density increased due to the exposure to moisture.
To evaluate the effect of salt exposure on the samples, Figure 18 shows the results of the mechanical compressive strength tests of the samples in comparison with the results of the samples with 28 days of curing. This improvement in the compressive strength of the mixtures could be attributed to the additional hydration due to the presence of water, as mentioned in the study by   [6]. They used the same temperature and time interval as in the salt exposure test, achieving a significant increase in compressive strength after salt exposure. The authors attributed this increase in additional hydration due to the presence of water and weight gain due to salt crystallization within its pores, suggesting that small pores facilitated salt crystallization. In the case of the HS-L-A mixture, it presented a loss of 31.66% of its resistance and did not present any change in appearance or visible discoloration. In this composition, the percentage of ashes was higher combined with hydrated lime, and exposure to moisture decreased resistance, otherwise, with the other mixtures. Table 5 presents a consolidation of all the results, assigning an evaluation from 1 to 8, with 1 being the best result and 8 the worst result, to determine which of the mixtures had the best results in the mechanical and durability tests.  In the case of the HS-L-A mixture, it presented a loss of 31.66% of its resistance and did not present any change in appearance or visible discoloration. In this composition, the percentage of ashes was higher combined with hydrated lime, and exposure to moisture decreased resistance, otherwise, with the other mixtures. Table 5 presents a consolidation of all the results, assigning an evaluation from 1 to 8, with 1 being the best result and 8 the worst result, to determine which of the mixtures had the best results in the mechanical and durability tests.  Compressive strength performance  3  2  4  8  5  6  7  1  Thermal coefficient  8  7  1  2  4  6  3  5  Thermal Resistance  8  7  2  1  4  5  3  6  Coefficient of water absorption  by capillarity  6  7  5  8  1  3  2  4 Freezing and thawing resistance 6  The mixture with the best performance in the different tests was HS-L-A, with a replacement of hemp by SCB of 25% and also a replacement of lime by SCBA of 25%, with the better results mainly in mechanical performance and in the behavior against environmental agents such as temperature difference and exposure to rain and salts.

Tests H-L-ref HS-L HS-B HS-A HS25-L-B-A HS50-L-B-A HS75-L-B-A HS-L-A
It could also be observed that the substitution of hemp by SCB only showed the best results with the addition of SCB ash.
Additionally, considering the results of the mixtures with 25, 50, and 75% of SCB, it was also noted that there was no direct relationship in the results with the percentage of SCB used. Table 6 shows a comparison of the best performing HS-L-A mixture with some benchmark studies. Although it cannot be directly compared this study, the new composite had good performance, considering that it took two agro-residues and tried to reduce the use of lime.  [3], and Sinka et al. (2014) [63]. In the case of the other authors, they had additional components to their mixtures such as paper pulp and borax. Comparing the mixture of the new composite with the mixture of only SCB with lime, the results were very close; therefore, it is necessary to analyze its advantages with other parameters that contribute to improve both compositions at the same time.
Within the thermal parameters, the new composite has good performance, with an approximate improvement of 22.44%, even in the case of other materials in the mixtures such as paper pulp from Eire's study, and the lime-and natural-cement-based mixtures of Delannoy with a coefficient between 0.10 and 12 W/m • C [60]. In the case of Araújo's results, they were evaluated in the form of plates, with a thickness of 0.0173 m, which makes them not very comparable.
In the case of the durability tests, as the Salim study refers to, it is an area with little attention, except for freezing and thawing. The few studies that have analyzed hemp concrete suggest that it works well in most cases, it can be seen that the loss of density in general is low, and, in this study, the loss was higher by 1.36% than Walker et al. (2014) [6] study. Thus, the recommendation that hemp concrete should be reinforced against freezing and thawing is valid.
In the salt exposure tests, which were evaluated under the same conditions in both studies,   [6] mixtures gained better compressive strengths in some compositions and to a lesser extent in mixtures with SCBA. In this case, as we are comparing as the HS-L-A mixture in this test, it did not perform well and presented a loss of 31.66%. In   [6] study, it gained 16.21% performance, and it is an area for further research to check for a trend.
After 50 ageing cycles in Zerrouki's study [62], the compressive strength decreased by 1.35 to 0.35 MPa, corresponding to 26.76% with the metakaolin mix. In this study, the mix HS-L-A that reduce from 0.39 to 0.32 MPa, it had a reduction of only 14.70%, considering that the samples were subjected to degradation by simulated rainfall for 100 years.

Conclusions
The mixtures with the best mechanical performance were those with 25% aggregate (H-75%, SCB-25%) and 75% binder. An increase of more than 50% in SCB addition reduced compressive strength, as did a decrease in binder percentage to less than 65%. As well as sugar, bagasse ash significantly increased the compressive strength by almost 40%. This confirms the findings of other studies that a decrease in fiber percentage with an increase in binder improves mechanical strength and that SCBA increases the level of pozzolan in the composition. In addition, there was a reduction of 25% in the amount of hydrated lime used since this percentage was replaced by SCBA, which contributes to the sustainability of the composite.
The incorporation of SCB to hemp concrete (HC) to form a new composite confers new properties that improve durability conditions, but they must be in a percentage range in which the SCB does not exceed 50% in the composition.
The SCBA improved the water absorption performance, which allowed good results after the durability tests, for example, improving the resistance to freeze-thaw cycles by 400% more than HC.
The mixtures have good thermal resistance and conductance, considering the function of this material to other non-structural wall materials. The best performing mixtures were HS-B and HS-A, due to the low density of the samples.
It is worth highlighting the direct benefits, as no additional treatment is applied to this natural aggregate SCB, of agro-industrial waste, highly available from the sugar production process. This can improve the mechanical strength and durability conditions of the widely studied hemp concrete without any additive that was not natural.
The improvement of the durability of this lime-based composite is something that allows building owners and builders to increase their confidence in lime-and plant-based composites. However, for future use, it is considered that this composite still needs to be further studied before to be applied.
It is believed that this composite will be able to be used under the same conditions as hemp concrete, using a rendering for protection against the action of water and reducing water absorption. Structurally, a structural material will be necessary, and this composite could be used as non-structural masonry, in blocks or cast and compacted in situ.

Recommendations
For researchers intending to conduct research using this type of material, it is recommended to develop the following topics:

•
Develop microscopic studies to better understand the adherence and reactions between the materials. • Try different percentages of sugar cane bagasse ash.

•
Study the effects of other binders, such as hydraulic lime and metakaolin, in combination with SCBA ash. • Study methods to standardize the compaction process to assess the influence of the compaction process on the mechanical and thermal strength of the mixtures.

•
To carry out pre-treatment procedures of sugarcane bagasse ash to reduce the effects of high water absorption and to be able to control the amount of water in the mixture.

•
Develop tests with real scale walls with blocks or cast and compacted in situ.