Effect of Nano Additive on Mechanical Properties of Natural Fiber Reinforced Soil

ABSTRACT Organic soil is a poor bearing medium and is susceptible to uncharacteristic volume change and settlement behavior. Geotechnical design of substructures in organic soil is often of high economic value to account for its weak geotechnical properties. Fiber reinforcement improves the strength of the soil. However, fibers act as a mechanical reinforcement in soil mass rather than enhancing the intrinsic nature of the soil. But chemical additives alter its behavior, thereby improving its strength. This study investigates the strength and hydraulic conductivity (HC) of low plastic organic silt reinforced with banana fiber and nano-silica. Banana fiber was added in dosages of 0.25%, 0.50%, 0.75% and 1%. The optimum fiber dosage was 0.75%. Nano-silica was added to the 0.75% banana fiber reinforced soil at dosages of 0.2%, 0.4%, 0.6% and 0.8% to improve the strength of the reinforced soil. The effect of aging on HC and unconfined compressive strength (UCS) was studied for a period of 90 d. Results showed that nano-silica and banana fiber beneficially complemented each other in the enhancement of geotechnical properties. Nano-silica increased the strength of fiber-reinforced soil, and banana fiber averted the brittle failure nature imposed by the nano-silica. Also, it improved drainage characteristics through aggregation action.


Introduction
Numerous techniques can be adopted to improve the properties of the soil and includes a variety of options like densifying the soil using compaction techniques or installation of stone columns; draining the water to facilitate consolidation of soil using vertical drains; reinforcing the soil using geosynthetics or fibers; and by the inclusion of additives to the soil such as cement, lime, flyash, polymers and other strength-enhancing materials. The choice of such methods depends on the extent of ground reported that both micro-silica and nano-silica contributed to strength improvement. Kannan and Sujatha (2022) noted that 0.6% nano-silica exhibited a tendency to improve the strength of organic silt by 2.21 times after 180 d due to viscous CSH gel formation. However, they also reported brittle behavior in the treated soil with aging. A thorough literature search clearly indicates that nano-silica can improve soil strength at a relatively lesser dosage but impart brittle failure in the treated sample (Changizi and Haddad 2017;Kannan and Sujatha 2022). Changizi and Haddad (2015) used 1% nano-silica and 0.3% polyester fiber and improved the shear strength by 190%. Boz et al. (2018) infused lime amended clay with basalt and polypropylene fiber. A maximum strength improvement of 8.5 times was achieved after curing for 90 d on 0.75% of 19 mm basalt fiber and 9% lime. Choobbasti, Samakoosh, and Kutanaei (2019) combined carpet waste fibers and nano-calcium carbonate and observed that 0.6% carpet waste fiber and 1.2% nano-calcium carbonate increased the UCS by 50%. Gobinath et al. (2020) showed that the addition of 1% sodium silicate and 0.5% banana fiber improved the UCS by 445%.
Banana fibers are obtained from the pseudo stem of the banana tree and are used for various industrial applications in making paper, natural filters, textiles and rope. They are lightweight with high strength, durable, have less elongation tendency (Adeniyi et al. 2021;Bharathi, Vinodhkumar, and Saravanan 2021), eco-friendly and biodegradable in contrast to synthetic fibers. India is a leading producer of banana at global scale and contributes by nearly 20.08% of worlds banana production. Nano-silica has proved to be an effective additive that can give enhanced strength at a relatively lesser dosage than conventional additives. Research on fibers and nano-silica blended soil are trending in recent times with very less attention being paid to the choice of natural fibers (Cui et al. 2018). The present study investigates the choice of banana fiber and nano-silica to enhance the strength of low plastic silt with organic content. The study reports the effect of the combined treatment on decisive geotechnical parameters such as compaction characteristics, UCS & HC of the fiber-reinforced nanosilica treated soil. The results are supplemented with micro-morphology observations and study on chemical changes using SEM, FESEM, XRD and FTIR to understand the mechanism of modification of geotechnical properties.

Materials and methods
Soil was excavated from 1 m deep trenches made in the agricultural fields at Ariyalur, Tamil Nadu avoiding topsoil and agricultural waste. The soil was black in color, and had a characteristic pungent odor indicating the presence of organic content. Organic content was determined based on loss on ignition method (ASTM D2974 2020). Soil had a organic content of 13.6%. The soil is classified as low plastic silt with organic content (OL) based on the Unified Soil Classification System (USCS). XRF analysis (Table 1) indicated the presence of abundant silica and noticeable calcium oxide indicating the presence of calcium content in the soil. Calcium content was of particular interest as the soil was extracted from an area with abundant limestone mines (Mohan et al. 2021).
Banana fibers of 1 mm average diameter and tensile strength 715 MPa was purchased from Fiber Region, Chennai, India. Banana fibers used for the study were cut to 10 mm length as authors Estabragh, Ranjbari, and Javadi (2017) suggested that fiber length upto 10 mm aids in the increase of UCS but lengths beyond 10 mm causes bundling effect while mixing and impacts the strength adversely. Nano-silica with an average particle size less than 17 nm, 99.88% purity and a specific surface area of 202 m 2 /g was procured from Astraa Chemicals, Chennai, India. Table 2 lists the properties of the soil, banana fiber & nano-silica. Specifications for banana fiber and nano-silica are provided by the supplier.

Experimental investigation
Trial investigations at dosages of 0.25%, 0.50%, 0.75%, 1.00% and 1.25% were conducted to determine the banana fiber dosage and was fixed as 0.25%, 0.50%, 0.75% and 1.00% by dry weight of the soil based on UCS test. UCS tends to decrease beyond 0.75% addition of banana fibers. Nano-silica stabilization was adopted for the optimum fiber dosage of 0.75% and nano-silica content varied as 0.2%, 0.4%, 0.6% and 0.8% after trial testing in the dosage range of 0.2% to 1.00% as UCS registered a decrease beyond 0.6% nano-silica addition. The test plan of the geotechnical investigation is presented in Table 3 and tests were conducted in triplicate to ensure the repeatability of the results. The soil sample was dried at a temperature of 100°C in a thermostatically controlled oven to remove the field moisture, rammed and sieved to the required particle sizes as per the specifications of the experimental investigation. Treated banana fibers were thoroughly hand-mixed to ensure uniform spreading in the soil. In case of the combination study involving banana fiber and nano-silica, the soil was dry-mixed with the predefined quantity of nano-silica manually, followed by the addition of fibers to prevent improper spreading of nano-silica if added along with fibers. The soil was initially divided into ten parts with appropriated quantity of nano additive and dry mixed for uniform blending and were then combined and mixed for ten minutes to achieve a homogeneous mixture after which banana fibers were added.
Microscopic images obtained through a handheld digital microscope with a complementary metaloxide semiconductor sensor having a focus range of 15 mm to 45 mm was used to study the surface cracks developed in the failed sample. Micrographs of soil were also obtained from Tescan Veega 3 scanning electron microscope (SEM) and Zeiss Gemini field emission scanning electron microscope (FESEM) to study the surface morphology after treatment. X-Ray Diffraction (XRD) analysis for a 2θ range of 20° to 80° was carried out to investigate the formation of new compounds using Bruker X-ray diffractometer. Perkin Elmer's FTIR spectrometer was used for analyzing the modifications in the functional group after treatment in the soil. Light compaction tests according to ASTM D698 (2012) were conducted on samples after curing them in airlock bags for a period of two hours. Cylindrical samples of 38 mm diameter and 76 mm height were prepared for unconfined compression (UCC) test at their maximum dry unit weight (MDUW) to determine the strength of modified soil (ASTM D2166 2006) after curing in airlock covers till their respective curing periods, namely 0d (two hours after sample preparation), 7d, 14d, 28d, 56d and 90d. The samples were loaded at a strain rate of 1.25 mm/min. Sample preparation of banana fiber reinforced soil (BFRS) and nano-silica stabilized banana fiber reinforced soil (NBFRS) is illustrated in Figure 1. Hydraulic conductivity (ASTM D5856 2015) of BFRS and NBFRS was investigated for the same curing periods as that of strength. Samples were in continuous saturation for the entire test period to evaluate the prolonged effects.

Compaction characteristics
Compaction characteristics of soil, BFRS and NBFRS were studied to determine the MDUW and OMC (Figure 2). The MDUW of BFRS decreased from 16.8 kN/m 3 for soil to 15.75 kN/m 3 at 1.00% banana fiber addition. The marginal decrease in MDUW can be attributed to the two reasons -(i) replacing heavier soil with lighter fibers and (ii) resistance offered by fibers to compaction effort. A similar trend in MDUW was observed by authors Soltani, Deng, and Taheri (2018) and Murthi, Saravanan, and Poongodi (2020) on polypropylene fiber-reinforced expansive soil and black cotton soil. The OMC was not affected by the addition of banana fibers as the treated fibers do not tend to absorb water.
The soil reinforced with 0.75% banana fiber had an MDUW of 15.95 kN/m 3 and an OMC of 17.5%, as observed from Figure 2a. The addition of nano-silica reduced both MDUW and OMC of the fiberreinforced soil. Figure 2b shows that the MDUW and OMC decreased to 14.6 kN/m 3 and 15.0%, respectively, after treating the reinforced soil with 0.8% nano-silica. The addition of nano-silica along with fiber caused the absorption of water due to the presence of nano-silica particles, owing to its hydrophilic nature and higher specific surface area (Mirzababaei et al. 2021). Further, nano-additives initially filled the voids and then prompted the aggregation of soil particles (Changizi and Haddad 2017), forming a flocculated structure preventing further compression of soil mass at a significantly lesser water content. A similar trend on the reduction of both MDUW and OMC with an increase in dosage of additives was observed by Jasim and Cetin (2016) in clayey silt treated with sawdust, where the sawdust filled the voids and absorbed the moisture from the soil, causing a drop in MDUW and OMC.

Failure strain and failure pattern
The organic soil failed at a strain of 4.61% exhibiting a small lateral expansion in the sample, indicating a ductile failure mode. The addition of fibers produced a gradual failure with improved post-peak strength. For all the fiber-reinforced samples, the stress corresponding to a strain of 10% was selected as the UCS (Sujatha et al. 2018) as the stress-strain response was gradual. Failure mode of soil, nano silica-treated soil and NBFRS are shown in Figure 3. A trial study on samples treated with nano-silica showed that the samples failed at a strain of 4.28%, 4.28%, 2.96% and 2.96% corresponding to the dosages 0.2%, 0.4%, 0.6% and 0.8% respectively with slight bulging accompanied by diagonal cracks (Figure 3b), corresponding to a semi-brittle shear failure. Literature shows that with aging, samples failed completely with strong signatures of brittle nature (Changizi and Haddad 2017) indicating that nano-silica increased the resistance to deformation causing the soil to fail at a comparatively lesser strain. The addition of fibers in the nano-silica treated sample altered the brittle behavior of soil as observed from the stress-strain response with a relatively gradual and smooth post-peak response (Figure 4). Figure 3c shows the NBFRS sample at 0d, exhibiting a combination of vertical tensile cracks and bulging. Although surface cracks were visible, the fibers bridged the gap (Changizi and Haddad 2015;Sujatha et al. 2018). The magnified image ( Figure 5) clearly shows that the fiber is embedded in the soil matrix, thereby establishing a firm grip. The tendency of the fibers to hold soil particles in the matrix on the application of load to failure and beyond results in a gradual failure response of the treated soil.

Unconfined compression strength
The organic soil has a UCS of 172.4 kPa at a water content of 17.5%. At 0.25% and 0.50% fiber reinforcement, the strength reduced to 120 kPa and 154.97 kPa, respectively, lesser than that of the organic soil. The reason for the reduced strength could be attributed to the fact that at lower doses, the fiber-soil contact is minimal and weak; hence, the fiber net cannot be formed efficiently, resulting in poor performance (Gao et al. 2015). The strength of the soil increased to 201.01 kPa with a 0.75% fiber addition but then again reduced to 185.02 kPa at 1.00% banana fiber reinforcement. At higher fiber dosages, fiber balling/clumping could have caused the strength reduction (Sujatha et al. 2018). Hence, 0.75% banana fiber by dry weight of the soil is the optimum fiber dosage. This dosage was used in the NBFRS. Fiber reinforcement enhanced the bonding in the  soil matrix at the fiber-soil interfaces, leading to strength improvement. Beyond optimum dosage, adding fibers in excess made the soil lose its cohesive strength (Puppala et al. 2000), owing to more fiber per unit volume. This clearly indicates that fiber reinforcement does not alter the chemical nature of the soil and the strength improvement was through mechanical reinforcement only. Controlled curing conditions adopted for the study ensured negligible variations in the moisture and microstructural state of the fiber-reinforced soil and therefore effect of aging on strength was ignored for the BFRS. Strength of soil treated with 0.2% to 0.8% nano-silica after 2 hours of curing showed that the UCS increased to 211.01 kPa, 216.07 kPa, 207.06 kPa and 190.81 kPa for the dosages from 0.2% to 0.8%, respectively. The UCS of NBFRS increased upto 0.4% nano-silica and decreased thereafter ( Figure 6). The strength increased with aging for both BFRS and NBFRS. At 0d and 90d of curing, strength improved to 220.94 kPa and 270.35 kPa, respectively, at the optimum dosage of 0.4% nano-silica and 0.75% fiber-reinforced soil.
A strength improvement by nearly 57% than the organic soil and 34.50% than the fiber-reinforced soil was observed on combining nano-silica stabilization and fiber reinforcement.

Failure modulus
Failure modulus shows the ease or ability of the material to stretch or deform and is calculated as the ratio of stress to 10% strain. The soil possessed a failure modulus of 3739.70 kPa. Reinforcing soil with banana fiber resulted in a decrease in the magnitude of the modulus as 1216 kPa, 1570.36 kPa, 2036.90 kPa and 1874.87 kPa at dosages ranging between 0.25% and 1.00%. During the trial study, nano-silica treated soil showed higher modulus of 4930.14 kPa, 5048.36 kPa, 6995.27 kPa and 6446 kPa for the dosages ranging from 0.2% to 0.8%, respectively. Failure modulus of BFRS and NBFRS increased with aging (Table 4) indicating the strength gain and resistance to deformation with ageing. Figure 7 shows the mechanism of strength change with various dosages of fiber and nano-silica. The addition of fiber upto optimum dosage bridged the voids in the soil matrix; however, further addition of fibers beyond the optimum dosage caused separation in the soil matrix, and fibers failed to compensate the cohesive strength of the soil when they occupied the clay's space in case of BFRS. Similarly, up to optimum dosage, the nano-silica filled the voids in the soil mass at initial curing periods, but with time became a viscous gel (Changizi and Haddad 2017); however, beyond the optimum dosage nano-silica caused aggregation in the soil particle increasing the void spaces in case of NBFRS. Despite the presence of fiber, the aggregated soil mass, therefore, was not able to attain strength gain. Also, Changizi and Haddad (2017) reported that the presence of nano silica reduces the distance between the clay particles and improves the interfacial friction in the soil particles leading to higher strength. The presence of banana fibers in soil matrix causes the viscous gel to adhere to the fiber and thereby improves the interfacial strength of the soil (Changizi and Haddad 2015). XRD pattern (Figure 8) reveal that the organic soil has quartz, traces of calcium carbonate, silicon dioxide, clay minerals like halloysite and saponite and NBFRS exhibited calcium silicate hydrate (CSH) peaks indicating the gel phase in the treated soil, causing strength enhancement. Authors like Pastor et al. (2019) observed that the reaction of calcium and silica ions to form CSH is a long-term slow process and this explains the slow rate of strength improvement with aging. The viscous gel formed induced a higher degree of aggregation at dosages beyond optimum values, thereby reducing strength (Changizi and Haddad 2017). FTIR spectra of the soil and treated soil are shown in Figure 9. The intensity variation around 1033 cm −1 in the organic soil and treated soil hints the presence of alumino-silica lattice of clay minerals, such as kaolinite, smectite, illite (Tinti et al. 2015). This result augments the presence of saponite (a clay mineral that belongs to the smectite group) as identified from XRD. Also, it indicates the stacking of soil hinting the formation of hydration products (Lv et al. 2018). The change in the shape of spectral response around 3400 cm −1 confirms the stretching of the OH bond, indicating the adsorption of water molecules (Vakili et al. 2020). This phenomenon signifies the water adsorption involved in the reduced OMC and the CSH gel formation phase.

Hydraulic conductivity (HC)
HC is an important geotechnical parameter that ensures the suitability of the soil for various geotechnical applications like earthen dam constructions and landfill liners. The organic soil has a HC of 8.8 × 10 −8 cm/s. The addition of banana fiber and nano-silica increased the HC of the treated soil, indicating the enhanced drainage properties of the treated soil. However, the HC decreased with aging. Figure 10a shows that the inclusion of fiber reduced the HC upto 0.75% banana fiber but HC increased at dosages higher than 0.75%. HC of 0.75% BFRS are 2.68 × 10 −7 cm/s and 5.15 × 10 −9 cm/s at 0 and 90 d respectively. Fiber reinforcement bridges the voids in the soil leading to reduced HC, but beyond optimum dosage, more fibers are available per unit volume, causing increased fiber-soil contact regions, creating more flow paths and increased HC (Divya, Viswanadham, and Gourc 2018).
The HC of NBFRS decreased up to a dosage of 0.4% nano-silica addition and increased beyond the optimum dosage of 0.4% nano-silica (Figure 10b). At 0.4% nano-silica, HC at 0d and 90d curing period are 5.85 × 10 −7 cm/s and 5.74 × 10 −8 cm/s respectively. Aggregation increased with the dosage of nano-silica (Ahmadi and Shafiee 2019). Mirzababaei et al. (2021) suggested that the addition of nanoparticles would cause a flocculated structure in the soil mass as an effect of aggregation and this flocculation causes more voids in the soil. The voids at saturated conditions allow more water to pass through them, causing an increase in HC. This also explains the more permeable nature of the NBFRS sample than the BFRS samples. However, with time, the viscous gel formed tends to reduce the ease of water flow, resulting in a slightly lesser HC with aging. Figure 11a points that the organic soil has lesser voids, but in NBFRS void spaces have increased due to particle aggregation (Figure 11b). Although the fibers bridge the gap between the aggregated soil mass, the agglomeration led to void formation in the soil (Changizi and Haddad 2017), facilitating easy drainage of water. SEM images could not show the presence of CSH due to their lower magnification, hence FESEM micrographs were used to visualize the viscous gel formation. The FESEM images of the treated soil in Figure 12 show the CSH gel formation caused by the nano-silica addition in the soil after 28d of curing confirming the changes in strength and HC of NBFRS.

Conclusion
A combination of an additive like nano-silica with fibers not only enhanced the strength but also provided appreciable post-peak behavior. The results of the experimental investigation showed that reinforcing soil with banana fibers reduced the MDUW and OMC. OMC decreased negligibly. Strength improved appreciably by 16.6% for BFRS at 0.75% banana fiber dosage and exhibited gradual failure post-peak. On nano-silica stabilization to BFRS a considerable increase in strength by 56.82% was observed in comparison with virgin soil. Also, 34.50% increase was noted in comparison with 0.75% BFRS at 0.4% nano-silica dosage. HC showed a decreasing trend upto the optimum dosage and increased on further addition of fiber and nano-silica. Also, with aging HC decrease for both BFRS and NBFRS. After 90d curing, HC decreased by 88.35% in case on NBFRS compared to that of BFRS. The changes in strength and HC can be attributed to the formation CSH gel as observed from chemical analysis and surface morphology studies. The study advocates random fiber inclusion in additive stabilized soils showing brittle failure to improve the post-peak failure characteristics and the choice of natural fibers like banana, jute, sisal, areca, coconut, etc to aid sustainable development as they provide the option for minimal intervention to the soil environment.