Dynamics of Biochar-Silty Clay Interaction Using In-House Fabricated Cyclic Loading Apparatus: A Case Study of Coastal Clay and Novel Peach Biochar from the Qingdao Region of China

Biochar has been recently investigated as an eco-friendly material in bio-engineered slopes/landfill covers. A majority of recent studies have focused on analyzing water retention behavior while very few have examined dynamic behavior (i.e., cyclic loading due to earthquake, wind, or wave) of biochar amended soil. As far as the authors are aware, there is no study on the dynamic behavior of biochar amended soils. Considering the above mentioned study as a major objective, field excavated soil was collected and mixed with in-house produced biochar from peach endocarps, at three amendment rates (5%, 10%, and 15%). The un-amended bare soil and biochar amended soil were imposed to a cyclic load in a self-designed apparatus and the corresponding stress-strain parameters were measured. Dynamic parameters such as shear modulus and damping ratio were computed and the results were compared between bare and biochar amended soil. Furthermore, the residual cyclic strength of each soil types were correlated with an estimated void ratio to understand the interrelation between dynamic loading responses and biochar amended soils. The major outcomes of this study show that the addition of biochar decreases the void ratio, thereby increasing the shear modulus and residual cyclic strength. However, the modulus and strength values attenuates after 15 cycles due to an increase in pore water pressure. In contrary, at higher amendment rates, Biochar Amended Soils (BAS) forms clay-carbon complex and decreases both shear modulus and residual cyclic strength.


Introduction
The conversion of organic bio-waste material into an energy has received significant attention as a sustainable alternative in various applications. The bio-derived resource materials serves as a solution for environmental management to scale down global threats such as Greenhouse gas emission, climate change, and waste management [1,2]. Further, the conversion of bioresource materials reduces the dependency of other resource materials and promotes the direction towards a bio-based

Soil Properties and New Biochar Production
The soil used in this study was collected from the coastal region of Qingdao city, Shandong province, China. It should be noted that the site was amenable to many offshore constructions and the offshore structures were susceptible to dynamic loads. The collected soil was characterized as per the American Standard of Testing and Materials (ASTM) provisions. The soil predominantly consists of clay (39%) trailed by silt (33%) and sand (28%) [36], thus the soil could be termed as silty clay as per the USCS classification [37]. The consistency limits of the soil were attained as a liquid limit value of 42.7% and plastic limit value of 27.5% [38]. The soil was also characterized for its compaction parameters using the modified standard proctor test [39]. The optimum moisture content and maximum dry density was found out to be 14.7% and 1.78 g/cm 3 respectively.
Peach endocarp wastes were collected from the local site and were pyrolyzed using the pyrolyzer. The whole pyrolysis process was executed in the absence of oxygen at a temperature of 400 • C for 3 h, which in general is termed as slow pyrolysis (Figure 1a) [40]. The produced biochar from the feedstock was crushed to powder and analyzed for its surface properties such as surface morphology and the functional groups. Figure 1b portrays the surface properties of the biochar, where the morphology and the functional groups were analyzed using Field Emission Scanning Electron Microscope (FE-SEM) and Fourier Transformation Infra-Red (FTIR) spectroscopy respectively. The figure shows that the Peach Endocarp (PE) biochar, being a plant-based feedstock possessed intrapores on the entirety of its surface. This high density of intrapores could be attributed to the thermal degradation of the simple biopolymers such as cellulose and hemicellulose, which degrades faster than the complex lignin [32,41,42]. The PE biochar was also enriched with the functional groups and had major hydrophilic functional group i.e., hydroxyl (OH − ), which was indicated by a prominent band around 3500 cm −1 wavelengths. The biochar also possessed other aromatic and aliphatic functional groups such as C-H, C-O, C-N, C=O, and so on, as represented in the figure. Addition to the surface parameters, PE biochar was examined for its elemental composition and molar ratios. The values obtained were presented in Table 1 with carbon, nitrogen, and hydrogen compositions.

Experimental Methodology
The dry soil was mixed with PE biochar at three different amendment rates (5%, 10%, and 15%) in weight to weight composition. The amended soils were characterized for its basic geotechnical properties and the respective results are listed in Table 2. Monotonic direct shear tests were conducted for the un-amended bare soil and the BAS at the compaction state of 0.9 MDD, where the test results could also be considered as a verification index for the designed constant stiffness direct shear apparatus. The monotonic direct shear test was carried out at a shear rate of 2 mm/min and normal load of 40 kN/mm 2 . Furthermore, the bare soil and BAS were subjected to cyclic loading at the similar compaction state using the designed constant stiffness direct shear apparatus. The pictorial representation and the sample preparation for cyclic loading are picturized in Figure 2. The shear rate and normal load in the case of cyclic loading was kept at a rate of 5 mm/min and 40 kN/mm 2 respectively. Sustainability 2020, 12, x FOR PEER REVIEW 5 of 17 the bare soil and BAS were subjected to cyclic loading at the similar compaction state using the designed constant stiffness direct shear apparatus. The pictorial representation and the sample preparation for cyclic loading are picturized in Figure 2. The shear rate and normal load in the case of cyclic loading was kept at a rate of 5 mm/min and 40 kN/mm 2 respectively.

Design and Development of Cyclic Loading System
The constant stiffness direct shear apparatus is a large, constant stiffness loading setup, located in Qingdao University of Technology, Shandong province, China. This apparatus was used to analyze the cyclic stress behavior of bare soil and BAS owing to its major instrumentation advantages. The apparatus could be classified into four major groups and the specifications are given below.

Constant Stiffness Loading System
The constant stiffness loading system is composed of vertical screw group, upper fixing plate, spring group, and spring fixing plate.

1.
The vertical screw group consists of rounded thread rods, rigidly connecting the operating platform and the entire loading system at all the four corners. Each rod measures a diameter of 20 mm and a height of 800 mm; 2.
The upper fixing plate is attached to the vertical screw group at all the four corners by means of screw and nut. The upper fixing plate has dimensions of 840 mm length and 540 mm breadth, while the thickness is maintained to 20 mm; 3.
The upper fixing plate supports the spring group, where the springs enclosed the steel columns of 20 mm dia. The spring group consists of 8 standard springs which are divided into 2 equal pairs. The distance between two adjacent springs are maintained to 135 mm, both lengthwise and breadthwise; 4.
The spring group firmly holds the spring fixing plate, where the dimensions of the plate is confirmed to a length of 700 mm and breadth of 300 mm. The thickness of the spring fixing plate is also maintained to 20 mm as in the case of upper fixing plate.

Interfacial Shear Simulation System
The interfacial shear simulation system consists of two parts, the upper and lower shear box.

1.
The upper shear box is a closed box, having a dimension of 700 mm * 300 mm * 300 mm (length * breadth * height). The front face of the upper shear box is detachable by a steel plate of 20 mm thickness, and has a plexiglass window of size 500 mm * 210 mm; 2.
The lower shear box is positioned just below the upper shear box and is a "U" groove made on the cut box. The lower shear box measures a dimension of 500 mm*300 mm*100 mm. The lower shear box is flexible to move back and forth during the shearing process.

Roller Bi-directional Transmission System
The roller bi-directional transmission system consists of platform, ball screw, CNC motor, and bi-directional control parts.

1.
The platforms are made of steel and supports the entire instrument in both horizontal and vertical direction. The vertical platform's size is 1500 mm*500 mm*40 mm and the size of horizontal platform is 500 mm*500 mm*40 mm in the order of length, breadth, and thickness respectively. Both the platforms are strengthened by connecting the ends using a trapezoidal rib plate; 2.
The ball screw is composed of screw, nut ball, pre-stressing sheet, and dust proof device. The ball screw translates the rotational motion to linear motion with little friction and enables the shear box to move back and forth. The device uses Mishmi SFK_R00802 ball screw which facilitates the shear box to move 10 cm in a linear direction; 3.
The CNC motor is operated by three major components i.e., frequency conversion motor, reducer, and frequency convertor device. The frequency conversion motor has the motor capacity of 2 kW with speed limit as 1500 rpm. The frequency of the motor is ranging between 5 Hz and 50 Hz. The shear rate of the reducer could be adjusted between 0.15 mm/min and 15 mm/min using the frequency convertor device. The frequency convertor is designed for self-adopting circuit, which senses the starting and termination of the shearing process automatically; 4.
The bi-directional control parts have stroke switch and a scale. The stroke switch is directly connected to the frequency convertor and are provided at either sides of the shear box. This arrangement facilitates the shear box to reverse the direction after the specified displacement.

Data Acquisition System
The fundamental data collected for this experiment were normal stress and shear load at the shear plane, using a micro-earth pressure gauge and S-type load cell sensor respectively.

1.
The shear load was determined using a "S" type load cell, which possess a spring element and 4 strain gauges in a wheat stone bridge formation. The deformation in the spring element is picked by the strain gauges and converts them into electric signal. The electric signals are further connected to DY220 high-precision weighing display controller, to obtain the shear load value; 2.
The normal stress in the shear test was measured using two CL-YB-2 resistance strain type force sensors which are symmetrically arranged between the spring group and the loading stress. The data obtained from the sensor is logged to the computer using DH3816N static strain test system.

Determination of Dynamic Loading Parameters
Shear modulus and damping ratio are efficient and important parameters to estimate the dynamic loading responses of soil, as these parameters are directly related to the soil deformation properties [44,45]. These parameters can be determined from the hysteresis loop obtained from the cyclic shear test. Figure 3 represents a typical asymmetrical hysteresis loop, which portrays the conventional method of calculating shear modulus and damping ratio [46].
From the equations presented in the figure, shear modulus for each cycle can be calculated from the secant elastic modulus (E sec ) and Poisson's ratio (ϑ). The Poisson's ratio was considered to be 0.5 as per the recommendations in Rollins et al., 1998 [47]. Similarly the damping ratio for each cycle is computed using the area of the hysteresis loop and the area of the right triangle that forms under the maximum shear stress and maximum shear strain. Since the loop obtained is asymmetrical, the area of enclosed loop (A L ) is calculated as suggested by Kreyszig, 2010 [48]. From the equations presented in the figure, shear modulus for each cycle can be calculated from the secant elastic modulus (Esec) and Poisson's ratio ( ). The Poisson's ratio was considered to be 0.5 as per the recommendations in Rollins et al., 1998 [47]. Similarly the damping ratio for each cycle is computed using the area of the hysteresis loop and the area of the right triangle that forms under the maximum shear stress and maximum shear strain. Since the loop obtained is asymmetrical, the area of enclosed loop (AL) is calculated as suggested by Kreyszig, 2010 [48]. = 1 2 ∈ * ∈ * + ∈ * ∈ * + ⋯ + ∈ * ∈ * (1)

Estimation of Void Ratio
The correlation between void ratio and cyclic stress behavior are similar to the influence of fines content. This unique relationship is clearly exemplified in several studies [49][50][51], affirming that decrease in void ratio increases the cyclic shear strength. In order to validate this statement, the void ratio was estimated from the following set of equations [52] and are associated with residual cyclic strength.
For partially saturated soil (three phase soil system): Since, weight=volume*density and = Figure 3. Schematic representation for shear modulus and damping ratio determination in an asymmetrical hysteresis loop.

Estimation of Void Ratio
The correlation between void ratio and cyclic stress behavior are similar to the influence of fines content. This unique relationship is clearly exemplified in several studies [49][50][51], affirming that decrease in void ratio increases the cyclic shear strength. In order to validate this statement, the void ratio was estimated from the following set of equations [52] and are associated with residual cyclic strength.
For partially saturated soil (three phase soil system): Since, weight=volume*density and γ s = Gγ w Dividing the numerator and denominator by V v , we get: Given that, degree of saturation, s = V w /V v and void ratio, e = V V /V S : Similarly w = W w W s Following the similar steps as in the case to derive Equation (1): se = wG; γ b , γ s , γ w represents bulk density, density of solids, and density of water; W w and W s denotes weight of water and weight of solids; V v , vs. and V w denotes volume of voids, solids, and water individually; and G, e, s, and w represents specific gravity, void ratio, degree of saturation, and water content respectively. From the known values of w and G, the multiple "s·e" was found with respect to Equation (6). It is followed by the determination of void ratio using Equations (2)-(5) with known values of γ b , γ w , and G.

Stress-Strain Response on Monotonic Loading
The stress strain response provides the basic understanding of BAS on static loading, which is depicted in Figure 4. It can be observed from the graph that the peak shear stress in both BS+5% PE and BS+10% PE increases with respect to the bare soil. This is mainly attributed to the porous nature of the biochar particles. During static loading, the porous biochar particles is expected to undergo extensive particle rearrangement and facilitates an interlocking between the grains and biochar, restricting the free gain movement [53].  Despite the positive result at the first two cases, there is a contrasting trend in BS+15% PE, where the peak shear stress value is slightly decreased with respect to the bare soil. This peak value reduction could be ascribed to the formation of clay-carbon complexes at higher amendment rates [54]. The formation of clay-carbon complex was also observed in previous studies, where this complex reduces the interlocking mechanism between the biochar and soil grains [55]. In terms of numerical expressions, the peak shear stress value was increased with respect to the bare soil by 30% and 45% for BS+5% PE and BS+10% PE respectively. For BS+15% PE, the peak shear stress value was decreased to 8% with respect to the bare soil. Figure 5 shows the stress-strain response of bare soil and BAS under cyclic loading. The stress-strain response for all the soil specimen follows an asymmetrical hysteresis loop trend and the hysteresis curves are smooth. Despite the positive result at the first two cases, there is a contrasting trend in BS+15% PE, where the peak shear stress value is slightly decreased with respect to the bare soil. This peak value reduction could be ascribed to the formation of clay-carbon complexes at higher amendment rates [54]. The formation of clay-carbon complex was also observed in previous studies, where this complex reduces the interlocking mechanism between the biochar and soil grains [55]. In terms of numerical expressions, the peak shear stress value was increased with respect to the bare soil by 30% and 45% for BS+5% PE and BS+10% PE respectively. For BS+15% PE, the peak shear stress value was decreased to 8% with respect to the bare soil. Figure 5 shows the stress-strain response of bare soil and BAS under cyclic loading. The stress-strain response for all the soil specimen follows an asymmetrical hysteresis loop trend and the hysteresis curves are smooth. In order to compare the loading behavior between bare and BAS precisely, shear modulus and damping ratio values were computed and plotted against a number of cycles of loading ( Figure 6). From the figure, it could be seen that there is a decrease in shear modulus of all soil types with an increase in the number of cycles. This reduction in shear modulus is evident due to an increase in the pore water pressure of compacted soil under cyclic loading [56]. Whilst, when comparing the results between the individual soil types, the shear modulus for all BAS increased with respect to the bare soil. This increment could be seen only in the first 15 cycles, later the modulus values were almost similar with the bare soil. On the other hand, there was not a significant change in the damping ratio, when the soil samples were compared. In order to compare the loading behavior between bare and BAS precisely, shear modulus and damping ratio values were computed and plotted against a number of cycles of loading ( Figure 6). From the figure, it could be seen that there is a decrease in shear modulus of all soil types with an increase in the number of cycles. This reduction in shear modulus is evident due to an increase in the pore water pressure of compacted soil under cyclic loading [56]. Whilst, when comparing the results between the individual soil types, the shear modulus for all BAS increased with respect to the bare soil. This increment could be seen only in the first 15 cycles, later the modulus values were almost similar with the bare soil. On the other hand, there was not a significant change in the damping ratio, when the soil samples were compared.

Dynamic Loading Characteristics
Before devising a concluding remarks from the attained results, it is important to comprehend the influence factors of shear modulus and damping ratio. According to Hardin and Drnevich, 1972 [57], the major influence factors for these dynamic loading parameters were the number of cycles of loading, pore water pressure, and void ratio. Considering these factors, it is clear that the addition of biochar altered the void ratio and increased the soil dynamic modulus by maintaining the aspects of damping ratio. Before devising a concluding remarks from the attained results, it is important to comprehend the influence factors of shear modulus and damping ratio. According to Hardin and Drnevich, 1972 [57], the major influence factors for these dynamic loading parameters were the number of cycles of loading, pore water pressure, and void ratio. Considering these factors, it is clear that the addition of biochar altered the void ratio and increased the soil dynamic modulus by maintaining the aspects of damping ratio.

Interrelation between Residual Cyclic Strength and Void Ratio
The residual cyclic strength measured at each cycle for bare and BAS is presented in Figure 7(a) and the estimated void ratio for the compacted soil specimens are graphically represented in Figure 7(b).

Interrelation between Residual Cyclic Strength and Void Ratio
The residual cyclic strength measured at each cycle for bare and BAS is presented in Figure 7a and the estimated void ratio for the compacted soil specimens are graphically represented in Figure 7b.
From the residual cyclic strength, it could be seen that the strength increased in BS+5% PE and BS+10% PE was mainly observed in the first 15 cycles, then the residual cyclic strength levels off and resembled bare soil. Alternatively, the void ratio values decreased with an increase in amendment rates. This clearly indicates that the decrease in void ratio increased the cyclic residual strength by means of better interlocking between the biochar and soil grains in the first 15 cycles. Thereafter, the residual cyclic strength attenuated and yielded a similar strength to the bare soil. This is logically assumed that after certain number of cycles, capillary pressure inside the pores of the biochar builds up pore water pressure and attenuates the residual cyclic strength. This phenomenon is supported by the fact that capillary pressure is indirectly proportional to the diameter of the pores [58]. Nonetheless, it is also clear that the residual cyclic strength of BS+15% PE decreased with respect to the bare soil. This is again attributed to the clay-carbon complex between biochar and soil grains, which disrupts the interlocking mechanism, neglecting the void ratio decrement [54]. From the residual cyclic strength, it could be seen that the strength increased in BS+5% PE and BS+10% PE was mainly observed in the first 15 cycles, then the residual cyclic strength levels off and resembled bare soil. Alternatively, the void ratio values decreased with an increase in amendment rates. This clearly indicates that the decrease in void ratio increased the cyclic residual strength by means of better interlocking between the biochar and soil grains in the first 15 cycles. Thereafter, the residual cyclic strength attenuated and yielded a similar strength to the bare soil. This is logically assumed that after certain number of cycles, capillary pressure inside the pores of the biochar builds up pore water pressure and attenuates the residual cyclic strength. This phenomenon is supported by the fact that capillary pressure is indirectly proportional to the diameter of the pores [58]. Nonetheless, it is also clear that the residual cyclic strength of BS+15% PE decreased with respect to the bare soil. This is again attributed to the clay-carbon complex

Conclusions
It is important to emphasize that the results shown is limited only to the biochar produced from peach endocarp wastes. The results showed that the addition of biochar decreased the void ratio, which facilitated the BAS to produce an interlocking mechanism between the biochar and soil and increased the shear modulus and residual cyclic strength. However, this mechanism was effective only for the first 15 cycles and then the shear modulus and residual cyclic strength attenuated due to an increase in pore water pressure, caused by capillary pressure in the biochar pores. On the contrary, beyond certain amendment rate, BAS forms a clay-carbon complex, which disrupted the interlocking mechanism. Therefore the study concluded that biochar could be considered as a potential material for landfill liners and pre-soil amendment in offshore constructions where the dynamic loads are prone, instead of using environmentally hazardous additives.

Conflicts of Interest:
The authors declare no competing conflict of interest.