Experimental Investigation of Mechanical Behaviors of Fiber-Reinforced Fly Ash-Soil Mixture

In recent years, applications of different types of solid waste in fiber-reinforced soil are developed to improve the strength of soil. 1is study presents an experimental investigation of mechanical properties of polypropylene fiber-reinforced fly ash-soil mixtures. A series of direct shear tests and unconfined compression tests were carried out. 1e effects of fly ash content and fiber content on compaction characteristics, shear strength, strength parameters, and unconfined compressive strength of the reinforced soil are investigated and discussed. Results reveal that when the fly ash content of the specimen exceeds 20% and the curing period exceeds 14 days, specimens become more brittle in the unconfined compression tests. It can be deduced that 30% fly ash and 1% fiber provide the optimum content, and the inclusion of fiber reinforcement has positive benefits on the mechanical properties of the reinforced soil to a certain extent.


Introduction
Fly ash, a waste generated in the process of burning coal to produce energy in thermal power plants, has caused a lot of pollution to the environment. e large increase of fly ash production and its environmental protection treatment have become a global concern [1]. At present, many scholars are engaged in utilizing fly ash in various aspects of geotechnical applications [2]. Mixing fly ash and fine-grained soil to improve its engineering properties has been widely recognized and applied. e application of the reinforced soil method in geotechnical engineering has a long history, including in construction, roadways, railways, dikes, stabilization of soil slope, and improvement of soft soil. Certain desired properties such as load bearing capacity, shear strength, and permeability of soil can be improved by various techniques. One of the most commonly used techniques is the use of geosynthetics to reinforce the soil. Latha and Varman [3] conducted large triaxial tests to investigate the effect of the quantity of geotextile on the static and cyclic load response of sand. It can be observed that the inclusion of geotextile layers improves the peak shear strength, axial strain at failure, and stiffness under static conditions, and higher dynamic moduli can be exhibited in reinforced specimens compared to the unreinforced sand under cyclic conditions. Abdi and Arjomand [4] conducted pullout tests aiming at studying the interaction of clays reinforced with geogrids embedded in thin layers of sand. Results revealed that encapsulating geogrids in thin layers of sand under pullout conditions enhances pullout resistance of reinforced clay. For the clay-sand-geogrid samples, an optimum sand layer thickness of 10 mm was determined, resulting in maximum pullout resistance, which increased with increasing confining pressure. Chen et al. [5] investigated the shear behavior of this geogrid-reinforced fill through large-scale triaxial tests. e shear deformation tends toward strainhardening behavior with an increase in the number of geogrid layers and the confining pressure. Geogrids significantly improve the apparent cohesive strength of coarsegrained soil. e number of geogrid layers influences the geogrid-soil interface's mobilization and the slip surface type. Rose et al. [6] and Park and Tan [7] individually incorporated the fiber into the retaining wall and subgrade filler, and the engineering properties of the mixed material were systematically studied. Lovisa et al. [8] studied the effect of water content on the shear strength of fiber-reinforced sand. e tests showed that the fiber can significantly improve the cohesion of sand under dry conditions. Sadek et al. [9] have investigated soil reinforcement using discrete randomly distributed fibers, and several models were suggested to estimate the improvement brought by fibers to the shear strength of soils. Anggraini et al. [10] carried out the indirect tensile test and unconfined compressive test to investigate the tensile and compression strengths of natural and treated soft soil.
e tensile strengths and compression strengths of compacted specimens of natural soil and lime and coir fiber treated soil were obtained. Results revealed that both tensile and compressive strengths increase with the addition of lime and coir fiber and by increasing of the curing time.
Wu et al. [11] investigated the mechanical properties of silty clay reinforced with discrete, randomly distributed sisal fibers using triaxial shear tests. e sisal fibers were cut to different lengths, randomly mixed with silty clay in various percentages, and compacted to the maximum dry density at the optimum water content. e results indicated that with a fiber length of 10 mm and content of 1.0%, sisal fiberreinforced silty clay is 20% stronger than nonreinforced silty clay. Gao et al. [12] performed a series of unconfined compressive strength tests on clay soil reinforced with basalt fibers under the condition of optimum water content and maximum dry density. Experimental results showed that basalt fibers can effectively improve the unconfined compression strength of clay soil. And, the optimal content and length are 0.25% and 12 mm, respectively. e results also showed that the basalt fiber-reinforced clay soil has the "poststrong" characteristic.
Because of few available literatures on the shear properties of fiber-reinforced fly ash-soil mixtures, effects of fly ash and fiber on the strength characteristics of fine-grained soil are still not clear and further research is needed. erefore, this paper focuses on the mechanical behavior of clay mixed with different contents of fly ash and fiber reinforcement. A series of direct shear tests and unconfined compression tests are carried out in this study. e influence of fly ash content and fiber content on the shear strength, strength parameters, and unconfined compressive strength of the fiber-reinforced fly ashsoil mixture is thoroughly investigated.

Testing Materials.
e clay was obtained from a road construction site in Hongshan District, Wuhan. e fly ash was collected from a power plant in the Wuhan City, and it belongs to Class C fly ash. e physical parameters of clay and fly ash were obtained by laboratory tests shown in Table 1. e polypropylene fibers were used in this study, and the average dimensions are length 12 mm and diameter 0.023 mm as shown in Figure 1(a). e behavior parameters are shown in Table 2.

Preparation of Specimens.
In the tests, the performance of a total of 20 stabilized soil mixes was investigated by varying the percentage of fly ash and polypropylene fibers. For each mix, two specimens were prepared. e soil was replaced by fly ash contents of 10%, 20%, 30%, 40%, and 50%, respectively, on dry weight basis. Further, four values of fibers (0%, 0.5%, 1.0%, and 1.5%) were considered for each of these mixes. e prepared mixtures were then stored in plastic bags for future use. e desired quantities of water were added to the mixtures as per the optimum water content and further mixed thoroughly to form a homogeneous mixture. e composite was used to estimate the direct shear test and unconfined compression test.

Compaction Tests and Direct Shear Tests.
According to ASTM D698-12e2, the standard compaction test was carried out on fly ash-soil mixture to obtain the optimum water content and maximum dry density, and the content of fly ash varied from 0% to 50%. e direct shear test equipment is shown in Figure 1 e maximum vertical and horizontal loads of the equipment are both 700 kN, and the size of specimen is 504.6 mm in diameter and 400 mm in height. e direct shear test was performed under a strain-controlled condition, and the shear rate was maintained at 1 mm/min during the test. rough a series of tests, the optimum content of fly ash and the effect of fiber on the reinforcement effect of the fly ash-soil mixture were studied. e specimens were compacted in three layers. e shaving treatment was needed between each compacted layers. Fly ash-soil mixture was filled in the upper and lower shear boxes. e ribs are placed at the shear plane of the upper and lower shear boxes, and the ends are fixed to prevent slippage during the shearing process. And, the test was performed under vertical normal stresses of 100 kPa, 200 kPa, and 300 kPa. When the shear strain reached 15%, the test was terminated.

Unconfined Compression Tests.
e unconfined compression test equipment is shown in Figure 1(c). For the unconfined compressive strength test, a metal mold with an inner diameter of 39.1 mm and a height of 80 mm was used and a detachable collar was attached at both ends to prepare a cylindrical specimen. e clay was uniformly mixed with desired amount of fly ash and fibers, and the fiber content varied from 0.5% to 1.5% (the fiber content was a percentage of the dry weight of the fly ash-soil mixture). e mixture was placed in a mold and, in order to ensure uniform compaction, the specimen was statically compressed on both ends until the required size of the specimen reached. e specimens were then demoulded with a hydraulic jack, wrapped in a sealed polyethylene bag, and cured in a desiccator for 1 day, 7 days, 14 days, and 28 days. After the required curing periods, specimens were tested for unconfined compressive strength using a strain rate of 1.2 mm/min. Figure 2 depicts the standard compaction curves of six different fly ash-soil mixtures (the y ash content varies from 0% to 50%). Figure 3 shows a comparison of the optimum water content and maximum dry density of the y ash-soil mixture with di erent y ash contents. It can be clearly seen from Figure 3 that as the y ash content gradually increases, the maximum dry density of the y ash-soil mixture is reduced from 1.52 g/cm 3 to 1.42 g/cm 3 . On the other hand, Figure 3 also reveals that as the y ash content increases, the optimum water content of y ash-soil mixture increases from 22.7% to 25.9%. Notably, the optimum water content of y ash-soil mixture with 30% y ash content reaches the maximum value, and the maximum dry density reaches the minimum value. Figure 4 shows shear stress-strain curves for y ash-soil mixture specimens with a y ash content of 0% (pure soil), 10%, 20%, 30%, 40%, and 50%, respectively. It was found that the shear strength of the y    ash-soil mixture specimen reached the maximum when the y ash content was 30%. For example, when the normal stress is 300 kPa, the shear strength of the y ash content of 0%, 10%, 20%, 30%, 40%, and 50% of the y ash-soil mixture is 183 kPa, 198 kPa, 212 kPa, 266 kPa, 238 kPa, and 229 kPa, respectively. It can be seen from the gure that the loss of residual shear strength of the y ash-soil mixture decreases with increasing of the y ash content under the same normal stress, which indicates that the increase of the y ash content is bene cial to transform the shear properties of the y ashsoil mixture specimen from brittle to ductile. e relationship between the shear stress peak value and y ash content under di erent normal stresses is shown in Figure 5. e shear strength of the specimen with 30% y ash content reached the maximum value, which indicates that the optimum y ash content is 30%, so this content is selected to further study the e ect of ber on the shear characteristics of the reinforced soil. e shear stress-strain curve is shown in Figure 6. It can be seen from the gure that the inclusion of more ber contents signi cantly improved the shear characteristics of the specimens. When the shear strain reaches 15%, the shear stress of the specimen remains constant. In addition, there is no reduction of the shear stress, which indicates that the appearance of bers in the soil can prevent the reduction in peak shear stress to some extent. It can be also seen from Figure 6 that, for any ber-reinforced y ashsoil mixture, before the strain reaches 2%, the shear stress value of the specimen does not reveal the superiority of the  ber reinforcement. is can be due to the slippage of the ber under low strain condition [13]. As shown in Figure 6, the ber-reinforced y ash-soil mixture has better ductility and less loss of residual shear strength than unreinforced y ash-soil mixture. For large shear strains, ber-reinforced specimens could maintain shear stress with increasing deformation, which indicates that these materials have superior ductility.

Shear Strength and Strength Parameters.
e e ect of ber on the shear strength of ber-reinforced y ash-soil mixture can be evaluated by a strength ratio parameter R, and it can be de ned as where σ R f is the shear stress of ber-reinforced y ash-soil mixture at failure and σ f is the shear stress of y ash-soil mixture at failure. Table 3 and Figure 7 show the peak shear strength and strength ratios of the reinforced specimens with di erent y ash contents. It can be seen that, for the same ber content, the strength ratios of the ber-reinforced y ash-soil mixture specimens decrease with the normal stress varying from 100 kPa to 200 kPa, while increase slightly at 300 kPa. Moreover, it can be also seen that the shear strength of the reinforced specimens with di erent contents of ber under the same normal stress varies greatly. For example, under the normal stress of 100 kPa, the shear strengths of y ash-soil mixture reinforced with 0.5%, 1%, and 1.5% ber content are 129.5 kPa, 187.5 kPa, and 156.5 kPa, respectively. is can be suggested that when the ber content exceeds 1% in the y ash-soil mixture, the increasing amplitude in shear strength may decrease. e ber content plays an important role in increasing the shear strength of the ber-reinforced y ash-soil mixture. It can be seen from Table 4 that the cohesion c of y ash-soil specimen increases at rst and then decreases with increasing ber content. It is obvious that the contribution of ber content to the c value is signi cantly di erent. When the ber content increases from 0 to 0.5%, the c value increases by 29.37 kPa, and when ber content increases from 0.5% to 1%, the c value increases by 50.1 kPa although the ber increment is only 0.5%. It can be seen that the increase of c value is di erent with the same ber content increment so that the contribution of ber reinforcement is not linear with respect to c value, which is related to the reinforcement mechanism of ber and soil. e reinforcement mechanism of the ber-reinforced y ash-soil mixture in the direct shear test has two main aspects: the one-dimensional tensile action of a single ber and the three-dimensional tensile action of the ber web. Before the ber is pulled out, the surface of ber is surrounded by a large number of soil particles. Since the elastic modulus of ber is much higher than that of the soil, once bers are loaded at the same time, the inconsistency of the deformation will inevitably lead to a mutual displacement tendency between the ber and soil. e ber is in tension, and the interface force between the ber and soil particle is generated. Here, the value of interface force mainly depends on the interface friction and adhesion [14]. is interface force inevitably restricts the relative sliding of the ber so that the ber in the soil can withstand a certain tensile stress, thereby sharing the external load. In addition, due to the ber in tension, the bent portion of ber can also restrain the deformation of soil particles and improve the mechanical properties of the soil. is reinforcement of ber is mainly dominated by the one-dimensional tensile action of discrete ber. In another case, a large number of bers are randomly distributed in the soil and interwoven into a net. When one of the bers is in tension, the other bers must be forced together to form a three-dimensional forced net. us, the load is distributed to a wider area, which further improves the tensile e ect of ber. is reinforcement pattern of ber is mainly based on the three-dimensional tensile action of the ber web.
Di erent ber inclusions may a ect the main reinforcement pattern of ber. For low ber content, due to the large ber spacing, an e ective ber web cannot be formed, so the contribution of the ber to the c value mainly comes from the action of the discrete ber. As the ber content increases, more and more bers begin to interweave into a net. At this time, the contribution of the ber to the c value is superimposed on the three-dimensional tensile e ect. eoretically, more ber content in the soil can cause greater ber density on the shear plane and generate better performance of ber to bear the shear stress. However, this is not the case. e test results have shown that the contribution of ber reinforcement to the c value or other strength parameters has a critical amount. When exceeding the critical amount, the strength parameters may decrease with the increase in ber content. For example, Prabakar and Sridhar [15] found when the ber content exceeds 0.75%, the

Advances in Materials Science and Engineering
c value begins to decrease. Cai et al. [16] obtained the optimal ber content of 0.3% of the dry soil weight by the direct shear test, and the optimum ber length is 15 mm. e test results of Wu and Zhang [17] show that the shear strength of expansive soil increases with increasing ber content at rst, then decreases, and the optimum ber content is 0.3%. It can be seen that the general ber content threshold is greater than 0.3% and the optimum ber content in the direct shear tests of this study is 1%, which is consistent with the expected results.  It can also be seen from Table 4 that the increase in the internal friction angle of the ber-reinforced y ash-soil mixture specimens is not obvious and this phenomenon is almost consistent with the previous test results of Yang et al. [18]. However, in general, the internal friction angle of the ber-reinforced y ash-soil mixture is slightly higher than that of the plain soil, mainly because the majority of the ber arrangement on the shear plane is not parallel to the shear direction, which is bene cial to increase the occlusal friction action between soil particles and bers. Figure 8 shows the uncon ned compressive strength of plain soil and y ash-soil mixture with di erent y ash contents. It can be seen from the gure that the uncon ned compressive strength (UCS) of y ash-soil mixture is higher than that of plain soil under the same curing period. e curve shows a tendency to increase at rst and then decrease with increasing y ash content. Furthermore, after the shear stress peak, the curve of the y ash-soil is atter than that of the plain soil, and the specimen with 30% y ash content shows more ductility. e relationship between the peak uncon ned compressive strength of the y ash-soil mixture specimen and the curing period is shown in Figure 9. It can be clearly seen from the gure that the uncon ned compressive strength values of the specimens with 0% and 10% y ash contents are less a ected by the y ash content and the curing period. e uncon ned compressive strength of the specimen with y ash content exceeding 20% changes not signi cantly under the condition of curing from 7 days to 28 days. Among them, the uncon ned compressive strength of the specimen with 30% y ash content has greater improvement ratio with increasing curing period. e changing amplitude of uncon ned compressive strength (UCS * ) is de ned as follows:

E ect of Fly Ash Content and Curing Period.
where UCS clay represents the average value of the uncon ned compressive strength of the plain soil specimen, i.e., 223.47 kPa, and UCS mix represents the uncon ned compressive strength value of y ash-soil mixture, kPa. e magnitude of the increase in the uncon ned compressive strength value of the specimen is investigated, and the values are shown in Table 5.
A common trend is observed that the uncon ned compressive strength of y ash-soil mixture specimen increases with the y ash content and the curing period compared to pure soil, except for the 10% y ash content, which may be caused by the uneven distribution of y ash in the soil. For specimens with 30% y ash content, the uncon ned compressive strength values are greater than those of other y ash contents at 7 days, 14 days, and 28 days, respectively, except for 1 day. Table 5 also shows that, for the specimen with 10% y ash content, the increase of the uncon ned compressive strength of the specimen slightly uctuates with the increasing curing period. e occurrence of these uctuations is likely due to the heterogeneous distribution of wet y ash in the specimens.
In order to better evaluate the e ect of the curing period and y ash content on the ductility of specimen, the brittleness index I b can be de ned by where ε UCS presents the axial strain at UCS and ε UCS/2 presents the axial strain at UCS/2. e relationship between the brittleness index I b value and curing period is plotted in Figure 10. It can be seen from Figure 10 that as the y ash content and curing period increase, the brittleness index I b gradually decrease. e lower the I b value is, the greater the brittleness of the specimens is. It can be indicated from the decrease of I b value that the failure mode of specimens changes from ductility to brittleness. When the y ash content of the specimen exceeds 20% and the curing period exceeds 14 days, the I b values become close to each other.

E ect of Fiber Reinforcement.
e e ect of ber content (0.5%, 1%, and 1.5%) in y ash-soil mixtures after 28 curing days on uncon ned compressive strength is shown in  Advances in Materials Science and Engineering Figure 11. From the above analysis, it can be obtained that the specimen with 30% y ash content increases greatly in the uncon ned compressive strength compared with other specimens at the curing period of 7 days, 14 days, and 28 days, respectively. Hence, in order to further study the e ect of ber content on the UCS of specimens, the y ash content of the selected mixture was 30%. It can be seen from Figure 11 that the y ash-soil mixture with 30% y ash content exhibits brittle failure, and the specimen suddenly fails when the axial strain is less than 2%. However, the stress-strain curve of the ber-reinforced specimen shows ductile failure when compared with the unreinforced y ashsoil mixture. e axial strain of the specimen fail is between 3% and 4%, and the stress reduction of the ber-reinforced specimen after peak is smaller than that of the unreinforced specimen. It is found that when the ber content is 1%, the uncon ned compressive strength of the specimen reaches the maximum, but the residual strength after the peak is still lower than the specimen with the ber content of 1.5%. Adding ber to the plain y ash-soil can improve the uncon ned compressive strength of y ash-soil.
is is mainly because the specimen can bear the partial tensile stress when shear deformation occurs under the action of the axial pressure, due to the presence of the ber. is e ect limits the deformation of the specimen, and the more the ber content is, the greater the tensile stress that can be assumed, resulting that the uncon ned compressive strength of ber-reinforced y ash-soil is higher than that of unreinforced y ash-soil. With increasing the ber content, the uncon ned compressive strength of the specimen increases. After the initial cracks occur in the soil specimen, due to the existence of the ber, the further development of the cracks can be delayed. Consequently, the axial pressure that the soil can withstand does not suddenly decrease, so the ber-   Advances in Materials Science and Engineering reinforced y ash-soil exhibits fracture toughness, as shown in Figure 10. And, the tensile stress assumed by the ber mainly depends on the friction between the bers and the soil particles as well as the cohesion of the soil. Since the bers selected for the test are thinner and shorter and the surface of the ber is smooth, the tensile stress assumed by the ber mainly contributes to the improvement of soil cohesion.

Conclusions
In this paper, the mechanical characteristics of berreinforced y ash-soil mixtures were studied by standard compaction tests, direct shear tests, and uncon ned compression tests. e y ash contents were 10% to 50% by dry mass of the soil, and the ber contents were 0.5% to 1.5% by dry mass of the 30% y ash-soil mixtures. Based on the experimental results, the following conclusions can be drawn: (1) Compared to the unreinforced y ash-soil mixture, the shear strength of ber-reinforced y ash-soil mixture increases signi cantly. However, the inclusion of bers which is beyond 1% in the y ashsoil mixture may result in a decrease in shear strength.
(2) For the reinforced y ash-soil mixture specimens with di erent ber contents, the internal friction angle and cohesion increases with increasing ber contents. Compared with the unreinforced y ashsoil mixture, the cohesion and internal friction angle of the ber-reinforced y ash-soil mixture when ber contents vary from 0.5% to 1% are increased by 29.37 kPa to 79.47 kPa and 0.66°to 1.67°, which indicates that the ber content has more obvious e ect on the cohesion of soil. 30% fly ash + clay 40% fly ash + clay 50% fly ash + clay Pure clay 10% fly ash + clay 20% fly ash + clay Figure 9: Relationship between UCS and curing period for different proportions of y ash.   30% fly ash + clay 30% fly ash + clay + 0.5% fiber 30% fly ash + clay + 1.5% fiber 30% fly ash + clay + 1% fiber same curing period, and there is an obvious tendency to increase at first and then decrease when the fly ash content exceeds 10%. Moreover, after the shear stress reaches the maximum value, the curve of the fly ashsoil is flatter than that of the plain soil and the specimen with 30% fly ash content shows more ductility.
(4) With increasing fly ash content and curing period, the brittleness index I b value gradually decreases, which indicates the failure mode of specimen changes from ductility to brittleness. When the fly ash content of the specimen exceeded 20% and the curing period exceeded 14 days, the I b values become very close. (5) When the fiber content is 1%, the unconfined compressive strength of the specimen reaches a peak. But, the residual strength is still lower than the specimen with the 1.5% fiber content.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that they have no conflicts of interest.