3.1. Unconfined Compressive Strength Testing Results and Analysis
Through unconfined compression test and data processing, the relationship curve between compressive strength and material content was obtained, as shown in
Figure 2. It can be seen from
Figure 2a that: (1) When the content of sodium methylsilicate is fixed, the compressive strength of soil sample basically does not change with the content of lignin fiber. (2) With the increase of lignin fiber content, there was a small decrease, in which group a decreased by 6.57%, group B decreased by 4.38%, group C decreased by 1.93%, and group D decreased by 4.10%. (3) This is because the fiber itself is relatively soft and cannot play a role in bearing vertical pressure. In the meanwhile, with the addition of fiber, the workability of soil becomes poor and the compressive strength decreases slightly.
It can be seen from
Figure 2b that: (1) When the content of lignin fiber is constant, the compressive strength of soil sample is positively correlated with the content of sodium methyl silicate. (2) The compressive strength increases obviously when the content of sodium methyl silicate is 0%–0.3%, and the curve is steep. When the content is 0.3%–0.5%, the curve rises slowly. Therefore, it is considered that when the content of sodium methyl silicate is 0.3%–0.5%, the effect of improving the compressive strength of soil is the best.
On the whole, sodium methylsilicate can significantly improve the unconfined compressive strength of the site soil, while lignin fiber cannot directly improve the compressive strength of the soil, but its failure mode has changed. After the soil sample is damaged, it is broken but not scattered, which reflects that the fiber provides the compressive strength for the soil. Through the comparison of unconfined compressive strength of modified soil samples, it is concluded that the average compressive strength of group D-1 soil samples is the largest, which is increased by 31.04% compared with that of plain soil samples.
3.2. Direct Shear Testing Results and Analysis
Through direct shear test and data processing, the shear strength fitting curve of each group of soil samples can be obtained, as shown in
Figure 3. It can be seen from the figure that: (1) vertical pressure is an important factor affecting the shear strength of soil samples, and the shear strength increases with the increase of vertical pressure; (2) when the vertical pressure is the same, the shear strength increases with the increase of lignin fiber and sodium methylsilicate content.
It can be seen from
Figure 3a that, when other factors remain unchanged (vertical pressure 100 KPA, fiber content 0%), the shear strength increases by 6.45%, 10.32% and 13.00%, respectively, when the content of sodium methylsilicate is 0.1%, 0.3% and 0.5%, indicating that the shear strength increases rapidly when the content is 0%–0.3%, and the shear strength increases gently when the content is 0.3%–0.5%. It can be seen from
Figure 3b that the shear strength of soil samples increased more obviously when lignin fiber was added. When the vertical pressure and the content of sodium methylsilicate remained unchanged (the vertical pressure was 100 KPA and the content of sodium methylsilicate was 0%), the shear strength of soil samples increased by 2.95%, 14.48% and 26.24%, respectively, when the fiber content was 0.5%, 1% and 2%.
To sum up, the shear strength of D-3 soil sample is the largest. Therefore, when the content of sodium methylsilicate is 0.5% and the content of lignin fiber is 2%, the shear strength of the site soil is the best. Under the same load (taking 100 KPA as an example), it is increased by 48.15% compared with the plain soil sample.
According to the above shear strength fitting curve, the cohesion and internal friction angle of the soil sample can be calculated. According to the Mohr Coulomb theory, the intercept of the fitting curve in
Figure 3 is the cohesion, and the slope represents the internal friction angle. The curves of cohesion and internal friction angle with the content of modified materials are drawn according to the obtained data, as shown in
Figure 4 and
Figure 5.
It can be seen from
Figure 4 that: (1) With the increase of fiber content, the cohesion has been significantly improved, while the value of internal friction angle has not changed much. Fan Kewei et al. [
22] also obtained similar results when studying the strength improvement effect of fiber materials on soil. (2) The cohesion increases slowly with the change of fiber in the range of 0%–0.5% and 1%–2% but increases significantly in the range of 0.5%–1%. When the content of sodium methylsilicate remains unchanged (taking 0% as an example), the content of lignin fiber in the range of 0.5%, 1% and 2% increases by 10.65%, 38.61% and 53.86%, respectively, compared with that of plain soil. (3) As the lignin fiber itself is short in length and small in diameter, when the fiber content is relatively small (0.5%), the contact area between soil particles and fibers is small, which is shown by the small increase in cohesion. However, with the increase of the fiber content (1%), the fibers aggregate and bond a large number of soil particles, sharing the external load, which is shown by the significant increase in soil cohesion. When the fiber content continues to increase (2%), The water absorption characteristics of fibers will cause some fibers to cluster and cannot be uniformly dispersed into the soil, which will reduce the increase of fiber cohesion. (4) The influence of lignin content on the internal friction angle of soil is not obvious.
It can be seen from
Figure 5 that: (1) Sodium methylsilicate improves the cohesion and internal friction angle of soil samples. The 0.5% content of sodium methylsilicate in each group of soil samples improves the cohesion by 8.66%–24.25% and the internal friction angle by 12.30%–16.35%. (2) Sodium methylsilicate can effectively enhance the adhesion between soil particles, so that it can improve the mechanical properties of silt from two aspects of cohesion and internal friction angle, which is also consistent with the existing research results [
23,
24,
25]. (3) The combination of sodium methylsilicate and fiber makes the cohesive force and internal friction angle of modified soil sample D-3 reach the maximum, which are 50.60 kpa and 22.93°, respectively, which is 67.83% and 16.81% higher than that of plain soil. (4) Combined with unconfined compressive strength test, D-3 is selected as the best proportion of composite material. The soil sample has good compressive toughness, maximum cohesion and internal friction angle.
3.3. Dry Wet Cycle Testing Results and Analysis
It can be seen from
Table 4 that the surface change of plain soil sample (A-0) is obvious after dry wet cycle. After the second drying and wetting cycle, the surface of the plain soil samples began to be powdered and peeled. After 5 cycles, the soil samples began to dry and crack, and a small amount of shedding occurred at the corners, and the powdered peeling became more obvious. After 10 cycles, peeling, pulverization, dry cracking, edge and corner shedding are aggravated, and a few cracks appeared.
It can be seen from
Table 5 that composite material modified soil (D-3) is relatively less affected by dry wet cycle. After one cycle, the surface of the modified soil sample becomes smoother, which is equivalent to that the dry wet cycle removes the attachments on the surface of the soil sample, and more shows the film-forming effect of sodium methyl silicate on the soil surface. The surface changes of soil samples in subsequent cycles are not obvious, only a small amount of pulverization occurs after the 5th cycle, and the pulverization phenomenon is obvious after the 10th cycle. It shows that the soil modified by composite materials can effectively suppress the phenomenon of soil surface powdering, peeling and cracking caused by dry and wet cycles.
It can be seen from
Figure 6 that: (1) The mass loss of the plain soil sample (A-0) under the dry wet cycle is obvious. The surface of the soil sample changes little after the first two cycles, the soil falls off less, and the mass loss rate is low (below 1%). (2) When the number of cycles is between 2 and 5, the change range of the mass loss rate increases with the increase of the number of cycles. At this time, the overall stability of the soil is affected. The surface of the soil sample is obviously powdered, and the soil particles fall more. The mass loss rate reaches 5.11% at 5 cycles. (3) When the number of cycles is 5–10, the mass loss rate continues to increase slowly, indicating that the falling speed of soil particles becomes slow. After 10 cycles, the mass loss rate reaches 6.42%.
However, the composite material modified soil sample (D-3) is less affected by the dry wet cycle. After 10 cycles, the maximum mass loss rate is only 0.71%, which is about 89% lower than that of plain soil. It shows that the addition of composite modified materials can effectively reduce the spalling phenomenon of the surface soil under the dry wet cycle.
3.4. SEM Testing Results and Analysis
Figure 7 is the SEM image of sample A-0. It can be seen from the two groups of images in
Figure 7a,b that there are many cracks and holes of different sizes in the natural state of the plain soil sample, which significantly reduces the compactness of the soil sample structure. In
Figure 7c,d, it can be seen from the two groups of images that the surface of silt particles is rough and the gap between particles is large. The contact mode between soil particles is mainly point contact. The pores in the soil mass are mainly overhead pores with irregular shape, and there is basically no filler inside.
Figure 8 is the SEM image of sample D-0. It can be seen from the two images in
Figure 8a,b that the surface of D-0 soil sample is more dense than A-0, the surface of soil particles becomes relatively smooth, the contact mode between particles is mainly surface contact, and the number of macropores decreases significantly. In
Figure 8c,d it can be seen that there are many attachments on the surface of soil particles, and the lamellar structures are in parallel layers. The lamellar particles are in close contact in the form of face to face. The addition of sodium methylsilicate makes the surface of soil particles form a film, strengthening the connection between soil particles, which explains why sodium methylsilicate can effectively improve the mechanical properties of soil.
It can be seen from
Figure 9 that after adding lignin fiber, lignin fiber and surrounding soil particles are closely connected to form an overall spatial network structure. With the increase of fiber content, the fiber distribution is wider, and the interaction between fibers makes the overall stability of the network structure higher. When the soil is subjected to external force, the stress received by the soil particles will be transmitted to the fiber, and the fiber will transmit this part of the tensile stress to the surrounding structure, forming a three-dimensional stress structure and improving the connection force between the soil particles [
26,
27].
3.5. EDS Testing Results and Analysis
Due to the high accuracy of element plane and wide regional distribution, the element plane analysis method is selected for the test.
Figure 10,
Figure 11 and
Figure 12 show the selected representative areas.
By comparing
Figure 11 and
Figure 12 and
Table 6, it can be seen that due to the addition of sodium methyl silicate, the content of Na and Si elements in the soil sample increases slightly. This is because the sodium methyl silicate solution infiltrates into the soil, reacts with water and carbon dioxide, decomposes into methylsilicic acid, and rapidly forms a polymethyl siloxane film to cover the soil surface, resulting in the retention of Na and Si elements carried by sodium methyl silicate in the soil. It shows a small increase of these two elements.
It can be seen from
Figure 13 that after adding lignin fiber, the content of element C in sample D-3 is significantly increased, because the main elements contained in lignin fiber are C, h and O, so the content of element C is mainly increased.