Study on settlement characteristics of mudstone embankment

: It is in line with the concept of sustainable development to fill embankment with mudstone waste slag. However, the property of mudstone softening and disintegrating in water may affect the operation quality of the road. 2 Relying on the Hechang Highway project, the engineering properties of mudstone filler and the settlement 3 characteristics of embankment are studied by centrifugal model test and field test. The results showed that: under 4 Optimum moisture content, the settlement of mudstone embankment mainly occurred in the filling stage (accounting for 5 62.96% of the total), and it could reach stability after 24 months of operation. The embankment settlement in 6 cross-section showed the trend of "upward concave", and the settlement difference is not significant. Under continuous 7 rainfall condition, the settlement mainly occurred in the operation stage (accounting for 49.05% of the total), and it 8 could reach stability after 44 months of operation, in cross-section showed the trend of "M", the largest settlement at the 9 shoulder. The trend of the basement earth pressure and the settlement on top of embankment is basically the same. 10 Under continuous rainfall condition, the basement earth pressure is more concentrated toward the road centerline. It can 11 be seen that rainfall has a certain influence on the strength and stability of mudstone embankment. The above research 12 results provided a reliable basis for the design of embankment filled with mudstone waste. diffraction, disintegration test and point load test. The road performance of mudstone spoil was studied by compaction test, CBR test, large-scale direct shear test and

Mudstone is a kind of rock mass with high clay mineral content and low compressive strength, including iron, siliceous, 2 carbonaceous and silty. It is widely distributed in Sichuan, Chongqing, Guizhou, Yunnan and southwest Gansu. The 3 main characteristics of mudstone are low strength (5 < ≤ 15 ), easy weathering and easy disintegration 4 when exposed to water, which are important factors that lead to geological disasters and affect engineering safety. 5 For a long time, researchers have never stopped studying the disintegration mechanism, weathering characteristics, 6 mechanical properties and improvement effect of mudstone. In order to study the effects of dry-wet cycle, temperature, 7 mineral composition and external load on mudstone disintegration, a large number of laboratory tests, field tests and 8 theoretical analysis were carried out ( such as durability index, decay rate and surface energy were studied, and the classification method and standard of 11 mudstone disintegration were put forward (Zhu and Deng 2019; Shen et al.2020). The method of inhibiting the 12 disintegration of mudstone was studied, and the scheme of mixing seawater or calcium salt into mudstone to alleviate its 13 disintegration and improve the overall strength was put forward (Liu et al.2019). The lateral restraint was added to 14 restrain the water-rock interaction and water migration in mudstone, so as to control the disintegration expansion and 15 consolidation test. 1 The results showed that: the main mineral composition of mudstone is quartz, accounting for about 55%; Followed 2 by clay minerals (including illite, illite / montmorillonite mixed layer, chlorite, kaolinite, etc.), accounting for about 3 25%; Thirdly, calcite 11.7%, hematite 4.6%; It also contains a small amount of anatase and feldspar. A large number of 4 pores and micro-pores were developed in mudstone. The saturated uniaxial compressive strength is equal to 9.72 MPa. 5 According to "The Standard for Classification of Engineering Rock Mass (GB / T 50218-2014) ", it can be defined as 6 silty mudstone. After immersion in water, mudstone disintegrates strongly, and the anti-disintegration index is equal to 7 10.52%. Before and after disintegration, the mineral composition of mudstone hardly changed, but under the action of 8 water, the internal flaky crystal structure was tilted and curled outside, and the internal pores became larger, as shown in 9 According to the screening test results of samples before and after compaction, it can be seen that the initial 11 gradation of mudstone filler can be greatly changed by watering and compacting. The gradation changes of the samples 12 before and after compaction were as follows: before compaction, the particles with diameter D greater than 5mm were 13 too concentrated, the coarse particle content (P5) was 90%, the uneven coefficient (Cu) was 2.12, and the gradation was 14 poor, which could not meet the requirements of subgrade filler gradation in "Code for Design of Highway Subgrade". 15 After compaction, the coarse grain particle (P5) is 53%, the uneven coefficient (Cu) increases to 21.21, and the 16 gradation is good, reaching the maximum dry density. Therefore, it is of little significance to use the gradation before 17 compaction to control mudstone filler according to the codes. Before filling, it is necessary to determine the best 18 gradation of mudstone filler through compaction test. The optimum water content and maximum dry density of 19 mudstone filler in this project are 8.63% and 2.166g/cm 3 . 20 When the compaction degree of mudstone filler exceeds 93%, the CBR of the sample is greater than 12.0%, the 21 swelling in water is less than 1.39%, and the wetting deformation rate under load is less than 2.89%. The laboratory test 22 results showed that the indexes such as shear strength, CBR and compressibility of mudstone filler all met the 23 requirements of "Highway Subgrade Design Code (JTG D30-2015)"and "Highway Subgrade Construction Technical Code (JTG F10-2006)" for highway embankment filler. Embankment filling can be carried out directly when the 1 gradation is reasonable. 2

Model test scheme 3
According to the design principles, slope ratio, compactness and other requirements of the soft rock embankment in the 4 design codes, and in combination with the actual situation of the supporting project, the structural section of mudstone 5 embankment field test is designed as shown in Fig.4. 6 The centrifugal model test method was used to analyze the stability of embankment under the conditions of 7 optimum water content and continuous rainfall. The specific research contents include:  The similarity ratio (n) of model test was determined to be 100. The model is in the form of full-section 14 embankment, with specific dimensions of 245mm wide and 80mm high at the top, 1:1.5 gradient, 485mm wide and 15 360mm long at the bottom. 16 The test simulated three stages: filling stage, stabilizing stage and operation stage, and converted the pavement 17 structure layer and traffic load into an iron plate weighing 1.657kg. Under the condition of optimum water content, after 18 the simulated stabilizing stage, the equivalent load was applied to the top surface of embankment, and the simulated 19 operation stage was continued. Under the condition of continuous rainfall, combined with the meteorological data of 20 Chongqing from 1961 to 2018, the simulated rainfall was determined as follows: continuous rainfall of 8.4d days, total 21 rainfall of 53.8mm, and average daily rainfall of 6.4 mm. After a simulated filling stage and stabilizing stage, the 22 equivalent load is applied. Then, sprinkle water on the top surface of embankment and slope to simulate the rainfall 23 process, and carry out the next round of rainfall simulation after 24 hours. The cycle will be terminated after 8.4 times, and the operation stage simulation will continue. 1

Process of model test 2
The particle size range and dosage of mudstone filler required for model preparation are shown in Table 1. 3 In order to ensure the degree of compaction, the model was filled in four layers, with compaction first and slope 4 cutting later. Three parallel models were prepared for each working condition, totaling six. Displacement sensors, micro 5 earth pressure sensors and pore water pressure sensors were buried in the process of model making. The model making 6 and installation are shown in Fig.5. 7 In order to avoid the error caused by the over-consolidation of the model due to repeated starting and braking of 8 the centrifuge. Before the test, the embankment model was made at one time, and the layered filling was simulated by 9 applying different accelerations to the centrifuge during the test. It is assumed that the filling stage of embankment is 10 48d, and the stabilizing stage (self-weight consolidation stage from completion of filling to construction of pavement 11 structure layer) is 90d. The operation stage is initially five years after completion (1825d) (the specific time is 12 determined according to the change of settlement data during the test). According to the similar proportion relationship: 13 ＝ 1 2 , the loading state of each stage in the model test process is determined, as shown in Fig.6 . 14 According to the change of test data, under the optimal water content condition, the settlement of embankment 15 model was stable after the centrifuge operated for 157min (24 months after construction); under continuous rainfall 16 condition, the settlement was stable after 243min (44 months after construction). 17 It can be seen from Fig.8 that under the Optimum moisture content, the total settlement of the mudstone 23 embankment (M I) with a height of 8m is 8.18cm (5.15cm in the filling stage + 1.48cm in the stabilizing stage + 1.55cm 1 in the operation stage). Settlement mainly occurred in the filling stage, accounting for 62.96% of the total. Gravity of 2 pavement structure layers and traffic load in operation stage have little influence on mudstone embankment deformation, 3 and the ratio of post-construction settlement to embankment height is only 0.194%, which meets the requirements of 4 "Code for Design of Highway Subgrade" on post-construction settlement of highway embankment. 5

Analysis of model test results
Under the continuous rainfall condition, the total settlement of the mudstone embankment (MⅡ) is 12.68cm 6 (4.83cm in the filling stage + 1.63cm in the stabilizing stage + 6.22cm in the operation stage). Settlement mainly 7 occurred in the operation stage, accounting for 49.05% of the total. 8 It can be seen from Fig.8b that under the effect of rainfall, the mudstone embankment expanded by 2.19cm. Under 9 the effect of continuous rainfall, the expansion rate of embankment is only 0.27%, and the road diseases caused by 10 swelling of mudstone filler will not occur. The ratio of post-construction settlement to height of embankment under 11 rainfall is 0.778%, which still meets the requirements of design code. 12 In terms of settlement rate, the settlement rate is higher during the filling stage due to the increase of embankment 13 height. In the stabilizing stage, the compression coefficient of mudstone filler decreases, the compression modulus 14 increases and the degree of consolidation increases. It continues to settle under the action of its own gravity, but the rate 15 decreases significantly; In the operation stage, the settlement rate of embankment under the optimum water content is 16 very slow, and it reaches stable state after 24 months. Under the action of continuous rainfall, the filler on the top 17 surface and both sides of embankment softens and disintegrates, and the strength decreases. After a short expansion, the 18 settlement rate increases, but the duration is not long. The final settlement trend of mudstone embankment in the same cross-section under two conditions is shown in Fig.8. 5 Under the condition of optimum water content, the cumulative settlement of mudstone embankment along the 6 cross-sectional direction is "upward concave", which is large in the middle and small on both sides. The maximum 7 settlement of model MⅠin the same section is 8.18cm, which occurred at the center line of embankment. The minimum 8 settlement is 5.53cm, which occurred on the right shoulder. The maximum settlement difference of the same section is 9 2.65cm, with little difference. Under the condition of continuous rainfall, the cumulative settlement of mudstone 10 embankment along the cross-sectional direction is in the shape of "M". The maximum settlement of model MⅡin the 11 same section is 19.86cm, which occurred on the right shoulder. The minimum settlement is 10.12cm, which occurred in 12 the right half of the road, 6m away from the center line of the road. The settlement at the center line of the road is 13 12.68cm.The maximum settlement difference of the same section is 9.74cm, which is quite different. 14 There are two main reasons for the settlement deformation of mudstone embankment: one is the compression 15 deformation under the action of self-weight and upper load, and the other is the wetting deformation caused by the 16 softening and disintegration of filler when it meets water. By comparing and analyzing the settlement changes of the 17 two models, it can be seen that the strength of mudstone filler fully meets the requirements of load. Under the condition 18 of continuous rainfall, the filler on both sides of embankment has high softening degree and large deformation, and the 19 settlement is mainly wetting deformation. At the same time, due to the scouring action of rain, the fillers at the 20 shoulders on both sides will peel off and collapse without external restraint, resulting in a sudden increase in 21 deformation. After the experiment, the appearance of the embankment models is shown in Fig.9. 22 After the test, there was no obvious change in the appearance of embankment model MⅠ, only a small number of 23 tiny irregular cracks appeared on the surface. This is because the centrifuge rotates rapidly, which makes the filler on the surface of the model lose water and dry. However, obvious cracks appeared on both side slopes of model MⅡ, 1 which were not deep but distributed widely, especially at both side slopes. This phenomenon is due to the swelling and 2 disintegration of mudstone filler in slope after rainfall. Operation stage, the moisture on the embankment surface 3 evaporates, and the mudstone filler shrinks due to water loss, which leads to a large number of cracks on the surface. It 4 can be seen that under the condition of unprotected slope, continuous rainfall can easily lead to slope slip, surface 5 disintegration and peeling of mudstone embankment. Although the settlement can meet the requirements of the code, 6 the road diseases caused by slope damage can not be ignored. Therefore, in order to ensure the operation quality of 7 highway, it is suggested that the mudstone embankment slope should be covered and waterproof. 8

Analysis of basement earth pressure 9
Mudstone embankment is composed of loose particles, and its settlement process is accompanied by pore compression, 10 water discharge, stress change and so on. Now, the change of basement earth pressure in centrifugal model test is 11 analyzed.The stress change during the test was a dynamic process. Now, the average value of earth pressure in 12 stabilizing stage, operation stage and continuous rainfall operation period is selected for analysis, and the change trend 13 of earth pressure of mudstone embankment is determined as shown in Fig.10. 14 It can be seen from the above figure that the basement earth pressure in different engineering stages is generally 15 parabolic distribution along the cross-sectional direction, which is large in the middle and small on both sides. 16 According to the layout of the micro earth pressure sensors, the load generated by the fillers at the slope is small, and 17 the earth pressure borne by the basement is small. Within the width of pavement, the load generated by fillers is large, 18 and the soil pressure on the basement is large. The distribution trend of earth pressure within the width of pavement is 19 basically consistent with the settlement trend, and the load exerted by embankment on the foundation surface is not 20 evenly distributed along the cross-sectional direction. 21 The ratio of basement earth pressure to the corresponding upper load is shown in Table 2. Field test of settlement monitoring 10 It has good economic and social benefits to use mudstone waste slag as embankment filler. However, the engineering 11 properties of mudstone are greatly affected by the environment. Whether the embankment filled with mudstone meets 12 the requirements of stability needs to be determined by tests. Settlement monitoring is an effective method to study 13 embankment engineering characteristics and optimize design. By analyzing the monitoring data of embankment 14 settlement in different engineering stages, on the one hand, the construction speed of embankment can be reasonably 15 controlled to ensure the construction quality; On the other hand, it can predict the settlement law of embankment and 16 the final settlement after construction. 17 The field test was carried out in HC05 section of Hechang Highway. In this paper, the rationality of the design 18 scheme of mudstone embankment is analyzed based on the field settlement monitoring and indoor model test data . 19 Field monitoring test scheme 20

Layout of monitoring instruments and monitoring points 21
Two monitoring sections are set in the test section, the specific locations are K50 + 740 and K50 + 780. The number of 22 test instruments, test points and layout scheme of the two sections are shown in Fig.11. 23

Frequency of field monitoring
According to the construction schedule, the settlement monitoring frequency was determined as follows: 1 ⑴Filling stage. The monitoring frequency is once every 2 days. When the settlement of embankment is not more 2 than 10mm/d and the displacement change at the toe of slope is not more than 4mm/d, the next layer shall be filled. 3 ⑵Stabilizing stage. In the first month, the monitoring frequency of horizontal inclinometers, top piles are once 4 every 5 days. In the second month, is once every 10 days. The third month and beyond, is once every 15 days. Until the 5 settlement rate is less than 5.0 mm / month in two consecutive months, the next stage of construction can be carried out. 6 ⑶Construction stage of subbase course. The thickness of cushion layer, subbase layer and base layer is 20cm, the 7 monitoring frequency is once for each layer. 8 ⑷Operation stage. The monitoring frequency is once every 15 days in the first 3 months and once every 30 days in 9 the third to sixth months. In case of rainfall, increase the monitoring frequency. 10

11
The monitoring of horizontal inclinometers and side piles includes three stages: filling stage, stabilizing stage and 12 operation stage. The monitoring of embankment top surface includes two stages: stabilizing stage and operation stage. 13 In order to ensure the test progress, rainproof measures were taken for the embankment during the filling stage, without 14 considering the influence of rainfall. There were 7 rainfalls in the stabilizing stage. After the construction was 15 completed, the test section was used as an access road, and the settlement can be regarded as the settlement at the 16 operation stage, and five rainfalls occurred in the operation stage. 17

Analysis of settlement rate of embankment 18
The settlement changes monitored by three horizontal inclinometers with different buried depths in embankment are 19 shown in Fig. 12 ~ Fig.14 load on foundation settlement mainly occurred in the filling stage. When the filling height was 8.0m, the foundation 23 settlement was 10.22mm, accounting for 92.3% of the total, and the settlement rate was 0.21mm/d. Two monitoring points at the toe of the slope appeared slight uplift with the increase of embankment load, but the uplift amount was 1 very small, and the maximum value was only 1.5 mm. The settlement changed little during the stabilizing stage and 2 operation stage, and the accumulated settlement was 11.07mm after the 328-day on-site settlement monitoring. 3 According to the monitoring results, because of the high bearing capacity of the underlying bedrock, the foundation 4 settlement of the supporting project is not affected by various factors such as filler weight, construction machinery, 5 traffic load and external environment. The construction scheme of embankment filling directly can meet the settlement 6 requirements after removing the surface soil and compacting the original ground. 7 The No.2 horizontal inclinometer was buried 3.0m above the basement, and the monitoring content was the 8 settlement of mudstone embankment within 3.0m below and part of foundation. It can be seen from Fig. 15 that the 9 upper filler load is the main cause of settlement. The settlement during the filling stage reached 38.63mm, accounting 10 for 59.23% of the total, and the settlement rate was 1.29 mm/d. In the stabilizing stage, the increase of settlement at the 11 road centerline was 9.54mm in the first month, 3.71mm in the second month and 1.75mm in the third month. The Compare the meteorological data analysis during the field test. In the stabilizing stage, the settlement of 1 # and 11 # 17 monitoring points located at the slopes on both sides suddenly changed during rainfall, which increased by 15.72mm in 18 a short time. Settlement of 2 # and 10 # monitoring points also changed a little, increased by 5.63 mm. However, the 19 settlement of internal points of embankment was not affected by rainfall. Operation stage, the settlement of 20 embankment was not obviously affected by rainfall, including the slopes on both sides. It is because under the effect of 21 early rainfall, the filler at the slope has disintegrated into fine particles, part of which is washed away by rain, and the 22 remaining part forms a slope protection layer, which can prevent water from penetrating into the mudstone 23 embankment.
The No.3 horizontal inclinometer was buried 6.0m above the basement, and the monitoring content was the 1 settlement of mudstone embankment within 6.0m below and part of foundation. It can be seen from Fig. 16 that the 2 maximum settlement during the filling stage occurred at the road centerline, with a value of 42.39mm, accounting for 3 52.94% of the total, and the settlement rate was 3.53 mm/d. Compared with the No.2 horizontal inclinometer, the No. 3 4 horizontal inclinometer produces more settlement during the stabilizing stage and operation stage, and it takes longer 5 for the settlement to reach a stable state. In the stabilizing stage, the increase of settlement was 12.56mm in the first 6 month, 4.02mm in the second month and 2.30mm in the third month. As shown in Fig. 16, the settlement rate of No. 3 7 horizontal inclinometer increased to a certain extent in the early stage of operation. It can be seen that the increase of 8 pavement structural layers load and traffic load in operation stage have a greater impact on the upper part of 9 embankment. 10 According to the settlement changes of horizontal inclinometers at different depths of mudstone embankment, the 11 following conclusions are drawn: ⑴The upper embankment construction has different influences on the settlement at 12 different depths. The settlement range of different points in embankment is 2.58～7.06mm after each layer of filling is 13 completed. The greater the height, the greater the settlement. In order to ensure the construction quality, it is suggested 14 that the filling rate of embankment within 0 ~ 4m height should not exceed 2 layers per day, and within 4 ~ 8 m height 15 should not exceed 1 layer per day;⑵When the mudstone embankment with height of 8m enters the second and third 16 months of the stabilizing stage, the settlement rate at different depths is less than 5mm/ month, which meets the 17 construction standard of pavement structure layer. The construction time of pavement structure layer can be determined 18 as 3 months after the completion of filling, and other embankments with different heights can take this as a reference.⑶ 19 Traffic load and pavement structural layers load have great influence on the stability of the upper part of embankment. 20 In order to ensure the overall stability of embankment, the compactness of this part should be ensured during 21 construction.⑷After a rainy season, the overall strength and stability of mudstone embankment can meet the 22 requirements. However, due to the scouring of rain and its own softening and disintegration, the filling materials at the 23 slope has a large settlement. In order to ensure its long-term stability, measures such as clay wrapping and slope drainage can be adopted to reduce the impact of rainfall on mudstone embankment.