J. Cent. South Univ. (2019) 26: 3066-3086

DOI: https://doi.org/10.1007/s11771-019-4237-x

Enhancing performance of soil using lime and precluding landslide in Benin (West Africa)

Quirin Engelbert Ayeditan ALAYE¹, LING Xian-zhang¹, Yvette Sèdjro KIKI TANKPINOU²,Marx Ferdinand AHLINHAN², LUO Jun¹, Michel Hadilou ALAYE³

1. School of Civil Engineering, Harbin Institute of Technology, Harbin 150001, China;

2. National University of Sciences, Technologies, Engineering and Mathematics (UNSTIM),Abomey 133, Benin (West Africa);

3. Ministry of Public Works and Transport, 01 BP 351 Cotonou, Benin (West Africa)

Central South University Press and Springer-Verlag GmbH Germany, part of Springer Nature 2019

Abstract: The past decade has been characterized by the development of infrastructure in the main cities in West Africa. This requires more comprehensive studies of geotechnical properties of the soil in the region with an aim of creating sustainable development. This paper examined the performance of the soil in Benin (West Africa). In this research, three objectives have been adopted in-depth on the performance characteristics of West Africans soil and aim to (i) accessing characteristics of soil types in the region; (ii) assessing the performance of these soils with 2%, 3% and 5% of lime and (iii) characterizing landslide to evaluate the damage and potential instability. The methods used to examine these objectives are experimental tests according to standard French test. The particle size test, Proctor test, and Atterberg limits test which are physical tests and the mechanical tests such as dynamic penetration test, direct shear test, and oedometer test, were used to assess the first objective. The Proctor test and California bearing ratio test were examined for the second objective and geological, environmental, social and safety study of the river bank slide were evaluated for the third objective. This paper firstly reveals the unstable and stable areas in southern Benin (West Africa) with the presence of clays soil and gives an equation for predicting the unstable and stable area, and secondly shows that the proportion of percentage lime leading to the best performances varying between 2% and 3%. Finally, this paper shows that the sliding of a bank could be the consequence of the sudden receding water recorded in a valley.

Key words: soil characterization; soil performance; clay soil; lime; landslide

Cite this article as: Quirin Engelbert Ayeditan ALAYE, LING Xian-zhang, Yvette Sèdjro KIKI TANKPINOU, Marx Ferdinand AHLINHAN, LUO Jun, Michel Hadilou ALAYE. Enhancing performance of soil using lime and precluding landslide in Benin (West Africa) [J]. Journal of Central South University, 2019, 26(11): 3066-3086. DOI: https://doi.org/10.1007/s11771-019-4237-x.

1 Introduction

Soil contains solid particles composed of minerals, expanding at least ten times to its initial volume when coming in contact with water [1, 2]. For instance, soils continuously change their properties under stress or undergo various types of effects from the stressors [3]. The soil has an inherent capability to permit air and water to pass through its porous structure, providing a medium for plant roots to grow and prevent soil erosion [4, 5]. Nowadays, significant research has been done to interrogate the porous properties of soil, to limit the consequences caused by this phenomenon, representing a significant and growing proportion of entitlements related to the classification of natural disasters [6, 7]. To categorize the soil, the new approach has been adopted: using geological mapping to identify the soil’s characteristics by performing laboratory analysis [8-11]. At present, the diagnosis of inherent capability is realized from samples taken on-site and then laboratory analysis is conducted [12, 13]. For example, most of the laboratories direct shear tests (DST) were applied on samples with widths of approximately two to three inches [14]. However, soil samples collected at one-meter width were used in the analysis containing gravel particles [15]. In addition to laboratory tests, the compressibility characteristics of the soil are usually determined from consolidation tests [16]. General laboratory tests for measurement of the soil compression and consolidation characteristics are oedometer consolidation test, constant rate test (CRS) and Rowe cell test [17]. The procedures for these tests are fully described in Refs. [18, 19]. With the Rowe cell test, large soil samples were found to give higher and more reliable consolidation coefficient (c_v), especially under low stresses than the soil samples of conventional oedometer test [19]. Oedometer tests were applied on soil samples taken from the borehole [20].

The expansive soils greatly increase volume on contact with water and retract during drying up [21]. This shrinkage and swelling properties of soil create a natural risk of differential movement of land when subject to the variations in water content. To educate builders about shrinkage risk, a hazard map showing through probabilistic determination of the risk of shrinkage and swelling was introduced in France [22]. The infrastructure builds on expansive soils present enormous disorders, especially when no special provisions were made during their development. The particularities of expansive soils have long been known that there are multiple works and publications dedicated to expansive soils [23-33]. The geotechnical properties of the soil that can be used to determine the suitability of the soil to support engineering structures with regard to the soil’s stability to be assessed [34]. When the indicators are assessed, the soil is found to be unsuitable and the process of soil stabilization can be performed by altering the properties of the soil to make it usable [34].

Lime is one of the most common chemical stabilizers used in the process of soil stabilization [35]. Previous studies were concentrated on lime’s effect on the strength, compressibility, and swelling of expansive clays soils [36-38]. There are also some works investigating the effect lime on the hydraulic conductivity of soils. Conflicting results were obtained: the rise in hydraulic conductivity with lime content because of the particle reorganization by the flocculation effect [32, 38-40]; the decrease with the rise of lime content [41]. Other works also show that the hydraulic conductivity of cured soils increased until it reached a threshold value at the lime modification optimum (LMO) and it would decrease beyond the modification optimum (LMO) [38, 42]. The hydraulic conductivity of lime treated soils in the unsaturated state is also important for some case under the influence of the freezing and thawing of the pore fluid. The processes used in expansive soils curing were grouped into three categories: mechanical/chemical stabilization of base, geogrid reinforcement, and moisture control barriers [43].

Many researchers have investigated soil stabilization using lime by evaluating different mixing procedures using five types of soils. In some notable studies, two different mixes were performed using 2.5% and 5.0% lime respectively [44]. The results concluded that the lime content increased the soils liquid limit, plastic limit and decreased the plasticity index [45-47]. Meanwhile, another study demonstrated contradicted results that with the increase of lime content, the liquid limit also increases [48]. One researcher attributed these phenomena to the modification of the clay soils surface’s affinity for water precipitation action in the presence of hydroxyl ions. Therefore, the soil treated with lime experiences has increased in the liquid limit [49]. In the same context, other reported that the behavior of the variation of the liquid limit of soils treated with lime is dependent upon the type of soil [50, 51]. Nevertheless, the final results in these cases are reduced in plasticity index. Consequently, after the soil stabilization has been carried out, the soil becomes more workable material, more readily manipulated for the construction activities such as loading, leveling, and others.

In addition, the sensitivity of the soil strength to moisture is reduced. To evaluate the shear strength, some researchers have used the tri-axial test and indirect or flexural tensile strength test [52]. Limited studies have been addressed by the effect of lime on soil compressibility [53-55]. Related to the hydraulic conductivity of the soil, a number of studies show a great improvement of the soil when mixed with lime. On the other hand, some studies found that the soil permeability is significantly decreased with the increment in lime content [56]. Other researches also considered that during a year, there will be seasonal changes in the weather that may adversely affect the soil treatment with lime. The strength gain attributable to the soil treatment measures the soil’s capability to the changing weather [57]. One researcher performed a compressive strength test on immersed and non-immersed samples that had been subjected to lime treatment. The results indicated that the ratio between immersed and non-immersed soil strengths ranged from 0.7 to 0.85, which was significantly higher [58]. The above findings concluded that both swell potential and swell pressure can significantly be reduced in expansive soils that have been treated with lime [59]. The swell potential decreases proportionally with plasticity index. More specifically, the characteristic of the reduced swell potential was seen to be a function of a reduction in the thickness of the diffused double layer [60]. Another research direction was the study of soil compressibility before and after lime treatment by taking into consideration of the initial moisture content of the compacted sample and the time taken for the completion curing process [61]. The samples were treated with 3% lime content [61]. The test procedures were carried out using standard and modified versions of Proctor tests. The samples moisture content was prepared in such a way that it would reflect three different tests scenarios: optimum water content, optimum dry and optimum wet. Further procedures were made by the researcher in addition to the tests performed with an oedometer. He developed a delayed procedure, which involved tests with a constant curing time of 7 d and 28 d [61]. His results buttressed other research findings as to how lime-treated samples exhibit lower compressibility. The soil compaction responded particularly well with decreased compressibility, leading to dynamic compaction over consolidation. Moreover, the samples tested using delayed procedure showed more compression than those tested using the standard procedure, concluding that lime addition did not significantly impact curing time. The remarkable decrement in compressibility related to the lime addition was associated with the short-term reaction.Furthermore, the Pozzolanic reaction had the limited influence [61]. There was a causal effect on the addition of lime to the soil and the reduction of the compression index with an increase in pre-consolidation stress and consolidation coefficient. Many views considered the bond formation between the particles of the soil as the reason for the change in features [62].

Benin (West Africa) is a sub-Saharan country in the West Africa, where the problem of the performance of the soil affects a specific area in the southern part. This area is located in the south of the country, highly affected by the shrinkage problems of soil and landslide. In this area, the soil in the depression of the Lama was developed from the Eocene formations. The known and studied area is limited by the longitudes 1°7'and 2°05'and latitudes 6°6'and 7°'[63]. In the others parts of this area, the rocks are essentially alluvial deposits of clayey and sandy-clayey in nature. These alluvial deposits are sometimes based on kaolinite clayey subordinate to sandstones and iron micro-conglomerates of the Pliocene-Pleistocene and upper Miocene [64]. This systematically looks at dimensions of the problem. Firstly, it establishes the type of soil and its soil structure and secondly it goes on to discuss the performance of the identified soils to achieve the physics tests (particle size test, Proctor test, and Atterberg limit test) and mechanical tests (direct shear test, dynamic penetration test, and oedometer test) are conducted on different kilometric points of ground site at different levels. This paper also provides the use of the pavement structure of the clayey sandy that has undergone treatment by using hydrated lime and the characterization of a landslide in order to evaluate the damage and propose prevention measures in the areas of potential instability.

2 Materials and methods

2.1 Characterizations of soil

The drilling was done at 4 m from the main axis at kilometric points 0+550, 0+650; 0+750 and 0+850. Each sample per borehole and per layer was put into bags of jute and marked according to the number of the borehole, the kilometric point, the samples depth and the date of the collection before being moved to the laboratory for soil description, soil identifications, and testing. The particle size test (PST) was carried out according to French standard (NF)P94-056 [65, 66]; the Proctor test (PT) was done according to French standard (NF) P94-093 [67]; Atterberg limits test (ALT) was carried out according to NF P94-051 [68]. Dynamic penetration test (DPT) was estimated according to French standard (NF) P94-114 [69]; direct shear test (DST) was carried out according to French standard (NF) P 94-071-1 [70] and finally, oedometer test (OT) was done according to French standard (NF) P 94-090-1 [71]. The position and depths of each test are given in Table 1.

Table 1 Tests carried out at each position and depth

Regarding the particle size test, a large quantity of material was put in the oven for 24 h. The sample was placed on the sieve with the largest diameter followed by sieving on each sieve until no grain or chippings could pass through. The aggregates retained were weighed as soon as the operation was carried out, and then the sieve of the last sieve used was weighed before checking whether the sum of the weights obtained (cumulative percent retained + last sieve) did not differ by more than 2% of the initial weight of the sample.

To know the limit of liquidity by Atterberg limit, a sample was taken, then kneaded after being washed and dried in an oven at 50 °C. At first, a sample of 70 g was taken in the cup, then all (materials and cup) was put on the base before being grooved. The determination of the blows counts from which one has 1 cm closure was made, then from that point, it was determined; the other points of ways did not exceed 35 blows after five (05) tries. In a second step, a sample of 10 g of each material (where the 1 cm closure was observed) had been taken and put in the oven to know the water content (W). This liquidity content was determined at 25 blows. To know the limit of plasticity, sampling was taken, dried in an oven at 50 °C for 24 h. After being rolled on a marble plate until 3 mm in diameter to observe the cracks, 10 cm of this sample was taken and put in an oven to determine the dry weight and to know the water content (liquidity limit) followed by the determination of the plasticity index.

In addition, to the Proctor test, a large quantity of the material was taken after quartering, followed by a sampling of 6 kg of this material to determine the water content of the soil specimen. Each 6 kg was wetted with a predetermined percentage of water i.e. 2%, 4%, 6%, 8% and 10% before mixing to the point of obtaining a homogeneous mixture. A first layer was put in the mold and compacted with 55 blows, so on up to five (05) layer before weighing then the whole mold and material were aimed to know the maximum dry density and the optimal water content. It is important to notify that the instrument used in the dynamic penetration test is the dynamic penetrometer PDB, with rods of 3 m in length. The rods were held vertically on the soil and then the dynamic penetrometer was actuated. At each kilometer point, the test was done in a maximum of 5 points each at a distance depth of 0.2 m and the blows count to evolve was mentioned. The minimum result of the blow counts known from the 5 tests tried at each position makes it right to calculate the dynamic resistance following the Eq. (1) and then a safety coefficient (Ω) equal to 20 was applied to this breaking strength to know the allowable stress Eq. (2).

                        (1)

                                   (2)

where q_d is dynamic resistance; m₁ is mass of hammer; m₂ is a mass of a rods; H is the height of the fall; e is average penetration per blow; A_e is cross-section area of the point; g is the acceleration due to gravity; σ is the allowable stress; Ω=20 is safety coefficient.

In addition, the direct shear test (DST) was done on the samples with a section 36 cm² to determine the shear behavior and the shear strength of the samples. The shear strength equation used for saturated soils was expressed as a linear function of effective stress:

                          (3)

where τ is shearing resistance; c_uu is unconsolidated undrained cohesion; φ_uu is unconsolidated undrained of internal friction angle; λ is normal stress on the failure plane.

Considering oedometer test, the compression index and swelling index were determined by reading the difference between 100 and 1000 kPa on the second and third tangent line respectively to curve whereas the compression consolidation was determined by Eq. (4). Also, during the permeability test, 160.6 and 316 kPa of pressure load were applied on the specimen to determine the permeability coefficient by Eq. (5).

                                (4)

k=c_v×m_v×γ_w                                               (5)

where c_v is consolidation coefficient; c_c is compression index; e_o is voids ratio; k is permeability coefficient; m_v is the coefficient of the volume compressibility; γ_w is the unit weight of the water.

2.2 Improvement of clay soils

The samples were taken from the towns of Avrankou, Allada, Bopa and Bohicon in the southern parts of Benin (West Africa). Apart from the classical identification tests, the Proctor test (PT) and California bearing ratio (CBR) tests were done by following Belgian standards EN13286-2 [72] and EN13286-50 [73] respectively. These standards recommended among others such as Proctor test (PT), California ratio (CBR) at five (5) points of variable water contents in three (3) layers of fifty-six (56) blows per layer. Three (3) dosages were retained for the curing of hydrated lime materials, for instance, 2%, 3% and 5%. After the use of the results of this step, the samples taken from Avrankou town were retained by adding 2% and 3% hydrated lime according to French standards NF P94-093 [67] for Proctor test (PT) and according to NF P94-078 [74] for California bearing ratio (CBR). The compression and tension tests were done on an improved sample, namely compression strength and tension strength at 7 d air bath, then compression strength and tension strength at 3 d air bath and 4 d water bath, finally compression and tension at 28 d in air. This approach was helped for the improvements from the curing of clay soil (called red clay) by the addition of hydrated lime which will favor the rational use of this improved soil in a pavement layer.

In addition, the following approaches were addressed: firstly, to know the bearing of the capacity which is a variance of the California bearing ratio (CBR). It expresses the value of the California bearing ratio (CBR) punching measured neither overload nor immersion on a soil sample compacted to the normal Proctor energy. This parameter evaluated the overall bearing capacity and the ability to be compacted and adjusted than to support the traffic on the later site. Secondly, the California bearing ratio (CBR) test was determined after 4 d of immersion. These two parameters were obtained by carrying out the Proctor test and California bearing ratio (CBR) test, which was used in the verification of the bearing capacity criteria such as California bearing ratio (CBR) value of immersed sample during 4 d over California bearing ratio (CBR) and no immersed sample must be superior or equal to 1.

2.3 Investigation and causes of landslide

To investigate the measurement of the geographical orientation of the cracks at the top of the bank, the sliding plans, the direction of the sliding traces on the sliding plans and the direction of propagation of the slide were done with the help of the compass. Then the height of the vertical offsets along the shear plans was measured using a ribbon, allowing diagnosing the causes of the landslide. The distances from the pavement to the river flow channel were measured with a ribbon. Similarly, structural measurements of the fault plans observed in the underlying clayey were carried out using a compass equipped with a clinometer. The environmental damage of the slip was estimated.

3 Results

3.1 Characterizations of soil

3.1.1 Particle size tests

The results of the particle size tests (PST) confer clarity classification of samples are analyzed in Figure 1 by showing the percentage finer by weight versus grain diameter.

The soil with grain diameters d<0.006 mm has 0% passing aggregates at all the kilometric points conferring zero presence of fine silt. The soil having 0.006 mm≤d<0.01 mm has 0% of passing aggregates at all kilometric points, while the soil having 0.01 mm≤d<0.02 mm contains up to 5% of the passing aggregates. Thus, it is only 5% medium silt. From 0.02 mm≤d<0.06 mm diameter, the soil at the kilometric points 0+750 and 0+850, has smaller aggregates than the soil at the kilometric points 0+550 and 0+650. Thus, this soil presents between 5% and 50% aggregates of coarse silt.

Figure 1 Percentage finer by weight versus grain diameter (F is fine; M is medium; C is coarse)

The soil with diameter 0.06 mm≤d<0.2 mm is considered sand. The soil at the kilometric point 0+750 is thinner than all aggregates from other positions with a percentage of passers-by ranging from 50% to 95%, while for the soil at kilometric point 0+650 the aggregates are less fine with a percentage of passers-by between 42% and 80%. However, for soil at the kilometric points 0+550 and 0+850, the aggregates are intermediate with a ratio of the passers between 45% and 85%. For 0.2 mm≤d<0.6 mm, the soil at kilometric point 0+750 presents more medium aggregates than other positions with a percentage of passers-by between 95% and 99%, while this soil at the kilometric point 0+650 presents fewer medium aggregates with a percentage of passers-by between 80% and 90%. In addition, this soil at the kilometric points 0+550 and 0+850 has some average medium aggregates whose percentages of passers are between 85% and 93%. For 0.6 mm≤d<2 mm, the soil at kilometric point 0+750 presents more smaller aggregates, when it is in the vicinity of diameter d≈0.6 mm, while soil at kilometric point 0+650 has fewer smaller aggregates.

3.1.2 Proctor and Atterberg limits tests

The results of Proctor test and the Atterberg limits tests (ALT) are shown in Table 2 and Figure 2.

Table 2 Properties of sample soil according to Proctor and Atterberg limits tests

Figure 2 Casagrande plasticity chart

The samples with high plastic silts are under the line A (Figure 2). The visual soil description defines that the soil of the borehole consists of less plastic sandy clay for kilometric points 0+550, 0+650, 0+750 and 0+850, conferring very significant results.

3.1.3 Dynamic penetration tests

The samples are analyzed according to the blow counts, breaking strength and allowable stress soil.

The similarity between the geotechnical soil profile and blow counts versus depth for the following kilometric points 0+550; 0+650; 0+750 and 0+850 are shown in Figure 3.

At the kilometric point 0+550 as shown in Figure 3(a), the upper layer of the subsoil consists of loose fine sand with a blow N₁₀<4. From a depth distance of 1 m, the fine sand becomes denser, and the dense state is found at a depth of 2 m. At the kilometric point of 0+650 as shown in Figure 3(b), the water depth is about 3 m. The water lies on a peat of soil layer, inter-bedded by clay with soft consistence and correlates well with the recorded blow counts between 0 m and 7.6 m depth. At the kilometric points of 0+750 as shown in Figure 3(c), the water depth is about 5 m, followed by a very soft-to-soft silt up to 13.5 m depth with zero recorded blow counts.

Figure 3 Geotechnical soil profile and blow counts versus depth:

The silt layer overlays a stiff clay layer with inter-bedded sand with a mean blow count about 10. At kilometric points of 0+850 as shown in Figure 3(d), the water depth is approximately 1.3 m followed by loose silt layer of 4.5 m. The silt layer lies above a soft-to-stiff clay layer. In general, the boreholes are comparable with the recorded blow counts from the standard penetration tests.

The similarities between the geotechnical soil profile and the dynamic resistance versus depth for the following kilometric points of 0+550; 0+650; 0+750 and 0+850 are shown in Figure 4.

The dynamic resistance recorded is generally low or zero at the kilometric points of 0+650 and 0+750 as shown in Figures 4(b) and (c) with high intermittent values of 0.2 m to 15 m depth. Then the low-average kilometric points 0+550 and 0+850 from 0.2 m to 15 m depth are shown in Figures 4(a) and (d). The recorded contrary values are ranged from 0 kPa to 1095 kPa at the kilometric points of 0+550 and 0+850 as shown in Figures 4(a) and (d), then from 0 to 375 kPa at the kilometric points of 0+650 and 0+750 as shown in Figures 4(b) and (c) up to 15 m depth.

The similarities between the geotechnical soil profile and the allowable stresses versus depth for the following kilometric points 0+550; 0+650; 0+750 and 0+850 are shown in Figure 5.

At the kilometric point of 0+750 as shown in Figure 4(c), Figure 5(c) and from 0 m to 13.5 m, the soil has dynamic resistance with zero allowable stress and then from 13.5 m to 20 m depth, the dynamic resistance and allowable stress were increased to reach the maximum values of 7500 kPa and 375 kPa respectively. The soil is resistant at this depth. At the kilometric point of 0+650, from 7.5 m to 16 m depth as shown in Figure 4(b),Figure 5(b), the soil has the dynamic resistance with zero allowable stress; the maximum values of dynamic resistance and allowable stress were found at 12-14 m depth. So the soil is resistant between 12 m to 14 m. At the kilometric point of 0+550 as shown in Figure 4(a), Figure 5(a) and from 0 m to 16 m depth, the soil presents three peaks of dynamic resistance and allowable stress peaks between 0 m to 3 m, 7.5 m to 9 m and from 12 m to 13 m. So the soil is resistant at these three depths than the others. At the kilometric point of 0+850 as shown in Figure 4(d) and from 0 m to 3 m depth, the dynamic resistance is zero. This soil presents two peaks of dynamic resistance and allowable stresses peaks between 3 m and 4 m, 6.5 m at the kilometric point of 0+850 as shown in Figure 5(d). So the soil is more resistant from 3 m to 4 m depth and from 6.5 m to 9 m than the others.

Figure 4 Geotechnical soil profile and dynamic resistance versus depth:

Figure 5 Geotechnical soil profile and allowable stress versus depth:

3.1.4 Direct shear tests

To evaluate the short-term behavior of the soils, the soil samples are collected and tested at the kilometric points of 0+650, 0+750 and 0+850. The first objectives are analyzed according to the shearing resistance (τ) versus normal stress on the failure plane (λ). Thus, the results of the direct shear test (DST) which is mechanical test are shown in Table 3 and presented in Figure 6.

Table 3 Properties of sample of soil according to direct shear test

The soil at the kilometric points of 0+650, 0+750 and 0+850 as shown in Figure 6, its state has constant shearing resistance (τ) or unconsolidated undrained cohesion (c_uu) of 19 kPa, 28 and 40 kPa when the unconsolidated undrained internal friction angles (φ_uu) are 29°, 15.1° and 15.2°, respectively. During the test process, no drainage of the water is observed during the application phase of the normal stresses (λ) and the shearing resistance (τ) phase itself. The c_uu and φ_uu at the kilometric points of 0+650, 0+750 and 0+850 respectively are different to zero, then clear evident is that there is an ability of a coherent soil with the existence of a shear stress. At the kilometric point of 0+650 as shown in Figure 6(a), all parts having cohesion greater than 19 kPa with the internal friction angle bigger than 29° are regarded as an unstable area. At kilometric point 0+750 as shown in Figure 6(b), all parts having c_uu greater than 28 kPa with the φ_uu greater than 15.1° are also considered the unstable area. At the kilometric point of 0+850 as shown in Figure 6(c), all parts having cohesion greater than 40 kPa with the internal friction angle greater than 15.2° are regarded as an unstable area.

Figure 6 Shearing resistance versus normal stress:

3.1.5 Oedometer tests

The samples are analyzed according to the voids ratio versus vertical effective stress. Thus, results of the Oedometer test (OT) are shown in Table 4 and presented in Figure 7.

Following the oedometer test, Figure 7 and Table 4 show a e₀>0.5, low compressibility coefficient (C_c) value between 0.09 and 0.16 and very low swelling coefficients (C_s=0.01).

Figure 7 Voids ratio versus vertical effective stress:((1)-Tangent line to curve; (2)-Tangent to linear portion of curve in virgin compression range; (3)-Tangent to linear portion of curve in virgin swelling range; (4)-Intersection of lines (1) and (2) (vertical effective stress at point 4 equals pre-consolidation stress))

Table 4 Properties of soil sample according to oedometer test

3.2 Improvement of clay soil using lime

The results are shown in Figure 8 and Table 5. The lime hydrate treatment has brought a substantial improvement at the California bearing ratio (CBR) of soil samples from the various sites as shown in Figure 8. The soil samples taken from Avrankou town are shown in Figure 8(a). The physical properties test of improved soil sample have CBR_max=157, W=15% and P_I=19% against the physical properties test of natural samples with 30 California bearing ratio (CBR), thus improving by 3% of hydrated lime. Considering the samples taken from Allada town as shown in Figure 8(c), for the improved soil sample, the CBR_max=140.5, W=12.3% and P_I=21.50% against the natural samples of CBR_max=14.5. This performance relates to the addition of 3% of hydrated lime. The improved soil samples taken from Bopa town as shown in Figure 8(b) have CBR_max=127, W=12.6% and P_I=11% against natural soil sample of CBR_max=11. This performance relates to the addition of 5% hydrated lime. Finally, the improved soil samples taken from Bohicon town as shown in Figure 8(d), have CBR_max=150, W=17.1% and P_I=19.5% against the natural sample of CBR=12.5. This performance regards the addition of 5% of hydrated lime.

Figure 8 California bearing ratio versus water content:

After the above analysis, the bearing capacity which is a variance of the California bearing ratio index is presented in Table 5 and then the results of the physical tests are presented respectively in Figure 9.

The treatment of clayey soil with hydrated lime has led to great improvement in their characteristics (Table 5, Figures 8 and 9). The analysis of these different results made it possible to envisage a thorough study with 2% and 3% of hydrated lime on the sample, as presented in Table 6. The site chosen for this study is in Avrankou town.

From Table 5, Figures 9(a) and (b), the verification of the soil bearing capacity of soil was made. In our case study, California bearing ratio (CBR) value of immersed soil samples during 4 d over the California bearing ratio (CBR) value of zero soil sample immersed should be greater than or equal to 1.

As shown in Figures 9(a) and (b) and Table 5, the dosage of 3% hydrated lime, California bearing ratio (CBR) value of immersed sample during 4 d over California bearing ratio (CBR) value of zero immersed sample is 110.5/91=1.21, verifying the criterion. In addition, from Table 6, the CBR at 95% of the modified optimum proctor shows the minimum value of 30% recommended for foundation layer.

3.3 Causes and propagation of landslide

An unstable environmental studied slide area was Oueme valley near the Sakete plateau (Benin (West Africa)). The river has meandered with a dissymmetrical profile having an almost vertical, sandy, convex bank (straight bank on the opposite side to the pavement) and a medium to the steep concave bank, clayey (bank on the side of the pavement). (Figures 10 and 11).

In the bend, erosion affects the concave bank due to the current of the surface while a bottom current tends to the deposit on the convex bank material transported from upstream. More specifically, the landslide occurred in the curve on the left bank of the river. The slope is medium to strong. There is very close proximity to the pavement with the boundary of the channel which shows the low flow of the river. The channel distance during low flow periods from the river to the pavement is low and varies from 15 m to 90 m. In this area, the rocks are essentially alluvial deposits of clayey and sandy-clayey in nature. (Figure 12).

Table 5 Properties of samples of soil improved at different percent lime

Figure 9 CBR versus water content of sample:

Structurally, kaolinite clayey are foliated and fault plans oriented from North-South (NS) to North-Northeast (NNE), South-Southwest (SSW) and from East-Northeast (ENE), West-Southwest (WSW) to East-West (E-W) with strong dips (50°) to sub-vertical dips are observed (Figure 12). Sandstones and micro-conglomerates are presented as the blocks from decimetric to metric dimension, affirming the existence of faults in the river valley that affected upper Miocene and Pliocene- Pleistocene sediments. It could lead to reactivation of the crystalline basement faults creating the zones of weakness and having controlled the structuring of the river valley, reflecting a neo-tectonic activity.

Table 6 Properties of sample of soil according to compressive and tension strength tests

Figure 10 Partial view of Oueme river in area affected by slide of bank

Figure 11 Interpretative scheme of helical current in bend of Oueme river in area affected by slide of bank adapted to Degoutte [75]:

Figure 12 Structure of clayey by faults on left bank of Oueme river and presence of iron sandstone blocks

The exploitation of the sand in the bottom of the river in the sector of the bend (Figure 13) leads to a depression of the riverbed. As a result, the height of the bank is increased and sliding stability may be reduced (Figure 14).

This was also exposed by DEGOUTTE [75], which shows the aggravating role of the bed penetration at high advantageous erosion phenomena material. The banks of a river are stable to erosion and sliding. Depression of the Oueme river bed is due to the removal of a large volume of sand in the bend (Figure 13). Therefore, it is an important aggravating factor of the river bank sliding.

Figure 13 Removal or extraction of sand from bottom of Oueme river

Figure 14 Excavating of riverbed sand and impact on bank slide [75]

The observed geological phenomenon is a mass sliding occurring at the bend of the Oueme river in Affame-Yovodonou on its left bank. In the field, shear plans oriented from North-Northeast (NNE) and South-Southwest (SSW) to Northeast- Southwest (NESW) and strongly inclined towards the west are observed, as shown by an arrow in Figure 15 and cause vertical movement of compartments.

Figure 15 Visible sliding plane at bank of river near Yovodonou stream

The height of the vertical displacement of the compartments (dislocation) varies from 1.5 m to 6 m in the sector of the bridge erected on the stream of the Oueme river called Yovodonou. The sliding striations are inclined at 40° southwest and indicate an overall displacement area towards the southwest bank of the river bed, lying at the beginning of the slide. The propagation of the breakage energy was translated on both sides of this area at the beginning and in particular in the South, by slightly sliding from 0.2 m to 0.4 m displacement. Failure plans oriented from North-South (N-S), Northeast (NE)- Southwest (SW) to East-West (E-W) and the sub-vertical to vertical is visible at the top of the bank. Likewise, the arc cracks, concentric and parallel slits from 6 m to 8 m at the top of the bank (Figure 16(a)) and on the pavement (Figure 16(b)) are observed.

Considering the environmental aspect such as the risks of break and the evaluation of the damages has been noted as the cracks on the pavement and at the top of the bank which leads to the modification of the local topography of the soil. For the first aspect of the cracks on the pavement and at the top of the bank in its propagation and the sliding energy has caused cracks and sliding on the pavement of the Akpro Missrete-Dangbo-Adjohoun-Kpedekpo road at the height of the kilometric point of 39+00, this led to the non-practicability of the pavement in this area (Figure 17).

Figure 16 Cracks observed on bank top (a) and cracks on pavement (b)

Figure 17 Cracks on pavement at kilometric point 39+00

Similarly, concentric cracks sometimes accompanied by sliding or subsidence are observed at the top of the bank. For the second aspect relating to the modification of the local topography of the soil, the area of the bridge of Yovodonou with maximum sliding has led to a modification of the local topography by creating very high slopes of the bank (Figure 18).

Figure 18 High slope of left side bank of Oueme River in vicinity of Yovodonou bridge

4 Discussions

The classification of the samples was done based on the particle size method, Casagrande plasticity chart and following by the system AASHTO [76]. The Casagrande plasticity chart (Figure 2) and a classification of samples were made. Depending on the system AASHTO [76], the classification of samples made from the results of particle size (from 80 μm) and Atterberg limits tests (ALT) revealed that all the samples studied were organic silt clays. Following Figure 4, the soil has not fine silt, and only 5% medium silt presents between 5% and 50% aggregates of coarse silt. So from 16 m to 20 m of depth, the soil is moderately impermeable. Concerning the sand, the soil has not enough fine sand, enough medium and coarse sand so from 16 m to 20 m of depth, the soil is moderately permeable. Therefore, the soil is little deformable from 16 m to 20 m of depth.

The Atterberg limit test (ALT) shows that for all studied soil samples, the percentage of the liquid limit is between 66% and 76% with the percentage plasticity index between 30% and 36%. Figure 5 shows that at all kilometric points, the studied samples which are very plastic silt and very plastic organic soils were deformable due to the degree of the plasticity from 16 m to 20 m in depth.

No tested points offered apparent refusal to the dynamic penetration up to the depth limit tests (15 m). The cutting of the soils deduced over 15 m depth revealed that the soils in this place are made of peaty, sandy and clayey. At the kilometric point of 0+550, the soil presents the dynamic resistance from 0 m to 3 m and from 6 m to 14 m which is considerable and from 3 m to 6 m is low. At the kilometric point of 0+650, the soil presents an important dynamic resistance and increase when entering from 7.5 m to 14 m of depth. At the kilometric point of 0+750, the dynamic resistance is the least but increases when entering from 13.5 m to 14 m. At the kilometric point of 0+850 as shown in Figure 4(d), the soil presents the dynamic resistance from 7 m to 9 m, which is an important dynamic resistance and the least from 3 m to 4 m and from 9 m to 14 m. So, following Figure 4, for the soil at different kilometric points and at a different depth, dynamic resistances increase according to the depth.

Considering the allowable stress (Figure 5), we noticed the same characteristics with the dynamic resistance as described above.

Based on the direct shear test (DST) carried out from 16 m to 20 m of depth, the increased distance proportionally increases, with the cohesion value with a reciprocal decrease of internal friction angle. Since the un-drained and un-consolidated cohesion (c_uu) varies at the same time with the internal friction, then there is compensation. Hence the unstable area remains constant. At the kilometric point of 0+650 as shown in Figure 6(a), the c_uu=19 kPa with φ_i=29° define limits of constraint if τuu=19 kPa and 100 kPa≤λ≤ 300 kPa, then the soil is stable at this kilometric point. At the kilometric point of 0+750 as shown in Figure 6(b), the c_uu=28 kPa with φ_uu=15.1° define limits of constraint, if τuu=28 kPa and 100 kPa≤λ≤300 kPa, then, the soil is stable at this kilometric point. At the kilometric point 0+850, as shown in Figure 6(c), the c_uu=40 kPa with φ_uu=15.2° define limits of constraint if τuu=40 kPa and 100 kPa≤λ≤300 kPa, then the soil is stable at this kilometric point. Following Figure 6, at all kilometric points, if τ≤c_uu, the soil is stable; if τ>c_uu, the soil is:

Stable if ;

Instable if

where τ is shearing resistance; c_uu is unconsolidated undrained cohesion; φ_uu is unconsolidated undrained angle of internal friction angle; λ is normal stress on the failure plane.

The changes in permeability resulting from compression are drastic for fibrous peat soils [77]. From Table 4, the values of the permeability coefficient obtained in the present study have shown reciprocal proportion between pressure and variation of permeability coefficient (k), implying k(160.6 kPa)=2.4×10^-10 m/s and k(316 kPa)=1.2× 10^-10 m/s at the kilometric points of 0+650 as shown in Figure 8(b). Whenever the pressure is low at the kilometric point that means the permeability coefficient is high and vice versa. Outside the soil kilometric point of 0+850, the soils at the kilometric points of 0+550, 0+650 and 0+750 is moderately compressible from 16 m to 19 m of depth. Pre-consolidation stress (σ') at kilometric points of 0+550 and 0+750 is shown in Figures 8(a) and (c), with 50 kPa <σ'<100 kPa. Therefore, in accordance with the Terzaghi theory [78], the soil is mid-consistent from 16 m to 19 m of the depth unlike the soils of the respective kilometric point of 0+650, 0+850 as shown in Figures 8(b) and (d) of which 100 kPa <σ'<200 kPa. This revealed that at these kilometric points the consistent soil is from 18 m to 20 m. These characteristics were confirmed by the low permeability of the soil at all studied kilometric points. Therefore, it shows that the soil is stable or little deformable from 16 m to 20 m.

Concerning the improvement of the clay soil, the above-outlined approaches have made it possible to the better understanding of the various improvements brought by the treatment of clay soils by adding hydrated lime, which promotes a rational use of the improved soil in the layer of pavement. This study focused on the soil sample taken from Avrankou town, which confirmed the substantial improvement of the studied soil characteristics after incorporation of the hydrated lime. The value of California bearing ratio (CBR) at 95% of the modified optimum proctor which shows the minimum value of 30% is recommended for the foundation layers (121 and 118 respectively for 2% and 3%). The compressive and tension strengths are satisfactory for the use as a foundation layer. The degradation index C'/C which makes it possible to appreciate the sensitivity of the mixtures to water is also acceptable: 0.41 and 0.58 respectively for 2% and 3% of hydrated lime. In view of the requirements, the value of C'/C>0.50 is desirable and considered the criterion of acceptability concern.

Concerning the unstable area, the causes of the sliding were the parallel and concentric cracks at the top of the Oueme river bank. The plans and traces visible sliding at the top of the bank confirm that it is a mass sliding at the level of the bend of the river. This sliding occurs during the period of receding water following a flood event due to heavy rainfall in northern Benin (West Africa) and whose alert had been given to the Oueme river by the warning system set up by the Government. This bank slide could be the consequence of the sudden recede registered in the Oueme river valley. Indeed, due to the natural morphological evolution of the river, the left bank in the concave area of the bend of the river on the side of the pavement formed the deposits of low-draining sediments essentially consisting of clayey to clayey-sandy nature of prey to erosion. The geometry of the bank shows a medium to the strong slope. During the flood, the water of the river saturates the clayey soil and exerts a stabilizing pressure on the bank. The brutal receding water constituted an unfavorable circumstance for the holding of the left bank. The pressure exerted by the river channel on its left bank decreased roughly and the driving forces due to the weight of the lands above the potential sliding breaking surface outweighed the resistant forces due to friction along the breaking surface. The bank has therefore slide and the sliding energy had spread on both sides where slide created cracks in the bend at the top of the bank sometimes accompanied by less important displacement. These cracks over 800 m in length at the top of the bank cause degrading in the inter-state pavement in the bend of the river, conferring the potential sliding surfaces during subsequent receding water as revealed by DEGOUTTE [75]. The lands will be able to slide again along these plans as soon as their cohesion becomes insufficient [79, 80] and when a sliding has occurred, it can trigger new landslides by regression [75]. The sliding of the bank on the side of the pavement is thus caused by the set of factors, such as the erosion on the surface by the hydrological currents of the river, the underground action of infiltrated rainwater (more or less slow processes of dissolution or internal erosion increasing the cracks), and the fluctuation of groundwater levels. Nevertheless, the action of the tectonics should not be occulted in the triggering of the bank sliding in this area because the breaks are the zones of weakness which can favor the displacement of the grounds. In addition, the parallelism between the orientation of the cracks measured on the bank along the Oueme river is the clayey and the shear planes of the sliding which should not be neglected.

The causes of sliding include certain risks in the neighboring populations and the environmental factors. The non-perception or ignorance of the existence of these risks by the populations results in a recklessness or a lack of civic responsibility which is translated by the anarchic extraction of sand on the bottom of the bed of the Oueme river in the area of the curve leading to crack of area and increased vulnerability of structures such as the pavement and the bridge, whose proximity to the river bed (from 15 m to 90 m) is not taken into account.

By interacting the analysis of the landslide, it could be notified that the causes of the sliding allow the driving forces due to the weight of the lands above the potential breaking surface of the landslide to override the resistant forces due to friction along the breaking surface. The bank has therefore slipped and the energy of the sliding propagates on both sides of the area of the beginning of the sliding creating cracks in the curve at the top of the bank. These observed concentric cracks are the result of landslide or subsidence of the soil modifying the local topography of the soil and consequently degrading the infrastructure such as the pavement which is in the area of the river.

5 Conclusions

This paper is based on the enhancing performance of Southern Benin (West Africa) soil in West Africa using lime and precluding the landslide soils. According to the particle size test, this soil is not very deformable because of the dimensions of the particles or aggregates that make the layers ranging from 0.01 mm to 2 mm. Based on the Atterberg limit test, this soil is deformable due to the very plastic silt and very plastic organic soils. This plasticity is due to the ratio of the liquid limit between 50% and 100% and a percentage of plasticity index between 0% and line A of the plasticity diagram. The dynamic penetration test shows that resistance of this soil ranging from 1000 kPa to 22500 kPa increases with the depth whereas, through the oedometer test, it was found that this soil is moderately compressible because of the low compression index between 0.09 and 0.16, then alternate, mid-consistent and consistent because of the very low permeability coefficient between 1.2×10^-10m/s and 1.4×10^-10 m/s under a 316 kPa of load pressure, and between 1.9×10^-10 m/s and 2.3×10^-10 m/s under a 160.6 kPa of load pressure. Following the direct shear test, apart from the unconsolidated undrained cohesion (c_uu) and unconsolidated undrained internal friction angle (φ_uu), which make it possible to know the stable areas and unstable areas through the coulomb straight line, this study allows us to obtain a formula that determines the stable and unstable areas as a function of the normal stress (λ), shearing resistance (τ), unconsolidated undrained cohesion (c_uu), unconsolidated undrained internal friction angle (φ_uu).

Considering the second objective of this paper, it has been discovered that in general, the values of the California bearing ratio (CBR) index of the clay soil are less than 20 despite the liquid limits and plasticity index are high. With the addition of hydrated lime, the California bearing ratio (CBR) index experienced a very appreciable improvement. Also, it was found that a significant decrease of the plasticity index (P_I). The performances recorded after incorporation of 2% and 3% hydrated lime are above the recommended values for the materials destined for the foundation layers. Apart from the California bearing ratio (CBR) index, all other parameters have reached the thresholds generally recommended for improved base layer materials and California bearing ratio (CBR) of at least 160 is required in the laboratory for the base layer. The proportion leading to these performances varies between 2% and 3%. Viewing all the above, the study recommends that the clay soils (red clay in the case of the second objective studied), treated with hydrated lime can be used for the foundation layers where the California bearing ratio (CBR) index required is greater than or equal to 30.

Concerning the third objective, the geological and environmental studies carried out have shown that the sliding of a bank could be the consequence of the sudden receding water recorded in a valley. However, the removal of sand from the bottom of a riverbed would be an aggravating factor. It is, therefore, necessary and urgent to carry out a detailed geological and geomorphological study for the localization and identification of areas of potential instability presenting a great significant risk of breaking soil in the localities and the regular monitoring of the sector of a river. In view of the above regarding the sliding of the soil, the following recommendations may be implemented below:

1) Conduct a geological and geomorphological study to identify and analyze sliding on a bank and around the river areas of potential instability of rock masses presenting a significant risk of rupture.

2) On the basis of the scientific data from the studies, propose the state or government to take preventive measures against the risks factors to which the populations are exposed. For example in the realization of systems or protective works (rigid screens or deformable, energy-dissipating screens, wall protection), carry out regular and periodic monitoring of the sector.

3) Take conservatory measures by prohibiting unstable areas. After sensitization, the occupation of the areas avoids risk and taking sand from the river bed.

4) Conduct intensive reforestation at the embankment side of the pavement to reduce long-term shocks caused by hydraulic currents and erosion. However, this efficiency is limited and forest protection is not absolute according to their age, their nature, their position. Because more or less violent floods can uproot them and carry them away.

5) Avoid staying, occupying and withdrawing the sand in the bend of the river at the level of the area at risk.

6) Help the town hall to identify areas at risk.

Nomenclature

ALT

Atterberg limits test

CBR

California bearing ratio

C

Compressive strength at 7 d, kPa

C′

Compressive strength at 3 d air and 4 d in water, kPa

A

Cross-section area of cone, m²

d

Diameter, mm

DST

Direct shear test

DPT

Dynamic penetration test

e

Average penetration per blow

ENE

East-northeast

E-W

East-west

g

Acceleration due to gravity, m/s²

H

Height of fall, m

k

Permeability coefficient, m/s

P_I

Plasticity index, %

L_L

Liquid limit, %

m

Mass of hammer, kg

m′

Mass of rods, kg

NNE

North-northeast

NS

North-South

OT

Oedometer test

OC

Organic content

PST

Particle size test

PT

Proctor test

SSW

South-southwest

T

Percentage of ratio, %

W

Water content, %

WSW

West-southwest

Greek symbols

γ_s

Unit weight of solid grains, kN/m³

γ_d

Unit weight of dry soil, kN/m³

γ

Total unit weight of soil, kN/m³

σ

Allowable stress, kPa

σ′

Pre-consolidation stress, kPa

q_d

Dynamic resistance, kPa

c_uu

Unconsolidated undrained cohesion, kPa

φ_uu

Unconsolidated undrained angle of internal friction, (°)

λ

Normal stress on failure plane, kPa

Ω

Safety coefficient

τ

Shearing resistance, kPa

C_v

Consolidation coefficient

C_c

Compression index

e_o

Voids ratio

m_v

Coefficient of volume compressibility

γ_w

Unit weight of water, kN/m³

C_s

Swelling index

ρ

Density dry, kg/m³

References

[1] LAMBE T W, WHITMAN R V. Soil mechanics SI version [M]. John Wiley & Sons, 2008.

[2] HILLEL D. Fundamentals of soil physics [M]. Academic Press, 2013.

[3] BROWN S, LUGO A E. Rehabilitation of tropical lands: A key to sustaining development [J]. Restoration Ecology, 1994, 2(2): 97-111.

[4] BATEY T, MCKENZIE D. Soil compaction: Identification directly in the field [J]. Soil Use and Management, 2006, 22(2): 123-131.

[5] CARTER M R, GREGORICH E G, ANDERSON D W, DORAN J W, JANZEN H H, PIERCE F J. Chapter 1: Concepts of soil quality and their significance [J]. Developments in Soil Science, 1997, 25(97): 1-19.

[6] MILETI D. Disasters by design: A reassessment of natural hazards in the United States [M]. Joseph Henry Press, 1999.

[7] KUKUKA A. Natural disasters and FDI inflow in the developing countries of South-Eastern Asia [J]. Prace Naukowe Uniwersytetu Ekonomicznego we Wrocawiu, 2014(370): 208-216.

[8] LAUBACH S E, OLSON J E, GROSS M R. Mechanical and fracture stratigraphy [J]. AAPG Bulletin, 2009, 93(11): 1413-1426.

[9] FARIFTEH J, FARSHAD A, GEORGE R. Assessing salt- affected soils using remote sensing, solute modelling, and geophysics [J]. Geoderma, 2006, 130(3, 4): 191-206.

[10] EDITION S, BUOL S, SOUTHARD R, GRAHAM R C. Morphology and composition of soils [M]// Soil Genesis and Classification. 6th ed. 2011.

[11] NICHOLSON G A, BIENIAWSKI Z T. A nonlinear deformation modulus based on rock mass classification [J]. International Journal of Mining & Geological Engineering, 1990, 8(3): 181-202.

[12] LIU R H, GRODZINSKI P, YANG J, LENIGK R. Self-contained, fully integrated biochips for sample preparation, PCR amplification and DNA microarray analysis [M]// Integrated Biochips for DNA Analysis. 2007.

[13] HAYES T, HASTON K, TSUI M. Atrazine-induced hermaphroditism at 0.1 ppb in American leopard frogs (Rana pipiens): Laboratory and field evidence [J]. Environmental Health Perspectives, 2003, 111(4): 568-575.

[14] SIVAKUGAN N, DAS B M, LOVISA J. Determination of c and φ of rocks from indirect tensile strength and uniaxial compression tests [J]. International Journal of Geotechnical Engineering, 2014, 8(1): 59-65.

[15] NICKS J, ADAMS M. Large-scale direct shear testing of common open-graded aggregates [C]// Geo Congress, 2014.

[16] TERZAGHI K, PECK R B, MESRI G. Soil mechanics in engineering practice [M]. John Wiley & Sons, 1996.

[17] CHU J, BO M W, CHANG M. Consolidation and permeability properties of Singapore marine clay [J]. Journal of Geotechnical and Geoenvironmental Engineering, 2002, 128(9): 724-732.

[18] ARNOLD P. Manual of soil laboratory testing. Vol. 2. Permeability, shear strength and compressibility tests [J]. Geofisica Internacional, 1995, 65: 339-340.

[19] MCCARTNEY J S, ZORNBERG J G. Centrifuge permeameter for unsaturated soils. II: Measurement of the hydraulic characteristics of an unsaturated clay [J]. Journal of Geotechnical & Geoenvironmental Engineering, 2010, 136(8): 1064-1076.

[20] LEFEBVRE G M B, PHILIBERT A, HORNYCH P. Evaluating K_o in champlain clays with hydraulic fracture tests [J]. Canadian Geotechnical Journal, 1991, 28(3): 365-377.

[21] ALAOUI A, LIPIEC J, GERKE H H. A review of the changes in the soil pore system due to soil deformation: A hydrodynamic perspective [J]. Soil & Tillage Research, 2011, 115(5): 1-15.

[22] VINCENT M, PLAT E, LE ROY S. Cartographie de l’aléa retrait-gonflement et plans de prévention des risques [J]. Revue Francaise de Géotechnique, 2007, 120–121: 189-200. (in France)

[23] GHASABKOLAEI N, CHOOBBASTI A J, ROSHAN N. Geotechnical properties of the soils modified with nanomaterials: A comprehensive review [J]. Archives of Civil and Mechanical Engineering, 2017, 17(3): 639-650.

[24] AMIEJUS J, DIRGLIEN N, NORKUS A. Comparison of sandy soil shear strength parameters obtained by various construction direct shear apparatuses [J]. Archives of Civil and Mechanical Engineering, 2014, 14(2): 327-334.

[25] MYNAREK Z, GOGOLIK S, POTORAK J. The effect of varied stiffness of soil layers on interpretation of CPTU penetration characteristics [J]. Archives of Civil and Mechanical Engineering, 2012, 12(2): 253-264.

[26] SNETHEN D R. Characterization of expansive soils using soil suction data [C]// American Society of Civil Engineers Amsterdam: ASCE. 1980: 54-75.

[27] LYTTON R L. Foundations on expansive soils [M]. Elsevier Science Publishers B. V, 1988.

[28] MOUROUX P, MARGRON P, PINTE J C. La construction économique sur sols gonflants [M]. Editions du Bureau de Recherches Geologiques et Mineres, 1988: 14. (in France)

[29] AL-RAWAS A A, MCGOWN A. Microstructure of Omani expansive soils [J]. Canadian Geotechnical Journal, 1999, 36(2): 272-290.

[30] DJEDID A, BEKKOUCHE A, MAMOUNE S M A. Identification et prévision du gonflement de quelques sols de la région de Tlemcen (Algérie) [J]. Bulletin-Laboratoires des Ponts et Chaussées, 2001, 233(4375): 67-76.

[31] KAZMIERCZAK J B, MAISON T, LAOUAFA F. Un nouveau dispositif pour la caractérisation du retrait et du gonflement des sols argileux [J]. Revue Francaise de Géotechnique, 2016(147): 1. (in France)

[32] TRAN T D. Role de la microstructure des sols argileux dans les processus de retrait-gonflement: de l’échelle de l’éprouvette à l’échelle de la chambre environnementale [D]. Paris, ENMP, 2014.

[33] KIKI T, CISS I, FAVRE J-L. Plans d’expériences pour la stabilisation d’une argile gonflante du Bénin [J]. Revue Francaise de Géotechnique, 2006(114): 33-41. (in France)

[34] ATTOH-OKINE N. Lime treatment of laterite soils and gravels-revisited [J]. Construction and Building Materials, 1995, 9(5): 283-287. (in France)

[35] MALLELA J, QUINTUS H V, SMITH K L. Consideration of lime-stabilized layers in mechanistic-empirical pavement design [R]. The National Lime Association, 2004, 200:1-40. (in France)

[36] LE RUNIGO B, CUISINIER O, CUI Y J. Impact of initial state on the fabric and permeability of a lime-treated salt under long-term leaching [J]. Canadian Geotechnical Journal, 2009, 46(11): 1243-1257. DOI: 10.1007/978-3-642-31116- 1_5.

[37] LOCAT J, BERUBE M A, CHOQUETTE M. Laboratory investigations on the lime stabilization of sensitive clays: Shear strength development [J]. Canadian Geotechnical Journal, 1990, 27(3): 294-304.

[38] LOCAT J, TREMBALY H, LEROUEIL S. Mechanical and hydraulic behaviour of a soft inorganic clay treated with lime [J]. Canadian Geotechnical Journal, 1996, 33(4): 654-669.

[39] NALBANTOGLU Z, TUNCER E R. Compressibility and hydraulic conductivity of a chemically treated expansive clay [J]. Canadian Geotechnical Journal, 2001, 38(1): 154-160.

[40] AL-MUKHTAR M, KHATTAB S, ALCOVER J F. Microstructure and geotechnical properties of lime-treated expansive clayey soil [J]. Engineering Geology, 2012, 139: 17-27.

[41] TRAN T D, CUI Y J, TANG A M. Effects of lime treatment on the microstructure and hydraulic conductivity of Héricourt clay [J]. Journal of Rock Mechanics and Geotechnical Engineering, 2014, 6(5): 399-404.

[42] BOARDMAN D, GLENDINNING S, ROGERS C. Development of stabilisation and solidification in lime–clay mixes [J]. Geotechnique, 2001, 51(6): 533-543.

[43] LUO R, PROZZI J A. Combining geogrid reinforcement and lime treatment to control dry land longitudinal cracking [J]. Transportation Research Record, 2009, 2104(1): 88-96.

[44] PARSONS R L, JOHNSON C P, CROSS S A. Evaluation of soil modification mixing procedures [R]. Kansas Department of Transportation, 2001.

[45] ESTABRAGH A, RAFATJO H, JAVADI A. Treatment of an expansive soil by mechanical and chemical techniques [J]. Geosynthetics International, 2014, 21(3): 233-243.

[46] STOLTZ G, CUISINIER O, MASROURI F. Weathering of a lime-treated clayey soil by drying and wetting cycles [J]. Engineering Geology, 2014, 181: 281-289.

[47] CHINKUYU A J, KANWAR R S. Effect of lime application on the movement of atrazine and nitrate-nitrogen through undisturbed-saturated soil columns [J]. Water Air & Soil Pollution, 1999, 115(1-4): 371-384.

[48] OSINUBI K J, IJIMDIYA T S, NMADU I. Lime stabilization of black cotton soil using bagasse ash as admixture [J]. Advanced Materials Research, 2009, 62-64(2): 3-10.

[49] CHESHOMI A, ESHAGHI A, HASSANPOUR J. Effect of lime and fly ash on swelling percentage and Atterberg limits of sulfate-bearing clay [J]. Applied Clay Science, 2017, 135: 190-198.

[50] SAHOO J P, PRADHAN P K. Effect of lime stabilized soil cushion on strength behaviour of expansive soil [J]. Geotechnical & Geological Engineering, 2010, 28(6): 889- 897.

[51] SIVAPULLAIAH P V, SRIDHARAN A, RAMESH H N. Strength behaviour of lime-treated soils in the presence of sulphate [J]. Canadian Geotechnical Journal, 2000, 37(6): 1358-1367.

[52] LITTLE D A, JR R B R, MARTENS D C. Lime-stabilized and chemically fixed sewage sludges as lime amendments [J]. Bioresource Technology, 1991, 37(1): 93-102.

[53] RAJASEKARAN G, RAO S N. The microstructure of lime-stabilized marine clay [J]. Ocean Engineering, 1997, 24(9): 867-878.

[54] TREMBLAY H, LEROUEIL S, LOCAT J. Mechanical improvement and vertical yield stress prediction of clayey soils from eastern Canada treated with lime or cement [J]. Canadian Geotechnical Journal, 2001, 38(3): 567-579.

[55] RAO S M, SHIVANANDA P. Compressibility behaviour of lime-stabilized clay [J]. Geotechnical & Geological Engineering, 2005, 23(3): 309.

[56] JAWAD I T, TAHA M R, MAJEED Z H. Soil stabilization using lime: Advantages, disadvantages and proposing a potential alternative [J]. Research Journal of Applied Sciences, Engineering and Technology, 2014, 8(4): 510-520.

[57] AL-AMOUDI O S B, KHAN K, AL-KAHTANI N S. Stabilization of a Saudi calcareous marl soil [J]. Construction and Building Materials, 2010, 24(10): 1848-1854.

[58] THOMPSON M R. Deep-plow lime stabilization for pavement construction [J]. Journal of the Transportation Engineering Division, 1972, 98: 311-323.

[59] PETRY T M, LITTLE D N. Review of stabilization of clays and expansive soils in pavements and lightly loaded structures—history, practice, and future [J]. Journal of Materials in Civil Engineering, 2002, 14(6): 447-460.

[60] ROGERS C D F, GLENDINNING S. Improvement of clay soils in situ using lime piles in the UK [J]. Engineering Geology, 1997, 47(3): 243-257.

[61] TEDESCO D V. Hydro-mechanical behaviour of lime- stabilised soils [D]. Universita Degli Studi Di Cassino Facolta Di Ingegneria, 2006.

[62] AL-KIKI I, AL-ZUBAYDI A, AL-ATALLA M. Compressive and tensile strength of fibrous clayey soil stabilized with lime [J]. AL Rafdain Engineering Journal, 2012, 20(2): 66-77.

[63] VOLKOFF B. Carte pedologique de reconnaissance de la Republique populaire du Benin a 1/200,000; feuille d' Abomey [M]. Travaux et Documents de l' Office dela Recherche Scientifique et Technique Outre-Mer no. 66, 1976. (in France)

[64] GLODJI L A, BASCOU J, YESSOUFOU S. Relationships between deformation and magmatism in the Pan-African Kandi Shear Zone: Microstructural and AMS studies of Ediacaran granitoid intrusions in central Bénin (West Africa) [J]. Journal of African Earth Sciences, 2014, 97: 143-160. (in France)

[65] AFNOR N. 94-056. Sols: reconnaissance et essais. Analyse granulométrique: méthode par tamisage à sec [M] Normalisation Francaise, 1996. (in France)

[66] AFNOR NF 94-056. Sols: reconnaissance et essais. Analyse granulométrique : méthode par tamisage à sec après lavage [M]. Normalisation Francaise, 1992. (in France)

[67] AFNOR NF P94-093. Détermination des caractéristiques de compactage d’un sol, essai Proctor normale, essai Proctor modifié [M]. Normalisation Francaise, 1993: 14. (in France)

[68] AFNOR NF P94-051. Sols: reconnaissance et essai de détermination des limites d’Atterberg [M]. Normalisation Francaise, 1993: 15. (in France)

[69] AFNOR NF P94-114. Sols: reconnaissance et essais- essai de pénétration dynamique type A [M]. Normalisation Francaise, 1990. (in France)

[70] AFNOR NF P 94-071-1. Sols: reconnaissance et essais- Essai de cisaillement rectiligne à la boite-Partie 1: cisaillement direct [M]. Normalisation Francaise, 1994. (in France)

[71] AFNOR NF P 94-090-1. Essai oedométrique - essai de compressibilité sur matériaux fins quasi saturés avec chargement par paliers [M]. Normalisation Francaise, 1997. (in France)

[72] AFNOR EN 13286-2. Mélanges traités et mélanges non traités aux liants hydrauliques-Partie 2: méthodes d’essai de détermination en laboratoire de la masse volumique de référence et de la teneur en eau-Compactage Proctor [M]. Normalisation Francaise, 2010. (in France)

[73] AFNOR EN 13286-50. Unbound and hydraulically bound mixtures-Part 50: Method for the manufacture of test specimens of hydraulically bound mixtures using Proctor equipment or vibrating table compaction [M]. Normalisation Francaise, 2004. (in France)

[74] AFNOR NF P94-078. Sols: reconnaissance et essais-Indice CBR après immersion. Indice CBR immédiat. Indice portant immédiat-Mesure sur échantillon compacté dans le moule CBR [M]. Normalisation Francaise, 1997. (in France)

[75] DAMS I C on L Dams F C on L, DEGOUTTE G. Petits barrages : recommandations pour la conception, la réalisation et le suivi [M]. Cemagref Editions, 1997. (in France)

[76] ASTM D3282-09. Standard practice for classification of soils and soil-aggregate mixtures for highway construction purposes [M]. West Conshohocken: ASTM International, PA, 2009.

[77] DHOWIAN A W, EDIL T B. Consolidation behavior of peats [J]. Geotechnical Testing Journal, 1980, 3(3): 10.

[78] TAVENAS F, BRUCY M, MAGNAN J-P. Analyse critique de la théorie de consolidation unidimensionnelle de Terzaghi [J]. Revue Francaise de Géotechnique, 1979(7): 29-43. (in France)

[79] JAMES GLASTONBURY, FELL R. Geotechnical characteristics of large slow, very slow, and extremely slow landslides [J]. Revue Canadienne de Géotechnique, 2008, 45(7): 984-1005. (in France)

[80] GUETTOUCHE, SAID M. Modeling and risk assessment of landslides using fuzzy logic; Application on the slopes of the Algerian Tell (Algeria) [J]. Arabian Journal of Geosciences, 2013, 6(9): 3163-3173.

(Edited by YANG Hua)

中文导读

石灰加固贝宁(西非)土的性能及滑坡防控

摘要：过去数十年，西非主要城市的基础设施建设快速发展，亟需对该地区的岩土特性进行全面研究，以实现可持续发展。本文研究了西非贝宁土的特性。在本研究中，对西非土的性能特征进行了三方面深入研究：1)获取该地区土壤类型的特征；2)评价了石灰掺量为2%、3%和5%时土的性能；3)描述滑坡特征，以评估其破坏和潜在的不稳定性。根据法国标准，采用试验方法进行研究，开展颗粒分析试验、击实试验、液塑限试验等物理试验和动力触探试验、直剪试验、固结试验等力学试验以评估目标一，开展击实试验和加州承载比试验以评价目标二，同时对河岸滑坡的地质、环境、社会和安全等进行评估。本文首先揭示了贝宁南部(非洲)存在黏土的稳定/不稳定地区，并给出了预测稳定/不稳定地区的方程。其次，研究表明石灰占比在2%~3%之间时土体可达到最佳性能。最后，本文指出河岸滑坡可能是由山谷中的突然退水所致。

关键词：土壤特性；土体性能；黏土；石灰；滑坡

Foundation item: Project(41627801) supported by the National Major Scientific Instruments Development Project of China; Project(41430634) supported by the State Key Program of National Natural Science Foundation of China; Project(2016YJ004) supported by the Opening Fund for Innovation Platform of China; Project(2016G002-F) supported by the Technology Research and Development Plan Program of China Railway Corporation

Received date: 2018-01-07; Accepted date: 2019-07-03

Corresponding author: Quirin Engelbert Ayeditan ALAYE, Doctoral Candidate; Tel: +86-15776602579; E-mail: alaye47@yahoo.fr; ORCID: 0000-0002-5476-9929