Role Of Clays Addition On The Stability Of Collapsible Soil Selected From Mosul City

Volume changes of a collapsible soil selected from Hawy Al-Kanissa in Mosul city were studied. The Influence of addition of (5, 15 and 30) % of low (CL) and highly plastic (CH) clayey soil were studied. Many variables as: the initial dry unit weight, water content and applied stresses on volume change were considered. State diagrams representing gradual volume change due to these variables were also presented and discussed. The main results were showed that the collapse potential (Cp) increased continuously for samples mixed with (CL) soil while there were a maximum value at (5%) of (CH) soil. At low unit weight, (Cp) for samples mixed with (CL) reached a highest value at certain stress, then started to decrease, while it increased continuously using (CH). (Cp) of soils containing (CH) was found to be less than that mixed with (CL) soil. Concerning soil volumetric changes using state diagrams presentation, volume change for undisturbed natural soil at low moisture content was found less than that in disturbed samples, but the inverse was obtained at saturation. Besides that, volume changes increased with the increase of the clay additive in different manner for (CL) and (CH) soils.

. . 2. When water content remains constant and the applied stress change. 3. When water content and applied stresses are changing together.
The state diagrams recommended by Andri silvan et al (1981) [17] showed different ways to represent these diagrams. Then Al-khashab and Bahhe (1999) [3] studied the effect of particle finesse on these diagrams, then Khattab (2004) [2] used these diagrams to represent different state of compaction for soil.
This investigation was performed on collapsible alluvial soils selected from Hawy Al-Kanissa destrict, which are the most commonly occurring in Mosul city.
The main objective of this paper is to experimentally define the effect of the addition of different types and amounts of natural clayey soils to collapsible soil on their collapse potential under the same identical condition; as well as, the effect of applied load and degree of saturation on volume change of the same types of prepared soil, using the "State Diagrams".

2-Laboratory Work and Test Procedure
Disturbed and undisturbed soil samples were obtained at depth (1.0 m) from Hawy Al-Kanissa, located at the right bank of Tigris river in Mosul city. Undisturbed samples were taken directly by pushing the consolidation rings perpendicular on the top of perfectly leveled natural soil. They were placed on small glass plate to prevent disturbance, sealed and kept in a desiccator to insure that there is no loss in their natural water contents. The standard and modified compaction tests indicated that Hawy Al-Kanissa soil has a maximum dry unit weight of (16.48 and 17.81) kN/m 3 respectively as shown in Fig. (1).
Two types of soil were obtained from Al-Gara / Al-Anbar governorate and Hay Al-Shohadaa/ Mosul city at depth (1.0 m) and used as clay additive to Hawy Al-Kanissa soil with different amounts under the same identical condition. The engineering properties for all natural soils were listed in Table (1). Hawy Al-Kanissa soil characterized by low dry unit weight (12.75 kN/m 3 ) with high initial void ratio ( o e =1.054), as well as, the soil is non plastic with low liquid limit ( % 24 w L ). Therefore, Hawy Al-Kanissa soil classified as a collapsible soil according to [9], [10]. The properties of the clayey soils were shown in Table (1) and Fig. (2). The soils constituents, listed in Table (1), showed that the major clay minerals were smectite, kaolinite and mixed layers from illite and smectite respectively.

Methodology:
For the purpose of this research, the laboratory tests for the following studies were conducted: 2-1 Comparison study of collapse potential "Cp" of soil, with different types and amounts of clay soils under the same identical condition: conventional collapse tests were conducted on natural undisturbed as well as artificially prepared soil samples under different dry unit weights and applied stresses.    [12]. While for compacted soil at different dry unit weights mixed with different types and amounts of clayey soil, the double oedometer test was used only. For single test, applied loads on specimens with in-situ water content (w/c=7%) with loading increment for each hour up to (100 kPa) to represent the expected load that the collapsible soil can support. Then, water added, the deformation caused by this inundation was recorded after (24 hour). Then, load increments proceeded up to (800 kPa).
For double oedometer test, two "identical" soil specimens are placed in oedometers. One specimen is tested at natural in-situ water content, which is generally quite low. The other specimen is fully saturated before the beginning of the test, stresses were applied to keep volume constant that is refer to swelling pressure for specimens, results are listed in Table (2), and then subjected to an identical compression test.

2-2 Studying the volume change for collapsible soil using state diagrams:
Undisturbed and disturbed samples with and without different percentages of clay additives were prepared in the same manner described in sec. (2-1-1). At each dry unit weight the compaction water contents are (7, 15, 20)% and saturated samples. For unsaturated specimen, cells were kept and sealed with Aluminum foil paper and then with several plastic bags to prevent evaporation of water during the test. Water content during the test was verified periodically through additional cells prepared for this reason. Then, conventional Oedometer tests were conducted by applying load increment (12-800) kPa. The moisture contents under each applied load for each sample were found from identical specimen. Specific volume which represent (volume (cm 3 ) corresponds to 100 gm of dry soil, V=100/ d ) was calculated under each applied load. The correlation between final water content and volume represent "State Diagram".

3-1 Effect of clay additive on the index properties of Hawy Al-Kanissa soil:
The effect of clayey soil (5, 15, 30)% on the index properties of Hawy Alkanissa soil were presented in Table (3). As expected an increase in liquid limit, plastic limit, plasticity index and linear shrinkage were noted from low to high value with percentage of clay additive as shown in Table (3). The results showed a higher increase in index properties with addition of (CH) soil than that with (CL) soil. This could be attributed to mineralogical composition of (CH) soil which consists of mixed layers from montmorillonit and illite which have high water adsorption ability comparing with kaoline [1].

3-2-1 Undisturbed sample:
A. Single test, typical curve of the test is plotted in Fig. (3) and collapse potential "Cp" is calculated as (11.89%). Fig. (4). Two stress versus strain curves were generated, one for the partially saturated soil and the other for the saturated soil. For a given applied stress n , the strain offset between the two curves is called collapse strain for that stress level. Hence, collapse strain for undisturbed samples was (10.46%) under applied pressure of (100 kPa). The above results showed a very limited difference compared with that of single test. Same conclusion was obtained by [6], [13].
Results from double oedometer tests showed that undisturbed samples had collapse potential (Cp=3.23 %) under very light stress (12 kPa). This indicates that the soil had the ability to collapse under its own weight only as a result of the reduction in the initial water content of the honeycombed structure of this soil.

3-2-2 Disturbed sample:
To show the effect of disturbance on collapse potential of Hawy Al-kanissa soil, samples prepared by static compaction method to an identical condition of initial dry unit weight and water content, The methods mentioned in sec. (3-2-1) were followed. Results showed that the soil had collapse potential of (6.8, 4.1)% respectively under applied stress (100 kPa) Results from single and double oedometer tests showed that the collapse potential of undisturbed samples was more than that of disturbed samples.  Mixed with 30% of (CH) Clayey Soil .

3-3 Effect of clay addition on volume change of collapsible soil:
The results of collapse potential for the compacted sample with different percentage of clay additive were presented in Table (4) and Figure (5) (a-d), at different stresses and low dry unit weights (12.75, 14.23) kN/m 3 . It is observed that, in general, the collapse potential (Cp) increases continuously with increase of (CL%) soil, while (Cp) reached a maximum value at (CH) soil contents of (12%). This observation can be explained by the fact that a portion of the fine-grained soil (clay fraction in this case) acts as a bonding material for large-grain particles. At low water contents, the clay particles at the contacts provide shear strength that resists deformation. The higher clay content, the higher the resistance to densification, (i.e higher void ratio). However, when the soil inundated, the clay binder is partially or fully broken, lubricated and thus facilitating their slipping and sliding against each other [3], [6]. However, this phenomena depends on the type and amount of the clay additive, hence, when a highly plastic clayey soil is used, collapse potential decreases at clay content more than (12%) as a result of swelling which reduces the void ratio for specimen. Consequently, collapse potential for highly plastic soil is less than that of low plastic soil.
Collapse potential was affected by the applied stresses, the greater the pressure, the greater the collapse, but, collapse was reached highest value at certain pressure where the dry specimen is mixed with (CL) soil with loose structure which breaks down, beyond which the collapse was reduced. However, this behavior depends on the type of the clay additive as the dry specimen with (CH) soil are more stable under applied pressure as shown in Table (4).
Samples prepared at high dry unit weights (15.69, 16.68) kN/m 3 , (Cp) decreases considerably for compacted soil with and without clay additives, at certain stress. The denser soil has lower initial void ratios, which leads to a lesser collapse upon wetting. In the other hand, the use of increased relative compaction resulted in a significantly increase in swelling of the soil during inundation. The swelling pressure increased with the increase of the amounts of two types of clayey soil. (i.e greater plastic soil, the greater swelling pressure) as shown in Table (2). The state diagrams that described the correlation between the final water content and volume was presented Fig. (6). These diagrams are affected by many factors namely: initial water content, applied stresses, initial dry unit weight as well as the amount and clay types. It can be noticed that there are gradual decreases in volume with the increase of water content and applied stresses. Compressibility for undisturbed natural soil at low moisture was less than disturbed samples due to bonding force between these particles. But the inverse was obtained at saturation as a result of weakening and breakdown of these bonding agents. The increase of the initial dry unit weight leads to the convergence of the stress paths from each other, and hence, a decrease of compressibility at any certain value of the applied stress and water content was obtained.
The effect of the amount and types of clay on these diagrams were showed in Figures (7-11) and Tables (5 and 6), at any value of dry unit weight, water content and applied stresses. It could be seen that the stress paths diverge from each other with the increase of the amount of clayey soil. This is because clay particles help the coarse grain soil to slide and slips on each other, therefore, the compressibility increased. For the soil mixture with (CL), the volume change was more than that with (CH) soil. This could be attributed to the lesser cation exchange capacity for the natural soil mixed with (CL), which helps to reduce the binder force between particles.

4-Conclusion
1-Collapse potential increased continuously for samples compacted at a low unit weight, mixed with the percentages of (CL), while there was a maximum collapse potential occurred at a 5% of (CH) soil.
2-Generally, collapse potential increased with applied stresses using the two clay additives with all placement conditions with the exception of samples compacted at a low unit weight. In this case, where collapse potential reaches a highest value at a certain stress.
3-Collapse potential of soils containing (CH) clays was less than that mixed with (CL) soil.
4-Compressibility for undisturbed natural soil at low moisture content was less than that of disturbed samples, but the inverse was obtained at saturation. 5-Compressibility increases with the increase of the clay additives, where the reduction in volume for soil mixed with (CL) was more than that mixed with (CH).