Preloading Effect On The Strength of Cement Mass Mixing Treated Salty Sand

Existing problematic sub-layers in mixing technologies are a challenge, and for the rst time, the effects of salt sub-layers in mass mixing technology have been investigated in this study for sandy salt in the southwest of Iran. This paper discusses the inuence of adding various cement contents, Aw, and imposing different preloading values on the salty sand soil. First, salt and sand samples were dried, then, 90 % sand was mixed with 10% salt. After that, 30 % water was mixed thoroughly with the mixture of salty sand to obtain slurries. Cement slurry at a water-cement ratio (w/c) of 0.6 was then added to the sample and thoroughly mixed. The amount of cement in a slurry form that was added to the salty soil was 2, 4, 6, 8, and 10% by mass of dry soil. Each treated soil preloaded by 0, 9, and 45 kPa. After 120 days, the unconned compressive strength of the sample was determined. Furthermore, by Scanning Electronic Microscope, SEM, the microstructures of treated samples were analyzed. At the end Unconned Compression Strength, UCS, test results normalized to the non-preloaded treated soil. By increasing cement content, the effect of preloading in increasing UCS will decrease. In the SEM images for Aw=2%, the effect of preloading indicates porous shape for non-preloaded samples. Vice versa by Aw=8%, porous shape in the SEM images will disappear. In the end, treatability studies of pure salt in the thick layer have been investigated.

content of the mixture, and the organic content of the intrinsic soil (G.M. Filz, 2015, Jacobson, 2005, Shiells, 2003, Bruce, 2000, Takenaka, 1995. The reaction between soil and studied by various researchers. Van Lier et al. (1960) concluded that the solubility of silica is greatly increased in the presence of sodium chloride. The chloride ions react with the Ca 2 + and Al 3 + ions to form hydrocalumite Ca 2 .Al(OH) 6

Effects of the Soil Mixing Installation Process, and Thickness of Salt Layer on Homogeneity
Homogeneity of the nal soil mixture related to the thickness of the salt layer, Hs, and soil mixing execution method. Generally, dry and wet soil mixing, and mass mixing method used in soil mixing technologies. As depicted in Fig. 1b, in the mass mixing method if there is a salt sub-layer until the depth of 8 m, the salt sub layer will mix totally with above and lower soft soil. Due to the operation issues, using preloading embankment in the mass mixing method is essential.
On the other hand in deep soil mixing, due to the thickness of salt sub-layer, Hs, injection method (dry or wet), number of shafts, blade types, and installation process the homogeneity of mixing soft upper and lower soil with salt layer have been differentiated.
In the case of a very thick salt sub-layer the binder agent will mix only with salt and soft upper and bottom soil have not been mixed with salt layer (Fig. 2. a or c). In another word to attain a uniform mixture of soil-salt-binder in a very thick salt layer withdrawal by continuous upstroke by stepped or even full restroking is needed ( Fig. 1 d, e).
In the thin salt layer, partial restroking ( Fig. 1 b) has produced a more uniform mixture of soil-salt-binder.
The best installation method to attain optimum uniform soil-salt-binder is full depth restroking (Fig. 2 e) which produces more spoil (Topolnicki, 2004).

Experimental soil improvement base on mass mixing by preloading
In highly soft soil, mass mixing is a very well method for soil improvement, especially for road constructions. In Southwest of Iran, there are some problematic soils in which construction projects such as road and foundation had encountered serious damages. Soil replacement for constructing these projects is very expensive, but the mass mixing method by improving existing soil is more cost-effective.
Currently, two primary mass stabilization methods can be chosen from, namely, stabilization in layers and stabilization by the blocks. The optimum method chosen depends on the soil type, and the application of the ground improvement project. With stabilization in layers, the soil is simultaneously mixed and moved towards the excavator as illustrated in Figure 1b. Once the mixture of soil-binder has been completed to the proper depth in front of the excavator, the excavator moves forward on top of the accomplished mass mix and the process is repeated. This method can only be used with soils that are strong enough to bear the weight of the excavator immediately after being mixed.
The general principle of the mass stabilization method is presented in Figure 1b. With the current equipment, the attachment of a mixing unit to an excavator allows for carrying out stabilization to the depth of 7-8 meters, providing the conditions are favorable (ALLU, 2015). A system of pressure feeder injects the binder(s) through the hose directly to the mixing drums of the mixing unit. The rotating drums mix the binder(s) into the ground, and consequently homogenize the soil. Mixing process is executed by moving the mixer unit vertically from the ground surface to the desired depth, as well as laterally.
Depending on the geometry of the site plan, the work area divides into blocks, or area, of equal size. Most of the time, work proceeds from block to block, with the size of a block between three to ve meters square. A working platform is constructed after the completion of a block(s) to enable the excavator to move on to the site. The working platform also serves as a primary compaction embankment. After accomplishing the ground stabilization, an embankment as a preloading layer is also constructed. The allowable strength of the treated ground is usually attained over 1-3 months.
The main object of this research is experimental investigating feasibility use of treating loose sand with salt sublayer in the southwest of Iran for road and foundation construction in mass mixing method with a preloading embankment. Due to loose soil, the mass mixing method is applicable just by block type. After conducting mass mixing, the salt layer will mix with loose sand, and uniform salty sand will produce. In loose sand stabilization with salt layer, block type of mass mixing with preloading embankment have been applied. Therefore, this embankment has the function of preloading for mass. In this research, the preloading value chose 9 kPa and 45 kPa which is equal to 0.5 m to 2.5 m embankment height with a speci c gravity of 18 kN/m 3 . For applying surcharge load steel rods have been applied.

Geology and geomorphology
This area belongs to Zagros zone. There are some active faults such as Aghajari fault (in 70 km distance), Ahwaz fault (in 60 km distance), and white fault (in 80 km distance). This area is located in a low possibility zonation of occurring earthquake with 0.2 g, according to the Iranian earthquake code. This area is located near the Persian Gulf and is categorizes as lowland regions. Also, this lowland area is lled with soil, and now it has an elevation in the range of 1.3-3.1 m. In addition, it has two major estuaries with the name of Zangi Estuary (800 m width) and Jafary Estuary (400 m width). The general pro le of soil in this area includes soft soil, sludge, and salt until 10 m depth, then medium-stiff soil to 20 m depth, and nally compacted sand.
NaCl Salt in this area exists in 2 forms. The rst one is a mixture of salt and soil. The second one is a pure salt layer. The underground water table is in the range of 0.2 to 3.4 m in winter.

Field site investigation
For this study, 41 boreholes by drilling method and continuous coring with a depth of 11 to 25 m used for determining geotechnical site investigation in the Special Economic Petrochemical Zone, southwest of Iran. Boreholes describe as BH101 to BH141 are illustrated in Fig. 3.

Pressuremeter tests
This eld test is very useful for determining the stress-strain behavior of soil. It is performed with a radial expandable probe inside the borehole. The type of pressuremeter used in this study is P.B.P. (Pre-Bored Pressuremeter) with the name of Ménard Pressuremeter. Table 1 illustrate the pressuremeter test results.    Figure 4 illustrates the salt layer in the depth 4-5 m below the ground. Also, gure 5 depict sieve analysis of sand material.
In the mass mixing method this salty layer will mix with above, and below the soil and a combination of treated salty soil with cement will produce.
There are Interlayers of salt beneath the ground in the southwest of Iran. Shallow salt layers are not compact, but deeper salt converts to salt rock with a density in the range of 20-25 kN/m 3 and compression strength of more than 15kPa. In this research for investigating the effects of salt on soil, treatability studies performed. As illustrated in the Scanning Electron Microscope images, Fig. 6, shallow salts are completely in the crystal form, which are soft soils, but deeper denser salts have less crystal shape.
In soil stabilization of various soil types in the southwest of Iran, type two Portland cement has been used.
General properties of this type of Portland cement with a speci c gravity of 31.5 kN/m3 are presented in Table 4. to simulate the water content increase taking place in a mass mixing process in the eld. Cement slurry at a water-cement ratio (w/c) of 0.6 was then added to the sample and thoroughly mixed. The amount of cement in a slurry form that was added to salty soil was 2, 4, 6, 8, and 10% by mass of dry soil. The mixture was then placed in 5 layers into PVC molds 35 mm in diameter and 70 mm in height for the uncon ned compression test. To remove entrapped air bubbles, the placement of the samples into the molds was accompanied by a tapping action around the molds. Each treated salty sand preloaded by 0, 9, and 45 kPa. Steel rods were used for imposing preload to the salty sand. Curing periods of 120 days were chosen to determine the long-term effect.
Uncon ned compression tests were performed on the samples at an axial deformation rate of 1 mm/min. Figure  7, illustrates apparatuses used for preloading salty sand.

The Effect of Preloading on Salty Sand
Three types of salty sand with preloading values equal to 0, 9, and 45 kPa prepared for conducting UCS tests. As an interesting depiction in Fig. 8, at the top of the treated salty sand without preloading, salt crystals were composed. This phenomenon according to the capillary migration of water and salt (Jia et. al., 2020). When the preloading value is equal to zero, salty water comes up to the surface of the treated salty sand and bypassing time these salt crystals compose. On the other hand, by 9 and 45 kPa preloading value, these salt crystals were eliminated.
The crystal form of salt in Fig. 8 occurs on the surface of the ground, and it decreases the durability of the concrete foundation in real projects. This a good practical result that with a small amount of preloading in the real soil mixing project, the durability of the concrete foundation has been increased in salty soil. Figures 9, 10, 11, 12, and 13 illustrate the UCS results of treated salty sand with 2, 4, 6, 8, and 10 % of cement content, respectively. Even at the lowest cement content of 2%, the increase in the strength and the stiffness is evident (Fig. 9). As depicted in Table 5, by increasing the preloading value, UCS results will increases. Furthermore, the UCS results with preloading are 483.87 to 2174.8 kPa by cement content is in the range of 2-10%. By normalizing UCS values for all samples to the UCS value of zero preloading ones, Fig. 14 will derive. This gure illustrates that by increasing Aw, the effect of preloading in increasing UCS will decrease.
In the SEM images for Aw=2%, Fig.15-c, the effect of preloading indicates the existence of porous shape in the morphology of non-preloaded treated salty sand. The difference between Fig. 15-a-b and Fig. 15-c in producing more porosity by eliminating the preloading factor on the treated salty sand in the mass mixing method is obvious.
On the other hand, treated soil by A w =8%, Fig. 16, porous shape in the SEM images were disappeared. Also, by increasing cement content, the effect of preloading on increasing UCS will decrease.

The Effect of Cement on Pure Salt
When the thickness of the salt sub-layer is high, then the mixture is just salt-binder. Therefore, a higher amount of cement as a binder should be used. A mixture of salt with Aw=20% and 10 % water content makes a material with higher mechanical properties. UCS test results depicted that by increasing curing time from 7 to 120 days the UCS will increase. It is interesting that using the high amount of Portland cement, Aw=20%, causes to have salt-cement material which has UCS higher than 1500 kPa after 120 days. Therefore the big challenges for the treatability potential of pure salt and cement are positive. Figure 17 illustrates the UCS results of pure salt with Portland cement.
Interestingly, in the SEM images of salt mixture with cement after 28 curing days, salt crystal has been disappeared (Fig. 18).

Conclusions
This By increasing preloading value and cement content, the uncon ned compressive strength of treated salty soil in the southwest of Iran was increased.
Salt crystals produce and cover the top surface of treated salty sand samples which are not preloaded. The reason for this phenomenon is the capillary effect of treated soil and accumulating salty water on the top surface of the treated soil sample. This phenomenon may produce a more severe corrosive environment for deep soil mixing columns that are connected to the concrete foundations.
In loose ground for conducting mass mixing method, preloading is essential for preparing a suitable platform for construction equipment. Due to the surcharge load, and by preloading treated salty sand, salt crystals on top of soil samples were eliminated. It is the bene t of the mass mixing method on deep soil mixing columns in the soft ground with salt layers.
SEM images revealed that in treated salty sand by 2% cement content, and without preloading, the morphology of treated soil has more porosity. On the other hand, by increasing cement content this porous shape has been disappeared.
The big challenges for the treatability potential of pure salt and cement in the ground improvement methods for PETZONE are positive. Therefore mixing technologies such as deep soil mixing (dry or wet method) and mass mixing have good reliabilities even in the thick layers of salt sub-layers.