Correlation between Plasticity Index and Methylene Blue Value to Determining Soil Classification

Most soil classification systems developed for geotechnical engineering purposes are based on simple index properties such as grain size distribution and plasticity. In Indonesia, the use of the methylene blue value, VBS, is still rarely used, compared to Atterberg limits that commonly used to determine the value of the plasticity index, PI. Therefore, this study aims to gain a correlation between PI and VBS. Coarse-grained soil with fines (clay and silt) contents varies from 12.36% to 29.45% are used of this study. The linear correlation between PI and VBS produces a linear equation with a coefficient of determination, R2 = 0.854 for 0.62 < VBS < 2.39. The PI obtained from the correlation to VBS that combined with the sieve analysis results can then be used for soil classification based on the Unified Soil Classification System (USCS).


Introduction
Soil classification is used to determine certain soil types that suitable for an application of the geotechnical construction, such as: embankment, subgrade material, and drainage material. Varies type of soils with similar properties can be classified into classes or groups and subclasses according to soil behavior. For geotechnical engineering purposes, most of soil classification systems that developed are based on simple index properties such as grain size distribution and plasticity. Plasticity index, PI, is obtained from the Atterberg limits test, which is the difference between liquid limit, LL, and plastic limit, PL [1]. When carrying out LL and PL tests, man-mad error that easily occurred [2], can impact on the subjectivity of PI values. To solve this condition, other parameters from other methods that are universal, fast and reliable are needed as a comparison for the determination of PI [3]. Another parameter for comparison with PI is methylene blue value (valeurs de blu du sol, VBS). VBS is obtained from methylene blue test [4], [5], [6]. This test was developed in France, where initially, this test was carried out to determine the suitability of granular material in making concrete by detecting clay content from granular material [7]. The spot method is more common for the test and is a simplified type of titration technique [7], [8]. This study aims to obtain a correlation between the PI and VBS. The results of these correlations can be used to determine soil classification.

Method
The standard test for sieve analysis is based on ASTM D 422. Determination of Atterberg limits (LL and PL) refers to the standard ASTM D 4318. The standard for methylene blue test is based on NF P 94-068. The test results are then analyzed to obtain a correlation between PI and VBS. The soil classification is based on the Unified Soil Classification System (USCS) [9], [10].

Plasticity Index
The plasticity index, PI [1] is calculated using the following equation: Test of LL and PL are shown in Figure 1.
(a) (b) Figure 1. (a) Test of liquid limit, LL (b) Test of plastic limit, PL

Methylene blue value (Valeurs de blue du sol, VBS)
The methylene blue test standard refers to NF 94-068 [4] or NF EN 933-9 [5]. Figure 2 shows the methylene blue test equipment and Figure 3 shows the flow chart of the methylene blue test with the spot method conducted in this study.
Otcu et.al [12] stated the correlation between the plasticity index and methylene blue value expressed in grams of dye per kilogram of the 0/2mm fraction, MB [5]

Results and discusion
The study of plasticity index and methylene blue value for soil classification are carried out on data obtained from laboratory test results of soils originating from areas in North Maluku, Indonesia. Table 1 shows the results of a sieve analysis of those data. It can be seen from Table 1 that the coarsegrained soil have fines (Silt and Clay) contents varies from 12.36% to 29.45%. Following the completion of sieve analysis, methylene blue test was conducted to get VBS. Also, Atterberg limits test (LL and PL) were conducted to obtain PI. Table 1 also shows the VBS and the PI of the samples. A plot of PI vs. VBS is presented in Figure 4. Based on the data given in Figure 4, a linear correlation between the PI and the VBS can be expressed by the following equation: With coefficient of determination, R 2 = 0.854.
The results of the Atterberg limits test are strongly dependent on the technician. For example, to decide the number of blows to close the groove in the sample to obtain the liquid limit. Furthermore, the pressure on the operator palm hand, while roll the sample on a glass plate, affect the results, which is the plastic limit.
On the other hand, compare to the Atterberg Limits test, the methylene blue test with spot method is less dependent on the technician. This is because the procedure for conducting the methylene blue test is more natural to perform.   Table 2 shows the PI values obtained directly from the Atterberg Limit test and those obtained based on the correlation with VBS (Equation 5). The absolute difference between these two PI is also shown in Table 2. It can be seen from Table 2 that the average of this absolute difference is only 2.9%. Therefore, one can say that the methylene blue test can be used as an alternative method to obtain the PI of soil. The PI value obtained from the correlation to VBS combine with the sieve analysis results (Table 1) is used to classify soil sample according to the USCS. The results are given in Table 3. It can be seen from Table 3 that for the whole tested coarse-grained soil, there are four Group Name and four Group Symbol.

Conclusion
Plasticity index, PI is one of the most important soil parameters for conducting soil classification based on the Unified Soil Classification System (USCS). PI can be obtained through correlation to VBS, which is the result of a methylene blue test. Compare to the Atterberg limits test, the procedure for conducting the methylene blue test is more straightforward. The results of this study show that for VBS value between 0.62 and 2.39, the correlation can be expressed by a linear equation with a coefficient of determination R 2 = 0.854. The PI obtained from the correlation to VBS combine with the sieve analysis results can then be used for soil classification based on the USCS.