Volume replacement methods for measuring soil particle density

Abstract: The measurement of soil particle density is of great importance for the study and application of soil and water relationships. This paper proposed a volume replacement method for accurate measurement of soil particle density. Two approaches were used. In the first approach, wet soil samples were used. In wet soil sample method, a wet soil sample of unknown volume and mass was placed into a constant volume container (CVC) and then it was weighted. The CVC was filled up with water to displace the air-filled space in the soil and then the container was weighted to determine the replaced air-filled volume and to determine the volume and mass of initial moisture and soil particles. After that the soil sample was oven-dried to determine the volume and mass of soil particles so that soil particle density (SPD) was calculated. In the other approach, dry soil sample was used. In this method, the soil was completely dried before measuring the mass of soil particles. The soil sample was put into the CVC before it was filled up with water to replace the air-filled volume. The replaced volume was computed from the volume of water used to fill the container. Four soil samples were used for the experiments: a clay loam from Yangling of Shaanxi Province (31.80% clay, 62.80% silt and 5.40% sand particles), a silty loam from Beijing (16.75% clay, 53.00% silt and 30.25% sand particles), a black soil from Jilin Province (12.00% clay, 86.78% silt and 1.22% sand particles) and a red clay from Jiangxi Province (47.60% clay, 45.00% silt and 7.40% sand particles). Soil samples were prepared at 5 initial soil water contents as 10%, 15%, 20%, 25%, 30%. The two approaches discussed above were used to determine the SPD, and then the measured results were compared with the results measured by the pycnometer and volumetric flask methods. Detailed laboratory experiment procedures were made to illustrate the measurement and computational procedures. The results showed that the SPDs of the four soils were 2.6576, 2.6301, 2.6411 and 2.6687 g/cm3 respectively, by using wet sample method, and measured SPDs were 2.6557, 2.6320, 2.6423 and 2.6660 g/cm3 by using dry sample method, when dried for 72 h at 105℃. Oven-dry of 8 h produced measurement errors in SPDs were 1%-2%, indicating SPDs measured by the volume replacement method with a CVC have high accuracy by extending the drying time. Data points of SPDs measured with volume replacement method by using a CVC was located close to the 1:1 line to indicate the feasibilities of these two proposed methods. Errors in SPDs between wet sample method and dry sample method were less than 0.5%. Generally, in view of measuring accuracy and operating procedures, the newly-suggested methods developed in this study performed better than the conventional methods such as the pycnometer and volumetric flask methods.