Particle size distribution in soils and marine sediments by laser diffraction using Malvern Mastersizer 2000—method uncertainty including the effect of hydrogen peroxide pretreatment

PURPOSE: Methods for particle size distribution (PSD) determination by laser diffraction are not standardized and differ between disciplines and sectors. The effect of H₂O₂ pretreatment before a sonication treatment in laser diffraction analysis of soils and marine sediments was examined on soils with less than 1% C and some marine sediments. MATERIALS AND METHODS: The method uncertainty for particle size analysis by the laser diffraction method using or not using H₂O₂ pretreatment followed by 2 min ultrasound and 1-mm sieving was determined for two soil samples and two aquatic sediments by analyzing ten replicates on a Malvern M2000 instrument. The carbon content was in the normal range for upland soils 0.1–0.9% C, but one of the aquatic sediment samples had a high carbon content (16.3% C) for which the H₂O₂ pretreatment was not feasible. RESULTS AND DISCUSSION: The effect of H₂O₂ pretreatment on the PSD was small and not significant. The standard deviation (std) in particle size fractions increased with particle size. PSDs and std for some samples were presented for future reference. Similar to other studies, the content of clay and silt (by sieving/hydrometer, SHM) was lower by the laser diffraction method (LDM) than by the hydrometer method. To avoid confusion, LDM results for the < 2-μm fraction should not be reported as a clay fraction to be used in texture classification. The absolute standard deviation of within-bin class was well below 1% in the colloidal fractions (using 2, 6, 8, and 20 μm bin delimitations) and below 0.2–3% in the sand fractions. CONCLUSIONS: The LDM method requires new calibrated transfer functions for classification of soil textural class due to deviating results for colloidal fractions. It can be used without H₂O₂ pretreatment on soils containing less than about 1% C, but results should be corrected for 1–2-mm coarse sand content if higher than, e.g., 5%. If the analysis is not referring to a particular volume of sample, the correction for 1–2-mm material can be omitted.