Particle size distribution analysis of sand-sized particles by laser diffraction: an experimental investigation of instrument sensitivity and the effects of particle shape

Laser diffraction is now widely used for particle size distribution analysis of sediments and soils. The technique can be very precise, and offers advantages of speed and cost over many other methods when used to analyse mixtures of sand, silt and clay. This study presents the results of an experimental investigation that examined the sensitivity of the Beckman-Coulter LS230 instrument to mixtures of different grain populations and differences in particle shape. The instrument was found to have high sensitivity to coarse particles in a finer matrix (detection threshold 1–2%), but much lower sensitivity to finer particles in a coarser mixture (detection threshold 12–17%). Experiments using near-spherical ballotini showed that laser analysis provides very similar values to dry sieving for the mean, median and mode, but for a range of natural sand samples values for the mean, median and mode were offset by 8–21%, with an average of ca 15%, compared with sieving. Analysis using a Beckman-Coulter RapidVUE instrument, which provides both size and shape information, provided support for the hypothesis that the differences between laser analysis and sieving are partly attributable to the effects of particle shape. However, an additional factor is the way in which the laser software interprets the optical diffraction data. The software predicts a high degree of log-normality in the size distribution, such that highly skewed, truncated or bimodal samples are poorly represented. Experiments using sieved fractions of ballotini indicated that, even with near-perfectly spherical particles, the particle size distribution predicted by the laser software includes a relatively large percentage of particles outside the sieve class limits.