Two types of injectors are described for introducing solid samples as slurries in flow analysis systems. A time-based and a volume-based injector based on multitube solenoid pinch valves were built, both can be characterized as hydrodynamic injectors. Reproducibility of the injections of dispersed solids (< 150 pm) was tested with several concentrations of slurry samples up to 30 mg/ml; the injected volume was 1 ml. For both injectors dye and slurry samples could be injected with good precision (relative standard deviation for the peak area less than 2%). Peak detection was performed turbidimetrically. Data analysis and operation of the injectors were automated. The usual peristaltic pumps in flow analysis are normally not capable of handling slurries of the type investigated, therefore a valveless piston pump was used instead.

In order to achieve a reproducible on-line pretreatment of slurry samples in flow-injection analysis the dispersion behaviour of these samples has to be examined. This paper reports the effects of flow rate and some configurations of reaction manifolds. The reaction manifolds were a straight tube and two helically coiled tubes with different aspect ratios and spatial orientation. The slurry samples consisted of spheres of different densities suspended in a Triton X-100/water solution, the particle diameters (up to 75 ~tm) were relatively large compared to the internal diameter of the tubing (0.75 ram). For the straight tube theoretical calculations were found to give a reasonable impression of the experimental flow behaviour of the spheres. Statistical moments were used to characterize the slurry peaks. For tightly coiled tubes it was found that the retention time of a slurry sample depends on the particle density. The spatial orientation of the coils mainly affects the retention of high density particles.

In order to achieve a reproducible on-line pretreatment of slurry samples in flow-injection analysis the dispersion behaviour of these samples has to be examined. This paper reports the effects of flow rate and some configurations of reaction manifolds. The reaction manifolds were a straight tube and two helically coiled tubes with different aspect ratios and spatial orientation. The slurry samples consisted of spheres of different densities suspended in a Triton X-100/water solution, the particle diameters (up to 75 ~tm) were relatively large compared to the internal diameter of the tubing (0.75 ram). For the straight tube theoretical calculations were found to give a reasonable impression of the experimental flow behaviour of the spheres. Statistical moments were used to characterize the slurry peaks. For tightly coiled tubes it was found that the retention time of a slurry sample depends on the particle density. The spatial orientation of the coils mainly affects the retention of high density particles.