Table 1.
Strength exponents for clay flocs from the literature.
Fig 1.
Schematic diagram of the floc breakup facility.
Table 2.
Siphon flow conditions used in the floc breakup experiments.
Fig 2.
Microscope images of the particles and processing steps used to derive particle measurements.
(a) Primary particles flowing from the pump, (b) flocs formed in the aggregation tank, (c) floc image after the thresholding step, (d) final binary image for particle processing (background shading in c and d added for uniformity), (e) perimeters of each floc and (f) fitted ellipses from the imageJ particle analysis overlaid on the microscope image. Scaling of all images is the same.
Table 3.
Summary of experimental tests with floc population stability and resulting strength exponents.
Fig 3.
Size-spectra of particle volume for the primary particles and flocs in the aggregation tank measured with the LISST-100X.
Consistency of floc size distribution during experimentation is shown with measurements obtained nine hours apart during continuous facility operation. Error bars represent one standard deviation.
Fig 4.
Particle size spectra measured in the aggregation tank for all eight experimental runs.
Symbols consistent with Table 3. Error bars represent one standard deviation.
Fig 5.
Normalized volume spectra measured with the LISST compared with spectra obtained from the microscope images.
Fig 6.
Plot of in-water light attenuation measured with the ac-9 spectrophotometer as a function of wavelength.
Fig 7.
The difference in particle volume between the aggregation tank and the primary particle distribution.
This figure illustrates the calculation of the volume of primary particles removed from the population (SVag) and the floc volume formed due to aggregation (FVag).
Table 4.
Floc population characteristics for the unmixed (left, white) and mixed (right, grey) aggregation tank conditions.
Fig 8.
Size spectra of particle volume for the unmixed aggregation tank conditions.
Data is plotted for (a) Laminar flow, (b) turbulent flow with ε = 0.10 m2/s3, and (c) turbulent flow with ε = 2.15 m2/s3 conditions. Spectra for the primary particles, flocs in the aggregation tank, and flocs in the breakup tank are plotted in each case.
Fig 9.
Size spectra of particle volume for the mixed aggregation tank conditions.
Data is plotted for (a) laminar flow, (b) turbulent flow with ε = 0.10 m2/s3, and (c) turbulent flow with ε = 2.15 m2/s3 conditions. Spectra for the primary particles, flocs in the aggregation tank, and flocs in the breakup tank are plotted in each case.
Fig 10.
Change in largest floc size with turbulent dissipation rate in the siphon tubes.
Symbols are consistent with Table 3.
Fig 11.
The difference in particle volume after flowing through the siphon tubes.
Data for (a) an unmixed aggregation tank condition and (b) a mixed aggregation tank condition.
Fig 12.
Response of the spectral slope of attenuation with changes in the particle volume-weighted mean diameter after flowing through the siphon tubes.
Fig 13.
Plots of largest stable floc size vs. dissipation rate in the siphon tubes.
(a) Unmixed aggregation tank conditions and (b) the mixed aggregation tank conditions are shown. Symbols consistent with Table 3.
Fig 14.
Floc strength curves plotted with lengthscales obtained from the LISST and from the image analysis for an unmixed aggregation tank.
Fig 15.
Effects of particle lengthscale on the breakup trends.
(a) Floc strength curves plotted with lengthscales obtained from the LISST and from the image analysis for a mixed aggregation tank condition. Kolmogorov lengthscale of the siphon flow plotted as the solid black line. (b) D95 of the Feret-diameter particle spectra normalized with the Kolmogorov lengthscale of the siphon flow and plotted against the dissipation rate in the siphon tubes. Symbols in (b) consistent with Table 3.