Determining Flocculation Efficiency

Problem with Previous Efficiency Factor

The efficiency value assumed at the beginning of Spring 2008 was 0.5. This means that we assumed it would take only two collisions for a particle of clay to stick to a particle of aluminum sulfate. Using the MathCAD program, we determined that this efficiency value was inaccurately high by inputting the dimensions of the Fall 2007 Demo Plant, and comparing the MathCAD output for horizontal length of maximum floc size with the actual horizontal length of maximum floc size we observed. We found that assuming an efficiency of 0.5, the program reported an optimal horizontal length about 2 times smaller than what we observed.

This issue was addressed this semester in the experiment described below.

Procedure

We observed flocculation in the demo plant in order to determine at what point in the length of the flocculator the flocs reached a maximum size before beginning to break up. These flocs were also measured in order to be able to more accurately represent in the flocculation model the maximum floc size that it is possible to achieve in the plant. Using the location of the largest flocs as an estimate of the optimum plant length, we set the maximum floc size to the size of the measured flocs and changed the value of the efficiency factor in the flocculation model until the design length of the plant model created by the flocculator design program was consistent with the observed occurrence of maximum floc size.

Results and Discussion

The maximum flocs were observed at a distance of 28 cm from the flocculator inlet. After that point in the plant, they ceased to get any bigger, and eventually began to break up from too much shear. The size of the largest flocs was about 0.7 mm, so that value was set as the maximum floc size in the flocculation model. When various efficiencies were used in the model, the design program returned a design length of approximately 28 cm for a value of 0.2, and so we have determined that the flocculation efficiency is equal to this value.

Figure 1. Maximum Floc Formation

Originally, we assumed that problems with the Fall 2007 Demo Plant were attributed to the dimensions of the channels, including channel width and baffle spacing. These properties were predetermined by the corrugated plastic material that was used. Thus, we originally planned to obtain different materials from McMasterCarr.

However, once we adjusted the efficiency to a more realistic value, we found that the channel dimensions were not that far from ideal. Therefore, the same materials were used for the Spring 2008 Demo Plant that were used for the Fall 2007 Demo Plant.