Smectite was weighed and added to a 25 mL glass Erlenmeyer flask (Kimble KIMAX Flask with ST16 Glass Stopper, Graduated, 26,600) based on the amounts shown in Table 1. 18.5 mL of deionized water was then

added to the flask, which was sealed using a glass stopper. The mixture was vortexed for 5 s and the resulting mixture was left for 24 hours to allow the smectite to swell and minimize its phosphate sorption. To this mixture 0.5 mL of added a 10 ppm concentration. The mixture was vortexed for 5 s. To this mixture 1 mL of the polymer solution was added to make the final polymer concentration of 100 ppm and a final volume 20 mL. The mixture was then vortexed again for 5 s, a stir bar was added to the flask, and sealed with a glass stopper and then placed on a stir plate (JEIO TECH, Multi Channel Stirrer, Model: MS-53M) at 250 RMP for 24 hours. Two control experiments were examined. The first control was to examine the amount of removed by the smectite. This control was prepared as above but with the addition of 1 mL of water in place of the polymer. The second control contained only, no smectite, and 1 mL of water was added in place of the polymer. After 24 hours, the stir plate was stopped, the stir bars were removed and 8 mL of solution were placed in a centrifuge tube. The solutions were then centrifuged (Eppendorf Centrifuge 5810) at 3500 RPM for 15 min to remove any solid material that might be present. The supernatant was then filtered through a 0.2 μm syringe filter and analyzed for using a Dionex ICS 2100 ion chromatography system (Suppressor, AERS 500 2 mm, Guard and Analytical Column, AG18, AS18) with a 25 μm injection loop, running at a flow rate of 0.25 mL/min in gradient mode (23 mM KOH for 13 min then 30 mM KOH for 12 min) resulting in an elution time for phosphate of 23 min.

150 mg of smectite was weighed and added to a 25 mL glass Erlenmeyer flask (Kimble KIMAX Flask with ST16 GlassStopper, Graduated, 26,600). 18.5 mL of deionized water was then added to the flask, which was sealed using a glass stopper. The mixture was vortexed for 5 s and the resulting mixture was left for 24 hours to allow the smectite to swell and minimize its phosphate sorption. To this mixture 0.5 mL of a spike added to obtain the various concentrations listed in Table 1. The mixture was vortexed for 5 s. To this mixture 1 mL of the polymer solution was added to make the final polymer concentration of 100 ppm and a final volume 20 mL. The mixture was then vortexed again for 5 s, a stir bar was added to the flask, and sealed with a glass stopper and then placed on a stir plate (JEIO TECH, Multi Channel Stirrer, Model: MS-53M) at 250 RMP for 24 hours. Two control experiments were examined. The first control was to examine the amount of H₂PO₄^? removed by the smectite. This control was prepared as above but with the addition of 1 mL of water in place of the polymer. The second control contained only, no smectite, and 1 mL of water was added in place of the polymer. After 24 hours, the stir plate was stopped, the stir bars were removed and 8 mL of solution were placed in a centrifuge tube. The solutions were then centrifuged (Eppendorf Centrifuge 5810) at 3500 RPM for 15 min to remove any solid material that might be present. The supernatant was then filtered through a 0.2 μm syringe filter and analyzed for as described above.

150 mg of Smectite was weighed and added to a 25 mL glass Erlenmeyer flask (Kimble KIMAX Flask with ST16 GlassStopper, Graduated, 26,600). 18.5 mL of deionized water was then added to the flask, which was sealed using a glass stopper. The mixture was vortexed for 5 s and the resulting mixture was left for 24 hours to allow the smectite to swell and minimize its phosphate sorption. To this mixture 0.5 mL of added to obtain a 10 ppm concentration. The mixture was vortexed for 5 s. To this mixture 1 mL of the polymer solution was added to make the final polymer concentration of 100 ppm and a final volume 20 mL. The mixture was then vortexed again for 5 s, a stir bar was added to the flask, and sealed with a glass stopper and then placed on a stir plate (JEIO TECH, Multi Channel Stirrer, Model: MS-53M) at 250 RMP for the allotted amount of time described in Table 1. Two control experiments were examined. The first control was to examine the amount of removed by the smectite. This control was prepared as above but with the addition of 1 mL of water in place of the polymer. The second control contained only, no smectite, and 1 mL of water was added in place of the polymer. After the allotted amount of time, the stir plate was stopped, the stir bars were removed and 8 mL of solution were placed in a centrifuge tube. The solutions were then centrifuged (Eppendorf Centrifuge 5810) at 3500 RPM for 15 min to remove any solid material that might be present. The supernatant was then filtered through a 0.2 μm syringe filter and analyzed for as described above.

The first experiment conducted was to determine the effect of increasing the amount of smectite added to the mixture. It was assumed that as more smectite was added less of the polymer would be available to interact with the H₂PO₄^? in solution, as it would be adhered to the surface of the smectite particles. The results initially support this assumption as the polyacrylamides generally removed less with additional smectite added up to 100 mg (Figure 2). The smectite control also showed an increase in adsorption from 4% to 10% at 100 mg. The nonionic polyacrylamide MAGN removed an additional 5% of the from solution above the amount removed by the smectite alone, up to 100 mg of smectite added. The amount of removed by MAGC and A14 generally trended downward with the addition of smectite up to 100 mg with A14 removing the most of all the polyacrylamides at 28%. Generally, the polyacrylamides increased the amount of removed with increased smectite added, up to 35% for A14 at 500 mg of smectite added. PDM diverged from these trends by increasing the amount of removed up to 100 mg of smectite added, followed by a slight decrease in the removed with increasing smectite added.

The results from the polymers up to the 100 mg smectite addition was likely caused by saturation of the negative charge of the clay by the positive charge of the polymer. When there is a high enough concentration of the cationic polymer flocculant compared to the solid material present the charge of the solid material can become negated, thereby changing net charge of the clay-polymer complex [6] [14] [15] . This has the effect of impairing the polymers ability to form flocs and settle out of solution. The net result of this oversaturation of solid material is dispersion as the now neutral or positively charged smectite-polymer complexes repel each other. Given the smallest amount of smectite added (25 mg to 100 mg), less of the polymer was available to interact with in solution so the amount of removed with increasing smectite weight added was observed. The results from additional amounts of smectite above 100 mg up to 500 mg trended toward more removed from solution, however, they were also within error of each other. The slight increase in removed from solution with increasing smectite additions is likely due to sorption of to the smectite rather than to the polymer. Given these results, 150 mg of smectite was chosen as the amount of smectite that would be added to the following experiments and remain constant as described in Table 1.

As the solution concentration of increased from 1 ppm to 100 ppm the general trend was an increase in the amount of phosphate removed from solution for all the polymers tested as well as the smectite control (Figure 3). Generally, increasing the amount of one reactant will increase the probability of interaction with the second reactant until the second reactant is completely consumed. In the concentration range of 1 - 50 ppm, the polymers continued the trend of increasing the amount of removed from solution with increasing positive charge density of the polymer. For the concentration range of 50 - 100 ppm the polyacryla- mides removed the same amount of as smectite alone and were within error of each other. PDM with 100% positive charge density consistently removed more from solution than the other polymers tested. At 100 ppm PDM removed almost 9 ppm of, however, the error associated with the PDM results began to increase as the concentrations increased above 50 ppm. From the results of this experiment a concentration of 10 ppm was chosen for use in the next study to observe the effect that the amount of time the polymer mixtures were stirred for would have on the amount of removed from solution.

The third and final study examined the effect of mixing time on the amount of removed from solution. Increasing the amount of time tested the progress of the interaction of with the polymers. In general, the results again show that an increase in positive charge density of the polymer results in more removed from solution (Figure 4). The smectite control experiment showed an increase in sorption from ~2% to ~7% removed in the first two hours. After this the smectite drifted toward more removed from solution up to ~10% at 36 hours. The nonionic polyacrylamide MAGN effectively followed the results from smectite alone, removing only what the smectite had sorbed. The MAGC and A14 followed similar trends up to 18 hours with A14 removing more from solution than MAGC. After 18 hours, A14 and MAGC gen-

erally removed similar amounts of. PDM increased its sorption up to ~35% at 10 hours and maintained that level of sorption, within error out to 36 hours. In all cases, including the smectite control, the amount of removed remained constant after the maximum was reached suggesting an irreversible binding.