1) dripping into a drain line gives more contact time to precip PO4 before it hits the mechanical filter. Dosing into a side stream would yield similar results.
2) ideally you shouldn't strive to dose LaCl2 into the display tank in order to react with PO4 there. Since LaCl2 depletes alkalinity I prefer to couple the treatments with water changes in order to bring alkalinity back up ASAP.
1) I'm not quite clear here. As I understood it you were dripping into the drain and catching the resulting precipitate immediately in a 10um filter bag. The flow rate is quite fast, and any unreacted LaCl2 goes thru the bag. Do you have another mechanical filter after that? Are you basically dripping slowly enough that you feel the LaCl2 has had time to react completely with PO4? How are you controlling dripping this stuff (I'm sure this one's been answered)?
Selectivity: I just browsed a study (Journal of Environmental Sciences 20(2008) 670"“674) they used a zeolite matrix with LaCl2. They flowed through a solution with:
20°C, pH 6.0, 1.5 mg/L phosphate concentration in the presence
300 mg/L potassium sulfate, calcium
bicarbonate, and sodium chloride concentrations, respec-
tively.
They found excellent selectivity for removal of phosphate under these conditions, with carbonate, sulfate and chloride adsorption following in that order. Side note: In the limit of high pH (10.5) the substrate released 90% of the phosphate that it had adsorbed. Under these conditions phosphate is in a monodentate form that does not form a lanthanum salt.
Another study found removal efficiency of less than 10% of phosphate under conditions of pH 8, and found 99% removal at ph 4. Because this stuff is cheap, the only concern this would bring is that of residual unreacted LaCl2 in the aquarium- potentially up to 90% of what you put in
side note: Also, it turns out if you can make a matrix with zeolite carrying the LaCl2, the zeolite removes ammonia (yay?)
This looks like it's gonna be big! It would be great if we could lower the pH before a LaCl2 column, then increase it again after removal of the precipitates. hmmmm.