Activated-carbon, oxidation, conventional-treatment, filtration, and m
embrane studies are conducted to determine which process is best suite
d to remove the herbicide glyphosate from potable water. Both bench-sc
ale and pilot-scale studies are completed. Computer models are used to
evaluate the results. The activated-carbon results show that glyphosa
te adsorbs very strongly in distilled water, but has a much lower capa
city in Ohio River water. The jar-test studies with an alum coagulant
show that as turbidity is removed, so is glyphosate. The majority of t
he glyphosate removal occurs as turbidity is reduced below 2 nephelome
tric turbidity units (NTUs). Powdered-activated-carbon treatment is in
effective. Ultrafiltration membranes and 0.45 mum filters do not remov
e glyphosate in Ohio River water even though the effluent turbidity is
reduced below 0.2 NTU. The oxidation results indicate that glyphosate
is easily destroyed by chlorine and ozone. Chlorine dioxide, permanga
nate, and hydrogen peroxide are less successful. These conventional-tr
eatment and adsorption results are confirmed by pilot-scale studies.