Availability of fluoride to plants grown in contaminated soils

Two pot experiments were carried out to study uptake of fluoride (F) in clover and grasses from soil. Fluoride concentrations in Trifolium repens (white clover) and Lolium multiflorium (ryegrass) were highly correlated with the amounts of H2O- and 0.01 M CaCl2-extractable F in soil when increasing amounts of NaF were added to two uncontaminated soils (r=0.95-0.98, p<0.001). The amounts of H2O- or 0.01 M CaCl2-extractable F did not explain the F concentrations to a similar extent in Agrostis capillaris (common bent) grown in 12 soils (Cambic Arenosols) collected from areas around the Al smelters at Ardal and Sunndal in Western Norway (r=0.68-0.78). This may be due to variation in soil pH and other soil properties in the 12 soils. Soil extraction with 1 M HCl did not estimate plant-available F in the soil as well as extraction with H2O or 0.01 M CaCl2. Fluoride and Al concentrations in the plant material were positively correlated in most cases. Fluoride and Ca concentrations in the plant material were negatively correlated in the first experiment. No consistent effects were found on the K or Mg concentrations in the plant material. The F accumulation in clover was higher than in the grasses. The uptake from soil by grasses was relatively low compared to the possible uptake from air around the Al smelters. The uptake of F in common bent did not exceed the recommended limit for F contents in pasture grass (30 mg kg(-1)) from soil with 0.5-28 mg F(H2O) kg(-1) soil. The concentration in ryegrass was about 50 mg F kg(-1) when grown in a highly polluted soil (28 mg F(H2O) kg(-1) soil). Concentrations in clover exceeded 30 mg F kg(-1) even in moderately polluted soil (1.3-7 mg F(H2O) kg(-1) soil). Liming resulted in slightly lower F concentrations in the plant material.