142 R. T. Prentki et al. 



pore waters of Barrow sediments and soils. These calculations indicated 

 that the phosphate concentrations existing in solution were too low to have 

 been set by solubility criteria and, therefore, must have been set by 

 sorption phenomena. 



Other investigators have generally found that sorption rather than 

 discrete mineral formation is the primary means of phosphate fixation 

 within sediments, at least in lakes without intense calcium carbonate 

 precipitation (Syers et al. 1973). Li et al. (1972) found that phosphate that 

 could be exchanged with ^"P04"~ constituted 14 to 43% of inorganic 

 phosphorus in the sediments of four lakes, which is about the same range 

 found for Barrow sediments and soils (see Table 4-18). Williams et al. 

 (1971a, b, c, 1976) concluded that sorption to hydrated iron oxides was 

 primarily responsible for phosphate fixation in aerobic and anaerobic 

 sediments of both calcareous and noncalcareous lakes. Two earlier but 

 similar studies (Frink 1969 and Harter 1968) considered aluminum to be 

 more important than iron in phosphate fixation, but this conclusion 

 depended upon sequential NH4F and NaOH extractions releasing only 

 aluminum-bound and iron-bound phosphorus, respectively. We and 

 Williams et al. have found this assumption to be incorrect. 



Phosphate Sorption to Sediments 



In a sorption system, the relationship between water concentration, 

 sorbent, and sorbed pool can usually be described by one of three 

 mathematical functions (Freundlich, Langmuir, or Temkin isotherm) 

 which are derived from different assumptions about the energy with which 

 individual ions or molecules are held (Trapnell 1955). The applicability of 

 these three isotherms to phosphate sorption has been discussed by Bache 

 and Williams (1971) and need not be repeated here. 



For the pond sediments, Prentki (1976) concluded that the Temkin 

 isotherm best modeled the relationship between DRP and sorbed 

 phosphate over the environmental range of concentrations. The Temkin 

 isotherm may be represented by the equation (Bache and Williams 1971): 



XX^-' ={RTb')i\ngC), 



where 



X = sorbed phosphate per unit sorbent 



Xrr, = sorption maximum 



C = equilibrium DRP concentration 



R = gas constant 



T = temperature in K, and 



b.g = constants. 



