CONCENTRATIONS OF STABLE ISOTOPES OF CESIUM AND STRONTIUM IN FRESH-WATERS IN NORTHERN ENGLAND AND THEIR EFFECT ON ESTIMATES OF SORPTION COEFFICIENTS (K-D)

Concentrations of stable isotopes of cesium and strontium were measure d in twenty lakes in northern England. Stable Sr concentrations ranged from 8-214 mu g L(-1) and showed a good linear relationship to Ca + M g concentrations. These observations are consistent with previous obse rvations from waters with no major sources of strontium minerals in th e catchment. Stable Cs concentrations were at least 3 orders of magnit ude lower generally less than the 12 ng L(-1) limit of detection. Thes e data increase the number of stable Cs concentrations reported in the literature by a factor of 2. A critical analysis of literature data o n stable Cs concentrations indicated a range of <12-60 ng L(-1). Unlik e the case for Sr, no relationship was found between Cs and its major competitor, potassium. An ion exchange model was used to derive equati ons for radiocaesium and radiostrontium sorption coefficients, K-d, in the presence of both a competitive major ion and non-trace concentrat ions of the relevant stable isotope, i.e., stable isotope concentratio ns at which significant competition with the major ion occurs. Since t he competition for sorption sites by stable Sr in natural freshwaters never exceeded 0.26% of the competition from Ca, the normal, simplifyi ng assumption of trace Sr levels (i.e., concentrations which are too s mall to influence the chemistry) was confirmed as applicable in the io n exchange model of Sr sorption. This was also found to be true in lab oratory studies when carrier was added with radioactive isotopes. The assumption of trace concentrations was also confirmed for stable Cs in natural waters. However, the very high selectivity coefficient of Cs, relative to potassium, resulted in the assumption being breached duri ng the measurement of K-d in the laboratory when stable carrier was pr esent. K(d)s measured using non-carrier free isotopes were underestima ted by up to a factor of 5. A very good relationship was found between log K-d and the potassium concentration (R(2) = 0.67 when corrections for the presence of stable carrier were made). The relative importanc e of other competing monovalent and divalent cations (Na+, K+, Mg2+, C a2+) was investigated using modifications of the ion exchange definiti on of K-d. In the natural waters we studied, Mg2+ had an effect on the K-d(Sr) equivalent to up to 30% of the Ca effect, whereas ammonia and potassium showed very little effect on the K-d(Sr). In our samples, N a influence never exceeded 50% of the Ca effect on K-d(Cs), but in est uarine and marine samples, the effect would obviously be greater. Calc ium had no effect on K-d(Cs) and Na+ and NH4+ had only a minor effect in our samples. Copyright (C) 1997 Elsevier Science Ltd.