Development and evaluation of a diffusive gradients in a thin film technique for measuring ammonium in freshwaters

• A new DGT technique, using Microlite PrCH, was developed and evaluated for monitoring NH4–N in freshwaters.
• PrCH-DGT is suitable for a wide range of pH conditions (3.5–8.5) and ionic strengths (up to 0.012 mol L−1 NaCl).
• PrCH-DGT showed a good linear uptake for NH4–N over 72 h in synthetic freshwater.
• During field deployments CDGT values with DBL matched well with the average grab sample values.

A new diffusive gradients in a thin film (DGT) technique, using Microlite PrCH cation exchange resin, was developed and evaluated for measuring NH4–N in freshwaters. Microlite PrCH had high uptake (>92.5%) and elution efficiencies (87.2% using 2 mol L−1 NaCl). Mass vs. time validation experiments over 24 h demonstrated excellent linearity (R2 ≥ 0.996). PrCH-DGT binding layers had an extremely high intrinsic binding capacity for NH4–N (∼3000 μg). NH4–N uptake was quantitative over pH ranges 3.5–8.5 and ionic strength (up to 0.012 mol L−1 as NaCl) typical of freshwater systems. Several cations (Na+, K+, Ca2+ and Mg2+) were found to compete with NH4–N for uptake by PrCH-DGT, but NH4–N uptake was quantitative over concentration ranges typical of freshwater (up to 0.012 mol L−1 Na+, 0.006 mol L−1 K+, 0.003 mol L−1 Ca2+ and 0.004 mol L−1 Mg2+). Effective diffusion coefficients determined from mass vs. time experiments changed non-linearly with electrical conductivity. Field deployments of DGT samplers with varying diffusive layer thicknesses validated the use of the technique in situ, allowed deployment times to be manipulated with respect to NH4–N concentration, and enable the calculation of the diffusive boundary layer thickness. Daily grab sample NH4–N concentrations were observed to vary considerably independent of major rainfall events, but good agreements were obtained between PrCH-DGT values and mean grab sample measurements of NH4–N (CDGT:CSOLN 0.83–1.3). Reproducibility of DGT measurements in the field was good (relative standard deviation < 11%). Limit of detection was 0.63 μg L−1 (equivalent to 0.045 μmol L−1) based on 24 h deployments.

Figure optionsDownload as PowerPoint slide