Are lipid soluble metal complexes important in aquatic environments?

Trace metals are normal constituents of natural waters and may be present in a wide range of physico-chemical forms or species. The bioavailability and toxicity of trace metals is dependent on their speciation. In laboratory experiments it has been demonstrated that the toxicity of some trace metals to aquatic organisms is related to the activity of the free metal ion rather than the total metal concentration, however, in a small number of studies it has been shown that toxicity and bioavailability of lipid soluble metal complexes may exceed that of the equivalent free metal ion species. Some toxicity data has shown that these lipid soluble metal complexes can be up to 25 times more toxic than free metal ions. LMSC are chemical complexes containing a metal and a biotic ligand. The nature of the LMSC means that it readily traverses membranes. LSMC are of great importance yet there is currently no method that has been developed that has the ability to accurately quantify the small concentrations likely to occur in the environment. The aim of this study is to firstly develop a method for the determination of ultra-trace (ng/L) concentrations of lipid soluble cadmium, copper, nickel, lead and zinc complexes in water. Waters were extracted with 1-octanol, a solvent with a similar dielectric constant to that of cell membranes. A preconcentration factor of 50 was achieved by extraction of 250mL of water into 10 mL of octanol followed by back extraction into a final volume of 5 mL of acidic matrix. Metal concentrations were subsequently determined by Inductively Coupled Plasma Mass Spectrometry (ICPMS). Detection limits (3σ) for the method were 0.002, 0.003, 0.003, 0.001, and 0.011 μg/L for Cd, Cu, Ni, Pb and Zn respectively. The detection limits of the analytical method were lower than those previously reported in the literature. The developed method was used to conduct a survey of aquatic environments from the Sydney, NSW region. Three locations were surveyed, Centennial Park, Homebush Bay and the Cooks River. Detectable concentrations of lipid soluble Cu, Pb and Zn were found in all three locations, whereas lipid soluble Cd and Ni were only detected in the Cooks River. Lipid soluble Cd concentrations ranged from 0.002 to 0.004 μg/L, lipid soluble Cu concentrations ranged from 0.003 to 0.533 μg/L, lipid soluble Ni concentrations ranged from 0.006 to 0.025μg/L, lipid soluble Pb concentrations ranged from 0.002 to 0.087 μg/L and lipid soluble Zn concentrations ranged from 0.020 to 0.499 μg/L. The results were comparable to previously reported lipid soluble metal complex concentrations in environmental samples. The developed analytical method was also used to determine the concentration of lipid soluble metal complexes in mine waste water. Xanthates such as potassium amyl xanthate (PAX) are commonly used in mining processes and are expected to be common in mine waste waters. Potassium amyl xanthate is known to form neutral metal complexes. Detectable concentrations of lipid soluble Cu, Ni, Pb and Zn complexes were measured in the waste water. Lipid soluble Cu concentrations ranged from 70.5 to 84.1 μg/L, Ni concentrations ranged from 0.256 to 0.264 μg/L, Zn concentrations ranged from 0.044 to0.050 μg/L and a single Pb concentration of 0.006 μg/L was recorded. Of these lipid soluble metal complexes, only Pb and Ni concentrations represented a considerable proportion (a proportion greater than the limit of detection of the method) of the total dissolved Pb and Ni concentration. The octanol/water partition coefficients of a number of neutral inorganic complexes were measured to assess the environmental significance of these complexes. In total, 14 complexes were tested. CdCl₂, CuCO₃, Cu(OH)₂, NiCl₂, NiCO₃, Ni(OH)₂, PbCl₂, PbCO₃, Pb(OH)₂, PbSO₄, ZnCO₃ and Pb(OH)₂ all returned low octanol/water partition coefficients (<0.2) indicating that if these complexes are present in natural waters they are unlikely to be lipid soluble. Octanol/water partition coefficients of 3.28 and 0.20 were calculated for the HgCl₂ and B(OH)₃ neutral complexes. The partition coefficients were comparable to quoted literature values. This result indicates that if present in natural waters the HgCl₂ and B(OH)₃complexes may have the ability to passively diffuse across cell membranes. While a robust method for measuring LSMC concentrations has been developed, the entrainment of colloids within the octanol extracts could be of significant concern. Based on analysis of Al and Fe within the back extracts (used as an indication of the entrainment of colloids within the back extracts) it was determined that, in a number of samples, the entrainment of colloids could be greater than the extraction of LSMC. This was of particular concern when assessing the concentration of lipid soluble Cd, Cu, Ni, Pb and Zn. The use of octanol filled dialysis cells could potentially address these concerns in addition to allowing the concentration of lipid soluble metal complexes to be determined in sediment and biota. This thesis also highlights the frequency of LSMC in the environment. Future work should also be performed to determine the concentration of lipid soluble metal complexes in a wider range of aquatic environments, both saline and freshwater. Analysis of a wider range of waste waters from industries that used synthetic organic ligands should also be included in this investigation. The stability of both organic and inorganic lipid soluble complexes in laboratory prepared waters and natural waters should be assessed to gain a better understanding about the environmental significance of these complexes. Finally toxicity testing should be performed using a range of lipid soluble metal complexes with the measurement of actual concentration rather than relying on nominal concentrations of lipid soluble metal complexes in test solutions. This may allow for more accurate toxicity data to be generated.