Speciation using computer

Kounaves and Zirino [ 145] studied cadmium-EDTA complex formation in seawater using computer-assisted stripping polarography. They showed that the method is capable of determining the chemical speciation of cadmium in seawater at concentrations down to 10 8 M. 

Brugmann [784] discussed different approaches to trace metal speciation (bioassays, computer modelling, analytical methods). The electrochemical techniques include conventional polarography, ASV, and potentiometry. ASV diagnosis of seawater was useful for investigating the properties of metal complexes in seawater. Differences in the lead and copper values yielded for Baltic seawater by methods based on differential pulse ASV or AAS are discussed with respect to speciation. 

Kennedy, S. K., Walker, W and Forslund, B. (2002). Speciation and characterisation of heavy-metal contaminated soils using computer-controlled scanning electron microscopy. Environ. Forensics 3,131-143. 

CCM) (Stumm et al., 1970, 1976, 1980 Schindler et al., 1976), the triple-layer model (TLM) (Davis and Leckie, 1978, 1980 Davis et al., 1978 Hayes and Leckie, 1987 Hayes et al., 1988), and the 1 pK basic Stem model (Bolt and van Riemsdijk, 1982 Van Riemsdijk et al., 1986, 1987). The application of many of the commonly used computer models in the determination of the speciation in solution phase has been dealt with exhaustively by Lumsdon and Evans (1995). 

Many chemical models have been developed over the years for problems solving chemical speciation. A comprehensive survey of many of the models and detailed analysis of computer modeling in relation to trace element speciation has been by made by Waite (1989), with particular reference to MINEQL, and by Lumsdon and Evans (2002), with particular reference to MITEQA2. Soil solution has significant amounts of DOM, which is poorly quantified and understood. The trace element speciation information obtained using computer modeling with data on complexation constants of metal—DOM complexes has to be substantiated with data on free metal ion species obtained from sophisticated physical techniques. 

Normally, XAFS is considered a bulk sample technique which means that it averages the speciation of the element of interest across the sample as a whole thus, multiple species can be present. In these mixed systems the average XAFS spectrum is additive in nature, i.e. it is a combination of XAFS spectra of each individual species detected (Kelly et al. 2008). Due to this additive nature a common approach in XAFS studies of natural samples involves modeling the unknown sample using computed fractions of representative standards which give the best fit to the unknown sample. The approach commonly used is linear combination fitting (LCF) which involves performing least-squares fits of the sample... 

Barnett, M.I., J.R. Duffield, D.A. Evans, J.A. Findlow, B. Griffiths, C.R. Morris, J.A. Vesey and D.R. Williams, 1989, Modelling bioavaiiability as a function of speciation using physicochemical data and computers, in Nutrient Availability Chemical and Biological Aspects, eds D. Southgate, I. Johnson and G.R. Fenwick (Royal Society of Chemistry, London) pp. 97-99. 

Twiss, M., Errecalde, O., Fortin, C., Campbell, P., Jumarie, C., Denizeau, F., Berkelaar, E., Hale, B., and van Rees, K., Coupling the use of computer chemical speciation models and culture techniques in laboratory investigations of trace metal toxicity, Chem Spec Bioavailab, 13 (1), 9-24, 2001. 

Although computer programs are now used to perform speciation calculations, examining how these calculations are performed provides important insights into the limitations of the model predictions. Thus, we will step through a small part of the calculation used to generate the results presented in Figure 5.4, which represents the iron... 

Various munerical techniques are used to indirectly obtain solutions to large systems of equations with too many imknowns to solve explicitly. One approach is to solve the equations iteratively. This is done by first assuming that all of the anions are unbound and, hence, their free ion concentrations are equal to their total (stoichiometric) concentrations. By substituting these assumed anion concentrations into the cation mass balance equations, an initial estimate is obtained for the free cation concentrations. These cation concentrations are substituted into the anion mass balance equations to obtain a first estimate of the free anion concentrations. These free anion concentrations are then used to recompute the free cation concentrations. The recalculations are continued imtil the resulting free ion concentrations exhibit little change with further iterations. The computer programs used to perform speciation calculations perform these iterations in a matter of seconds. 

Although the emphasis in this article has been on the discussion of toihniques and methods that can be used in the laboratory for the identification of species, increasing importance is being attached to computer simulation of trace element speciation. The reason for this increased interest could be attributed in part to the availability of relevant experimental data which could be used in developing the required models. However, computer simulation comes into its own when the species are so imstable that separation techniques cannot be applied and/or the detection systems do not have the required sensitivity. 

Computer simulation has been used to predict the speciation of various trace elements during chelate therapy > and in total parenteral nutrition... 

One other serious criticism regarding the data on Cu speciation is the neglect of the cysteine present in blood plasma. Cu11 and cysteine undergo a facile redox reaction (Chapter 20.2). Since the reaction is irreversible, no quantitative thermodynamic quotient is available for use in the computer calculations. Another assumption often made is that the overwhelming concentration of other amino acids may prevent cysteine coordination and, as a result, stabilize the Cu11 state. Recent studies show that this assumption is totally unjustified48 and so the dilemma still has to be resolved. 

Raman and infrared vibrations are mutually exclusive and consequently use of both techniques is required in order to obtain a set of vibrational bands for a molecule. The advent of powerful computer-controlled instrumentation has greatly enhanced the sensitivity of these vibrational spectroscopies by the use of Fourier transform (FT) techniques, whereby spectra are recorded at all frequencies simultaneously in the time domain and then Fourier transformed to give conventional plots of absorbance versus frequency. The wide range of applications of FT Raman spectroscopy is discussed by Almond et al. (1990). Specific examples of its use in metal speciation are the observation of the Co-C stretch at 500 cm-1 in methylcobalamin and the shift to lower frequency of the corrin vibrations when cyanide is replaced by the heavier adenosyl in going from cyanocobalamin to adenosylcobalamin (Nie et al., 1990). 

Inorganic speciation in solution can also affect the mobility of metal ions (Doner, 1978). The formation of an ion-pair with Cl can more than double the mobility of Cd in the presence of 200molm 3NaCl. At the same chloride concentration, however, the mobilities of Cu2+ and Ni2+ are only increased slightly (5-10%), presumably because of very weak complexation with Cl. This mechanism could increase the leaching of Cd from saline soils but it may not be effective in non-saline soils because the ratio of the total concentrations of Cd Cl must be >1 106 before >50% of total Cd is complexed by Cl (estimated using the computer model TITRATOR (Cabaniss, 1987), which considered the chloro and hydroxy complexes of Cd at pH 5.0 and a total Cd concentration of 0.1 mmolm-3 equilibrium constants were taken from Lindsay (1979)). 

Computer simulation is now used extensively as a tool to help to understand and predict the transport of radionuclides through environmental systems. Most models relate to waste disposal and are based on measured parameters such as water movements, salinity, suspended load and the radionuclide concentration in the solute, suspended particulate matter and bottom deposits. Comparatively few attempts appear to have been made to include chemical speciation into this type of model, presumably because of the added complexity involved. Some modellers have attempted to take into account the characteristics of the major chemical phases such as those present in different particles or coatings (e.g. Martinez-Aquirre et al., 1994). Others have noted the importance of including details of particular chemical species present in industrial waste releases when constructing models to predict dispersion (Abril and Fraga, 1996). 

For experimental studies, a chemical thermodynamic modelling approach could theoretically reduce unnecessary experimental effort and hence the overall cost of a research project. Once experiments are underway, the computer simulation should also offer valuable assistance in the interpretation of results. Modelling techniques with particular reference to radionuclide speciation have been discussed by Cross and Day (1986) who pointed out that models are only as good as the thermodynamic data upon which they are based. For example, formation constants (a prerequisite for chemical modelling) are invariably generated in idealised laboratory conditions and their use seldom reflects the natural environment... 

Since equilibrium constants are defined by ion activities, which are defined by their concentrations and coefficients (see equation 4.15), they do not include ion pairing or complexation effects. In a multi-ion and multiligand solution, where ion pairing is common, it is necessary to use thermodynamic equilibrium constants to convert the ion-pair concentrations to concentrations of free ions. This equilibrium constant (Kc) is defined by concentrations, making it useful to compute ion speciation. The thermodynamic equilibrium constant (K q) used in calculating Kc is based on the following conditions I = 0 m, 25°C and 1 atm. Thus, Kc is defined by the following equation ... 

Chemical speciation in soil solutions and other natural waters can be calculated routinely with a number of software products offered in a variety of computational media.27 30 Five examples of these products are listed in Table 2.5. They differ principally in the method of solving the chemical equilibrium problem numerically, or in the chemical species and equilibrium constants considered, or in the model used to estimate single-species activity coefficients. Irrespective of these differences, all the examples follow a similar algorithm ... 

---