Seawater test case

The results of the river water and seawater test cases computed by the aqueous models listed in Table I are summarized in Tables IV-X. Tables IV and V compare selected major and minor species computed for the river water test case, and Tables VI and VII make a similar comparison for the seawater test case. Table VIII compares activity coefficients computed for the major species in seawater and Table IX and X tabulate saturation indices for selected minerals in the river water and seawater test... 

Saturation Index for Selected Minerals in Seawater Test Case... 

Ba, Sr, and B. Consistency between programs was evaluated by comparing the log of the molal concentrations of free ions and complexes for two test solutions a hypothetical seawater analysis and a hypothetical river water analysis. Comparison of the free major ion concentrations in the river water test case shows excellent agreement for the major species. In the seawater test case there is less agreement and for both test cases the minor species commonly show orders of magnitude differences in concentrations. These differences primarily reflect differences in the thermodynamic data base of each chemical model although other factors such as activity coefficient calculations, redox assumptions, temperature corrections, alkalinity corrections and the number of complexes used all have an affect on the output. 

One approach to determine the reliability of geochemical codes is to take well-defined input data and compare the output from several different codes. For comparison of speciation results, Nordstrom et al. (1979) compiled a seawater test case and a river-water test case, i.e., seawater and river-water analyses that were used as input to 14 different codes. TTie results were compared and contrasted, demonstrating that the thermodynamic databases, the number of ion pairs and complexes, the form of the activity coefficients, the assumptions made for redox species, and the assumptions made for equilibrium solubilities of mineral phases were prominent factors in the results. Additional arsenic, selenium, and uranium redox test cases were designed for testing of... 

Collectively, the programs mentioned above represent the "state of the art" in the calculation of the equilibrium distribution of species in aqueous systems. As a means of examining the consistency of these programs, two test cases (a dilute river water and an average seawater analysis) were compiled and mailed to more than fifty researchers who have been active in the field of chemical modeling. These test cases may overlook many of the features of specific programs, but they provide a common basis by which most of the programs can be... 

Table II gives a general description of the program features such as total number of elements, aqueous species, gases, organic species, redox species, solid species, pressure and temperature ranges over which calculations can be made, an indication of the types of equations used for computing activity coefficients, numerical method used for calculating distribution of species and the total number of iterations required by these models for each of the two test cases. The chemical analyses for the two test cases are summarized in Table III. The seawater compilation was prepared in several units to assure consistency between concentrations for proper entry into the aqueous models.

In another example, five test cases were computed by PHREEQE and EQ3/6 and the same thermodynamic database was run for each program (INTERA, 1983) to test for any code differences. The five examples were speciation of seawater with major ions, speciation of seawater with complete analysis, dissolution of microchne in dilute HCl, reduction of hematite and calcite by titration with methane, and dedolomitization with gypsum dissolution and increasing temperature. The results were nearly identical for each test case. Test cases need to become standard practice when using geochemical codes so that the results will have better credibility. A comparison of code computations with experimental data on activity coefficients and mineral solubilities over a range of conditions also will improve credibility (Nordstrom, 1994). 

Because multiple samples will normally be exposed in a seawater test program, specimens should be labeled in a maimer whereby the sample can be easily identified during sample preparation and evaluation. ASTM G 52 recommends several methods of such specimen labeling including the use of coded drilled holes, edge notches, corrosion-resistant tags, and stamped numerals. In some cases, the use of stamped... 

Saponification Paints are most commonly used to protect steel from corrosion by seawater in marine applications and soil in the case of buried structures. Additional protection is often supplied by the application of cathodic protection to the steel. Any paint coating used in conjunction with cathodic protection must be resistant to the alkali which is produced on the steel at defect sites in the coating. The amount of alkali generated depends on the potential to which the steel is polarized. Some paint binders such as alkyds and vinyl ester are very susceptible to saponification, and should not be used on cathodically protected structures. Cathodic disbondment testing should be undertaken if the relevant information is not available. 

Comparative tests between HSI and HSCI in seawater at 93° C and 10-8Am showed consumption rates of 8-4kg A y and 0-43 kg A y , respectively . These figures show that the consumption rate of HSI when used in seawater without the addition of chromium may approach that of steel, but because of the very deep pitting and its fragility, it is in most cases inferior to steel. However, in fresh waters HSI has a far lower corrosion rate than steel. The consumption rate of HSCI freely suspended in seawater in the current density range 10-8 to 53-8 Am increases from 0-33 kg A y at 10-8Am to 0-48 kg A" y at 53-8Am Direct burial in seawater silt or mud will also increase the consumption rate, with values of 0-7kg A y at 8-5 Am increasing to 0-94 kg A " y at 23-4 Am . 

Samples were processed in clean rooms in the shore laboratory within 30 min of sampling. Results indicated (i) the feasibility of inter-calibrating using the enclosure approach (ii) the availability of chemical techniques of sufficient precision in the case of copper, nickel, lead, and cobalt for sampler intercomparison and storage tests and (iii) a problem in sub-sampling from the captured seawater in a sampler, and the difficulty of commonly used samplers to sample seawater in an uncontaminated way at the desired depth. 

The first aim of this work was to study the influence of an unwashed membrane filter on the cadmium, lead, and copper concentrations of filtered seawater samples. It was also desirable to ascertain whether, after passage of a reasonable quantity of water, the filter itself could be assumed to be clean so that subsequent portions of filtrate would be uncontaminated. If this were the case, it should be possible to eliminate the cleaning procedure and its contamination risks. The second purpose of the work was to test the possibility of long-term storage of samples at their natural pH (about 8) at 4 °C, kept in low-density polyethylene containers which have been cleaned with acid and conditioned with seawater. 

In a related case, FT-30 membrane elements were placed on chlorinated seawater feed at OWRT s Wrightsville Beach Test Facility. Flux and salt rejection were stable for 2000 hours at 0.5 to 1.0 ppm chlorine exposure. Chlorine attack did become noticeable after 2000 hours, and salt rejection had dropped to 97 percent at 2500 hours while flux increased significantly. Long term laboratory trials at different chlorine levels led to the conclusion that the membrane will withstand 0.2 ppm chlorine in sodium chloride solutions at pH 7 for more than a year of continuous exposure. 

A group of saline (2.5-14.5g Cl/1) and warm (28°C-40°C) springs and wells in the eastern part of the study area were suspected by earlier workers to contain a seawater component. This was tested and confirmed by Cl-<5180 relations, and the percentage of seawater intermixed in each case could be calculated (Fig. 9.27). 

The precision of the technique for seawater analysis as presented in the literature (i, 5) tends to be considerably better than we have observed here. The values obtained in other papers were for duplicate analysis of the same sample and were most likely extracted sequentially from the same bulk sample and analyzed one directly after the other. This was not the case here because the data analyzed in this paper were not generated specifically to analyze the ultimate precision of the technique. Line water samples run normally were as a rule interspersed throughout the test samples. A number of water samples would be drawn at the start of an experiment and stored unacidified in 4-1. polypropylene bottles. Over the course of up to 6 or 8 hr, extractions would be performed so that difiFerences in trace metal concentration might be expected between replicates run early and late in the experiments. This factor, which allows for significant adsorption and/or desorption of trace components, could readily explain our high standard deviations. We feel that this approach is valid to determine the precision of the technique in the field where non-optimum conditions often occur and where the factor of time between sampling and analysis is often an uncontrollable variable. It is likely that the actual precision of this technique in the field lies between those values calculated here and elsewhere (1,5). 

A sulfonated polysulfone membrane with commercial potential has been developed in the form of a hollow fiber by workers at Albany International Corporation.86 87 In a 5,000-hour test on 3,500 ppm brackish water at 400 psi, this membrane exhibited 98% salt rejection at 1 gfd. Flux and salt rejection remained constant even with addition of 100 ppm chlorine. In a 12,000-hour test on seawater at the Wrightsville Beach Test Facility, this membrane exhibited 98+% salt rejection and an average flux of 1.5 gfd at 1,000 psi operating pressure. Thus, it is possible to make a high rejection membrane from sulfonated polysulfone. Flux was rather low in this particular case, but suitable for hollow fiber membrane use. 

---