Devices & Hardware

A natural analogue of high-pH cement pore waters from the Maqarin area of northern Jordan: Comparison of predicted and observed trace-element chemistry of uranium and selenium

Description
A natural analogue of high-pH cement pore waters from the Maqarin area of northern Jordan: Comparison of predicted and observed trace-element chemistry of uranium and selenium
Published
of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  JOURNAL OF Contaminant Hydrology zyxwvutsr ELSEVIER Journal of Contaminant Hydrology 21 1996) 59-69 A natural analogue of high-pH cement pore waters from the Maqarin area of northern Jordan: Comparison of predicted and observed trace-element chemistry of uranium and selenium C.M. Linklater a, * Y. Albinsson b, W.R. Alexander ‘, I. Casas d, I.G. McKinley e, P. Sellin f a AEA Technology Decommissioning and Waste Management Hawell UK ’ Chalmers University of Technology Gothenburg weden ’ RWIG University of Berne Berne Switzerland ’ MBT Tecnologia Ambiental Barcelona Spain e Nagra Wettingen Switzerland ’ SKB, tockholm Sweden Received 17 December 1993; accepted 15 December 1994 after revision Abstract Current design concepts for low-/intermediate-level radioactive waste disposal in many countries involve emplacement underground in a cementitious repository. The highly alkaline groundwaters at Maqarin, Jordan, are a good analogue for the cementitious pore waters that will be present within such a repository. A geochemical modelling study of these groundwaters has been carried out in order to test the applicability of equilibrium models in geochemical programs and their associated thermodynamic databases in such hyperalkaline conditions. This was achieved by comparison of elemental solubilities and speciations predicted by the programs with observa- tions in the natural system. Five organisations took part in the study: ABA Technology, U.K.; Chalmers University of Technology, Sweden; MBT Tecnologia Ambiental, Spain; Nagra, Switzer- land; and SKB, Sweden. The modelling study was coordinated by the University of Berne. The results of the study showed good agreement between the predictions of the programs employed. Comparison of the observed solids with those predicted by the models has allowed limited validation of the databases. The results for U and Se are presented here. * Corresponding author. Elsevier Science B.V. SSDI 0166-3542 95)00033-X  60 C.M. Linklater et al. /Journal of Contaminant Hydrology 21 1996) 59-69 1. Introduction In the performance assessment of a radioactive waste repository, geochemical programs may be employed to contribute to an understanding of the expected behaviour of the radionuclides of interest in both the near field and the far field of the repository. An important constraint on the rate of release of many radionuclides present in radioactive waste is set by their very low solubility under the expected chemical conditions within the repository. The definition of solubility limits for performance assessment involves integration of field, laboratory and theoretical information. An important tool for selecting quantitative values for a performance assessment database is geochemical thermodynamic modelling. For a defined water composition, geochemical codes are used to calculate the solubility of a range of potential solubility limiting phases. In using such information to select solubilities for performance assess- ment, uncertainties must be evaluated in: 1) the reference aqueous conditions; 2) the thermodynamic database used; 31 the assumption that the system is in thermodynamic equilibrium. The thermodynamic databases upon which the programs depend are normally com- piled from data produced under controlled laboratory experiments. Extensive databases are available for most major elements, but less work has been reported for radionuclides of interest to a repository performance assessment, especially under the expected hyperalkaline conditions of a cementitious repository. The validity of a database may be uncertain for two reasons, the presence of inappropriate data and the omission of important data. Some problems associated with a database can be identified by intercom- parison of results of simulations run with different databases. Clearly, such tests do not identify errors in data which are common to the databases or data missing from all databases examined. A more rigorous test, which contributes to model validation, is the comparison of predictions with laboratory or field data. Laboratory experiments can be relatively well-defined but are frequently simplified significantly from the reference system of interest, and are inherently limited by the slow rate of equilibration of reactions involving the precipitation of mineral phases. Field data, either measurements of concentrations in natural waters or specific natural analogue studies, may represent systems which have reacted for much longer periods. However, the chemical conditions are generally less well-defined Bath et al., 1987; Read, 1990; Bruno et al., 1991,1992). The present study comprises Phase 2 of a geochemical modelling exercise forming part of a larger study of a natural analogue site at Maqarin, northern Jordan. Fig. 1 shows a map of the area. The site consists of intercalated bituminous limestone and marl which have been exposed to high-temperature metamorphism causing calcination of the limestone Khoury et al., 1992). The subsequent water-rock interactions lead to the circulation of highly alkaline groundwaters [chiefly due to buffering by portlandite, Ca OH12], and the formation of hydrated calcium sulphate minerals such as ettringite and thaumasite. The site forms a promising natural analogue for the hydration of cementitious materials which might form part of a radioactive waste repository. Many of the minerals contain concentrations of trace elements which can be measured by    I I I 1 I I I E2.29 E231 E235 SYRIA - N237 , i zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA  i \ i ; 5 i. i E A AnmA 3 .-_. / -._,_/ , a- Iy pf---q~ Fig. 1. Map of the Maqarin area, northern Jordan, showing the locations of the sampling sites. I = Eastern Springs, adit A6 (sample Ml); 2 = Eastern Springs, railway cutting (sample M3); 3 = Western Springs (sample M5). analytical techniques. The site therefore gives an ideal opportunity to study factors controlling the solubilities of these trace elements in a high-pH, cementitious system. Detailed hydrogeochemical characterisation of the site has been carried out e.g., Khoury et al., 1992; Milodowski et al., 1993). General geochemical modelling carried out during Phase 1 has been reported previously Alexander et al., 1992). The Phase-2 programme was focused more strongly towards investigation of the interaction of the hyperalkaline solutions and the marl. The present geochemical modelling exercise therefore benefits from the increased availability of detailed and relevant field data. 2. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA pproach to modelling of trace element chemistry Five organisations took part in the study: AEA Technology, U.K. AEA); Chalmers University, Sweden CTH); MBT Tecnologia, Spain MBT); Nagra, Switzerland Nagra);  62 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA .M. Linklater et al. /Journal of Contaminant Hydrology 21 1996) 59-69 Table 1 Details of the programs and associated databases used by the participating organisations Organisation Programme Database AEA Technology (AEA) HARPHRQ (Brown et al., 1990) HATCHES (5.0) a Chalmers University (CTH) PHREEQE (Parkhurst et al., 1980) HATCHES (2.0) a MBT (Barcelona) PHREEQE (Parkhurst et al., 1980) HATCHES (3.0) a Nagra RIPP2s b (Pate et al., 1993) NTB9118 (Pearson and Berner, 1991; Pearson et al., 1992) SKB EQ3/6(0288) SKB (Livermore Stage 47) a HATCHES is a thermodynamic database compiled by AEA Technology (Cross and Ewart, 1991) and released annually through the NEA. Versions 2.0, 3.0 and 5.0 correspond to the versions of the database released in 1989, 1990 and 1992, respectively. ’ A user friendly overlay for PHREEQE. and SKB, Sweden SIB). Details of the thermodynamic databases used by these organisations, along with the geochemical programs used, are given in Table 1. Table 2 Chemistry of the three waters used in the test cases: Ml, M3 (eastern springs) and M5 (western springs) Sample Temperature (“C) pH (field) pH (laboratory) Eh (field) (mV) Ml M3 M5 24.8 23.2 25.2 12.74 12.66 12.92 12.67 12.76 12.83 + 278 + 150 + 242 Major elements (mg L- ’ ): Ca Na K Cl SO, NO3 674 804 1,120 47.2 46.6 136 9.88 19.8 526 52.4 72.3 46.6 305 289 1,580 3.28 7.73 39.1 Trace elements (mg L- ‘): Mg NH, NO, Fe (total) Al Si 0.01 0.01 0.01 < 0.10 0.13 6.05 < 0.10 < 0.10 1.02 < 0.01 < 0.01 < 0.01 0.14 0.15 0.14 < 0.02 0.09 0.07 Others: TIC 4.83 5.51 5.97 TOC 3.20 1.35 6.38 TIC = total inorganic carbon; TOC = total organic carbon  CM Linklater et al. /Journal of Contaminant Hydrology 21 1996) 59-69 63 Table 3 Concentrations of the two elements of interest Element Ml M3 M5 U(ngL-‘) Se(pgL-‘1 2+1 16+1 37+2 114*1 206f3 1,315 f 21 The approach adopted in the work can be summarised as follows: 1) Groundwater compositions were specified, based on measured data, Table 2). The modellers were allowed to comment on the data e.g., with respect to charge balance, degree of saturation with respect to major minerals) but were required to carry out evaluation without alterations. The elements included in the exercise were U, Th, Ra, Se, Ni, Pb and Sn. As an illustration, only the results for U and Se are described here. 2) Modelling was carried out, based on assumed trace-element concentrations of 1 pg L-‘. Elemental speciation was predicted and the solubility limiting phase was identified. 3) Comparison of results was carried out to determine major database discrepancies. This procedure was a simple intercomparison between the databases used a type of verification). 4) Field concentrations of the trace elements of interest obtained by inductively- coupled plasma-optical emission spectroscopy and mass spectroscopy) were distributed to the groups Table 3). 5) A second phase of modelling was carried out, again of elemental speciation, solubility and selected appropriate solubility controlling solids. 6) Finally, the results were compared with highly detailed analytical data on the site mineralogy and approximate groundwater speciation separation of anionic, cationic and non-ionic, or neutral components). Anomalies between the code predictions and the analytical data provide guidance on the applicability of the models and databases to a cementitious repository environment and may additionally indicate areas for database improvement. 3. Results 3.1. Uranium The uranium results are given in Table 4. There are no great differences in predicted speciation between the different versions of the HATCHES databases, with UO, OH)i- dominating in all cases. Interestingly, the Nagra database indicates a prevalence of UO, OH), in all three waters, reflecting the different source of the uranium data in this database. All groups predict the existence of anionic species only. Field measurements are unavailable as the analytical method used was unable to characterise uranium speciation at such low concentrations. UO, OH), is omitted from the HATCHES databases as calculations with this species present have been found to predict high UO, OH), solubility at high pH-values. These predictions are inconsistent with labora-
Search
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks