| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a O.S. Qafoku, Pacific Northwest National Lab., 902 Battelle Blvd., P.O. Box 999, MSIN: K3-61, Interfacial Geochemistry Group, Richland, WA 99352
b Los Alamos National Lab., MS D469 EES-1: Geology and Geochemistry, Los Alamos, NM 87545
* Corresponding author (nik.qafoku{at}pnl.gov).
Received for publication January 3, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Abbreviations: DOE, Department of Energy • EDS, energy dispersive spectroscopy • EDXRF, energy dispersive X-ray fluorescence spectroscopy • FESEM, Field Emission Scanning Electron Microscopy • HLWF, high-level radioactive waste fluids • IS, ionic strength • QXRD, quantitative X-ray diffraction • SEM, scanning electron microscopy • XRD, X-ray diffraction
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
It is quite possible that, in addition to dissolution, precipitation of secondary phases may also occur when HLWF contacts the sediment. Both partial dissolution of the soil minerals present in the sediments and precipitation of secondary phases may result in a substantial alteration of the surface adsorption properties of the soil minerals that were initially present in the sediments. This should affect the ultimate fate of radionuclides and metal contaminants that are present in HLWF. In addition, different contaminants may also coprecipitate in the secondary phases that may possibly form in these systems. The extent of precipitation reactions in the Hanford sediment under extreme conditions of basic pH (
14) and high ionic strength (IS) is not known. The nature of secondary phases that may be formed is also not known and the presence of large amounts of Al and other chemical species in HLWF, like NO3, may also play a role in determining the type and morphology of the secondary precipitates that may form.
Precipitation reactions were investigated quite intensively during the process of producing gibbsite, in which Al-rich minerals in bauxite, as well as quartz and kaolin were dissolved in hot sodium hydroxide solutions (Zheng et al., 1997; Gerson and Zheng, 1997; Barnes et al., 1999a,b). These studies have shown that sodium aluminosilicates in the groups of zeolite, sodalite, and cancrinite were formed as a result of precipitation. Solution-mediated transformations of solid phases at 363 K followed the sequence: Aluminosilicate species 
Amorphours phase Zeolite (Linde A) Sodalite Carbonate–Cancrinite. Barnes et al. (1999a) found that the amorphous phase precipitated after 1 min. This precipitate transformed to zeolite after 2 h, which transformed to sodalite after approximately 2 d and ultimately to carbonate–cancrinite after 21 d. Bickmore et al. (2001) conducted experiments with quartz treated with simulated tank waste solutions, and they found that cancrinite precipitated onto quartz particles after 2 to 10 d.
The minerals in the sodalite and cancrinite groups are feldspathoids with a similar porous framework structure to zeolite (Hassan, 1997), but sodalite has a cubic crystal structure compared with cancrinite, which has an hexagonal crystal structure (Barnes et al., 1999a). Although the precipitated zeolites are likely to have different chemical formulae, sodalite and cancrinite, which are formed in Si, and aluminate-rich caustic systems have identical chemical stoichiometry (Na8Al6Si6O24X2), where X can be 1/2CO3, 1/2SO4, Cl, OH, or NO3 (Barnes et al., 1999a). The formation of nitrate–sodalite (Buhl and Lons, 1996) and nitrate–cancrinite (Bickmore et al., 2001) is possible in NO3–rich systems.
The transformation of the initial solid phase because of dissolution and precipitation under hyperalkaline and hypersaline conditions is an important area of research that warrants further investigation. To understand the extent and dynamics of dissolution and precipitation in these systems, batch reactors were used to investigate interactions between Hanford sediments and Al-rich, hyperalkaline, and saline solutions similar to some tank waste fluids. The morphology of the Al-bearing secondary phases was studied for different initial Al concentrations in the contact solution and different Si/Al aqueous concentration ratios created in the soil solution during the experiments because of dissolution and precipitation reactions. The solid phase transformations that occurred in the Hanford sediments when they were contacted with Al-rich, newly formed aluminosilicates (feldspathoids and other minerals) were characterized using different techniques.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Batch Experiments with Hanford Sediments
The full description of how the batch experiments were conducted is given in the previous paper (Qafoku et al., 2003). Briefly, we conducted two similar experiments with the same solutions. The first experiment had 20 treatments: 4 Al concentrations (0.055, 0.11, 0.165, and 0.22 mol L-1) by 5 time periods (3, 7, 14, 21, and 42 d), and the second one had 36 treatments: 4 Al concentrations by 9 time periods (20 min, 1, 2, 4, 8, 16, 24, 36, and 48 h). The background solution was 1 mol L-1 NaOH and NaNO3. Both these experiments are hereafter referred to as the Batch 1 study. Because of the coarse nature of these sediments (the fine fraction makes only 1–2% of the total sediment's mass) another similar study was conducted during this investigation to produce the necessary amount of posttreatment sediments needed for different analyses of the solid phases. A treatment with high concentrations of base and Al (4 mol L-1 NaOH, 0.88 mol L-1 Al, and 1 mol L-1 NaNO3) was also included in this latter experiment. The postexperiment analyses of the sediments were conducted after 3, 7, 14, 28, and 42 d of the experiment. These experiments were run inside a thermostat chamber where the temperature was kept at 323 K for the duration of the experiment. The sediments of all treatments were washed twice with DI-water over Buchner-funnel–like containers to wash out the excess salt, and were subsequently air-dried and used in postexperiment analyses.
Characterization of Precipitates
The air-dried reacted sediments were analyzed by powder XRD, ultra-high performance field emission scanning electron microscopy (FESEM), energy-dispersive X-ray spectroscopy (EDS), and energy dispersive X-ray fluorescence (EDXRF). The X-ray diffraction measurements of the sediments at the end of the Batch 1 experiments were performed using a Philips X'Pert MPD system that employs a PW3050/10 vertical theta goniometer and a PW3373 ceramic X-ray tube, with a Cu X-ray source operated at 45 kV and 40 mA (1.8 kW). Powdered specimens from the <53 µm fraction were mounted in cavity-type sample holders and measured across 2° to 60° 2
. The scan rate was 0.02°/2s, except for the untreated sediment's sample and the 21-d sample were scanned at 0.02°/5s and 0.02°/20s, respectively. The slower scan rate of 0.02°/5s was utilized for the untreated sample to firmly establish the baseline composition of the soil. The very slow scan rate of 0.02°/20s was utilized for the sample collected after 21 d of the experiment to increase the chances of observing the characteristic diffraction patterns of cancrinite and possibly sodalite. Data were analyzed using Jade (Materials Data, Livermore, CA) for matches with the Powder Diffraction File database (PDF-2, 2001 Release, International Centre for Diffraction Data, Newtown Square, PA).
Mineral abundances were determined by quantitative powder X-ray diffraction (QXRD) methods with a Siemens D-500 diffractometer using Cu–K
radiation and a Kevex PSi(Li) solid state detector. Data were obtained on samples that were ground to <5 µm in an automatic Brinkman Micro-Rapid agate mill. As only small amounts of sample were available, samples were not mixed with an internal intensity standard, such as corundum. The quantitative mineralogy was determined on a weight basis using the external-standard method of Chung (1974). The accuracy of the QXRD method, particularly for feldspars and zeolites, was improved by using multireflection rather than single-reflection methods. Typical two-sigma uncertainties in analyses of feldspar are ±3% (absolute) at 10% concentration and ±10% at 50% concentration; for zeolites, the uncertainties are ±1% at 10%, ±3% at 40%, and ±6% at 80%. The QXRD methods have been coded into a computer program (QUANT) that accounts for analytical problems important in the analysis of tuffs, altered tuffs, and soils: e.g., correction of clinoptilolite and opal-CT for mutual overlap and use of multiple reflections to improve results for analysis of complex tectosilicates. Further details of the basic method can be found in Bish and Chipera (1987) and the methods of multireflection analysis are described in Chipera and Bish (1995). A single mica standard was used for quantitative analyses, although SEM data show the existence of at least three different micas. The powder XRD data on nitrate–cancrinite were taken from the literature (Buhl et al., 2000).
A LEO 982 FESEM system with an Oxford ISIS energy dispersive X-ray microanalysis system, and with a SiLi detector was used for scanning electron microscopy and EDS analyses. This instrument is an ultra-high performance scanning electron microscope with a resolution of 1 nm at 30 kV and 4 nm at 1 kV. We used the microanalysis system to do both qualitative and quantitative elemental analyses of the secondary minerals that were formed during the experiment, and to identify the soil minerals onto which these secondary phases precipitated.
The EDXRF, which is a rapid and sensitive method for simultaneous analysis of up to 49 elements in solid samples, was used to determine the total amount of different elements in the sediments at the end of the experiment.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
(15 Å) disappeared in the latter pattern, which confirmed that smectite underwent dissolution in the first 3 d of the experiment, serving as an initial source for Si and possibly Fe released into the soil solution. This was also confirmed from the QXRD results conducted with the sediments treated with the 1 mol L-1 NaOH, 0.22 mol L-1 Al, and 1 mol L-1 NaNO3 solution. The most important change determined with QXRD was the significant decrease in the amount of smectite in the sample taken after 7 d of contact time (11.4%) as compared with the untreated sample (18.4%). The biotite content also decreased from 15.5% in the untreated sample to 11.8% in treated sample. It also appears that reaction changed the interlayer chemistry of smectite, collapsing some of the interlayers. Clinochlore was much more resistant to weathering and the characteristic maxima of clinochlore persisted in all the patterns. While other minerals that are present in the clay- and silt-size fraction may have also undergone dissolution, the diagnostic peaks of biotite appeared less intense in the treated materials as compared with untreated ones (Fig. 2). The scanning electron microscopy (SEM) micrographs that will be presented later in this paper showed clear signs of the OH attack on mica edges and the mica sheets were separated from one another (Fig. 5j).
|
), 3.6589 (24.31°2), and 3.2393 Å (27.51°2) are clearly more intense in the posttreatment sample. It should be added here, however, that some of the characteristic peaks of nitrate–cancrinite were missing or were less intense than the peaks reported for NO3–cancrinite, e.g., the peaks at 4.6906 Å (18.90°2), 2.7439 Å (32.61°2), and 2.4255 Å (37.03°2) (Buhl et al., 2000).
|
|
The results from the qualitative EDS analyses showed that almost all newly formed solid phases had similar chemical compositions; they were rich in Si, Al, O, and Na, and had Si/Al molar ratios of about 1. The EDS quantitative analysis confirmed that the secondary minerals were mostly formed on the surfaces of plagioclase with an average chemical composition of 72.67% Si, 15.91% Al, 6.22% Ca, 3.42% Na, 1.755% K, and 0.01% Fe. The presence of a Fe-rich precipitate with a smaller, ball-like structure (diam. < 500 nm) was also visible in Fig. 5g, but we were unable to confirm its identity with the EDS analysis because these particles were small.
When the sediments where initially treated with a 0.11 mol L-1 Al alkaline solution (Batch 1 study) the cancrinite balls were bigger and it appears that the cotton-ball structures of probably sodalite were also present in the sediments of this treatment after 42 d (Fig. 5h). The Fe-rich small balls were quite visible on the surfaces of the mica particles of this treatment (Fig. 5i). Because less Si was released on dissolution, much less cancrinite was formed in the 0.165 and 0.22 mol L-1 Al treatments after 42 d, as it is clearly shown in Fig. 5j and 5k. The cotton-ball structures were also present in these treatments, and a third mineral with a platy morphology, which could be either zeophyllite [Ca4Si3O8(OH, F)4 x 2H2O] or gibbsite, appears to be much more abundant in the 0.22 mol L-1 Al treatment (Fig. 5k) compared with other ones that had less Al in the initial contact solution.
Results from the Energy Dispersive X-ray Fluorescence Spectroscopy Analysis
The total Si content in the sediments after 42 d of contact time increased with Al concentration from 27.9 to 31.4% (Table 1). This is consistent with the results presented in the previous paper (Qafoku et al., 2003), where it was shown that Si release from the sediments was an inverse function of Al concentration in the soil solution. In the 0.055 mol L-1 Al treatment, some of the Si release on dissolution remained in the soil solution at the end of the experiment, because Al was almost depleted from the soil solution after 42 d. In addition, though the precipitation of alumino-silicate secondary phases enriched the solid phase with Al, resulting in an increase in the total Al percentage in the solid phase (Table 1), this also contributed to the decrease in the total Si percentage in the solids as a result of the dilution phenomenon. The lowest Fe, K, and Ca contents were found in the sediments that were treated with the 0.055 mol L-1 Al solution where the dissolution and precipitation were more intense, and their contents also changed slightly during the experiment.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Even though clearly seen in the SEM micrographs, the presence of cancrinite and possibly sodalite is not easily detected by XRD and not detected at all by QXRD, perhaps because their contents remain low (they represent a small mass fraction of the total sediment mass), and the detection limits for the QRXD analysis, given the above analysis conditions, are on the order of 1%. Also many diffraction peaks occur at the same d-spacing as other minerals present in much higher quantities, e.g., feldspars. Besides the patterns reported by Buhl et al. (2000) for cancrinite formed at a different temperature (473 K) from that in our experiments (323 K), we are not aware of other studies documented in the literature that would have been used to compare our results.
The SEM micrographs confirm the presence of secondary alumino-silicate coatings in the sediments by no later than Day 3 of the study. It is likely that nitrate–sodalite that is formed quite early in the experiments, i.e., in the first and second day, is then probably transformed to nitrate–cancrinite via a solution mediated phase transformation reaction, i.e., sodalite undergoes dissolution and reprecipitates in the form of nitrate–cancrinite (Barnes et al., 1999a). This is probably the reason why sodalite is not observed in the samples collected after 3 d in the 0.055 mol L-1 Al treatments. It appears that in other treatments sodalite does not undergo dissolution, probably because of the higher Al concentrations in the contact solutions of these treatments, which is shown in the previous paper to inhibit dissolution by decreasing the free OH concentration in the contact solution and by inhibition (Qafoku et al., 2003).
The aqueous Si/Al molar ratio appears to have affected significantly the morphology of the newly formed alumino-silicates. This ratio remains in the range 0.096 to 0.284 during the time period of 21 d in the Batch 1 study, which means that Al is present in much greater concentrations than Si. During this time nitrate–cancrinite's balls are quite visible as coatings on the sediments' surfaces. Barnes et al. (1999a) and Bickmore et al. (2001) conducted experiments with the Si/Al molar ratio of 0.0599 and 0.7875, respectively, and reported similar results. The situation changes after 21 d in our study, and the aqueous molar Si/Al ratio reaches the highest value of 14.17 after 42 d, as it is discussed in the previous paper (Qafoku et al., 2003). This change in the Si and Al composition of the soil solution appears to have drastically affected the morphology of the newly formed precipitates, from the ball-like to the rod-like structures (Fig. 4f and g). It is likely that the rod-like precipitates that are formed after 21 d cover the nitrate–cancrinite balls that are formed before 21 d because the balls are still visible.
Data from these experiments were modeled using GMIN, which is a chemical equilibrium program that uses a free-energy minimization approach (Felmy, 1995). The free-energy models incorporated in GMIN include the Pitzer equations that may be successfully used to model experimental data collected from experiments conducted with highly concentrated solutions. The aqueous phase of the low Al treatments (which had 0.055 and 0.11 mol L-1 Al in the contact solutions) was oversaturated with respect to cancrinite. This situation was different only in the treatment that had a 0.055 mol L-1 Al contact solution. After 42 d, the solution was undersaturated with respect to cancrinite in this treatment. The aqueous phase of the other treatments (which had 0.165 and 0.22 mol L-1 Al in the contact solutions) was oversaturated with respect to both cancrinite and gibbsite.
The presence of these secondary minerals in the after treatment sediments may have a significant effect in the immobilization and ultimate fate of radionuclides and contaminants, like Cs, Sr, and U, which are present in the vadose zone sediments at the Hanford Site. These minerals possess higher specific surface area than the minerals that are initially present in the Hanford sediments. Sodalite and cancrinite are both feldspathoids with structures similar to zeolite. The structures of the minerals in the cancrinite group are characterized by stacking of six-membered rings, which gives rise to large continuous channels that are amongst the largest channels known to occur in the mineral structures (Hassan, 1997). These are also called reservoir minerals because of their large cation exchange capacity (Buhl and Lons, 1996) and surface sorption capacities. In addition to providing the mineral with a higher specific surface area, the cages host a complex distribution of extra-framework cations and anions (Ballirano et al., 2000). It is possible, therefore, that contaminants and radionuclides may take part in the structure of these secondary phases when they are formed in the sediments. As an example, Se has been introduced into the cancrinite matrix by hydrothermal methods to form red-brown hexagonal crystals (Lindner et al., 1996).
The morphology of the newly formed solid phases appears to be a function of Si/Al molar ratio in the soil solution and microzones with different ratios probably occur in the sediments. It is quite likely, therefore, that the formation of morphologically different secondary phases will increase the physical heterogeneity in the Hanford sediments, which in turn may significantly affect contaminant mobility.
The presence of Fe precipitates is also suggested from the QXRD results and SEM micrographs, which indicates that Fe precipitates alone in pure Fe secondary phases. The relatively small balls of Fe-rich precipitates (diam. approximately 300–500 nm) are abundant on mica surfaces (Fig. 4j). Quite possibly Fe(II) is released from micas on their dissolution, and subsequently is oxidized to Fe(III) by residual O2 or any other oxidant, and then precipitates immediately under hyperalkaline conditions.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Related articles in JEQ:
This article has been cited by other articles:
|
|
 |
|
  K. Mashal, J. B. Harsh, and M. Flury Clay Mineralogical Transformations over Time in Hanford Sediments Reacted with Simulated Tank Waste Soil Sci. Soc. Am. J., March 1, 2005; 69(2): 531 - 538. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. P. Qafoku, C. C. Ainsworth, J. E. Szecsody, and O. S. Qafoku Aluminum Effect on Dissolution and Precipitation under Hyperalkaline Conditions: I. Liquid Phase Transformations J. Environ. Qual., November 1, 2003; 32(6): 2354 - 2363. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Soil Science Society of America Journal | Journal of Plant Registrations | The Plant Genome | |||
