Aqueous 99Tc, 129I and 137Cs removal from

0 downloads 0 Views 451KB Size Report
Jun 4, 2014 - Apak, R., Atun, G., Guclu, K., Tutem, E., 1996. Sorptive removal of cesium-137 and · strontium-90 from water by unconventional sorbents. 2.
Journal of Environmental Radioactivity 136 (2014) 56e63

Contents lists available at ScienceDirect

Journal of Environmental Radioactivity journal homepage: www.elsevier.com/locate/jenvrad

Aqueous 99Tc, 129I and 137Cs removal from contaminated groundwater and sediments using highly effective low-cost sorbents Dien Li*, Daniel I. Kaplan, Anna S. Knox, Kimberly P. Crapse, David P. Diprete Savannah River National Laboratory, Aiken, SC 29808, United States

a r t i c l e i n f o

a b s t r a c t

Article history: Received 27 January 2014 Received in revised form 13 May 2014 Accepted 14 May 2014 Available online 4 June 2014

Technetium-99 (99Tc), iodine-129 (129I), and cesium-137 (137Cs) are among the key risk-drivers for  environmental cleanup. Immobilizing these radionuclides, especially TcO 4 and I , has been challenging.  TcO 4 and I bind very weakly to most sediments, such that distribution coefficients (Kd values; radionuclide concentration ratio of solids to liquids) are typically 1  10 mL/g), I (Kd  1  10 mL/g), and Cs (Kd > 1  10 mL/g) and also demonstrated a largely irreversible binding of the radionuclides. Activated carbon GAC 830 was effective 5  3 at sorbing TcO 4 (Kd > 1  10 mL/g) and I (Kd ¼ 6.9  10 mL/g), while a surfactant modified chabazite 4 þ 3 was effective at sorbing TcO 4 (Kd > 2.5  10 mL/g) and Cs (Kd > 6.5  10 mL/g). Several sorbents were 5 K effective for only one radionuclide, e.g., modified zeolite Y had TcO 4 d > 2.3  10 mL/g, AgS had I Kd ¼ 2.5  104 mL/g, and illite, chabazite, surfactant modified clinoptilolite, and thiol-SAMMS had Csþ Kd > 103 mL/g. These low-cost and high capacity sorbents may provide a sustainable solution for environmental remediation. © 2014 Published by Elsevier Ltd.

Keywords: Technetium Iodine Cesium Organoclays Sorbents

1. Introduction Technetium-99 and iodine-129 are two of the three most common risk drivers (along with 14C) in low-level and high-level waste disposal sites and among the most common environmental contaminants at DOE sites. The worldwide inventory of Tc- and Ibearing nuclear wastes continues to increase rapidly due to the demand for more electricity and the need for nuclear power as an alternative energy source that emits less CO2 than fossil fuels. 99Tc and 129I can adversely enter groundwater and sediments from mismanaged wastes or through leakage from waste or spent fuel storage facilities. Technetium-99 is a long-lived radionuclide contaminant (t1/ 2 ¼ 0.22 million years) and very mobile in groundwater due to its existence dominantly as the anionic species, TcO 4 (Icenhower et al., 2010). The TcO 4 is not immobilized by most common minerals or inorganic sorbents because it is repulsed by their negative charge (Liang et al., 1996). The fate and mobility of 129I in environmental systems is especially complex due to its existence in several oxidation states (commonly as 1 (I), þ5 (IO 3 ), and 0 (organo-I)) * Corresponding author. Tel.: þ1 10 8037257520; fax: þ1 10 8037254704. E-mail address: [email protected] (D. Li). http://dx.doi.org/10.1016/j.jenvrad.2014.05.010 0265-931X/© 2014 Published by Elsevier Ltd.

and its tendency to form strong covalent bonds with natural organic matter (Kaplan et al., 2014). Adsorption behaviors of IO 3 and I onto sediments and some oxide/sulfide minerals can be quite different, and the IO 3 sorption is normally greater than I (Kaplan et al., 2014). Cesium-137 is a major radionuclide in spent nuclear fuel reprocessing, primarily due to its high fission yield. Radioactive Cs isotopes are commonly the most important risk drivers immediately after a nuclear accident, such as those that occurred at Chernobyl and Fukushima (Whicker et al., 2007). Cesium forms few inorganic or organic complexes in natural systems due to its hydrated nature and large ionic radius. Various sorbents are presently being used to remove radiological and non-radiological inorganic contaminants out of groundwater (Cundy et al., 2008; Guo et al., 2006; Misaelides, 2011). Unlike for organic contaminants that can be chemically degraded or converted to less toxic compounds, the intent of inorganic contaminant sorbents is to reduce the mobility of the contaminants and to reduce their bioavailability/toxicity. Some of the more common modes of immobilization is reductive precipitation with various iron phases (Cundy et al., 2008), sorption into high surface area clays and porous zeolites (Misaelides, 2011), co-precipitation into less mobile phosphate or sulfur phases (Guo et al., 2006), and more recently, partitioning to surface modified clays or

D. Li et al. / Journal of Environmental Radioactivity 136 (2014) 56e63 Table 1 Selected materials for

99

129  TcO I and 4,

Csþ sorption evaluation.

137

Sorbents

Acronym

Description

Manufacturer

Activated carbon 824 BC

824 BC

Charcoal House™ Brand, Crawford, NE

Activated carbon GAC 830

GAC 830

Bone char 824 BC, Brimac Carbon, coarse granular bone charcoal (8  24 mesh) produced by carbonization of selected grades of animal bone Norit® GAC 830, granular activated carbon produced by steam activation of select grades of coal Fish bone, phosphate mineral A biopolymer derived from crustaceans shells Todd Light™ Illite, a natural clay mineral ClayFloc™ 750, a bentonite organoclay based flocculant impregnated with a quaternary amine Organoclay MRM™ clay impregnated with a sulfur-containing organic compound A natural sulfide mineral Self-Assembled Monolayers on Mesoporous Silica (SAMMS) modified with thiol functionalization Engineered Herschelite KUR-EH, a natural chabazite mineral Herschelite KUR-CH, a surfactant modified natural chabazite mineral Herschelite KUR-SMZ, a surfactant modified clinoptilolite mineral CBV-780, a synthetic zeolite

Apatite II Chitosan Illite Organoclay OCB

OCB

Organoclay OCM

OCM

Argentite (AgS) Thiol-SAMMS

Chabazite Surfactant modified chabazite Surfactant modified clinoptilolite Modified zeolite Y

57

SM chabazite

organoclays (Yariv and Cross, 2002). For example, zeolite-based sorbents have been used to remove Co, Ni, Se, Sb, Tc, I, Cs and Sr from aqueous media (Bonhomme et al., 2010; Denton and Bostick, 2011). Organoclay OCM with a sulfur-containing quaternary amine was patented for Hg and As removal from water (Wang and Abraham, 2011). Organoclay OCB has been used as a sequestering agent for active cap remediation of metals and organic contaminants (Knox et al., 2007). The objective of this study was to evaluate several low-cost 129  sorbents for their effectiveness to bind aqueous 99TcO I and 4, 137 þ Cs . Many of these sorbents have been used for other nonradioactive contaminants. Therefore the goal of this study was more specifically to evaluate whether these environmentally benign  þ sorbents could be extended to effectively sorb TcO 4 , I , and Cs . As such, the scope of this study was not to determine the maximum sorption capacity, sorption kinetics, or to elucidate the sorption mechanism. While these are critical criteria for full evaluation of sorbents, they are the subject of on-going experimentation. 2. Materials and methods 2.1. Sorbent materials 129  A variety of sorbents were evaluated for 99TcO I , and 137Csþ 4, removal from the aqueous phases (Table 1). Tested sorbent materials included: 1) activated carbons (824 BC and GAC 830), 2) apatite (fish bone), 3) biopolymer (chitosan), 4) clay mineral (illite), 5) organoclays (OCB and OCM), 6) silver sulfide (AgS), 7) thiol-functionalized self-assembled monolayer on mesoporous silica (thiol-SAMMS), and 8) zeolites (chabazite, surfactant modified chabazite, surfactant clinoptilolite, and modified zeolite Y). These sorbents were tested because of their low cost (except for thiol-SAMMS) and expected 129  high capacity to sorb 99TcO I , and/or 137Csþ from low-level 4, nuclear waste streams and contaminated groundwater.

2.2. Radionuclide chemicals The radionuclide stock solutions were purchased from Eckert & Ziegler Isotope Products (Valencia, CA). The as-received stock

Norit America, Inc. (Cabot Corp.), Marshall, TX

PIMS-NW, Inc., Richland, WA AIDP, Inc., City of Industry, CA KentuckyeTennessee Clay Company, Nashville, TN Biomin Inc., Ferndale, MI

CETCO® Remediation Technologies, Hoffman Estates, IL Ward's Science, Des Moines, IO Steward Environmental Solutions, Chattanooga, TN

Kurion, Inc., Oak Ridge, TN Kurion, Inc., Oak Ridge, TN Kurion, Inc., Oak Ridge, TN Zeolyst International, Valley Forge, PA

solutions were 1.85  104 kBq/mL NH4TcO4 in H2O, 7.4 kBq/mL NaI in 0.1 M NaOH, and 4.07  104 kBq/mL CsCl in 0.1 M HCl. The 99Tc stock solution was diluted with deionized water to create an 18.5 kBq/mL 99Tc spike solution. The 129I stock solution was diluted with 0.1 M NaOH to create an 1.85 kBq/mL 129I spike solution. The 137 Cs stock solution was diluted with 0.1 M HCl to create a 0.204 or 2.04 kBq/mL 137Cs spike solution. 2.3. Contaminated sediment To demonstrate the effectiveness of selected sorbents for the treatment of contaminated sediments, Tc-amended Savannah River Site (SRS) clayey sediment was selected to perform a proof-ofconcept experiment. The SRS sediment has the following properties (Kaplan, 2010): pH 4.55 (measured for a suspension of 50 g sediment in 50 mL deionized water), 57% sand, 41% silt, 2% clay (pipette method), 1.2% organic matter (CHOS Niko Analyzer), 1.1 cmol/kg cation exchange capacity (un-buffered NH4Cl exchange), 1.6 cmol/kg anion exchange capacity (un-buffered NH4Cl exchange), 15.3 m2/g surface area (BET analyzer), and a clay-size mineralogy composed primarily of kaolinite, goethite, hematite, and quartz (based on XRD). To prepare the 99Tc-amended sediment, 0.4 mL of 1.99  104 Bq/mL 99Tc solution was added to 120 mL deionized water, which was then added to 900 g of the air-dried SRS sediment described above. The moistened sediment was mixed by shaking in a double baggie with plenty of air space. The Tc-spiked sediment was then air dried for two weeks to promote Tc and sediment contact. The sediment had a final 99Tc concentration of 8.66 Bq/g, a concentration well within the range of values reported in contaminated sites on the SRS (Denham et al., 2004). 2.4. Artificial groundwater (AGW) An artificial groundwater (AGW) solution was used in the batch sorption studies. Its chemical composition is based on the average groundwater composition reported in a survey of 26 uncontaminated wells located on the SRS. The AGW had a pH of ~6.0, electrical conductivity of 0.026 mS/cm, turbidity of 1.17 ± 0.07 >1.12 ± 0.01 N/A 14 ± 0 0±5 >2.56 ± 0.11 247 ± 59 >2.35 ± 0.01 0.6e1.8

Batch for I and Cs spiking

 105

 105  105

 104  105

pH

I Kd (mL/g)

Cs Kd (mL/g)

9.6 7.4 N/A 6.3 N/A 10.4 3.5 N/A 3.0 6.4 5.7 4.9 N/A

12 ± 5 >6.92 ± 0.22  103 N/A 88 ± 1 N/A >9.61 ± 0.63  103 >2.93 ± 0.04  104 >2.5  104 7±3 0±2 122 ± 13 6±7 N/A 0.3e0.9

21 ± 13 192 ± 50 N/A 134 >1.17 ± 0.39 >2.80 ± 0.57 >1.23 ± 0.11 N/A >1.70 ± 0.16 >4.0  104 >6.59  103 >6.09  103 N/A 10e50

 104  103  103  103

N/A ¼ not available. a Experimental conditions included: 2e4 replicates, ambient temperature, 10 g/L sorbent in artificial groundwater, working solution concentrations of 185 Bq/mL 99TcO 4, 18.5 Bq/mL 129I, and 2.04 Bq/mL 137Csþ, 7 day contact time, phase separation by settling and 0.2 mm filter.

very effective for Tc (Kd 1  105 mL/g) and I removal (Kd 6.9  103 mL/g), but less effective for Cs removal (Kd 200 mL/g). Surfactant modified chabazite was effective for Tc (Kd > 2.5  104 mL/g) and Cs (Kd > 6.5  103 mL/g) removal, but much less effective for I removal (Kd 120 mL/g). Meanwhile, several sorbents were tested for sorption of all three radionuclides, but were only effective at taking up one of the radionuclides. For example, modified zeolite Y was effective for removing aqueous 5 TcO 4 only (Kd > 2.3  10 mL/g). Sorbents effective for removing aqueous Cs only were natural chabazite (Kd ¼ 4  104 mL/g), surfactant modified clinoptilolite (Kd ¼ 6.1  103 mL/g) and thiolSAMMS (Kd ¼ 1.7  103 mL/g) (Table 2). A ranking of the sorbents by Tc Kd values was as follows: modified zeolite Y (>2.35  105 mL/g) z activated carbon GAC 830 (>1.08  105 mL/g) z organoclay OCB (>1.17  105 mL/ g) z organoclay OCM (>1.12  105 mL/g) > surfactant modified chabazite (>2.56  104 mL/g). A ranking of the sorbents by their I Kd values are as follows: organoclay OCB (>2.93  104 mL/ g) > argentite (AgS; >2.5  104 mL/g) > organoclay OCM (>9.61  103 mL/g) > activated carbon GAC 830 (>6.92  103 mL/g). The most effective sorbents for Cs were organoclay OCB (>2.8  103 mL/g), organoclay OCM (>1.23  103 mL/g), illite (IMt2; >1.17  104 mL/g), thiol-SAMMS (>1.70  103 mL/g), chabazite (>4.0  104 mL/g) and surfactant modified chabazite (>6.59  103 mL/g). The pH values at the end of the equilibration period are also presented in Table 2. No attempt was made to control the pH values in these batch suspensions. Consequently, the reported pH reflected the steady state pH between the un-buffered pH 5.5 AGW and the sorbent. The advantage of this approach is that it permits evaluating the various sorbents without altering the systems with strong acids/bases or buffer solutions that would influence radionuclide, surface, and background electrolyte speciation. The disadvantage of this approach is that if there is a target environmental condition of interest (e.g., a target pH or Eh level), then additional tests are required using an amended system to better mimic those conditions. Given the very important role that pH plays in aqueous speciation, surface site functionality, and number of sorption sites, it is expected that different Kd values would be obtained under different pH conditions. Therefore, the most effective sorbent for each contaminant can be selected for the specific contaminated systems of known geochemical conditions. Some previously tested TcO 4 sorbents that have shown promise included activated carbon (Gu et al., 1996), stannous apatite (Moore

et al., 2003), and stibnite (Peretroukhine et al., 2006). Anion exchange resins with a quaternary amine (e.g., Dowex SRB-OH) have been used to remove Tc contaminant from high pH raffinate (Liang et al., 1996). However, the anion exchange resins had only a modest TcO 4 loading capacity in the presence of competing anions, such as nitrate. More recently, chemical reduction has been studied to reduce TcO 4 concentration using zero valent iron (Darab et al., 2007), aqueous Fe2þ (Zachara et al., 2007), Fe2þ-bearing minerals or sediments (Peretyazhko et al., 2012), hydrogen sulfide (Liu et al., 2007), and iron sulfide (Watson et al., 2001). Microbial reduction (Abdelouas et al., 2005) and photo-reduction of TcO 4 by nano-size metal oxides (Burton-Pye et al., 2011) have also been reported. However, even if TcO 4 is immobilized to a more reduced form, TcO2$1.6H2O, its solubility is about 1.0  108 M (629 Bq/L) in groundwater, which greatly exceeds the Environmental Protection Agency (EPA) maximum contaminant levels (MCL) of 5.3  1010 M Tc (33 Bq/L). Thus, some researchers have proposed technologies 4þ for reducing TcO into 4 and subsequently encapsulating the Tc mineral phases (Um et al., 2012). However, such technologies would be difficult to implement in the field because of the need for sequential chemical additions and mixing. The Tc Kd values of selected sorbents (e.g., activated carbon GAC 830, organoclays OCB and OCM, surfactant modified chabazite and modified zeolite Y) were compared to those of various soils and sorbents reported in the literature (Table 3). The sorbent Kd values are much greater than the Kd values of most soils (Thibault et al., 1990), which are typically 1.08 >1.17 >1.12 >2.56 >2.35

    

105 105 105 104 105

7.5 10.5 3.5 8.7 3.8

Sandy soil Silt soil Clay soil Organic soil Fe filings Resin Bio-Rad AG-OH Activated carbon Lockit clayb Modified Lockitb CTMA apatiteb SM chabaziteb

0.01e16 0.01e0.4 1.16e1.32 0.02e340 1.0  104 6.07  104 6.37  104 1.52  104 2.17  103

N/A N/A N/A N/A 6.2 8.9 6.5 N/A N/A N/A N/A

Thibault et al. (1990) Thibault et al. (1990) Thibault et al. (1990) Thibault et al. (1990) Liang et al. (1996) Bostick et al. (1995) Gu et al. (1996) Denton and Bostick (2011) Denton and Bostick (2011) Denton and Bostick (2011) Denton and Bostick (2011)

>6.92 >2.93 >9.61 >2.5

   

103 104 103 104

7.4 10.4 3.5 N/A

Sandy soil Silt soil Clay soil Organic soil Chalcocite Cinnabar

0.04e81 0.1e43 0.2e29 1.4e368 1.38  103 3.08  103

N/A N/A N/A N/A

Thibault et al. (1990) Thibault et al. (1990) Thibault et al. (1990) Thibault et al. (1990) Balsley et al. (1998) Balsley et al. (1998)

>2.80 >1.23 >1.17 >1.70 >4.0 >6.59

     

103 103 104 103 104 103

10.4 3.5 N/A 3.3 6.4 5.7

Illite Mica Kaolinite Vermiculite Chabazitec Clinoptilolitec KCCF-chabazitec KCCF-clinoptilolitec

6.3e8.3  103 3.2e3.4  103 240e290 5.7e6.7  103 1.5  103 570 1.26  104 0.94  104

6.5 6.5 6.5 6.5 N/A N/A N/A N/A

Hinton et al. (2006) Hinton et al. (2006) Hinton et al. (2006) Hinton et al. (2006) Denton and Bostick (2011) Denton and Bostick (2011) Denton and Bostick (2011) Denton and Bostick (2011)

129

I Activated carbon GAC 830 Organoclay OCB Organoclay OCM Argentite (AgS)

137

Cs Organoclay OCB Organoclay OCM Illite Thiol-SAMMS Chabazite SM chabazite

a b c

N/A ¼ not available. Initial Tc spike ¼ 195 Bq/mL, CTMA ¼ a proprietary surfactant. Initial Cs spike ¼ 6.6  105 mol/L, KCCF ¼ potassium hexacyanoferrate.

et al., 2011) have been evaluated for the removal of 129I. Another approach to removing groundwater iodine has been to add stable iodine (127I) and then sparge ozone to promote the oxidation of I to volatile elemental iodine I0 (Denham et al., 2004). A demonstration of a radioiodine groundwater remediation technology is presently underway (Denham et al., 2010). AgCl particles were injected to the flow path of a 129I plume to promote the in situ formation of the sparingly soluble precipitate, AgI (Denham et al., 2010). To date, 129I groundwater concentrations in one of the four monitoring wells decreased by 55%. The other three groundwater monitoring wells did not show a decrease in 129I concentrations, which may be attributed to: 1) poor well location with respect to the 129I plume, and 2) a majority (~66%) of the 129I existing as IO 3 or organo-I species that likely do not respond to this technology. Also demonstrated in Table 3, some of the tested sorbents (e.g., activated carbon GAC 830, organoclays OCB and OCM, and AgS) had much greater Kd values for iodine removal than comparable soil Kd values (Thibault et al., 1990) or previously evaluated I sorbents, including chalcocite and cinnabar (Balsley et al., 1998). The illite mineral with wedged-shaped edge sites has long been known to selectively and nearly irreversibly sorb Cs (Francis and Brinkley, 1976). Since then, other Cs-sorbent materials have been investigated, including various clays and their modified derivatives (Galambos et al., 2012; Zachara et al., 2002), zeolites (Borai et al., 2009), ferrocyanides-based sorbents (Kopyrin et al., 1999), Ti and Zr hydroxophosphates (Bortun et al., 1993), crystalline silicotitanate (Mann and Todd, 2004), layered metal sulfide K2xMnxSn3xS6 (Manos and Kanatzidis, 2009), and ammonium molybdophosphate composites (Todd et al., 2002). Other inexpensive materials, such as humic acids (Celebi et al., 2009) and

coal ash fly (Apak et al., 1996), have also been demonstrated for Cs sequestration. As shown in Table 3, while the Kd values of selected sorbents (e.g., organoclays OCB and OCM, illite, thiol-SAMMS, chabazite and surfactant modified chabazite) are quite large, they are about the same as those previously reported for illite, mica, vermiculite (Hinton et al., 2006), chabazite, potassium hexacyanoferrate modified chabazite and clinoptilolite (Denton and Bostick, 2011). 3.2. Sorption of Tc, I and Cs by mixed sorbents To evaluate the synergistic effects of some top-performing sorbents for Tc, I and Cs removal from aqueous media, a batch experiment was conducted with mixed sorbents (Fig. 1). No synergistic effects were observed. This may in part be attributed to the fact that several of the individual sorbents were able to lower the aqueous radionuclide concentrations to near detection limits, permitting little opportunity to observe significant additional sorption attributed to the second sorbent. For example, activated carbon 824 BC was not effective for Tc or I removal. When it was mixed with each of organoclay OCB (Fig. 1A), organoclay OCM (Fig. 1B), chabazite (Fig. 1C) or surfactant modified chabazite (Fig. 1D), the Kd value of the resulting mixture was equal to or smaller than that of the stronger single sorbent. Similarly, when activated carbon GAC 830 was mixed with these sorbents (Fig. 1EeH), the Kd value of the resulting mixture was equal to or smaller than that of the higher performance component for either Tc or I removal. As shown in Fig. 1D, H and J, the Kd value of some mixed sorbents for Cs removal is higher than the corresponding individual sorbent component. It is unknown what is responsible for this synergistic effect on Cs removal. However, the mixed

D. Li et al. / Journal of Environmental Radioactivity 136 (2014) 56e63

61

sorbents had a larger Cs spike concentration than the single sorbents. Additional testing is warranted to evaluate this.

This indicated that I species sorbed on organoclay OCB are readily leached under acidic conditions.

3.3. Desorption of Tc, I and Cs from sorbents

3.4. Treatment of TcO 4 -contaminated sediments

In most engineering applications, both sorption and desorption processes play important roles in the separation of radionuclides from a waste stream or a contaminated system. The percentages of Tc, I and Cs desorbed from organoclays OCB and OCM, and surfactant modified chabazite are shown in Fig. 2. Tc desorption from the organoclays OCB and OCM was