RESULTS AND DISCUSSION
A: RESIN PERFORMANCE - BREAKTHROUGH OF Tc-99
A small-scale field demonstration of one of our best synthetic resins (RO-02-119, trihexylamine/triethylamine, +40/-60 mesh) was conducted from July 17 to October 15, 1997, at the Northwest Plume Pump-and-Treat Facility at the PGDP site. Two columns approximately 1 in. in diameter by 4 in. in length, each with a bed volume (BV) of ~50 cm3, were set up to run in parallel at the treatment facility. One column contained our synthetic resin RO-02-119, and the other column contained commercially available Purolite A-520E resin (sieved to +40/-60 mesh). Each of these anion exchange resins was initially in chloride form. The two columns were configured for upward flow of groundwater in parallel so that we could determine the relative effectiveness of the two resins in removing TcO4- from the contaminated groundwater under field conditions. The contaminated groundwater was diverted through a ¸-in. pipe from the main treatment facility (South wells) through the resin columns at an initial flow rate of ~300 mL/min (or ~6 BV/min) per column. Once the flow passed through the resin columns, it was diverted back to the equalization tank for treatment by the process system so that no liquid waste was generated from the field trial. Concentrations of TcO4- in the influent and the effluent were monitored periodically and analyzed at Oak Ridge National Laboratory. The detection limit for TcO4- is approximately 1 pCi/L (or ~6x10-5 parts per billion).
Results (Fig. 1) indicated that the column containing the synthetic resin (RO-02-119) performed extremely well with respect to both Tc-99 removal and hydraulic flow characteristics. Less than 3% breakthrough of Tc-99 was observed after approximately 3 months of operation. The breakthrough of Tc-99 appeared to increase relatively fast after ~600,000 BV groundwater had passed through the synthetic resin column. However, we had difficulty in maintaining a good hydraulic flow through the Purolite A-520E column. The flow was reduced from an initial flow rate of 300 mL/min to approximately 50ö60 mL/min after about 5 weeks of operation; the column became clogged and flow stopped on September 15, 1997. The maximum in-line pressure was approximately 12 psi, which varied slightly from 10ö12 psi throughout the test. Although less groundwater was passed through the Purolite column than the synthetic column, an early breakthrough of Tc-99 was observed in the Purolite column. At 1% breakthrough of Tc-99, approximately 580,000 bed volumes of groundwater had been treated by the synthetic resin whereas only about 105,000 bed volumes of groundwater was treated by the Purolite resin.
Results of TcO4- breakthrough from this field trial were consistent with our previous laboratory and field studies using Purolite A-520E resin and the same or related synthetic resins (Brown et al., 1995; 1996; 1997). Table 1 lists the laboratory determined K'd values for sorption of pertechnetate (6 mM) from a solution matrix containing NaCl, NaNO3, and Na2SO4 (each at 0.06 M) for some previously tested synthetic resins and the Purolite resin. The synthetic resins exhibited ~3ö5 times higher batch equilibrium K'd values and were therefore more selective for TcO4- than the Purolite resin. Laboratory column tests conducted with this same solution matrix at constant flow rate conditions also indicated a substantially retarded breakthrough (>10 times) of TcO4- in these synthetic resin columns than in the Purolite column (Fig. 2). To achieve a 10% TcO4- breakthrough, ~1540 BV test solution was passed through trihexylamine/tripropylamine (VP-02-217) resin whereas only ~ 50 BV test solution was passed through the Purolite column. In a small-scale field trial conducted in 1996 with groundwater from a monitoring well (MW 106) at DOE's Oak Ridge Reservation, columns containing VP-02-217 resin exhibited 10% pertechnetate breakthrough at about 127,000 BV in comparison with ~1500 BV in columns containing Purolite A-520E (Fig. 3). Note that a positive displacement pump was used in the 1996 small-scale field trial so that a constant flow rate (at ~30 mL/min or ~11 BV/min) was maintained throughout the experiment. Therefore, our previous laboratory studies and field trials indicated that the performance of our synthetic resins (RO-02-119 and VP-02-217) are superior to the commercial Purolite resin in terms of both TcO4- selectivity and hydraulic characteristics.
Currently, PGDP is running two Purolite treatment columns (~65 ft3 or ~478 gal per column) at a flow rate of ~125 gal/min per column. At this flow rate, ~137,400 bed volumes of groundwater are treated per column per year. We estimated that, by using the same configuration, our synthetic resin column would operate for ~4.2 years before a breakthrough of 1% TcO4- would occur, while the Purolite resin column would operate for ~0.8 years (or 9ö10 months) to achieve 1% breakthrough of TcO4-. A significant cost saving may therefore be anticipated by using our synthetic resin assuming the resin cost is about twice as much as the Purolite resin. Other cost savings may be indirectly related to the operational and maintenance costs.
The Purolite treatment columns at PGDP have been in service since the Northwest Plume Groundwater System plant operations began in August 1995. Operational and analytical data have shown that the Purolite is the most effective commercial resin when compared with four other commercially available resins that were tested for removal of Tc-99 to levels that meet plant effluent requirements. Analysis of influent Tc-99 concentrations, system flow rates, and effluent Tc-99 concentrations at one of the columns over a 13-month period indicated that the Tc-99 loading on the 65 ft3 of Purolite resin was ~0.19 Ci. No significant operational problems have been experienced while using the Purolite resin. However, analysis of spent resin removed during the last changeout revealed scaling and plugging of the resin surface. The materials responsible for the fouling were composed of such elements as calcium, silica, and manganese.
B: ANALYSIS OF RESINS
After the field trial at PGDP was completed, the resin columns were sectioned, and the amount of Tc-99 sorbed was analyzed as a function of column depth. These analyses were performed by directly adding the resin beads into a scintillation cocktail (Packard Ultima-Gold) and analyzing by means of liquid scintillation counting. However, it must be noted that TcO4- was so strongly adsorbed on the resin beads, we have not yet identified a good technique to desorb TcO4- from the bead surfaces. Thus, this analytical technique may have substantially underestimated the Tc-99 activity, particularly for those pertechnetate anions adsorbed in the internal surface areas of the synthetic resin beads. Therefore, results presented in Fig. 4 should be regarded as a minimum estimate of Tc-99 activity, which showed a general distribution of TcO4- adsorbed along the column. The actual amount of TcO4- retained by the synthetic resin could be much higher because it has a larger internal surface area than that of Purolite. As expected, a much higher Tc-99 activity was observed at the influent to the resin column on the bottom. The Tc-99 activity decreased rapidly from bottom to the top (or along the flow path).
The resin beads were also examined by scanning electron microscopy (SEM) and energy dispersive X-ray (EDX) analyses in an attempt to identify any surface adsorbed species or surface precipitates. SEM analysis (Fig. 5) indicated that the synthetic resin RO-02-119 appeared to be very porous and have a large internal surface area as expected for a macroreticular resin. On the other hand, the commercial Purolite A-520E resin beads appeared to have a smooth and relatively nonporous surface without a large internal surface area (Fig. 6). To further evaluate these differences, we also examined the untreated Purolite resin beads as shown in Fig. 7. Results with untreated beads were quite similar to results with resin beads that had been exposed to the contaminated groundwater. They also appeared to be rigid, and the surface was smooth. Literature from Purolite indicates the A-520E resin is a macroreticular material, but a high magnification SEM (75,000X) is required to see the secondary morphology of the beads. EDX analysis indicated that chloride was the only element that was detected on the untreated Purolite bead surfaces (since the resin was initially in chloride form).
SEM analysis revealed some obvious precipitates on the external surfaces of both the Purolite and the synthetic resin beads, but with more precipitates on the Purolite beads (Fig. 5,6). This difference is probably due to the fact that the RO-02-119 beads have a higher internal surface area, so the precipitates were dispersed onto a relatively larger surface area of the RO-02-119 beads. Furthermore, the Purolite bead surfaces appeared to be smooth and rigid so that the precipitates may have accumulated between the void spaces of the resin beads, and subsequently the flow channels between the beads became clogged. This may offer an explanation why the flow rate decreased rapidly and eventually stopped in the column containing the Purolite A-520E resin after being in the field for ~6 weeks. One of the Purolite columns in use at the PGDP Northwest Plume Treatment Facility also experienced a pressure drop during operation, and the facility personnel found it necessary to backflush the resin column once a year.
Our EDX analysis showed that the precipitates on the two types of resin beads composed primarily of silicate minerals. Some other minor constituents such Ca, Fe, and Mg were also identified by EDX analysis. These constituents could have co-precipitated along with the silicate minerals; however, they may also result from an accumulation of colloidal clay particles that were present in the groundwater. An acid extraction analysis of both types of resin beads (Table 2) indicated the presence of the Ca, Fe, and Mg metal cations. Silica in the acid extractant was not determined because dilute acid could not give a complete extraction of Si. However, it is of great significance to note that the Purolite resin column retained 3 times more sulfate (SO4-2) than did the RO-02-119 column although about three times less groundwater was passed through the Purolite resin column. These results indicate that the Purolite resin has a higher affinity for SO4-2 than the synthetic resin RO-02-119 does and that the TcO4-/SO4-2 selectivity afforded by Purolite A520-E is lower than that afforded by RO-02-119. The presence of large quantities of SO4-2 in the Purolite resin (or an enhanced surface concentration of SO4-2 ) was important because the presence of this anion (as well as dissolved silica) may have facilitated the formation of mineral precipitates (with Al, Ca, or Mg). The hypothesis that SO4-2 anions are preferentially retained by the Purolite resin was independently tested by conducting the SO4-2 breakthrough experiments in the laboratory with both resins used in the field trial. These laboratory experiments (with 1 mM sodium sulfate) revealed that SO4-2 breakthrough occurred much earlier with the synthetic resin (240 bed volumes to produce 50% breakthrough) than with the Purolite resin (460 bed volumes to produce 50% breakthrough). However, RO-02-119 resin adsorbed more NO3- anions than did the Purolite (Table 2) because NO3- is similar to TcO4-, both of which are less hydrated than SO4-2 anions.
C: FURTHER FIELD DEMONSTRATIONS
The resin RO-02-119, which used in this field trial at PGDP, was prepared in the laboratory of Professor Spiro Alexandratos at the University of Tennessee in Knoxville. While the performance of this resin looks promising in relation to the resin in current use, this material is not commercially available. At present, we can only estimate the cost of this resin in the quantities that would be required for practical applications. We would like to conduct a larger pilot-scale field demonstration (at ~1 gal/min flow rate with a 0.5 ft by 1.5 ft column) using a bifunctional resin similar to RO-02-119 (made to our specifications by a commercial vendor). This demonstration would begin in the spring of 1998 and run for six or more months. This field demonstration would reveal practical processing information needed for the design of an ideal TcO4--selective anion exchange resin. Furthermore, the pilot synthesis by a commercial vendor would help us better estimate the cost of the synthetic resin in bulk. Our synthetic resin RO-02-119 is at least five times more selective for TcO4- and has better hydraulic characteristics than the Purolite resin. These features of the synthetic resin should significantly increase the life time of the treatment columns by as much as five times longer and reduce the overall operational and waste management costs.
LITERATURE CITED
- Ashley, K.R., J.R. Ball, A.B. Pinkerton, K.D. Abney, and N.C. Schroeder. 1994. Sorption behavior of Tc-99 on Reillex-HPQ anion exchange resin from nitric acid solution. Sol. Ext. Ion Exch. 12:239-259.
- Brown, G.M., L.M. Bates, P.V. Bonnesen, B.A. Moyer, S.D. Alexandratos, L.A. Hussain, V. Patel, L. Liang, and R.L. Siegrist. 1995. Selective resins for sorption of technetium from groundwater, FY 1995 letter report. Oak Ridge National Laboratory. September 27, 1995.
- Brown, G.M., P.V. Bonnesen, D.J. Presley, L.M. Bates, B.A. Moyer, S.D. Alexandratos, V. Patel, R. Ober, B. Gu, and L. Liang. 1997. Resins for selective sorption of technetium from groundwater. American Chemical Society National Meeting. San Francisco, April 133-17, 1997.
- Brown, G.M., D.J. Presley, P.V. Bonnesen, L.M. Bates, B.A. Moyer, S.D. Alexandratos, L.A. Hussain, V. Patel, B. Gu, L. Liang, and R.L. Siegrist. 1996. Column tests of resins for selective sorption of technetium from groundwater. Oak Ridge National Laboratory. July 31, 1996.
- Gu, B. and R.K. Schulz. 1991. Anion retention in soil: possible application to reduce migration of buried technetium and iodine. NUREG/CR-5464. U.S. Nuclear Regulatory Commission. Washington, DC 20555.
- Gu, B., J.A. MacDonald, J.L. Clausen, and J.F. McCarthy. 1994. Assessment of the Influences of Groundwater Colloids on the Migration of Technetium-99 at the Paducah Gaseous Diffusion Plant Site in Paducah, Kentucky. ORNL/TM-12747. Oak Ridge National Laboratory.
- Gu, B., K.E. Dowlen, L. Liang, and J.L. Clausen. 1996. Efficient separation and recovery of technetium-99 from contaminated groundwater. Sep. Technol. 6:123-132.
- Liang, L., B. Gu, and Y. Yin. 1996. Removal of technetium-99 from contaminated groundwater with sorbents and reductive materials. Sep. Technol. 6:111-122.
- Pourbaix, M. 1966. Atlas of Electrochemical Equilibria. Pergamon Press, Oxford, U.K.
Table 1. Sorption of TcO4- by Purolite A-520E and three synthetic resinsa
|
Resin |
TAECb (meq/g) |
1 h K'd (mL/g) |
24 h K'd (mL/g) |
168 h K'd (mL/g) |
Comment |
|
Purolite A-520E (-16/+50 mesh) |
2.8 |
9,400 |
12,800 |
10,400 |
As purchased |
|
Purolite A-520E (-40/+60 mesh) |
2.8 |
NDc |
13,600 |
13,600 |
Sieved |
|
RO-02-119 (-40/+60 mesh) |
2.53 |
14,400 |
38,500 |
50,100 |
Hexyl/Ethyl |
|
VP-02-217 (-60/+200 mesh) |
2.06 |
17,900 |
48,400 |
56,300 |
Finer size |
|
RO-02-74 (-40/+60 mesh) |
3.04 |
14,600 |
37,800 |
46,000 |
Hexyl/Ethyl |
|
RO-03-124 (-40/+60 mesh) |
3.15 |
18,300 |
27,200 |
30,800 |
Hexyl/Ethyl |
- The groundwater test solution contained 0.06M NaCl, NaNO3, and Na2SO4; all K'd values have a ±5% uncertainty.
- TAEC = total anion exchange capacity.
- ND = not determined.
Table 2. Analysis of major cations and anions (mg/g resin) following the acid extraction of resin beads a
|
Synthetic resin RO-02-119 |
Commercial resin Purolite A-520E | |||
|
Sulfate (SO4-2 ) |
10.20 |
32.84 | ||
|
Nitrate (NO3-) |
15.52 |
10.68 | ||
|
Calcium (Ca) |
0.201 |
0.103 | ||
|
Iron (Fe) |
0.031 |
0.029 | ||
|
Magnesium (Mg) |
0.037 |
0.013 | ||
|
Manganese (Mn) |
0.003 |
0.002 |
- Synthetic resin column = 18.8 g; Purolite resin column = 18.3 g.
Fig. 1. TcO4- breakthrough curves for the synthetic resin RO-02-119 and Purolite A-520E resin using PGDP groundwater. C/C0 is the ratio between the effluent TcO4- concentration and the inlet TcO4- concentration.
Fig. 2. TcO4- breakthrough for selected resins using groundwater test solution (with 0.06M NaCl, NaNO3, and Na2SO4) at about 33 mL/min (or ~11 BV/min) flow rate.
Fig. 3. TcO4- breakthrough curves for the synthetic resin RO-02-217 and Purolite A-520E resin in a test using groundwater from monitoring well 106 on the Oak Ridge Reservation.
Fig. 4. Retention of Tc-99 at different depths of the resin columns.
Fig. 5. Synthetic resin RO-02-119 beads after treatment with PGDP groundwater.
Fig. 6. Purolite A-520E resin beads after treatment with PGDP groundwater.
Fig. 7. Untreated Purolite A-520E resin beads.
Provided by Oak Ridge National Laboratory's Chemical Sciences Division
Rev: October 20, 2005