*Obtained from reference 3.

A cursory survey, in light of the demand for the above applications, revealed that there are only a few commercial sources for these useful radionuclides and their availability is limited. Hence, by recovering these radionuclides from contaminated wastes, not only would the sites be cleaned up, but also some of the cleanup cost would be offset by marketing the valuable radionuclides.

Recent developments in membrane technologies (e.g., nanofiltration and reverse osmosis) have resulted in the recovery of metals from wastewater streams. These membrane technologies are also currently used to treat wastewater streams containing low-level radioactive compounds. Hence, there is a potential to recover the radionuclides through membrane technologies.

Based on the above reasons and the initial Phase-I feasibility study, a bench-scale investigation (Phase-II) was proposed to determine/demonstrate the recovery and enrichment of radionuclides (including ⁹⁰Sr and ¹³⁷Cs) from 105-N Basin wastewater at the Hanford Site.

The objectives of the Phase-II bench-scale study were as follows:

• To prove the concept of membrane separation to remove radioactive metals or metal complexes from the 105-N Basin wastewater.
• To select appropriate membrane (nanofiltration and/or reverse osmosis) to perform the desired separation.
• To generate sufficient quantity of filtrate/permeate through a bench-scale membrane separation equipment for evaluating the separation efficiency.
• To estimate the potential for recovering radionuclides from 105-N Basin water.
• To prove the feasibility of enriching/concentrating the radionuclides (e.g., ¹³⁷Cs and ⁹⁰Sr) and reduce the radioactivity of treated water.
• To support National need for new sources of medical isotopes.

The 105-N Basin, containing about 3.8 million liters of liquid waste, was chosen as a source for supplying samples of wastewater for this bench-scale study. The liquid waste contains several radionuclides and nonradioactive inorganic salts. A detailed characteristic of this liquid is presented in Table II. Concentrations of radionuclides ¹³⁷Cs and ⁹⁰Sr are shown in bold figures.

Prior to delivery to the laboratory, samples of water collected from the 105-N Basin would be diluted to reduce the level of radioactivity and would then be filtered through a cartridge filter to prevent any carryover of suspended solids from the 105-N Basin.

A high-pressure cell test unit, which is intended for use in determining feasibility of a prospective use for membrane separation technology, was selected for this bench-scale testing.

As shown in Fig. 1, the unit consisted of a feed tank, two stainless-steel pressure test cells, a high-pressure P.D.-type pump, and associated valves and piping. The unit was designed to operate up to a maximum pressure of 70 bar and up to a maximum temperature of 90°C with operating pH limits of 1-13.

The stainless-steel pressure test cells were so designed that the feed water, when delivered through the top plate, would pass through the selected membrane, permeate collected through the bottom plate, and concentrate from the top plate. Collected concentrate then would be returned to the feed tank for the next cycle. Upstream of each test cell, a back-pressure regulator was installed. The unit was also equipped with relief and bypass valves for safety and flexibility. A schematic of the unit is shown in Fig. 2.

A brief overview of a typical bench-scale experiment to recover the radionuclides is provided below.

At first, a known quantity of the water sample is filtered (through a micro-filter cartridge) to remove any solid residues. The solids-free water, thus, is used for subsequent experiments. The activity (alpha, beta, and gamma) of the water is tested. The concentration of radionuclides, pH, conductivity, and total dissolved solids (TDS) of the water are also analyzed.

The solids-free water is then treated with appropriate macro-molecular chelating agents. After chelation, the water is subjected to membrane separation. Depending on the nature of chelates, appropriate membranes (either nanofiltration or reverse osmosis) are used for separation. The permeate from the membrane filtration device is analyzed for residual radioactivity (alpha, beta and gamma), radionuclides, pH, conductivity, and TDS.

The radionuclides are concentrated gradually by adding appropriate amount of source water and extracting permeate from the membrane filtration system. This process continues until a desired concentration of radionuclides is reached.

During this Phase-II bench-scale testing, the following procedure was utilized.

• At first, the high-pressure cell test unit was assembled and tested using a dilute sodium chloride solution. The rejection efficiency was determined by comparing the conductivities of test NaCl solution and the permeate from reverse osmosis.

A series of measurements revealed that the first membrane, designated as "AD," showed 97% rejection and the second membrane designated as "AE," showed a 99% rejection. Hence, the second membrane was chosen for the study.

• Initially, 2,500 mL of 105-N Basin water was treated with suitable macro-molecular complexing agent to increase the molecular weight of the radionuclides.

This water was treated with a high-rejection reverse osmosis membrane and 1 L of permeate was extracted (first permeate). At this point, the concentration of radionuclides in the concentrated solution is expected to be about 1.67 times the initial concentration.

• Then, 1 liter of 105-N Basin water (treated with macro-molecular agent) was added to the remaining concentrate. At this point, the expected concentration of radionuclides is about 1.4 times the initial concentration.
• From the above solution, 1 liter of permeate is extracted through the reverse osmosis membrane (second permeate). At this point, the concentration of radionuclides in the concentrated solution is expected to be about 2.33 times the initial concentration.
• Then, 1 liter of 105-N Basin water (treated with macro-molecular agent) was added to the remaining concentrate. At this point, the expected concentration of radionuclides is about 1.8 times the initial concentration.
• From the above solution, 1 liter of permeate is extracted through the reverse osmosis membrane (third permeate). At this point, the concentration of radionuclides in the concentrated solution is expected to be about 3.0 times the initial concentration.
• The experiment was concluded at the end of three-fold concentration.

At the end of the experiments, samples of 105-N Basin water, recovered concentrated solution, and permeate were sent to a laboratory for analyses. The parameters analyzed were as follows:

• pH
• Conductivity
• TDS
• Alpha activity
• Beta activity
• Gamma activity - scan
• 137Cs.

At the end of the experiments, the water samples, filters, and other equipment were subjected to proper decontamination/disposal procedures.

Standard quality assurance/quality control procedures were followed throughout the course of these experiments. The integrity of membranes was tested by means of testing the rejection on 2,000 mg/L of sodium chloride solution. With this sodium chloride solution, a rejection of >98% was achieved (indicating the membrane was intact and the performance of the membrane was as per the specification of membrane).

During the course of this test, the reverse osmosis membrane was operated at 21 bar. The pressure was maintained uniformly during the entire course of the experiment. The quality assurance/quality control tests were also conducted at 21 bar.

Table III summarizes the critical results of the bench-scale experiment. Based on these results, the following observations were made:

• Based on the volume of 105-N Basin water taken for the experiment (~4.5 L) and the permeate collected (~3.0 L), the volume of concentrate is ~1.5 L. This is equivalent to a three-fold concentration.
• Based on conductivity, the concentration factor is ~3.0. This corroborates with the volumetric concentration.
• Based on ¹³⁷Cs concentration, the concentration factor is ~2.97. This also corroborates with the volumetric concentration.
• Based on beta activity, the concentration factor is ~1.89. This is ~50% lower than expected from the volumetric concentration. This discrepancy may be due to the sensitivity of analysis for beta activity.
• The rejection (by membrane) for TDS, alpha activity, beta activity, and ¹³⁷Cs are 96.3%, 99.1%, 99.8%, and 99.4% respectively. This indicates that the membranes provide an excellent barrier to the radionuclides that are present in the 105-N Basin water.
• The permeate from the membrane system contained 3.77 pCi/L of alpha activity compared to 408.8 pCi/L of alpha activity in the feed; 1,530 pCi/L of beta activity compared to 891,200 pCi/L in the feed; and 715 pCi/L of ¹³⁷Cs compared to 119,300 pCi/L in the feed.

In addition to ¹³⁷Cs, the membrane rejected all the other radioisotopes (e.g., ⁶⁰Co, ¹⁵²Eu, ¹⁵⁴Eu, and ¹⁵⁵Eu) effectively.

Based on the preliminary experimental results and the actual wastewater characteristics of the 105-N Basin water, a preliminary economic analysis was made. A summary of this analysis is presented in Table IV.

Based on this analysis, it is possible to recover a quantity of ¹³⁷Cs worth 5.3 million dollars and a quantity of ⁹⁰Sr worth 1.1 million dollars from the 3.80 million liters of 105-N Basin water. After deducting estimated equipment and operating cost of 1.5 million dollars for the Pilot-Scale system, the net savings could be at least 4.9 million dollars. Thus, the recovery operation will be cost-effective and the expected pay-back period will be much less than a year.

It is important to note that the above estimates are highly conservative (particularly when it is assumed that the amount of recoverable radionuclides is only 10% and that the selling price is 10% of the commercial value of the radionuclides), as the actual expected recovery and the selling price will be much higher.

In addition to recovering radionuclides, this process will also result in removing radioactivity of the treated water and, hence, will reduce the overall cost of disposal.

The results of the Phase-II bench-scale tests indicate that the potential for recovering radionuclides from the 105-N Basin water is excellent. It is possible to recover >90% of the radionuclides by means of the above process. The resulting permeate (or treated water) has considerably lower levels of radioactivity compared to the 105-N Basin feed water, hence, the cost of disposal is minimized.

Based on the promising results, it is recommended that a Pilot-Scale test be conducted to further verify the effectiveness of the process and the operating cost of treating large volumes of various forms of contaminated waste, to recover substantial quantities of marketable radionuclides such as ⁹⁰Sr and ¹³⁷Cs. The results of the Pilot-Scale test, in turn, can then be used to develop a potential full-scale system for treatment of various forms of waste from Hanford as well as other DOE Sites for economic advantage.

1. Data Quality Objectives Summary Report for the 105-N Basin Liquid Disposition, BHI-00979, Rev. 0, March 1997.
2. Laboratory Analytical Results of Samples from Bench-Scale Experiments, SAF-B97-176, September, 1997
3. FFTF Medical Isotopes Market Study (2001 - 2020), Frost & Sullivan, 1997.

BACK