William J. Crooks III
Westinghouse Savannah River Company
Aiken, SC 29808

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

This report has been reproduced directly from the best available copy.

Available for sale to the public, in paper, from:  U.S. Department of Commerce, National Technical Information Service, 5285 Port Royal Road, Springfield, VA 22161,  phone: (800) 553-6847,  fax: (703) 605-6900,  email:  orders@ntis.fedworld.gov   online ordering:  http://www.ntis.gov/support/index.html

Available electronically at  http://www.osti.gov/bridge/

Available for a processing fee to U.S. Department of Energy and its contractors, in paper, from: U.S. Department of Energy, Office of Scientific and Technical Information, P.O. Box 62, Oak Ridge, TN 37831-0062,  phone: (865 ) 576-8401,  fax: (865) 576-5728,  email:  reports@adonis.osti.gov

The purpose of this work was to identify and quantify any flammable organic compounds remaining after digestion of ReillexÔ HPQ anion exchange resin in potassium permanganate solution. This information will be used by HB-Line to determine if the digested resin meets the Waste Acceptance Criteria for the Waste Tanks. Since the presence of volatile compounds is directly correlated with an increased flammability risk, digested resin samples were analyzed for volatile and semi-volatile organic compounds. Digested resin samples, prepared in 1998, were analyzed in the present work. In one test, a trace quantity of formic acid was identified. However, in all other tests no volatile or semi-volatile organic compounds were detected. The liquid product was a deep green caustic solution with a specific gravity of 1.127 g/cm³, contained 27% dissolved solids, with inorganic carbon and organic carbon concentrations of 1777 mg/mL and 670 mg/mL, respectively. The dissolved solids are expected to be manganese salts, sodium carbonate (source of inorganic carbon), and nonvolatile water-soluble polymer fragments (source of organic carbon). To confirm the absence of volatile organic compounds, a freshly-digested resin sample was prepared (at neutral pH) and analyzed. No volatile or semi-volatile organic compounds were detected as reaction products. This result suggests that volatile reaction products are not evolved during the chemical reaction of permanganate with resin. In conclusion, no volatile organic species [including volatile alcohols (excluding methanol because the Analytical Development Section (ADS) method can not identify methanol), aldehydes, ketones, acids, benzene, acetone, pyridine and styrene] were identified. Since volatile organic compounds were not identified, the permanganate-digested resin matrix should not present a significant flammability risk.

The purpose of this work was to identify and quantify any flammable organic compounds remaining after digestion of ReillexÔ HPQ anion exchange resin in a potassium permanganate solution. This information will be used by HB-Line to determine if the digested resin meets the Waste Acceptance Criteria for the Waste Tanks.

ReillexÔ HPQ is a strong-base macroporous anion exchange resin, and is sold by Reilly Industries Incorporated. It is composed of a copolymer backbone of 1-methyl-4-vinylpyridine (70%) and a divinylbenzene mixture (30%). Approximately 63% of the amine sites are methylated, forming pyridinium cations, which function as the principal anionic exchange sites. Unquaternarized amines protonate in concentrated nitric acid solutions, also resulting in anionic exchange sites. During the polymeric resin synthesis, the "divinylbenzene mixture" reactant is actually composed of 80% divinylbenzene positional isomers and 20% ethylvinylbenzene positional isomers. Based on the production recipe at Reilly Industries, Inc., the vacuum-dried nitrate form resin contains 62-68% carbon, 6-7% hydrogen, 11-12% nitrogen, and 14-19% oxygen.

HB-Line Phase II Start Up plans to use ReillexÔ HPQ anion exchange resin to purify plutonium or neptunium. After ReillexÔ HPQ anion exchange resin has completed its allowable service life in HB-Line, the spent resin will be digested using potassium permanganate, pH-adjusted by the addition of sodium hydroxide, and finally transferred to High Level Waste Tanks for disposal. Prior to the disposal of this material, additional characterization of this waste stream is required to address flammability issues arising from organic compounds present in the digested resin matrix.

Traditionally, spent H-Canyon resin has been digested using the plant potassium permanganate-alkaline digestion process that is outlined here. First, nitric acid is used to slurry the resin from the anion exchange column and flush it into a tank. Second, 50% sodium hydroxide is added to neutralize the acid and adjust the hydroxide concentration to 0.33 molar. Third, 1 kg per liter of resin is added as 6% potassium permanganate. Fourth, the resin is digested at 70 ^oC for 15 hours. Fifth, after the digestion interval is complete, 50% sodium hydroxide is added to the solution to adjust the concentration of NaOH to 1.2 molar in order to satisfy a waste acceptance criterion. Previous laboratory simulations of the ReillexÔ HPQ digestion process estimated the resulting waste volume for this process to be 43 liters of waste per liter of resin, but this waste volume will vary depending on actual plant processing conditions. Water-soluble organic degradation products in this waste are expected to be alcohols, aldehydes, ketones, and carboxylic acids. The motive of the present work is to identify and quantify any volatile and flammable components in the waste stream.

HB-Line will begin using ReillexÔ HPQ in January 2002 and will transfer approximately 500 gallons of digested resin per year (3000 gallons total) until March 2008, but the annual waste volume may vary depending on the resin change-out frequency.

All chemicals used in this study were reagent grade and used without additional purification.

The first sample to be studied was the product of the 15-hour test (1 kg KMnO₄ per liter of wet resin) from the initial permanganate digestion study (Walker, 1998), and the digestion treatment employed has been reported previously. Those treated samples (ReillexÔ HPQ lot #80302MA) were stored in plastic bottles with very little head space until they were used in this work. The reaction mixture was a deep green liquid containing a flocculent brown slurry of solid residue. No purple color was observed in the products, evidence that the permanganate anions were completely consumed. Three aliquots of the deep green liquid were removed for analysis without entraining any of the settled solid material from the bottom of the container. The first aliquot received no additional treatment (remaining caustic), the second aliquot was neutralized to pH 7 with a phosphate buffer (H₂PO₄^-/HPO₄^2-), and the third aliquot was acidified with 8 molar nitric acid to give pH <1 (upon acidifying, the deep green color changed to a pale yellow). Each liquid sample was analyzed for organic compounds using Volatile Organic Compound (VOC) Analysis and Semi-Volatile Organic Compound (SVOC) Analysis.

The second sample was prepared fresh by digesting 1.04 grams of ReillexÔ HPQ (lot #80302MA, converted to nitrate form, previously exposed to 100 MRad of gamma irradiation in the chloride form) in 9 mL of 0.39M potassium permanganate (approximately 5 kg KMnO₄ per liter of wet resin) for 7.2 hours at 87.5 ^oC (± 0.1) for 7.2 hours under 50 psig (nitrogen) in the sealed Parr bomb of an Advanced Reactive System Screening Tool (ARSST). The ARSST is a nearly adiabatic calorimeter that heats an open, well-insulated, 10-mL test cell within a sealed 350-mL Parr bomb. The Parr bomb was pressurized to 50 psig with nitrogen in order to minimize the endothermic process of evaporation of solvent (i.e. loss of the permanganate oxidant). The test cell mixture was stirred and heated at a constant rate of 1 ^oC per minute, then held at constant temperature as described above. The test cell mixture and the Parr bomb were monitored for temperature and pressure, respectively. At completion of the test, the reaction vessel was allowed to equilibrate to room temperature. The pressure in the Parr bomb was slowly vented through a charcoal tube (SUPELCO ORBO 32L, Lot # 1961). Then, a gentle stream of nitrogen was bubbled for 30 minutes through the product slurry, and the effluent gases were passed through the same charcoal tube. The charcoal tube was capped and evaluated by VOC Analysis.

No purple color was observed in the reaction products, evidence that the permanganate anions were completely consumed. The reaction mixture was a brown slurry of solid and a very pale brown liquid. In contrast, the caustic solutions from the 1998 resin digestion were deep green. The 1998 samples were stored for 3 years after the caustic. Under the caustic conditions, this color change is consistent with the formation of a chromophore resulting from hydroxide attack on the pyridine rings. Pyridine ring opening could lead to unsaturated carbonyl compounds that can be colored.

Volatile organic analyses were performed by GC-MS using the ADS method 2656 (Contract Laboratory Program SOW 7-93 for Volatile Organics). Samples were concentrated using an OI Analytical model 4460A Dynamic Headspace concentrator (Purge and Trap), using a three stage (10 cm Carbopack B / 6 cm Carboxen 1000 / 1 cm Carboxen 1001) trap. Separation was performed with a Hewlett Packard 5890 series II gas chromatograph on a 60m x 0.75mm VOCOL glass capillary column with 3 mm film thickness. Quantitation was performed with a Hewlett Packard model 5971 quadrupole mass spectrometer. A glass jet separator was inline prior to the inlet into the mass spectrometer. Internal standard and recovery surrogate compounds were added as specified in the Contract Laboratory Program for volatile organics (SOW 7-93). The mass spectrometer tuning was confirmed within 12 hours prior to each measurement using 4-bromofluorobenzene. Tuning verification was performed against CLP tuning requirements, specifically to optimize CLP requirements for high mass sensitivity. This method can not detect methanol.

Samples were extracted with methylene chloride and spiked with SVOC internal standards then analyzed by Gas Chromatography-Mass Spectrometry. GC/MS analysis was employed to identify organic compounds in the samples. The analytical separations were carried out on a Hewlett Packard 6890 gas chromatograph, equipped with a 30 m DB-5 column, with 0.25 mm diameter and 0.25 mm film thickness. Quantitation was performed using a Hewlett Packard 5973 mass selective detector. The mass spectrometer tuning was confirmed within 24 hours prior to each measurement using perfluorotributylamine.

The weight % (wt%) of dissolved solids was determined by the approved ADS method.

The total organic and total inorganic carbon was determined by the approved ADS method.

The specific gravity of the liquid was determined by the approved ADS method.

Three samples (basic, neutral and acidic matrices) were analyzed by VOC analysis. No VOC analytes were detected in the basic and neutral matrices. The acidic matrix contained 0.11 mg/L of formic acid (HCO₂H, boiling point 100-101 ^oC, flash point 68 ^oC), a concentration that is just above the detection limit. The detection limit for VOC analysis in this study was 0.05 mg/L. Additional samples were analyzed to confirm the essential absence of VOC analytes (one basic and one acidic matrix). In those tests, no VOC analytes were detected.

Three samples (basic, neutral and acidic) were analyzed by SVOC analysis. No SVOC analytes were detected. The detection limit of SVOC analysis in this study was 1 mg/L.

For the permanganate-digested mixture, the liquid portion contained 27 wt% of dissolved solids. Therefore, since no volatile or semi-volatile analytes were found, the dissolved solids are likely to be large (intermediate molecular weight) nonvolatile polymeric fragments.

For the permanganate-digested mixture, the liquid portion had a flash point greater than 100 ^oC.

The inorganic carbon determination was 1777 mg/mL and the organic carbon determination was 670 mg/mL.

For the permanganate-digested mixture, the specific gravity was 1.127 g/cm³.

The charcoal tube sample did not contain any VOC analytes. The detection method for this study was 0.005 mg/tube.

The ARSST was used to digest ReillexÔ HPQ (nitrate form) resin in 0.39 molar potassium permanganate. The addition of potassium permanganate to dry resin resulted in an exothermic reaction that warmed the sample. There was no visible evidence of gas evolution, and no odor detected. The warm permanganate-resin slurry was loaded into the ARSST, pressurized to 50 psig, and heated to 87 ^oC. After the temperature stabilized, the 87 ^oC temperature was maintained for 7.2 hours. Figure 1 shows that no net pressure increase occurred during the constant temperature digestion interval, suggesting that gases (i.e. volatile reaction products) are not evolved during the chemical reaction of permanganate with resin.

H-Canyon plans to digest Reillex HPQ based on the traditional H-Canyon potassium permanganate-alkaline digestion process. In this work, the digestion tests are a simplification of that historical process. Resin digestions were performed in 1998 (at pH 8) during the evaluation of the efficiency of the digestion process. Those digested samples were preserved in sealed plastic bottles for three years, and were analyzed in the present work. In addition, a freshly-digested resin sample (pH 7) was prepared in the present work, with deviations from the historical process. The difference in pH is not a critical factor for resin digestion efficiency. According to Snyder, ruthenium (as RuO₄) is evolved from acidic mixtures of digested resins that had used for processing short-cooled fuels. Ruthenium evolution was eliminated by maintain the digestion pH between 8 and 14. Since current processing plans are for long-cooled materials, ruthenium evolution is not an issue. A caustic pH is not required for process efficiency either. For example, Wehner showed that the MSA-1 resin-permanganate digestion rate decreases above pH 12 but saw no significant changes in efficiency of the degradation (measured as %C oxidized to CO₂) at pH 2.3, 8.2 and 12.5. Also, the higher concentration of permanganate used in the ARSST test (5 kg/L rather than 1kg/L) will increase the rate of reaction with resin, and increase the extent of reaction (i.e. more breakdown to smaller particles) relative to the historical resin digestion process. However, the digestion interval of the ARSST test was shorter (7.2 hours vs. 15 hours). Therefore, for the ARSST test, the efficiency of the digestion reaction at pH 7 and the use of a higher permanganate concentration for a shorter digestion interval provide an adequate simulation of the historical resin digestion.

The digestion products of ReillexÔ HPQ could contain neutral acidic functional groups (e.g. alcohols, aldehydes, ketones, amides), acidic (carboxylic acids, phenols), or basic functional groups (pyridine or amine derivatives). The analytical method used by ADS for the determinations of VOC and SVOC analysis employ helium gas purge/trap and solvent extraction, respectively, followed by analysis by gas chromatography. During the helium purge used in VOC analysis, the volatility of carboxylic acids and phenols is enhanced in acidified form (in acid solution) while the amine volatility is enhanced in the basic from (in caustic solution). Likewise, for extraction into methylene chloride in the SVOC method, the extraction of carboxylic acids and phenols is enhanced from acidic solutions while amines are best extracted from caustic solutions. Carboxylate salts and ammonium salts have high aqueous solubilities, and as a result, low volatility and low extractability into organic solvent. Therefore, to extract both acidic and basic products from the digested-resin mixtures, the pH of aliquots of the caustic solution was adjusted to give a neutral matrix (pH 7) and an acidic matrix (pH <1). During the sample purge, these methods will separate only volatile and semi-volatile organic compounds from the matrix. Large soluble polymeric fragments (acid, basic or neutral) are not expected to be volatile.

The only VOC analyte identified in the permanganate-digested resin (prepared in 1998) was formic acid (0.11 mg/L extracted from the acidified sample) although this result was not confirmed when a duplicate analysis was performed (no analytes were detected). If formic acid is produced during the permanganate digestion, the subsequent caustic adjustment converts the volatile carboxylic acid form into the nonvolatile carboxylate salt form. The 27 wt% of dissolved solids is clear evidence that the resin degraded into soluble fragments, but the VOC and SVOC results indicate that those fragments are not volatile. Although the VOC method can not identify methanol, if methanol were formed during the hot permanganate treatment, it is expected to readily oxidize to formic acid or carbon dioxide. In addition, the flash point of the aqueous solution was in excess of 100 ^oC, additional evidence of the thermal stability of the liquid.

The VOC analysis of the freshly digested resin (prepared in 2001) identified no VOC analytes, consistent with the preceding results for the previously digested resin (prepared in 1998). The ARSST reaction profile during permanganate digestion showed that heating at 87 ^oC for 7.2 hours under nitrogen backpressure (50 psig) did not cause a measurable pressure increase as would be expected if noncondensable volatile gases were evolved.

In summary, no volatile organic species [including volatile alcohols (excluding methanol because the ADS method can not identify methanol), aldehydes, ketones, acids, benzene, acetone, pyridine and styrene] were identified. The product liquid has a specific gravity of 1.127  g/cm³ and contains 27 wt% of dissolved solid.

1. In one VOA analysis, 0.11 mg/L of formic acid was identified in an acidified solution. However, this result was not confirmed. Nevertheless, the volatility of this trace quantity of a carboxylic acid will be eliminated in caustic solution.
2. Essentially no volatile organic species [including volatile alcohols (excluding methanol because the ADS method can not identify methanol), aldehydes, ketones, acids, benzene, acetone, pyridine and styrene] were identified by the VOC and SVOC analysis of ReillexÔ HPQ digested by the HB-Line permanganate method.
3. The liquid product was a deep green caustic solution with a specific gravity of 1.127 g/cm³, contained 27% dissolved solids, with inorganic carbon and organic carbon concentrations of 1777 mg/mL and 670 mg/mL, respectively. The dissolved organic material is likely to be large (intermediate molecular weight) polymeric fragments, which are not volatile or flammable.

1. M. Grenze, personal communication with W.J. Crooks III, Reilly Industries Inc., July 17, 2001.
2. M.D. Snyder, Dissolution of Ion Exchange Resins in Alkaline Permanganate, DP-717, July, 1962.
3. Technical Standard for Waste Disposal, DPSTS-221-HFR-500.
4. B.W. Walker, Permanganate Degradation of Reillex HPQ Ion Exchange Resin for Use in HB-Line, WSRC-TR-98-00235, Rev 0, December 21, 1998.
5. B.A. Davis, Interoffice Memorandum to I. A. Jilana, Acceptance of Degraded Reillex TM HPQ Ion Exchange Resin into Waste Tanks, HLW-STE-2001-00175, April 17, 2001.
6. J.A. Wehner, Permanganate Dissolution of Macroporous Anion Exchange Resin, DPST-88-569, May 19, 1988.