Environ. Sci. Technol. 1994, 28, 423-428
Thermochemical Nitrate Destruction John L. Cox,' Richard T. Hallen, and Mlchael A. Lllga
Pacific Northwest Laboratory, Richland, Washington 99352 The thermochemical destruction of nitrate was conducted in an aqueous media. Six reducing agents (ammonia, formate,urea, glucose,methane, and hydrogen)were mixed separately with 3 w t % NO3- buffered aqueous solutions at pH 13; ammonia and formate were also mixed at pH 4. Reactions were conducted in a batch reactor in the absence of air at temperatures of 200-350 "C and pressures of 600-2800 psig. Reaction times varied from 0.5 to 3.0 h. Both gas and liquid samples from the reaction were analyzed. The specific components analyzed were nitrate, nitrite, nitrous oxide, nitrogen, and ammonia. The observed order of nitrate reduction effectiveness in basic solution was formate > glucose > urea > hydrogen > ammonia = methane. Ammonia was more effective under acidic conditions than basic conditions. Formate was also effective under acidic conditions. Introduction
A variety of methods have been studied for removing nitrate from hazardous wastes, water, and process streams. Eliminating the nitratc.would in turn eliminate one of the major anthropogenic pollutants to our natural environment. However, the methods that have been examined suffer from one or more deficiencies, including the production of undesirable byproducts, and they are slow or do not provide complete removal. An ideal treatment method would be one that destroys the nitrate by converting it to molecular nitrogen and water without producing problematic byproducts such as ammonia or nitrogen oxides. The system described here destroys the nitrate within a reaction vessel by using lower temperatures than previouslystudied along with pressures adequate to maintain an aqueous phase. The rates of conversion shown by this system have demonstrated that it is a practical alternative for removing nitrate from hazardous waste streams. This paper provides the experimental methods used to test the system with six chemicalreducing agents and the results obtained. A short description of methods studied by others is also given for background. Background Methods previously examined for nitrate removal from aqueous media include ion exchange, extraction, membranes, distillation, and chemical and biological denitrification. Ion exchange, extraction, membrane, and distillation techniques leave the nitrate radical unchanged, which creates yet another seQaration and/or disposal problem. If the nitrate is going to be reused, a recovery operation is needed. For radioactive wastes, methods of nitrate reclamation and reuse are continuing to be investigated, but the use of nitrate in a form such as ammonium nitrate for fertilizer suffers from liability and product acceptance issues related to its being derived from a mixed waste. Use of nitrate in the form of nitric acid in nuclear fuel manufacturing and reprocessing faces a limited demand for nitric acid. 0013-936X/94/0928-0423$04.50/0
0 1994 American Chemlcal Society
Chemicaland biologics- nethods decomposenitrate into other chemical species; the nature of these species depends on the reaction conditions. Biological denitrification methods suffer from slow rates, pH sensitivity, and microbe poisoning by elevated levels of nitrates and a host of other compounds; the ammonia byproduct requires disposal. The method finds its greatest application in nitrate removal from dilute waste streams, such as those at sewage treatment plants or groundwater, where nitrate levels are usually less than 2000 ppm. A number of thermochemical methods have been examined for denitrification of dry nitrate wastes (1-3). Moderately high (300-500 "C) to high temperatures (>500 "C) can be employed to thermally decompose Nos- into Oz, Nz, and NO,. The amount of undesirable NO, formation is a function of temperature, heating rate, and co-reactant composition. Thermochemical methods of nitrate destruction can achieve denitrification at lower temperatures by employinga chemicalreducing agent such as Fez+ or ammonia. A variety of chemical reducing agents have been examined in aqueous media. Gunderloy et al. (4, 5 ) examined the denitrification of dilute nitrate ion solutions (44-220 ppm NO3-) under both basic and acidic anaerobic conditions at 85 O F for 48 h in the presence of ferric and cupric ion catalysts. The relative effectiveness of NO3reduction was Fez+ 2: iron powd-6r = NzHz > glucose >> CO 2: CHzO = carbon = SO2 = 0. Other investigators have reported the chemical denitrification of nitrate solutions with sugar (6),formaldehyde (9, formic acid (8), and powdered aluminum (9). Meile and Johnson (2) reported that denitrification of dilute solutions of nitric acid is achieved by refluxing the solution for 3 h in the presence of chemicalreducing agents. The nitrate destruction effectivenessof the reducing agents decreased in the order: formic acid (96 % > formaldehyde (54%) > sucrose (17%) > urea (8%) > ferrous sulfamate (6%1 > sulfamic acid (0 % ) = hydrogen peroxide (0%). The best nitrate destruction yields were obtained in a twostep process of (1)adding HzS04 to the waste and distilling off the resultant "03 and (2) refluxing the HN03 for 3 h with formic acid. If the HNO3 was not distilled out of the H2S04 mix, a nickel catalyst was required to obtain high destruction yields. However, high (4400 ppm) NO, gas was obtained in the off-gas. Meile and Johnson (2)also reported on high-temperature nitrate destruction in a high-temperaturefluid wall reactor. At temperatures of 2200 "C, 96% Nos- destruction with low NO, concentrations in the products was achieved in the presence of carbon black. In the absence of the carbon black, only 80 % nitrate destruction occurred. Slightly better nitrate destruction was obtained in the presence of carbon black in a molten salt reactor at 1000 "C, while only 25% destruction was observed with urea as the reducing agent in a muffle furnace at 400 "C. Experimental Methods The thermochemical reduction experiments were conducted behind a 114-in. barricade in a l-L batch reactor Environ. Sci. Technol., Vol. 28, No. 3, 1994
423
Acid Solution (pH = 0)
.
-0.272 I
I
"4-'
I
-1.275
-1.25
I &fi -1.41
-1.07
I
-0.79
Base Solution (pH = 14)
1
NH4OH
- - - I - -I - -0.1
-0.73
N2H4
NH20H
I1
-0.42
N2
-1.52
0.;
NO
N20
N20*
I I
-0.18
0.14
0.737 Oxidation State -3
-2
0.86
I I *:I:-N[
NOp-
-0.26 -1
0
ti
t2
13
t4
t5
Figure 1. Oxidation potential diagrams for nitrogen (ref 10, adapted by permlssion from Prentice Hall, Inc., Englewood Cliffs, Nj).
(autoclave) with temperature and pressure ratings of 450 "C and 5400 psig. Both gas and liquid sampling capabilities were incorporated in the reactor, enabling the samples to be collected and analyzed throughout the experimental run. The components analyzed in this system were Nos-, NOz-, N20, Nz, and "3. Experimental runs (A-F) were conducted using 3 wt % Nos- aqueous solutions mixed with a reducing agent (ammonia, A; formate, B; urea, C; glucose, D; methane, E; hydrogen, F) in a buffered aqueous solution. Strongly basic solutions were prepared with a NaOH-Na3P04 buffer; weakly basic solutions were prepared with a NaOHNH4HzP04 buffer, while H3P04 was used to prepare acidic solutions. The initial pH was measured, and the reactants were charged to the reactor, which was then flushed and pressurized with helium to 100 psig to provide a cushion gas. In those instances where a reactant gas, e.g., CH4 or H2, was employed, it was used in place of helium. The reactor was then heated to reaction temperature and allowed to react for a fixed period of time. The rapid stirring of reactants was maintained during heatup and reaction. Gas samples were drawn and analyzed a t predetermined intervals throughout the run. Following completion of each run, the reactor was rapidly Genched to room temperature with water via an immersed cooling c d . A final gas sample was taken, and the gas volume was measured by venting through a wet test meter. The contents of the reactor were then removed and weighed to permit overall material balances to be determined; the product was removed and stored in a tightly sealed glass vial in the refrigerator for subsequent analysis for nitrate, nitrite, and ammonia. Gas analyses were performed with a Model 158A Carle gas chromatograph equipped with an 8-ft 80% Porapak N-20% Porapak Q 50/80 mesh column; an 8-ft molecular sieve 13X 80/100 mesh column; a 6-ft 8% OV-101 on Chromosorb W 80/100 mesh column; a thermal conductivity detector; and helium and nitrogen carrier gases with 424
Envlron. Sci. Technol., Vol. 28,
No. 3, 1994
flow rates of 26 and 55 cm3/min, respectively. Analyses were performed a t a column temperature of 65 "C. This configuration allowed the measurement of hydrogen, nitrogen, nitrous oxide, carbon monoxide, carbon dioxide, and C1-C5 hydrocarbons. Nitric oxide, nitrogen dioxide, and ammonia could not be reliably analyzed in the gas phase with this gas chromatographic setup. The ammonia peak was broad and lacked quantitative reproducibility. Both NO2 and NO are reactive compounds that are scavenged by trace quantities of air in the system or introduced during sampling. The NOz, furthermore, is irreversibly adsorbed on many types of column packing. Chemical analysis for nitrogen compounds in the aqueous phase employed the Hach spectrophotometric methods Nitraver 5 and Nitriver 3 for Nos- and NOz-, respectively. An Orion Model 95-12 ion selective NH4+ electrode with a Beckman Model 3500 digital pH/mv meter was used for NH3 analysis.
Results and Discussion The variety of nitrogen redox chemistry in aqueous acid and base solution is illustrated by the standard oxidation potential diagrams shown in Figure 1(IO). While the halfreaction of the reduction of nitrate to nitrogen is favorable at any pH, the standard free-energy change in acid (144 kcal) (eq la) is much more than that in base (30 kcal) (eq lb).
+ 6H + 5e- = 1/2Nz+ 3H20 E" = 1.25 V (la) NO; + 3H20+ 5e- = 1/2N2+ 60H- E" = 0.26 V
NO;
+
(lb)
The reduction of nitrate to nitrogen is a five-electron process. Examination of Figure 1 reveals that in the reduction of nitrate to nitrogen, each one-electron reduction has a favorable free-energy change in acidic media
Table 1. Oxidation Potentials for Reducing Reagents (Data Compiled from Ref 10) couple
Eo, (V)'
E"b
0.20
(VIb
1.01
CO(NHz)z/COz,Nz -0.10 0.70 -0.16 0.66 CHdCOz -0.27 0.74 NHsINz 0 E",, standard oxidation potential at pH = 0. E%, standard oxidation potential at pH = 14.
(AG is negative). On the other hand, in basic solution, two of the one-electron reductions, NOs- to Nz04 and NO2to NO, have unfavorable free-energy changes (AG is positive). The standard redox potential for the N03-/Nz couple predicts that any reducing agent with an E" > -1.25 V in acidic solution and E" > -0.26 V in basic solution should provide favorable free-energy change for the redox reaction. The standard oxidation potentials for the investigated reducing reagents are listed in Table 1for both acidic (E",, pH = 0) and basic (EOb, pH = 14) solutions. All of these reducing agents in either acid or base have favorable thermodynamics for the reduction of nitrate to molecular nitrogen. Results from 16 experimental runs are summarized in Table 2. The pH values listed in Table 2 are those determined experimentally a t the beginning (pHi) and end (pHf) of the runs. The temperature, pressure, and time refer to those experimental parameters at run conditions. Time-zero corresponds to the time the reactor reached the indicated reaction temperature. The heatup time to reaction temperature was about 0.5 h for the lowtemperature (€250 "C) runs and about 1.0 h for the hightemperature (300-350 "C) runs. With the exception of runs A-1 through A-3, the reactant ratios were selected so that an excess of reducing agent was present to force the reaction to completion. The nitrate concentration was 3 wt %. With the exception of runs E-1 and F-1, the percent nitrate conversion reported in Table 2 is based on the amount of nitrate (NOs-) consumed during the reaction, which in turn was determined from the difference between the amount charged initially and that found in the recovered solution by the previously described spectrophotometric method. In runs E-1 and F-1, the conversion
is based on the amount of nitrogen (Nz) observed in the reaction products. Table 3 contains the total quantities of nitrogen compounds expressed in moles in the gaseous and liquid phase at the beginning and conclusion of each experimental run. The initial amounts are based on quantities of reagents used in the experimental runs determined by the weight of liquid and solid reagents and the volume of gaseous materials. The final amount of materials is based on analyses performed and material balance considerations. In the A series of runs, ammonia was used as the reducing agent. Variables examined were concentration, temperature, pH, and reaction time. The stoichiometry of the reduction reaction is given in eq 2 and 3 for acidic and basic conditions, respectively.
+ 3NO;
+ 2H' + 9Hz0
(2)
5NH3 + 3NO; = 4Nz + 30H- + 6H20
(3)
5NH,+
= 4N2
As shown in Table 2, the NO3- reduction in basic solutions was found to be a very sluggish reaction. A t pH = 12.5,the NOs- conversion was 0% (run A-1) after reacting for 3.75 h at 250 "C and only 9% (run A-3) after 6 h at 350 "C. Even at near-neutral conditions (pH = 7.8), the Nosconversion was only 1% after 2 h at 350 "C (run A-4). The slowness of nitrate reduction with ammonia in basic media may be due to the large activation barrier in the formation of N2O4 and NO as intermediate products, although the mechanism of the reaction is unknown. Also, the limited solubility of ammonia under basic conditions may have contributed to the poor nitrate conversion; Le., the ammonia concentration in solution was low. In acidic solution at pH = 4, the reduction with ammonia was comparatively facile as shown in run A-5 where 100% NO3- conversion was achieved in 2 h at 350 "C. Further experiments at pH = 4 (runs A-6 through A-8) revealed the effect of temperature on the conversion. Experimental NO3- conversions for 2 h reaction time were 13% at 250 "C, 51% a t 300 "C, and 100% a t 350 "C. The attainable reproducibility of the experimental runs is revealed by a comparison between runs A-5 and A-8. These runs, insofar as possible, were conducted under identical conditions. Total nitrogen material balances reveal 9.86 g of N charged and 9.61 g of N recovered in run A-5 and 9.40 g of N charged and 9.99 g of N recovered in
Table 2. Summary of Nitrate Reduction Experiments run no.
react,ants (mo1:mol)
pHi
pHf
temp ("C)
pressure a t temp (psig)
time a t temp (h)
nitrate conversion ( % )
A- 1 A-2 A-3 A-4 A-5 A-6 A-7 A-8 B-1 B-2 B-3
NHs:N03NHs:N03NH3:NOs2.4NH3:N032.4NH3:N032.4NHs:N032.4NH3:N032.3NHs:NOa3.OHC02-:N033.OHCOz-:N033.0HCOz-:N030.7urea:NOa0.4glucose:NO31.4CHd:N033.7Hz:N03-
12.5 12.5 12.5 7.8 4 4 3.9 4 3.9 13.2 13.2 12.7 13b 13.2 13.2
12.7 NM" 12.8 6.9 2.6 4.5 5.5 2.8 7.5 9.7 9.8
250 300 350 350 350 250 300 350 350 325 305 345 350 350 360
617 1220 2278 2484 2558 588 1264 2558 2852 1970 1294 2058 2323 2572 3102
3.75 3.00 6.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00
0 2 9 1 100 13 51 100 100 99 74 20 48 6 18
c-1
D-1 E-1 F-1
10.1
8.2 9.9 NM
NM = not measured. Estimated. Envlron. Sci. Technoi., Vol. 28, NO.3, 1994
425
Table 3. Summary of Nitrate Reduction Specificity initial (mol) run no.
NO3-
A- 1 A-2 A-3 A-4 A-5 A-6 A-I A-8 B-1 B-2 B-3
0.205 0.205 0.205 0.205 0.205 0.205 0.205 0.205 0.205 0.205 0.208 0.205 0.205 0.205 0.205
final (mol)
NOz-
Nos-
"3
Nz
N2O
unaccounted@
"3
0.205 0.213 tb 0.00 0.0043 0.193 -0.005 0.205 0.202 0.0007 0.00 0.0002 0.181 0.026 0.205 0.185 0.0039 0.00 0.0004 0.143 0.077 0.501 0.203 0.00 0.028 0.148 0.337 -0.186 0.499 0.0008 C 0.045 0.199 0.197 0.012 0.499 0.177 0.00 0.00 0.0046 0.677 -0.159 0.499 0.101 0.00 0.001 0.068 0.357 0.108 0.499 t 0.00 0.062 0.192 0.204 -0.007 0.00 t 0.00 0.0002 0.102 0.00 0.001 0.00 0.002 (b) 0.0002 0.046 0.00 0.111 0.00 (b) 0.054 0.0027 0.034 0.00 0.081 c-1 0.205d 0.166 0.00 t 0.0064 0.318 -0.087 D-1 0.00 0.099 (b) t 0.0050 0.00 0.097 E-1 0.00 (b) (b) t 0.0061 0.00 0.193 F-1 0.00 (b) 0.0002 (b) 0.018 0.00 0.169 Unaccounted nitrogen, mol of initial nitrogen less mol of final nitrogen. t = trace 50.01 g. Not determined. There is an additional 5.02 g of NH3 available if the urea is completely hydrolyzed. (1
0.00
t A d
0.0
0.50
1 .o
1.5
2.0
2.5
00
0 50
Figure 2. Timedependent gas composition for NOs- reduction with NH4+ at pH 4 and 350 O C (run A-8).
run A-8. Both runs proceeded with 100% nitrate conversion. Table 3 indicates that the major nitrogen products were Nz and N20, although as previously pointed out, the analysis for other gaseous nitrogen compounds (NO and NOz) was unreliable. The reactivity order of the nitrogen oxides, NzO < NO2 < NO, is consistent with the products observed; NO reacts instantly with air. Figure 2 shows the time-dependent production of N2 and N2O and the concomitant decrease of the helium cushion gas concentration with reaction time at 350 OC and pH 4. The production of N2 and NzO starts slowly, followed by a rapid increase in yield, and finally slows again toward the end of the reaction. The slowing of the reaction in the final stages is consistent with mass action effects,where the decrease in the concentration of reactants and the increase in the concentration of products retard the forward reaction. The slow initial stage of the reaction or induction period occurs in both acidic and basic solutions and has been observed by other investigators (11-14). While the induction period in acid is reported to be affected by temperature, concentration of reactants, and reactor walls,it is primarily a function of nitrous acid in the system where HCHO, HCOZH, and possibly other reducing agents are employed. The addition of sodium nitrite can shorten or even eliminate the induction period in the HCHO/HN03 and HC02H/HN03 systems (13). 428
Envlron. Sci. Technol., Vol. 28, No. 3, 1994
10
15
20
25
TIME HRS
TIME,HRS
Figure 3. Gas composition (run B-1) nitrate reduction with formate at pH 4 and 350 OC.
The B series of runs employed formate as the reducing agent. The stoichiometric reactions for both acidic and basic conditions are given below.
+
5HC02H 2NO;
+ 2Ht = 5C0, + N2 i-6Hz0
(4)
In contrast to the reduction with ammonia, formate produced high nitrate conversions in both acidic and basic solutions. At an initial pH of 13, nitrate conversions of 74% and 99 % were obtained by reaction for 2 h at 305 and 325 "C, respectively. At an initial pH of 4, a nitrate conversion of 100% was obtained by reacting 2 h at 350 OC. The product gas composition as a function of reaction time for the latter run (run B-1) is shown in Figure 3. Initially, CO, H2, and COz are the predominant gaseous products. As the reaction proceeds, the carbon monoxideshift reaction (HzO + CO = H2 + C02) producing C02 and H2 peaks, and the rate of nitrogen production increases. The only appearance of N0,'s was N20 at time-zero where the concentration was 0.5%. The implication is that since measurable quantities of the product gases from the reduction are present at time-zero, it is apparent that some reaction of the formate occurred in the 30-45 min required to bring the reactants from ambient to reaction
temperature (350 "C). At this temperature, some thermal decomposition of formate into hydrogen and carbon monoxide would be expected. Figure 3 shows very little N2 forming during this period, and in fact, the N2 production rate remains low until about 0.5 h into the reaction. A similar induction period was observed when ammonia was the reducing agent. The major difference observed between using ammonia and formate as reducing agents under acidic conditions is the amount of N2O formed. With formate there was no N20 in the final product gases under acidic conditions; with ammonia as the reducing agent, N2O was present throughout the 2-h run and accounted for 24 mol 5% of the product gases. As a final observation in the B series of runs, the reduction of NOS-by formate has a neutralizing influence on the reaction media, amounting to about 3.5 pH unit when starting from either pH 13 or pH 4. This change in pH in the absence of adequate buffering capacity by the reacting solution is predicted by eq 4 and 5. The C, D, E, andF series of runs employed urea, glucose, methane, and hydrogen, respectively, as reducing agents under basic conditions. The reaction stoichiometry under basic conditions is given in the following reactions. 5CO(NH2), + 6NO; = 5HCO;
+
5/3C6HI2O6 20H-
+ 8NO;
+ 7H20 + 8N2 + OH-
= 10HCO;
(6)
+ 6H20+ 4N2 (7)
5CH,
+ 8NO; 5H,
= 5HCO;
+ 2NO;
= N2
+ 6H20+ 4N2 + 30H-
+ 4H20 + 20H-
(8) (9)
The nitrate reduction results using urea are summarized in Table 2 under run C-1. A total of 20 % nitrate conversion was achieved after 2-h reaction at 345 "C and initial pH of 12.7. Carbon dioxide and small amounts of N20 were observed in the reaction products. Nitrogen gas was continuing to be produced when the run was terminated. No hydrogen or carbon monoxide was observed in the product gas, contrary to runs B-1 through B-3, where a carbon-based reducing agent (formate) was also used. Glucose (run D-1) proved to be amore effective reducing agent than urea. Under basic (pH = 13) conditions, 48% nitrate conversion was achieved after reacting 2 h at 350 "C. Gas chromatography revealed only trace amounts of N2O at the beginning of the run. The amount of nitrogen produced increased throughout the run. There were also carbon-based gaseous species produced, including carbon monoxide and carbon dioxide, as well as hydrogen, which would be expected products of the water gas shift reaction under these experimental conditions. For runs E-1 and F-1, where methane and hydrogen were employed as reducing agents, the nitrate conversion is based on product gas analysis. Methane produced a 6 % nitrate conversion to nitrogen, while hydrogen produced an 18% conversionto nitrogen. Only trace amounts of N2O were observed in the gaseous products for both reducing agents. In contrasting the different reducing agents, it is interesting that large amounts of NzO and N2 were produced from the ammonia-nitrate system under acidic reaction conditions, while N2O was nondetectable in the
final reaction product from the formate-nitrate system under both acidic and basic conditions. The effectiveness of nitrate destruction in basic solution roughly correlates with the standard oxidation potential of the reducing agent (Table 1). The best reducing agents are those with the highest oxidation potential: formate > glucose > urea > hydrogen > ammonia = methane. The lower than expected activity for hydrogen and ammonia could in part be due to the low solubility, i.e., concentration, of these reducing agents in the reaction medium.
Conclusions Moderately concentrated (3 w t % ) aqueous solutions of nitrate can be quantitatively destroyed thermochemically. The principal reaction products are nitrogen (N2) and water with a noncarbonaceous nitrogenous reducing agent (e.g., NHB) and nitrogen, water, and carbon dioxide with a carbonaceous reducing agent. The destruction is dependent on temperature, reaction time, pH, and the nature of the reducing agent. The effectiveness order of reducing agents in basic solution (pH 13) in this study was formate > glucose > urea > hydrogen > ammonia x methane. In acidic solution (pH 4), both ammonia and formate were found to be equally effective as reducing agents at 350 "C. The reaction stoichiometry in either acid or base predicted solution neutralization with the progress of reactions as was experimentally observed. That no NO2 or NO was observed as reaction products is consistent with the reactivity order NO > NO2 > N2O. These results, in turn, have demonstrated the feasibility of using thermochemical nitrate destruction for removing nitrate from hazardous waste streams.
Acknowledgments Pacific Northwest Laboratory is operated for the US. Department of Energy by Battelle Memorial Institute under Contract DE-AC06-76RLO-1830.
Literature Cited (1) Dotson, J. M.; Peters, T. E. U.S. Patent 3 862 296, 1975. (2) Meile, L. J.; Johnson, A. J. Waste Generation Reduction-Nitrates F Y 1982 Status Report; Technical Report No. RFP-3465, DOE/TIC-4500 (Rev. 72); Rockwell International: Golden, CO, 1982. (3) Meile, L. J.; Johnson, A. J. Waste Generation Reduction-Nitrates F Y 1983 Status Report; Technical Report No. RFP-3619, DOE/TIC-4500 (Rev. 72); Rockwell International: Golden, CO, 1983. (4) Gunderloy, F. C., Jr.; Fujikawa, C. Y.; Dayan, V. H.; Grid,
S. Dilute Solution Reactions of the Nitrate Ion as Applied to Water Reclamation; Technical Report No. TWRC-1; FWPCA: Cincinnati, OH, 1968. ( 5 ) Gunderloy, F. C., Jr.; Wagner, R. I.; Dayan, V. H. Deuelopment of a Chemical Denitrification Process; Technical Report NO. EPA 17010EEX/10/70; U S . Environmental Protection Agency, Water Quality Office: Washington, DC, 1970. (6) Bray, L. A. Denitration of Purex Wastes with Sugar; Technical Report No. HW-76973; Hanford Atomic Products Operation: Richland, WA, 1963. (7) Forsman, R. C.; Oberg, G. C. Formaldehyde Treatment of Purex Radioactive Wastes; Technical Report No. HW79622; Hanford Atomic Products Operation: Richland, WA, 1963. Envlron. Scl. Technol., Vol. 28, No. 3, 1994
427
(8) Bradley, R. F.; Goodlett, C. B. Denitration of Nitric Acid Solutions by Formic Acid; Technical Report No. DP-1299; Savannah River Laboratory: Aiken, SC, 1972. (9) Murphy, A. P. Nature 1991, 350, 223-225. (10) Latimer, W. M. The Oxidation States of Elements and Their Potentials in Aqueous Solution, 2nd ed.; PrenticeHall, Inc.: New York, 1952; p 104. (11) Cecille, L.; Kelm, M. In Denitration of Radioactive Liquid Waste; Cecille, L., Halaszovich, S., Eds.; Graham & Trotman: Norwell, MA, 1986. (12) Longstaff, J. V. L.; Singer, K. J. Chem. SOC.1954, 25042617.
428
Envlron. Scl. Technol., Vol. 28, No. 3, 1994
(13) Kubota, M.; Yamaguchi, I., Nakamura, H. J. Nucl. Sci. Technol. 1979, 16, (6). (14) Germain, M.; Bathellier, A.; Berard, P. In Proceedings of the International Solvent Extraction Conference;Society of Chemical Industry: London, 1974; pp 2075-2087.
Received for review May 18,1993. Revised manuscript received November 15, 1993. Accepted November 18, 1993.'
@
Abstract published in Advance ACS Abstracts, January 1,1994.
