SEPA United States Environmental Protection ‘Agency Research and Development Risk Reduction Engineering Laboratory Cincinnati, OH 45268 EPA(GOO/S2-91/046 Jan, 1962 Project Summary Bench-Scale Evaluation of Ammonia Removal from Wastewater by Steam Stripping G.B. Wickramanayake, D.P. Evers, J.A. Kittel, and A. Gavaskar irpose of the study summa- rized here was to generate laboratory data to support the development of ‘wastewater discharge standards for am- monia In nonferrous metal winning pro- cesses. Ammonia removal effected by steam stripping from synthetically com- pounded “wastewater” samples was studied to determine the Importance of factors affecting the rate of removal ‘and the degree of removal ‘The analyses of estimated Henry's Law constant and changes in ammonia solubilities indicated that adding caus- tic, as compared with adding slaked lime, resulted in higher Henry's Law constants and lower solubilities for the three waste streams studied. Although no significant variation of mass trans- {for coefficient (K) was observed when SO," concentrations were varied from '5,060 to 20,000 mg/L, K was the high- est for low SO,” wastewaters when pH was adjusted with NaOH. Results of the steam stripping study Indicated that varying chemical con- stituents such as SO," and the molal strength did not significantly affect the efficiency of ammonia removal. Pre- heating wastewater and operating the ‘stripping tower at high steam-to-waste- water flowrate ratios such as 4 Ibigal achieved higher removals (98.9% of more). Based on engineering unit pro- cass and operation requirements, the cost analysis indicated that lime ‘may be more economical than caustic for pH adjustment depending on waste ‘sludge characteristics and disposal re- quirement. This Project Summary was developed by EPA's Risk Reduction Engineering Laboratory, Cincinnatl, OH, to announce key findings of the research project that Is fully documented In a separate report of the same title (soe Project ‘Report ordering information at back). Introduction Extracting motal values from some ores requires the use of hydrometallurgical tech- niques that employ ammoniacallixiviants. The metal values are recovered from the Pregnant liquors, leaving an aqueous ‘wastewator residual high in dissolved sol- ids and ammonia, This wastewater ro- ‘quires treatment for removal of ammonia as well as dissolved solids to meet the discharge standards required under the Effluent Guidelines for the nonferrous met- ‘als industry. “Tho practical methods of removing am- monia-ritrogen from wastewaters include biological nitrification-denitrification, breakpoint chlorination, evaporation, re- fon exchange, air strip- ping, and steam stripping. The purpose of this study was to determine the useful- ‘oss of steam stripping for the removal of ‘ammonia from wastewaters generated in the metal smelting and refining industry. Approach ‘The overall objective was pursued as theo discrete tasks. Task 1 involved theo ralical and laboratory studs to determine the effec of wastewater compostion on ultram of ammonia (gas) and on water and gas-liquid mass tranelor rates, especially within the temperature range of B} Primo on Recycled PagerInterost in actual plant operations. In Task 2, laboratory tests wore performed in a ‘banch-scalo steam stripping apparatus pr cessing two representative synthetic wastewaters with pH adjustment by other 18 oF eaustic. Task 3 Involved the osti- ‘mation of the capital and operating costs associated with tho pH adjustment meth- ‘ods and the handling of the waste sludge formed. Disposal costs were (Ono objctho of tho fist task was to Investigate how the distribution of inor- Gani apectes and ‘changes in tempera: Ao et he aquirium ct ammonia oss) and io wactowatory Efecto aitront stoctgylos euch ao Har, Mgr Gar, SO; and Gen thotlonys Law conetant (1) tore ovata ie thooralical ety was ased on infomation gen in tho pub lished literature. he socond sspoct of his task ivolod cxpormonts to sady tho effec of varying 85. concentration” moll strength, and pit adsl mati [NaOH or Ga(0F), En the geo Wanslr tales (or ammonia Synths wastowater steam A {Table 1) was dosignod to havo a high SO fvel and molal strength, Waste stream & has @ tower 80," lovel than A, whore 38 both A and C have the same SO," level. Waste stroams B and C have the same molal strength and were brought to such condi- tions by adjusting Cl" concentrations. The ‘oxporiments were conducted in identical, ‘completely mixod batch reactors at tem: peratures near 90°C, The variation of ammonia concentration with time was stud- {ed after adjusting solution pH to 12 with NaOH or Ca(OH). Duting Task 2, a laboratory-scale steam stripping unit was designed and con- structed (Figure 1). The pH of two differ- ‘ont synthetic wasto streams, B and C (Table 1), was adjusted to 11.5 or greater ‘with eithor NaOH or CaO. Ammonia re- ‘moval was studied for the different pH adjustment methods and different steam- to-wastewater flowrates. The system was designed to pump a high pH synthetic wastewater influant to the top of the packing material in tho col- umn casing. As shown in Figure 1, the ‘wastewater flowed down over the packing material and exited through the offluent line at the column base. Steam was in- jacted into the column at the base of the packing material As the wastewater passed through the Jet of steam, it was heated to 100°C. Flowrates for the steam and/or the influent wastewater were var- ied to dotermine the differences in ammo- fla removal efficiency. The steam flowrate ‘was measured by condensing steam en- ‘ering the column and measuring changes in the flowrate of column offluent. The steam flow entering the column was con- ‘rolled using a valve/orifice unit. ‘A total of rine steam stripping experi- ments ware porformed to explore tho of fects of sulfate concentration and total ‘molal strength and to compare the use of NaOH (caustic) or CaO (lime) for pH ad- justment of the solutions. The extent of ‘ammonia removal was studied for differ- ‘ent steam-to-wastewator flowrates. In Task 3, the costs for chomical addi- tion, sludge removal, and thickening woro ‘estimated for four differant representative waste streams and for a varity of wasto- Water flowrates. Included in these esti- mates were costs for mixing facilities, chemicals, and sludge removal concen- trations, handling, and transport. Not in- cluded were cosis for chemical storage faciities, buildings, land, or sludge dis- posal—all of which are likely to vary con- Elderably from site to site, Results and Discussion Task 1: Theoretical and Laboratory Studies on the Equilibrium and Mass Transfer of Ammonia In Wastewater ‘The analyses of the estimated Henry's Law constant and changes in solubilities cof ammonia indicated that adding caustic, ‘28 compared with adding slaked limo, uted in higher Henry's Law constants ‘Table 1. Wastowator Charactoristcs for Gas Transfer Rate Experiments eee eee Commitee) Ghomical Specios Wastowater A NHN Mg? 5,000 200 Nar 19,927 sor 20,000 or 28,000 Wastewater B 5,000 200 11,999 5,000 28,000 5000 200 20,000 and fowar solubilities for the three waste streams considered. These effects can be attributed to tho relatively high ionic strength found in wastewaters when pH was adjusted with NaOH. For solutions B ‘and C, where the sulfate concentration ‘was varied without changing the solution Strongth (total number of moles), no sig- nificant difference in Henry's Law con- stant or solubilty could be seen for eithor ff the two pH adjustment methods. The highost Henry's Law constant and corr sponding lowest ammonia solubility were ‘observed In solution A, which had the highest molal strangth. Experimental studios to determine tho effects of dissolved spacies and pH ad- justment method on the mass transfer co- efficient (K) Indicated that those effects were relatively low and the maximum changes in mass transfer cocfficiont did ot exceed 25%, In two of the throo different solutions, however, the mass transfer coefficient was higher for the wasto sream where pH was adjusted using caus tic. For solutions with approximately the ‘same SO," concentration but differant mo- lal strength, the mass transfer coefficients ‘were comparablo when NaOH was used as the pH adjustment method. For Ca(OH), however, the K values were com- parable“for those where both the molal Strength and SO," level were diferent The overall analysts of data indicatod that, when compared with lime, addition of NaOH promoted ammonia removal for so- lutions with relatively low SO,- levels (6,000 mg/l). Task 2: Ammonia Removal Studies Using Steam Stripping Unit ‘The results of the steam stripping study ‘are summarized in Table 2. The data show that when the steam-to-wastewater flowrates were low (1.3 Ib/gal), the ammo- nla removal efficiency was as low as 99%. By increasing steam-to-wastewater flowrates to 3.8 lbigal, romovals of more than 99.9% were observed. The removal ‘ficiency in one study was improved by 2 percentage points when the temperature ‘of the influent waste stream was raised by ‘about 20°C. The addition of ime increased the temperature more than 10°C because hydration of lime is an exothermic pro- For wastewater with low initial SO," lovel (6,000 mgf.), ammonia removal was 3 percentage points higher when pH was ‘adjusted with caustic rather than wit lime. ‘Those observations agreed with the trends predicted from the estimated Henry's Law Constant and mass transfer coefficient. In~——>— steam out Heated Wastewater KRAAAAL -O a wot P ecine| cmon Sar Flowmeter Packing Materials 700 gal Wastowaler ator Tank Pump Main Steam Supply ‘Figure 1. Experimental setup fr ammonia removal by steam sipping. Table 2. Summary Results of Laboratory Study on Steam Stripping Wastewater Temp. °C pH Steam to ‘Adjistmont Column Column Wastowatar Ratio NH,-Neone, mgl Average NH, Wastewater Method __Influent __Effluent otgal) Tifloent Effluent Removal, % B(lowS0;) NaOH 26 101 19 5200 32-102 99.4 NaOH 26 tor 13 5.200 255-420 93.1 ‘C20 38 tor 19 5.100 188-288 953 . Gao 38 101 38 5,000 1216 99.7 (high $0) NaOH 26 tot 19 4750 121-198 26.9 NaOH 26 to1 38 4700 39-45 99.91 NaOH a7 11 19 $1950 20-80 98.8 Gao 39 401 19 3.950 87-92 g77 a0 39 tot 38 3.825 14-22 99.96 3 {kus covenNMeNT pmnTING OFFICE: 92 -casomunytho oxporiments conducted with wastewa- tara using highor initial SO," levels (20,000 mg/l), ammonia removal was slightly higher whon pH was adjusted with lime Instead of with caustic. These obser tions agrood with the conclusions reach from the corresponding mass transfor rat studies, which did not agree with the the Totical estimates of solubilities based on Honty’s Law constant. In summary, more than 99.9% of am- monla was removed by introducing high ‘stoam-to-wastewatorflowrates, such as 3.8 Ibigal. Varying chemical constituents such as SO," and the molal strength had only & litle effect on net NH, removal. Proho: Ing wastewaters and operating tho str ping tower at high temperatures by i ‘creasing tho stoam-to-wastewater flowrate ‘allo resulted in higha¥ rerroval efficierices. Task 3: Engineering Cost Estimates “The cost estimates for the chemicals and equipme ‘ammonia-bear {er botore stripping showed that limo can ‘be more economical than caustic. In ad- dition, tha most cost-effective method for disposal ofthe sludge solids generated is dewatering ina lagoon flowed by landfil Sisposal of the solds. Thore may be a Siiforont Set of cost-effective processes, however, when costs for land, transport, and handing ot age avanites of sludge are hig ‘Summary and Conclusions The analyses of estimated Henry's Law ‘constant and changes in. ammonia solu: bilties indicated that adding caustic rather 6G. B. Wickramanayake, D.P. Evers, J.A. Kittol, and A.R. Gavaskar are with Battelle ‘Mamorial Insiitue, Columbus, OH 43201-2693, ohn Bureklo is the EPA Project Officer see below). ‘The complete repar, entitled "Bench-Scale Evaluation of Ammonia Removal from Wastewater by Stoam Stripping, (Order No. PB91- 234 633/AS; Cost: $26.00; subject fo change) will be available only from: ‘National Technical information Service 5285 Port Royal Road Springfield, VA 22161 Telephone: 703-487-4650 The EPA Project Officer can be contacted at: ‘lisk Reduction Engineering Laboratory U.S. Environmental Protection Agency Cincinnati, OH 45268 than staked limo, would resutt in higher Henry's Law constants and lower solubi ties for the threo waste streams studied. ‘Although no significant variation of the mass transfor coefficient (K) was observed when SO," concentrations were varied from 5,000 to 20,000 mg/L, K was the highest for low SO," wastewaters. wh pH was adjusted with NaOH. Results of the steam stripping study indicated that varying the chemical con- stituents such as SO, and the molal strength did significantly atfoct 1 cioncy of ammonia removal, More amm: nia was removed (89.9% or moro) by pre- heating wastewater and by operating the stripping tower at high steam-to-wastewa- ter flowrates such as 4 Ib/gal. Based on ‘engineering unit process and operation requirements, the cost analysis indicated that lime may be more economical than caustic for pH adjustment. ‘The full report was submitted in ful ment of Contract No. 68-03-3248 by the ‘Columbus Division of Battelle Memorial Institute under the sponsorship of the U.S. Environmental Protection Agency. BULK RATE United States Center for Environmental POSTAGE & FEES PAID Environmental Protection Research Information EPA ‘Agency Cincinnati, OH 45268 PERMIT No. G-35 Official Business Ponatty for Private Uso $300 EPAVG00/S2-91/046