Document 6bz2a4dDKKvwpY30zk64qMo43

p* ni U I > OZONE/UV OXIDATION OF CHLORINATED COMPOUNDS IN WATER* By H. William Prengle, Jr., Charles E. Mauk, and Jack E. Payne Houston Research, Inc., 10600 Shadow Wood Drive, Suite 211 Houston, Texas 77043 ABSTRACT The presence of chlorinated compounds in source water as possible carcinogens Is of considerable current Interest; therefore, an examination of the oxidation of these corn** pounds is pertinent. Preliminary work on the ozone and ozone/UV oxidation rates, mechanisms, and Intermediate and final products for: pentachlorophenol, dichlorobenzene, dichlorobutane, chloroform, and polychlorinated biphenyls is presented. Oxidation rates are substantially enhanced by ozone/UV, as compared to ozone alone, and can be made to go approximately stolchlometrlcally. Analytical methods were used to track the disappearance of the compounds, ap pearance of intermediates, and the production of chloride ion and chlorine. For compounds with large UV absorption the chlorine atoms are released very rapidly to form chloride ion; whereas for low level UV absorbers dechlori nation occurs by oxidation releasing the atoms as chlorine and thence to hypochlorous acid. Pentachlorophenol is a strong UV absorber and releases chlorine by the former mechanism; whereas chloroform is a weaker absorber and both dechlorination mechanisms occur. Presented at the International Ozone Institute Forum on Ozone Disinfection; Chicago, Illinois, June 2-4, 1976. HONS 067324 (For Reference Only) INTRODUCTION CHARACTERIZATION OF COMPOUNDS OXIDATION OF THE INDIVIDUAL COMPOUNDS CONCLUSIONS LITERATURE CITED ' ACKNOWLEDGMENT ABOUT'. THE AUTHORS 2 3 6 19 20 21 21 Table 1. Table 2. RFI Values for Compounds Molecular Characteristics ' Figure 1. Pentachlorophenol UV-Spectrogram Figure 2. Destruction of Pentachlorophenol TOC Figure 3. ' Related Time Profiles -- Oxidation of : Pentachlorophenol Figure 4. Destruction of 1. Dichlorobenzene; i- 2. Dichlorobutane . I; , Figure 5. Destruction of Polychlorinated Biphenyls Figure 6. Destruction of Chloroform TOC . i _i 1 ! ' HONS 067323 .I I INTRODUCTION As s result of^ substantial recent Interest In the presence of chlorinated compounds, as possible or suspected carci nogens, in source waters and municipal water supplies^^ , Houston Research, Inc. Is currently Investigating the oxida tion of a number of chlorinated compounds by ozone, and ozone or oxygen with UV radiation. The work is a continua tion of previously reported work^'^'^ on the development of an advanced chemical oxidation water treatment system for the destruction of hazardous and refractory materials. The work reported herein examined oxidation rates, under a wide variety of conditions, and reaction mechanisms and products, for the following five compounds; Pentachlorophenol O-dichlorobenzene Dichlorobutane Chloroform ' Polychlorinated Biphenyls These were specifically chosen to Insure a wide spread of organic molecular types which may be encountered. A crucial aspect of the work relates to the question of what are the subsequent products as the oxidation proceeds from the initial species to completion? Do the Intermediates represent a greater potential hazard, and in particular, what happens to the chlorine atoms? In general, none of the five compounds are oxidized to the desired lower limits of concentration by ozone treatment alone, but require more severe treatment, viz ozone with UV. HONS 067326 3 CHARACTERIZATION OF COMPOUNDS Previously, ve have established a quantitative scale for refractory cospounds^^; the refractory index (RFI) was defined vhich measures the difficulty of oxidation of a given compound by ozone. Table 1 presents the latest list and indicates the five compounds relative to others that have been studied. It will be noted that the five compounds range from slightly refractory to highly refractory -- chloroform being the least refractory and polychlorinated biphenyls highly refractory. ' As a result of recent work, the correlation of the amount and the spectral distribution of the UV absorption, over the range 180-450 nm, with reaction characteristics and rates is emerging. Consequently, such information is Important in determining the particular UV radiators to be used in the reactor design. Table 2 presents some molecular characteristics of the compounds related to this matter. It will be noted that the Integrated specific absorbance values range from 10+'*' to 10+^ cm^/millimoles, and have peaks generally, at 180-240, 240-290, and 290-360 nm, which approxi mately correspond to the radiation ranges of commercially available UV radiators. MGNS 06732 7 4 ' TABLE 1. RFI VALUES FOR COMPOUNDS Compound .. 1) Potassium cyanide (KCN) 2) Phenol 3) Chloroform 4) Color (a-units) 5) Complexed Cd-cyanlde RFI* Value 0.41 0.44 0.53 0.66 0.96 Qualitative Scale Slightly Refractory (RFI < 1.0) - !1 6) Pentachlorophenol 7) Ammonium ion ~ 8) Simulated Medical Waste 9) Glycine 10) Palmitic Acid (NH^Salt) 11) Dlchlorobutane 1.6 8. 13. 19.7 27.3 56. Refractory (RFI - 1 -* 100) 12) Glycerol , 13) O-dichlorobenrene 14) Polvchlorinated Biphenyls 15) Ethanol 16) Complexed Ferrlcyanlde 112 113 200 (eat.) 245 270 Highly Refractory (RFI - 100 - 1000) -` . 17) Acetic Acid >1000 Very Highly Refractory (RFI > 1000) , *RFI 5 Bc Cl/2 a Ao ,_o _ . ; [B = c 0, 3 su pplied to t l/7> tl/7 5 tine for 1/2 conversion of A o;' A o = initial amount of compound.] MOMS 067328 i; r [ Compound 1) Phenol 2) fentachlorophenol 3) O-dlchlorobenzene 4) Dichlorobutane 5) Chloroform 6) Polychlorinated Biphenyls TABLE 2. MOLECULAR CHARACTERISTICS Structure C H OH CjCljOH C6H4C12 ckch2)4 Cl CH Clj C.H Cl.C.H Cl, 6` 56 c d Molecular Height Solubility,mg/t 94 *8.2xlC*(15*C) 266.5 80. 147 12 7 i 119.5 8.2x10^(20*0 321 5.6xl0~2(20*C) 0V-Specijlc Absorbance (cm /m mol) (\, nm) 1) 1.805xl0+3(180-240 nm) 2) 0. 543xIO+3(240-290 nm) 1) 41.5xl03(200-236 no) 2) 17.8x103(236-275 na) 3) 5.65x103(275-355 nm) 1) 1.70x103(180-240 nm) 2) 0.0419x10 (240-285 nm) 1) 0,203xl03(180-220 nm) 2) 0.0165x103(220-280 nm) 1) 0.058x102(180-220 nm) 2) 0.0028x10^(220-270 nm) 3) 0.018x103(270-350 nm) 1) 58x103(180-250 hi) HONS 0 6 7 3 2 9 6 ( r . ' '1 ' * |' OXIDATION OF THE INDIVIDUAL COMPOUNDS The experimental ozonation runs were carried out in stirred batch* continuously sparged gas* liquid phase reactors. The equipment and experimental procedures have been des cribed previously (4 * 5) . The ultraviolet input (UV) levels referred to are given as watts of useful UV light per liter. A stoichiometric limiting line represents the theoretical maximum rate at which a compound could be oxidized for the particular 0^ rate to the reactor. ( j 1 CC analyses were made to follow the disappearnce of the original species; TOC analyses to follow the destruction of Intermediate oxidation products; and turblmetrlc chloride analyses to determine whether the chlorine atoms went to chloride or some other form. The pH measurements followed the* formation and destruction of organic acids and the perma nent formation of hydrochloric and hypochlorous acids. ' Pentachlorophenol (PCP) r \ ^ Pentachlorphenol shows strong absorption of UV over a wide I range of wavelength^ as Indicated by Figure 1. This makes i It very susceptible to oxidation by ozone and typically !> the compound, as determined by gas chromatographic analysis, has reacted In less than 15 minutes residence time. The PCP ,' was charged to the reactor at a concentration level of ;70 mg/1. The further oxidation of intermediate oxidation Ip products, as Indicated by total organic carbon (TOC) analy sis, takes place less rapidly as shown by Figure 2. With the proper amount of UV input, the oxidation to completion _ proceeds essentially stoichiometrically. P MOMS 067330 l 7 A biorbane* FIGURE I. Pentachlorophenol UV-Spectrogram (7 .0 mg/t) MGNS 067331 T 8- 0 30 60 90 120 150 Time, Minutes FIGURE 2. Destruction of Pentachlorophenol TOC (initial PCP, 70 mg/1) HONS 067332 9 Beca use 0 f s1mllar cons lde r f i rst the oxld ati on 1 s first ale 1c ac id t OH o Oxygen o Oxygen 0 .COH c xCOH Bailey,(7')' has reviewed the ozone oxidation of phenol, and he concludes that the major pathway is through catechol to auconic acid. Since oxalic acid is found as an intermediate oxidation product. It appears that the ozone attacks muconlc acid at the double bonds, Following the carbon structure only, this gives. OH PHJJj OH 00 ni HOC-COH CO, As evidenced by TOC, phenol Is converted to CO^ by ozone-UV at essentially the stoichiometric rate. Because much of the conversion of pentachlorophenol to CO is also at essentially the stoichlometrlcally limited rate, there is justification for assuming a similar oxidation mechanism. But in addition, the pentachlorophenol by gas chromatograph ic analysis disappears much faster than does the TOC; at the same time, the pH of the water drops drastically and a chloride concentration corresponding to more than half the stoichiometric amount suddenly appears and then slowly Increases. With UV alone, the pentachlorophenol disappear ance, PH drop, and chloride appearance is extensive, but much slower than with ozone-UV; with UV alone. HONS 067333 10 - TOC disappearance is small. loss caused by UV, This implies an Initial chlorine OH Cl, Cl Cl '^x'Cl Cl hv h20 OH HOi^JiOH HoW^OH 5H+ + 5 Cl OH (Catechol Structure) followed by oxidation of the ring, OH HO|^ OH HoU^OH OH 00 HO COH 0, H II hoU^oh --HOC-COH --- CO, + H,0 (Oxalic Acid) 22 OH V * (Huconlc Acid Structure) It appears unlikely that the chlorine Is released from the ring as free chlorine which would produce hypochlorous acid In aqueous medium; whereas analyses of the apparent ozone content of the water gives no indication of transient hypo chlorite. Confirmation of this mechanism can be made by examining the time profiles in Figure 3. As the UV rapidly removes the chlorine atoms from the compounds, the PCP drops to zero very rapidly, and the chloride ion concentration rapidly reaches a plateau. The pH drops rapidly from 9.6 because of the formation of HCl and phenolic compounds of an acidic nature. Then the pH drops more slowly as the phenolics are converted to the more acidic organic acids. As the TOC shows the organic acids being destroyed, the pH rises to a plateau near 5, resulting only from the unoxldlzable HCl. Dlchlorobcnzene (DCB2) O-dlchlorobenzene is a strong absorber of UV, and by GC analysis disappears rapidly. When UV is used with ozone, L HOAS 067334 11 FIGURE 3. Related Time Profiles,- Oxidation of Pentachlorophenol. * (Re: Figure 2.) L HONS 067335 12 the chloride jumps to a value corresponding to about half of the chlorine in the molecule; suggesting an oxidation mechanism similar to pentachlorophenol, DCBZ was charged to the reactor at the 100 mg/ level, and by GC analysis was 957 destroyed in the first 15 minutes; destruction was complete by 30 minutes. The destruction of TOC from intermediate oxidation products was not as rapid, as shown in Figure 4, Dichlorobutane (DCB) With osone alone, the oxidation of 1,4 dichlorobutane proceeds relatively slowly and almost no chloride is formed. The molecule has a relatively low absorbance of UV, but when UV is used with ozone, the disappearance by GC Is twice as fast, and'- a stoichiometric production of chloride is obtained, suggesting the oxidation mechanism to be, C1(CH2)a Cl jpy* HC1 + H0(CH2)a OH -----HOOC (CH2> 2COOH ^ (dlhydroxybutane) (Succinic Acid) 3 3 . ----- HOOC-COOH -------------- *-+ C02 + H20 (Oxalic Acid) Dichlorobutane was charged to the reactor at the 50 mg/ level, and by GC analysis was destroyed with 0^-UV by 50Z at 15 minutes, and completely by 60 minutes. With 0^ alone, more than 60 minutes for 50Z destruction and more than 240 K0KS 067336 13 FIGURE 4. Destruction of 1. Dichlorobenzene (initial, 100 mg/i) . 2. Dichlorobutane (Initial, 50 mg/l). HONS 067337 14 minutes for complete destruction Is required. Destruction of TOC from dichlorobutane and intermediate oxidation products proceeds less rapidly, as shown In Figure 4. Polychlorinated Biphenyl The polychlorinated biphenyl (PCB) used was Aroclor 1254, essentially a mixture of isomers of pentachlorobiphenyl. In the environment, PCB degrades via successive dechlori nation steps before the biphenyl structure degrades, by a mechanism which can be postulated to be similar to that of pentachlorophenol, "~ Cl XX XX 03.hv X OH hv HC1 +X' H Ho/'roa XX H0\Al OH ^ (Subs. XX 0 XX q (Subs. Catechol) II tl HOC-COH (Oxalic Acid) CO 1 Muconlc Acid) In the above structures, for the commercial product, X Is undetermined and may be H, Cl, or OH. Figure 5 shows the disappearance of PCB by GC analysis. The original charge was at the 0.5 mg/i PCB level added to the water as a solution in methanol. The resulting dispersion in water was at about eight times the solubility of PCB in water. Since only the PCB In solution was being treated, the curve is also controlled by the rate at which'the PCB ie going into solution; even so, it will be noted that the rate of disappearance was fairly rapid! HONS 067338 15 FIGURE 5. Destruction of Polychlorinated Biphenyls (PCB) (initial, 0.5 mg/t) L HONS 067339 16 Chloroform Chloroform was charged to the reactor at the 600 tag/f level, and the disappearance was followed by GC. Using Oj-UV, disappearance was 50X in 60 minutes and 60Z in 120 minutes. Using 0^, disappearance was 50% In 75 minutes, and 70% In 120 minutes. Because of the volatility of the compound, a large portion of the loss results from the sweeping of the solution by the flow of gas'. It Is estimated that at least one fourth of the original charge was destroyed by O^-UV because the chloride content In solution rises to a level equivalent about a fourth of the chlorine in the original charge. The disappearance of TOC from the original compound and intermediate oxidation products is almost as rapid aa the disappearance of the chloroform. Implying few and fast intermediate oxidation steps. TOC disappearance Is shown in figure 6. The classical oxidation mechanisms for chloroform are <s>. -HO HCC13 olHiv> HC2C13 C13CH0 + H22 and for air with or without light^' , hcci3 ------- bco2ci3 ------- ci2 + co2 + HC1 ^ C0C12 + HC1 + 1/2 02 the carbonyl chloride (phosgene) decomposing thermally by. P HONS 067340 U I\ 17 - FIGURE 6. Time, Minutes Destruction of Chloroform TOC `(Initial chlororm, 600 mg/1) HONS 067341 18 C0C1 or by hydrolysis in water. coci2 + h2o CO + Cl 2 COj + 2HC1 For oxidation with O^-UV, chloroform by GC does not disappear abruptly, but Instead at a rate comparable to.the disappear ance of TOC. Unlike the aromatics, chloroform is not a strong absorber of UV, but without UV little chloride is formed, and with UV, only about a quarter of the chlorine atoms go to chloride; suggesting the mechanism to be. C12C0C1 H ^hv coci2 ++ U+ + Cl co2 + ci2 The question of whether the presence of phosgene represents ' a potential hazard arises. The compound is slightly soluble in water and hydrolyzes slowly. At low initial concentra tions of chloroform, it is unlikely that the solubility will be exceeded, and In the presence of ozone and UV, the compound should react very rapidly. MQMS 067342 19 - CONCLUSIONS Based on the preliminary investigation of oxidation of the subject compounds, it appears that: 1. Oj-UV oxidation is highly effective for chlorinated compounds in water; the original and partial oxidation species completely oxidize to stable forms, such as Cl", Cl2, C02, and H2<). 2. UV absorption produces activated species, free radicals excited state species, etc., from the organic compounds as veil as from ozone. UV absorption correlates with reactivity and reaction rate, e.g. aromatics absorb ore UV and are more reactive than aliphatic compounds. 3. Dechlorination occurs primarily by two mechanisms lead ing to: a) chloride ion, and b) chlorine ahd thence to hypochlorous acid. When UV absorption is substantial, initial rapid dechlorination occurs to form chloride ion, viz. pentachlorophenol, dichlorobenzene, dichlorobutane, and polychlorinated biphenyls. When UV absorp tion Is small, initially oxidation occurs setting free chlorine. The latter occurs to a lessor degree, but some compounds exhibit both mechanisms, viz. chloroform 4. As indicated previously^\ increased UV input is more effective than elevated temperature. 3, Performance of the Ozone/UV process, in a properly designed reaction system, is unequaled in accomplishing essentially complete removal of chlorinated compound from water; resulting in simple Innocuous species. HONS 067343 - 20- LITERATURE CITED 1. Beliar, T. A., Llchtenberg, J. J., Kroner, R. C., "The Occurrence of Organohalides in Chlorinaced Drinking Water," EPA-670/4-74-008, NERC, Cincinnati, Ohio, November 1974. 2. Stevens, A. A., Slocum, C., Seeger, D., Robeck, C.; "Chlorination of Organics in Drinking Water," pre sented at conference on Environmental Impact of Water Chlorination, at Oak Ridge National Laboratory, October 1975. 3. Coleman, E., Llngg, R., Melton, R., Kopfler, F.; "Occurrence of Volatile Organics in Five Drinking Water Supplies Using GCHS," presented at the First Chemical Conference of the North American Continent at Mexico City, December 1975. 4. Garrison, R. L., Mauk, C. E., Prengle, H. W., Jr.; "Advanced 0,-0xidatlon System for Complexed Cyanides," ProceedingsJof the First International Symposium, International Ozone Institute, p 551 ff, (1975). 5. Prengle, H. W., Jr., Heves, C. G., Ill, Mauk, C. E.; "Oxidation of Refractory Materials by Ozone with . Ultraviolet Radiation," presented at the 2nd Inter national Ozone Symposium, International Ozone Institute, Montreal, Canada, May 1975. 6. Prengle, H. W., Jr., Mauk, C. E., Legan, R. W., Hewes, C. G. , ,111; "Ozone/UV Process, Effective Waste water Treatment," Hydrocarbon Processing, 5i4/10), 82 (October 1975). 7. Bailey, P. S. , "Organic Groupings Reactive Toward Ozone Mechanisms in Aqueous Media," in "Ozone in Water and Wastewater Treatment," ed. by F. L. Evans, III, Ann Arbor Science (1972). 8. Neu, R., "Auto-oxidation of Chloroform," Pharmazie, 1, 251 (1948). ' 9. Chapman, A. T., "Peroxidation of Chloroform," J. Am. Chem. Soc. JW, 419 (1935). 10. Kawal, S.,""Decomposition of Chloroform," Takugaku . Zasshl >86 (12), 1125 (1966). HONS 067344 21 - ACKNOWLEDGMENT Data and discussion on pentachlorophenol were developed under U. S. Army Contract DAAG53-76-C-0089 with Mobility Equipment Research and Development Command (MERADCOM). Ultraviolet scans were made by Dr, R. A. Geanangel of the' University of Houston Chemistry Department. ABOUT THE AUTHORS 1. H. William Prengle, Jr. is a senior chemical process specialist with Houston Research, Inc., Professor of Chemical Engineering at the University of Houston, and a registered professional engineer. He received BS BS, MS, and D.Sc degrees from Carnegie Mellon University, and conducts research on Kinetics of Chemical Reactions In Liquid Phase, Thermodynamics of Solutions, and Remote Sensing of Pollutants from Emission Sources. He is a member of AIChE, ACS, WPCF, APCA and a'number of hono rary societies. 2. Charles E. Mauk is Manager of the Special Projects Section of Houston Research, Inc., and has been princi pal investigator for more than 12 years on a wide va riety of environmental research and development and aerospace projects. A registered professional engineer in Texas, he has BS, MS, and PhD degrees in Chemical Engineering from the University of Houston. He is a member, of ACS, AIChE, NSPE and Sigma Xi. 3. Jack E. Payne, Manager of Technical Services of Houston . Research, Inc., has 15 years of Industrial experience In technical and analytical services in chemical pro duction, research in chemical and photochemical reac tions, environmental monitoring and control, and industrial hygiene. He received a BS Chemistry degree from University of Texas, and did graduate work at Texas A&I and University of Texas. He is a member of Alpha Chi Sigma and the ASTM Committee on Standards for Dissolved Ozone. MGNS 067345