Document MQep2wMKb6dJZbgEK6Q0DemV

1 REFERENCE 9------- * PROGRESS RtPORT ORGANIC CHEMICALS DIVISION TECHNOLOGY DEPARTMENT row OCT AILS ftSftOH> NO. OATS Nell E, S. Thompson (D. K. Lynch) Aprl1-May 1972 Yes 7309003 6/13/72 1 VOCATION W. a. Krummrlch 7309003 rNOOl IHt III.- w; I Hi WGK LABORATORY PlEASC RETURN TMi dMWMM to the property of MONSANTO COMPANY. It oontclm CONFIDENTIAL INFORMATION whtoh mmt not to* _____ rxoUd 10 wmtoorlud pmom or tent outMOi the Company without proper authorization, the recipient`--rTii Mflinpiiniiii ______________________ DISTRIBUTION K. E. Boucher - WOK J. J. Burbage - AP95S 0. DeGarmo - T3B W. C. Engman - WGK 0. A. Hlppe - WGK WITH DETAILS *L. H. Horcher - WGK *H. J. Horner - WGK A. E. Lelfey - WGK *R. M. McCutchan - WGK W. R. Richard - T3A J. P. Quinn - T3B J. R. Savage - B2SL *E. E. Stewart - WOK Q. E. Thompson - T3E OnOJiCT TITLl ' Dept. 246 Aroclors. PCB Removal from HC1 Off Gas. MO.tCT Oi.BCVIVI Develop a process that will permit reduction of all chlorobiphenyls (mono poly) In the HC1 off-gas from the Aroclor Department such that when all the gas Is sewered the total chloroblphenyl content will be less than 2.5 lb/day. Zero lb/day would be preferred. Summary Both 1016 bottoms and chlorosulfonic acid have been found to be effective scrubbing liquids for the removal of biphenyl and chlorobiphenyls from Dept. 246 (Aroclors) off-gas HC1. By employing a 10 or 15-tray Olderahaw column and scrubbing _HC1 gas containing biphenyl and chlorobiphenyls at levels equivalent to the off gas from production units the total detectable polychloroblphenyl (PCB) found was well below the level required to meet the project objective. Of the two scrubbing liquids, the chlorosulfonic acid affords us a lower PCB level In the scrubbed gas. However, this potential benefit Is offset by the high cost of equipment and disposal problem associated with the used C/S acid. The use of 1016 bottoms on the other hand will require less expensive equipment, Is compatible with the Aroclor process and could provide an opportunity for some yield Improvement. No disposal problems are anticipated because the used 1016 scrubber can be further chlorinated to make Aroclor 1254 or 1260. - With appropriate carbon treatment, Industrial grade muriatic acid made with this scrubbed gas should meet the proposed PCB limit of 10 ppb. ft M A* V . o HONS 058293 Job No. 7309003 Page 2 June 13, 1972 Summary (Cont'd) Engineering data required to design a suitable scrubbing system using either C/S acid or 1016 bottoms was also generated during thlB study and Is communicated In the details section. II. Recommendation One practical way It appears of reducing the chloroblphenyl content in HC1 off-gas to meet environmental pollution standards and also provide an HC1 feed source for the adiabatic absorber Is to scrub the gas with either ehlorosulfonle acid or 1016 bottoms. It Is recommended therefore, that a scrubbing system be designed and Installed using one of these two materials as the scrubbing liquid. Nell E. S. Thompson Plant Process Research HONS 05829* Table of Contents I. Summary Page 1-2 II. Recommendation 2 III. Details A. Introduction 3 B. Experimental Work 1. HC1 gas Scrubbing Setup and Procedure 3-*t 2. Analytical Methods 4 C. Experimental Results and Discussion 1. Screening of Scrubbing Liquids 5 2. Chlorosulfonlc Acid Scrubber 5-6 3. 1016 Bottoms Scrubber 6-7 IV. References 7 HONS 058295 Job No. 7309003 Page 3 June 13, 1972 III. Details A. Introduction The HC1 off-gas from Department 246 (Aroclors) presently contains significant amounts of biphenyl, monoohloroblphenyls and polyohloroblphenyls (PCB). Calculation and laboratory analyses of sewer samples Indicate that If all the by-product HC1 were sewered the total PCB losses could be 15-30 lbs/day. Because of recent unfavorable reports on the negative ecological effects of PCB's and also from the standpoint of waste water treatment here at the Plant It Is a must to reduce the PCB losses to less than 2.5 lb/day. Zero lb/day are preferred. At the same time, however, it Is Important that whichever approach is undertaken to remove the PCB's the following faotors be given primary consideration: a. The process must permit the reduction of PCB in the HC1 gas to extremely low levels^<40 ppm. b. The process must be compatible with Dept. 217 process and the finished goods, chlorosulfonlc acid. c. Scrubbed HC1 gas be suitable as a feed source for the adiabatic absorber to produce industrial grade muriatic acid. In this study we plan to take advantage of the extremely low vapor pressures of the PCB's and thereby remove them from the gas stream by some kind of scrubbing system. B. Experimental Work 1. HC1 gas Scrubbing Setup and Procedure A diagramatlc sketch of the experimental gas scrubbing system is shown in Figure 1. Scrubbing experiments were carried out using 5, 10 and 15-tray Oldershaw columns. Hydrogen Chloride (HC1) gas was saturated with biphenyl and chlorinated biphenyls by scrubbing with MCS 1109 and off gas condensate at 50-60C and 45-50C respectively. The composition of these two scrubbing liquids are shown below: ' B10 0-CBP M-CBP P-CBP D1-CBP/II1B WCS 1109 2528..2640** 4.03* 14.66* 0.47* 0ff-0aB Condensate 24.74* 22.63* 171..2692** 33.72* HOMS 058296 Job No. 7309003 Page 4 June 13, 1972 Experimental Work (Cont'd) The composition of the organics In the gas scrubbed through MCS 1109 averages about 400 ppm biphenyl, 690 ppm monochloroblphenyl and <1 ppm polychlorobiphenyl (PCB). HC1 scrubbed with off-gas condensate contains about 250 ppm biphenyl, 107 ppm monochloroblphenyl and 150 ppm PCB. This composition compares with the typical plant off gas HC1 that is sent to Dept. 217 for C/S acid production or to the sewer. Oas feed rate varied between 500-600 cc/min. (1.1 SC PH) and liquid flow from 10-12 cc/min. (1.2-1.5 lb/hr.). The saturated HC1 gas was fed to the bottom section of the scrubbing column. The scrubbing liquid was fed to the top of the column. In some experiments the liquid effluent from the column bottom was recycled to the process. The feed gas stream to the scrubbing eolumn and the outlet gas from the column were analyzed for biphenyl, monoohloroblphenyls and polychlorobiphenyl (PCB) content by scrubbing a known weight of the HC1 gas with hexane. Sampling was done by first scrubbing the HC1 gas with hexane and then through an aqueous solution containing exactly 1 Mole of sodium hydroxide and phenolphthaleln Indicator. When the caustic scrubber changed from red to colorless the sampling was stopped. All the sample lines were heated to prevent any possible loss of PCB due to condensation or entrainment. 2. Analytical Methods The biphenyl and monochloroblphenyl content In the HCl gas were measured by a flame Ionization detector GLC. An Internal marker was added to the hexane scrubbed sample, which was then filtered through dry sodium sulfate and condensed to about 5 ml. This particular method is only accurate for biphenyl and monochlorobiphenyls only. The polychlorobiphenyl (PCB) content In the HCl was measured by an electron capture (E.C.) GLC. This method does not measure the biphenyl or monochloroblphenyl content. It Is therefore necessary to use both methods to determine the composition of the Impurities In the HCl off gas samples. It should also be pointed out, that the present PCB method does not differentiate the various Isomers, but merely indicates whether Aroclor 1221, 1242, 124B, 1254 or 1260 is present. Absolute precision and accuracy of the analytical methods used was not determined for this application. The results are Judged to be adequate for the purpose of .the design of the scrubbing equipment. Information concerning determining accuracy under these conditions will be obtained before design is complete. HONS 058297 Job No. 7309003 Page 5 June 13, 1972 C. Experimental Results and Discussion 1. Screening of Scrubbing Liquids _ Several possible scrubbing liquids were evaluated. These were selected primarily on the basis of being relatively cheap, avail able, good solubility for Chloroblphenyls, low pour points, low vapor pressure and compatible with Dept. 217 process and the finished chlorosulfonlc acid. Table 1 lists some of the various scrubbing materials tested using a 5-tray Oldershaw scrubbing column. Table 1 Scrubbing Liquid Result Oxo-alcohol bottoms HB-90 Aroolor 1292 Aroclor 1292 & 1259 1016 Bottoms Chlorosulfonlc Acid Not effective Effective Effective Effective Very effective Very effective On the basis of the results from these Initial screening runs It was felt that 1016 bottoms and chlorosulfonlc add had the most to offer, meeting two or more of the criterions specified. A more detailed study Involving these two liquids were then undertaken. 2. Chlorosulfonlc Add Scrubber Several runs were made with chlorosulfonlc acid using 5, 10 and 15-tray Oldershaw columns at 25C. In some cases the chlorosulfonlc acid was passed once through the column while It was recycled In others. Complete results of scrubbing study with chlorosulfonlc add are shown In Table II- The experimental data Indicates that the 5-tray column with no C/S acid recycle will essentially remove all the biphenyl and monochloroblphenyls and about 90* of the polyohloroblphenyls. The use of 10 and 15-tray columns were even more efficient, with the scrubbed gas having less than 1 ppm PCB. It appears that no additional benefits are obtained by going from a 10 tray to 15 tray oolumn. When the chlorosulfonlc acid was recycled to the process there was a slight Increase In the PCB level In the scrubbed gas. Recycling had no'effect on the biphenyl or monochloroblphenyl, which were still less than 0.1 ppm. These results Indicate that It is feasible by using a chlorosulfonlc acid scrubber the PCB level In Aroclor HC1 off gas can be reduced to less than 1 ppm. Figure II shows GLC chromatograms of feed gas before and after scrubbing with C/S add using 5, 10 and 15-tray Oldershaw columns. HONS 058298 Job No. 7309003 Page 6 June 13, 1972 Experimental Results and Discussion (Cont'd) It la suspected that the chlorbsiiflonlc acid Is acting more than Just a physical scrubbing system and Is actually' chemically reacting with the biphenyl and chloroblphenyls to form solid sulfonyl chlorides which remain the liquid C/S acid. Typical reactions might be as follows: Although the use of chlorosulfonlc acid did an excellent Job In reducing the PCB's there are a number of problems associated with Its use. These are listed below: 1. Expensive equipment. 2. Disposal of used scrubber liquid. 3. Safety of personnel. 4. System would have to be completely anhydrous, because of the violet hydrolysis of C/S acid with water. 3. 1016 Bottoms Scrubber The use of 1016 Bottoms Is probably the most logical choice for a scrubbing liquid. This material consists mainly of trl, tetra and penta-ohloroblphenyls. It is quite viscous at room temperature, but Is a easy flowing liquid at 40-45C. 1016 bottoms are also compatible with organics being removed from the off gas and with the Dept. 217 protess and finished goods C/S acid. HC1 gas saturated with off gas condensate was scrubbed with 1016 bottoms at 40-45C using 5, 10 and 15-tray Oldershaw column. Results of this series of experiments are shown In Tables III and IV and Figure III. Temperature has a very Important effect on the operating efficiency of the 1016 bottoms scrubber. The temperature must be high enough to allow easy flow of the normally heavy and viscous 1016 bottoms, but not too high as to significantly affect the vapor pressure of the contaminants being removed. The results In Table III show the significant Increase in biphenyl and monoehlorobiphenyl content in going from 42-80C. Care should also be exercised to maintain the temperature about 40C to keep the 1016 bottoms from possible solidifying. Laboratory work Indicated that 4o-45C was the preferred temperature to operate the scrubbing column. MONS 058299 Job No. 7309003 Page 7 June 13. 1972 Experimental Results and Discussion (Cont'd) The laboratory study also showed that the removal efficiency of biphenyl and monochloroblphenyl Increases with the number of trays employed In the scrubbing column. The 5-tray column removes about 95* of the biphenyl and about 80* of the monochloroblphenyls. Better than 99* of these contaminants are removed by the 10-tray column and there Is essentially a 100* removal with a 15-tray column. However, with regards to the removal of the polychloroblphenyls (PCB) the 5-tray column appears Just aB efficient as the 15-tray column. The saturated HC1 with an Initial concentration of 137 ppm PCB averages about 9 ppm PCB, after scrubbing and this value compares favorably with the calculated value of 11-18 ppm PCB for pure HC1 saturated with 1016 bottoms. If the Plant HC1 off gas can be lowered to about 9 ppm PCB and all this gas were sewered the total dally loss would be about 1 lb. versus the present 15-30 lt>s. It was observed that the 1016 bottoms used In the scrubbing experiments had a tendency to solldfy to a white mass at room temperature, while the unused portion still remained a clear heavy liquid. This white mass melted at about 26-30C to give the normal clear liquid. The unused 1016 bottoms also solidified on standing at room temperature when seeded with particles from the solid HC1 scrubbed bottoms. This problem can be eliminated in plant operations by keeping all piping and storage facilities above 40C. IV. References OR Notebook Pages 1923^6-192359. WOK Notebook Pages 17501-17515. Nell E. S. Thompson Plant Process Research dy MOM! 058300 Figure I - Schematic Diagram of Laboratory Scrubbing System HONS 0 5 8 3 0 1 -f 5, 10, or 15 TrayOldershaw 9---------------------- * Hexane Scrubber NaOH Scrubber h X9 pcrubbing Liquid 0 10 Tray Older shaw . HC1 gas (mcs liq 9 or pach gas Condensate MOMS 0 5 8 3 0 2 Figure 3 HC1 Scrubbed with 1016 Bottoms (Feed stream similar to Run 17 in Figure 2) Table XV. Aroclor HC1 Off Gas Scrubbing Data T. Effect or `rray trriciency 2.. 1016 Bottom Scrubbing Liquid 3.* Scrubber Temperature 40-45C o o 3r. 0 . No. of Trays 1 - .6 .7 ~ --Inlet HC1 Gas (com) F.I. Inalvsls S.fi .Analvs s BIB NO 10 PCB n?i 1254" PCB Total m 258 653 <1 107 - -94 . 32 "" -11 137 "~ Outlet HC1 Ga.i (mn'i_______________________ P.I Analvs sir NCBlf PCB fot^L Comments O X HC1 Scrubbed with MCS 1109 HC1 Scrubbed with Off Gas Condensate - t5 15 >5 411 653 <1 "" n "" R '- 45 258 107 - 94 55 - - 9 10 " * - - 0 10 " " - R 1 10 " " - " 3* 10 " " - " n 10 " " * 7 15 " " - " 8 15 _ I billeve this sample was contaminated -, 32 * " R " " " --- - - 20 57 <1 - 5 2 -T 1016 Bottoms once through Column - - - -w n -- 13 26 <1 - - - -m it 11 137 - - - - 4 7 11 . n " " tt 8 22 - - - - `- -- - Nor e Non t 8 8 it it ** 0.2 0.8 -' - - - -* it " " - - ft * -- " n- 0.7 3 " <.l <.l - - Nor e Non ( 11 Nor e Non s 36 --Nor e tfon - 2 11 Recycle 1016 (98 g HC1-- 590 g 1016) 3* " (253 g HCl- 590 g 1016) S' " (300 g H01 - 590 S 1016) 2 1016 Bottoms once through Column " <.1 <.1 _ Nor e Non ! 6 6 Recycle 1016 (126 e HC1-* 8o g lnifii