Document EdKRZDQzR6g9LqxnMxrRjo1gg

S . H. Vogel, D. K. Lync . Pentachlorophenol - D i o x i n & Related Proceso Maintenance Studies. MIC/TP&E/WGK PLANT PROCESS RESEARCH tCO./OIV./DEPTJ LOCAT IO N1 FINAL ..{TYPE OF REPORT) REPORT REPORT NO.: JOB/PROJECTNO.: DATE: 3996 02-760.03 7311A03 August 18 1976 TITLE: AUTHORS: Pentachlorophenol - Dioxin & Related Process Maintenance Studies. V. R-. Bennet, T. R. Berner D. R. Cova, D. R. Dill 0. Hicks, D. K. Lynch, G. V. Happes, J. P. Mieure, D. P. Roman, S, H. Vogel P. L. Wright. ABSTRACT: This report summarizes dioxin related research work carried out on Pentachlorophenol and other Chlorophenols during the period from early 1970 through May, 1976. The major subject areas covered are: Chlorophenol dioxin analyses, chlorination and stabilization studies -on ways to suppress dioxin formation, methods of purification, bloassay studies, and dioxin precursor chemistry. RESTRICTED REPORT: Qualified persons may obtain loan copies from Central Reports if approved by D. K. Lynch, D. P. Roman T Sli F. or G. J. Me Ewan. 3996 AUTHORS: TITLE: I/ RE?I. N' COPY NO.: C O M" P A N Y' C O N F I D E N T ! ^ This document is the property of Monsanto Company and the recipie responsible for its safekeeping and disposition. It contains CONFIDEN INFORMATION which must not be reproduced, revealed to i/nautho DISTRIBUTION COPY NUMBER 1-3 Technical Reports Library 4 5-6 Circulate to MICC Director of Technology, R&D Managers, and return to Central Reports R2C TSD Library WGK 7 Plant Process Research File 8-10 D. Danna EISA 11 D. R. Cova/J. . White T3C 12 T. W. Cummins WGK 13 R. W. Flint/W.L.Smull WGK 14 R. L. Barness/P.E.Helslr WGK 15 E. E. Hendrlkse B2NA 16 H. J. Horner WGK 17 J .C .Hume/D.Metherd` WGK 18 J. A. Labanosky WGK 19 D. K. Lynch 20 J .E .Maurer/T.N.Wallin 21 G .J .Me Evan/L.G .Scharp f 22 J.S.Metcalf/P.G.Hamner WGK E2NB NIA B2NK 23 J .P .Mieure/M.W.Dietrich/ 0. Hicks T2F 24 J .F.Quinn/R.E.Keller T3B 25 W.R.Richard/D.R.Dill 26 D.P.Roman/W.R.Bennet RlC NIA 27 5.H. Vogel WGK 28 P.L.Wright A2SC ABSTRACT ONLY ST. LOUIS R. W. Bucknell - B3NB C. F. Callis - T3B J. R. Darby - 1730 0. DeGarmo - B3NA M. W. Farrar - R2E H. C. Godt - TIB E. J. Griffith - NIA W. C. Hanmann - T3A W. K. Johnson - R2E R. E. Keller - TIB W. S. Knowles - T3A W. R. Knox - R4C F. P. LaBelle - F4WA R# L* Llss - NlA E. A. Matzner - NlA J. E. Maurer - EISA R. H. Munch - TIB H. K. Nason - G4ND W. B. Papageorge - B2SK T. M- Patrick - T3B J. H. Payne - NlA J. F. Quinn - T3B G. A. Rauh - B3SB W. R. Richard - T3A F. S. Riordan - E2NA W. D. Robinson - R3E C. W. Roos - B2S L. G. Scharpf - NlA R. J. Schatz - E3S R. Schuler/ F, C. Meyer - R4B M. J. Scott - NlA C. Y. Shen - NlA C. C, Sisler - R2E A. J. Speciale - C3N W. F. Symes - NlA P. A. Tierney - NlA Q. E. Thompson - T3E M. C. Throdahl - DID R. L.'Wasson - T3A J. C. Weber - B3S J. L. Wikoff - B2S B. L. Williams - RIB J. J. Zeman - B2S F. B. Zienty - T3A AKRON - 1010 A. Y. Coran W. 7. Vaughn R. W. Wise DAYTON - 1250 E. E. Hardy VIA D. DANNA- - E1S* MELBOURNE, AUSTRALIA MONTREAL, CANADA LOUVAIN-LA-NEUVE CENTRAL REPORTS LIBRARY M. E. Madden - R2B COMPANY CONFIDENTIAL When no longer needed, or upon request, return FINAL REPORT NO.3996 TABLE OF CONTENTS PAGE NO. INTRODUCTION ---------------------------------------- SUMMARY ~ -------------------------------------- CONCLUSIONS ----- :---------------------------------- RECOMMENDATIONS ---------------------------------------- PATENT STATUS ----------------------------- --------- - REFERENCES ----------------------------- ----------- NOMENCLATURE & A B B R E V I A T I O N S ------------- ------------ EXPERIMENTAL RESULTS & DISCUSSION -------------------- (Dioxin Analyses, Chlorination Studies, Penta Refining, BioAssay Tests) 1 1 2 3 A 5 9 13 A. BACKGROUND - THE DIOXIN STORY -------------------- 13 B. DIOXIN STUDIES - PENTACHLOROPHENOL/SANTOBRITE 1970 ------ 1. Product, Process Stream & Competitive Penta Dioxin Survey --- -- -- -- -- -- ---- ----- ------ 2 Discussion - Analytical Method Inadequacy - C y d l z a t i o n of Dioxin Precursors -- -- ------ -- --- ---- 13 13 1A C.* DIOXIN STUDIES - PENTACHLOROPHENOL/SANTOBRITE 1 9 7 1 ------- 1* Survey - Dioxins in Penta & Penta/Oil Solutions ----- 2. Chlorination Studies - Temp. Effect on Dioxin Formation ----------------------------------- -- -- --- 3. Chlorine - Initiator in Dioxin Formation -- ------- -- A. Methods of Refining - Feasibility Study -------- -- -- 5. Discussion - Dioxin Method of Analyses ------- -- --- 15 15 16 16 16 16 D. DIOXIN STUDIES - PENTCHL0R0PHEN0L 1972 ---------------- 1. Trichlorophnol Stability Studies ----- 2. Effect of Various Chlorination Conditions on Dioxin Formation -- -- ---- ---- -- ---------------------- 3. Distillation Process Development & Stabilization Studies ------------------------------- A. BioAssay - CEF on Typical W.G.K. Penta - Lot KM-261-- 5. W.G.K. Penta - Dioxin Analyses ---------------------- 6. Discussion - Hlghboiler & Dioxin Methods of Analyses- 17 17 17 18 18 18 19 E. DIOXIN STUDIES - PENTACHLOROPHENOL 1973 ----------------- 19 1. Introduction ---- ---- -------------------------- 19 2. Chlorination Studies - Alternate Catalysts, & Pressure Chlorination -----'--------------------------- 19 3. BioAssay - CEF & Chloracne Response - Lab Low Temp. & Distilled Penta ---------------------------------- 20 A. Distilled Penta a) Process Development ----------------------------- 21 b) Distilled Penta - Compatability in Santobrite Process- 21 c) Crude Penta X Pt. vs. Z Distillation Residue --- 22 d) Distillation Stabilizers/Corrosion Inhibitors--- 23 e) Distilled Penta Stability in Woodtreating Solutions- 28 FINAL REPORT NO. 3996 PAGE NO. E. (Cont'd.) 5. Recrystallization Studies. ------------ --------------- 6. Woodtreating Penta/Oil Solutions -- Dioxin Survey------ 7. Penta Esters - Dioxin Analyses ------ ----------------- 8. Dioxin Analyses a) W.G.K. Penta ------- *r-- ------------------------- b) Prill Fumes - Venturi Penta --------------------- c) Aqueous Santobrite - Alkali Insolubles ---------- d) Competitive Penta & Distilled Penta - Compariser.- 9. Improved GC Method - Dioxin Analyses -------- ---- ---- 28 29 30 31 31 32 32 33 F. DIOXIN STUDIES - CHLOROPHENOLS 1974/1976 1. Introduction ---------- -------------- --------------- 2. Plant Demo-Dual Circulation Pump Test --------------- 3. Dioxin Analyses s) W.G.K. Penta ------------------------------- -- -- ---------------- ---------- -- b) Monsanto Newport P e n t a ------ -- ------------------ c) Distilled Penta - Dow vs Monsanto---------------- 4. 5. 6. 7. 8. 9. 10. e) Competitive Penta - Domestic & Foreign ---------- ) Miscellaneous S a m p l e s ---------- ---------------- Plant Demo - Low Temperature C h l o r i n a t i o n ----------- BioAssay a) CEF & Chloracne Response on Low Temp. Plant Penta & Penta Oil Solution ----------------- ---- b) Quantitative Evaluation of CEF Data - Plant Penta From Low Temp. Demo. ' --- --------- -- Chlorination Studies a) Al/S - Catalysts, Catalyst Swamping, Catalyst Impurity Effects, & Penta Sulfur Analyses ------- b) H2O Content Before Catalyst Addition ----------- BioAssay - T E F on Parachlorophenol, 2, 4-Dichlorophenol; Lab Penta Al/S and Al/Te Catalysts, Relchhold Penta and hexa CDD, and TCDD Solutions ------------------------- Low Level Dioxin Analyses - Plant & Lab Penta (No Caustic Exposure) ----------------------------- Dioxin Precursors in Penta a) Base Catalyzed Ring Closure to Dioxins ---------- b) Significance of Precursor Cyclization ---*------- c) Methods of Analyses - Morpholine Treatment -- ---- d) Cyclization in NaOH vs Morpholine -- ------------ e) Cyclization Rate of Irgasan DP-300 vs. Dioxin Precursors in Technical Penta ----------------- f) Speculative Toxicity of Various Dioxin Precursors g) CEF Relation to Precursor Concentration in Penta Hexachlorobenzene In Penta ------------------------ 34 34 35 JJ 35 36 37 37 38 39 40 44 44 47 50 51 51 54 55 5S 58 60 61 61 G. CHLORINATED DIBENZ0DI0X1NS: COMPARATIVE TOXICOLOGY --- 62 H. PENTA TREATED WOOD - DIOXIN ANALYSES ------------------ 62 FINAL REPORT NO. 3996 PAGE NO. I. BIOASS! - CEF 6 REC DATA SUMMARY --------------------I J. "ACCURATE" DIOXIN & DIBENZOFURAN ANALYSES DATA SUMMARY ----------------------------------- K. AR&D SPECIAL STUDY - PLANT PENTA DIOXINS & "COOKED'' ALKALI INSOLUBLES-ANALYSES --------------------- L. SUMMARY -! DIOXIN METHODS OF ANALYSES DEVELOPED BY APPLIED SCIENCE AND PPT 1970 - 1976 ---------------- M. PENTA 33QDEGEADITION STUDY - PURDUE UNIVERSITY ---------- 63 66 66 68 68 DESCRIPTION OF RECOMMENDED PROCESS -------------------------- ANALYTICAL PROCEDURES 'TOXICITY & HAZARDS ATMOSPHERIC 6 STREAM POLLUTION ---------------------------- ACKNOWLEDGEMENTS 68 68 69 69 69 FINAL REPORT NO. 3996 APPENDICIES A. TSD Report No. 4242-2 "Pentachlorophenol/Santobrite Dioxin Study" 8/15/70, S. H. Vogel. B. MICC Progress Report No. 4, Pentachlorophenol/Santobrite Dioxin Studies. Job No. 090171 11/15/71, S. H. Vogel. C. MICC R&D Progress Report No. 1 Pentachlorophenol/Santobrite Quality Improvement & Maintenance - Job No. 7309103, S. H. Vogel 12/15/72. D. MICC R&D Progress Report No. 2 "Pentachlorophenol - Manufacturing Support" Job No. 7309103 S. H. Vogel 5/31/73. E. ITEMS: 1) WGK Penta - Dioxin Analyses (1972 - 1975) 2) Competitive Penta - Metals Analyses 3) Quantitative Evaluation of BioAssay Data From Low Temp. Plant Penta 4) a. Carnegie Mellon University "Rabbit Ear - Chloracne Test"Report No. CMU 74-41A Through G. b. Biotest Lab Report No. BTL 74-28. CEF on Plant Penta Std. Conditions - Lot No. OR 212180-1. 5) Special Study - AR&D Dioxin & "Cooked" Alk-Insol. Analyses on Plant Produced Penta - April 1976. 6) Applied Science - Dioxin Methods of Analyses 1970 - 1976. 7) WGK Plant Process Research - Dioxin/Furan Analyses in Technical Penta. Streamlined Method. June 1976. FINAL REPORT No. 3996 i. INTRODUCTION . In the late 1950's, a highly toxic compound, 2, 3, 7, 8-tcl rachlorod ib*.-,;/c-c dioxin (TCDD) was identified as an impurity in 2, 4, 5-trichlorophencxyacc i acid (2, 4, 5-T). This prompted concern for dioxins in other chlorinated hydro carbons. This concern extended to Pentachlorophenol (Penta) in the late 1960's when Penta was Implicated in the .accidental poisoning of millions of commerclaily raised chicks. Research on the dioxin subject, as it relates to Monsanto c^lorcph^nol products, began in 1970, and is still conipuir.g today. This leport summarire5 co: forc ings to date. The work was carried out by WGK Plant Process Reseaicn no MrCC R&D at G.O. Supportive work was also carried out by Applied Sciences Groups at G.O.. BioAssay testing was cone outside Monsanto at the direction of eui Corpor ate Medical Department. For purposes of clarity and continuity, details of the work are presented in a chronological format covering the period from early 1970 to Mld-1976. SUMMARY*1 The major areas studied and described in the report are as follows; 1. The amounts and kinds of dioxins found in Monsanto chlorophenol products and process streams are characterized. 2. Several approaches to Pentachlorophenol (Penta) and Santobrite Solution purification were studied. A process involving purification of Penta by distillation was developed. Optimum filtration parameters for dioxin removal from Santobrite solution were determined. 3. Molten Penta stabilization technology was developed and indirectly imple mented into the plant process. 4. Chlorination studies (Lab and Plant) aimed at dioxin suppression - o * carried out. 5. Supportive analytical procedures were developed by PFT and by Applied Science Section at G. 0. 6. Competitive Penta was routinely monitored for dioxin levels end oi.t t; characteristics. 7. Fata of dioxin* in Penta woodtreating processes and environment vc .? studied. 8. Santobrite and Santobrite Solution stabilities were stud ltd. 9. Biodegradation of refined and technical Penta uun cerilcd out i*y University Research. 2. FINAL REPORT NO. 3996 SUMMARY (Cont1d.) 10. Dioxin literature monitoring was conducted throughout the 1970 - 1976 time period and Is continuing. 11. Catomance Penta esters were analyzed for dioxins. 12. Dioxin precursor studies were carried out, and methods of analyses foe precursors veTe developed.* CONCLUSIONS Polychlorodibenzo-p-dioxins, polychlorinated dibenzofurans, and dioxin piecursorB (poly chioro-2-hydroxy diphenyl ethers) are present in technical Pentachlorophenol. These compounds are believed to be the cause of positive 'Chick Edema Factor" and Rabbit Ear chloracne response observed on Monsanto at well a* competitive technical grade Penta. Trace amounts of these impurities (but lower chlorinated) are also found in parachlorophenol, 2, 4-dichlorophenol, and Santophen-1. Presence of the highly toxic 2. 3. 7. 8 TCDD however, has not been confirmed in any of the W5K ehlore- phenol products MDL 0.1 ppm). Typical level8 of these impurities in WGK Penta are shown in the following table: CYCLIZABLE DIBENZODIOXINS DIBENZOFURANS DIOXIN PRECURSORS Tetrachloro penta n hexa ti hep ta it octa it 0.1 ppm ^0.1 5-20 200 - 300 400 - 1000 0.1 ppm .0.5 30 - 90 250 - 450 150 - 600 unknown ^ 0 . 5 ppm 300. + 2500 + 5000 + These impurities are chlorine oxidative by-products, formed primarily during chlorophenol chlorination. Molten Penta degradation can also be a mlnoi tourre of these impurities if the Penta is not stabilized with a chlorine scavtngei, such as parachlorophenol. There is no evidence of gross dioxin generation jn woodtreating processes. Accumulation of dioxins in the environment has not been fully studied but has not surfaced as a problem to date. Dioxin precursors can be converted to dioxins. A number of conditions such .* high temperature, photolytlc, oxidative, or alkaline media favor the cor.x*t. . One.or more of these conditions might be prevalent in bllogleal system., I>n. ingestion of dioxin precursors might produce the same pathogenic effect a* in gestion of dioxins themselves. BloAssay data to confirm the purported lu* i*. :*y of the precursors is lacking. Formation of dioxins and other by-products during chlorination are mode: <*r* Iy suppressed by using the lowest possible chlorination temperature and an aJumiin*;' sulfur catalyst system. Penta prepared under these conditions also bus redoo-d CEF activity. (Low temperature benefits have been demonstrated in the pi an the Al/S catalyst has only been demonstrated in the lab). FINAL REPORT NO. 3996 3. CONCLUSIONS (Cont'd .) Penta Is best purified by recrystallization or by distillation. A process to refine Penta by distillation has been developed in the laboratory. Dow Chemical Co. has introduced a refined Penta on the market and has discontinued production of technical grade Penta. Biodegradation studies by Purdue University indicate refined Penta biodegrades more rapidly than technical grade Penta. Lack of adequate analytical procedures for dioxin, dibenzofuran, and dioxin pre cursors in Penta has been a perplexing handicap from the onset of the dioxin problem in 1970. Real progress has been realized in this area in recent months however. Ve can now quantitate dioxins and dlbenzofurans in Penta by GC following a liquid chromatographic isolation procedure. Dioxin precursors can be measured by an indirect approach which involves cyclization of the precursors to dioxins. This test unfortunately is very time consum ing, and for this reason is not suitable for routine use without some refinement. Current research on the dioxin problem is primarily in two areas: 1) refinement of analytical methods, in hopes of developing procedures which are more adaptable for routine use and, 2) isolation and collection of the various families of im purities in Penta. Chick Edema Factor activity will then be determined on each family. The dioxin problem is currently under study by U. S., EPA and is also a concern to environmentalists in the U. S. and Western European countries. RECOMMENDATIONS In general terms, Monsanto should continue to study the dioxin problem as it re lates to safety and environmental aspects on all WGK chlorophenol products. In specific terms, better analytical procedures are needed for accurate and routine quantitation of all families of impurities in chlorophenols. These families in clude diphenylethers, hexachlorobenzene, dibenzodioxins, dibenzofurans, and dioxin precursors (2-HDPE's ). Secondly, accurate biassay data is needed on each of these families. * In the long run, providing there is justification through bioassay data or regu lating agency requirements, a program of impurity reduction in Penta should be initiated. Minor reductions can be made through process optimization. Major reductions would require refining. The current plant practice of sewering alkali insolubles associated with aqueous Santobrite solutions should be discontinued. Disposal by incineration is recom mended . Finally, in order to minimize worker exposure to dioxins and other toxic impuri ties in Penta, good housekeeping and adequate ventillation in working areas cannot be over emphasized. The same precautions apply to laboratory personnel when analyzing Penta. 4. FINAL REPORT NO. 3996 PATENT STATUS T h e following patent disclosures resulted from this work. 1. Disclosure No. D02-21-3971. Molten Fentachlorophenol stabilization by addition of salicylic acid, benzaldahyde, phenol, maleic anhydride or potassium iodide Dioxin build-up slowed or prevented by each compound. (S.H.Vogel) Status - Inactive 2 Disclosure No. 43-21-4196. Molten Fentachlorophenol stabilization by contact with HjS gas. Dioxin bulld-up slowed or prevented. (S.H.Vogel) Status - B Sept. 1974, Inactive Fab. 1976. 3* Disclosure No. 43-21-4197. Molten Fentachlorophenol stabilization by addition of nltroaromatics. Dioxin bulld-up slowed or prevented. (S.H.Vogel) Statua - Inactive Sept. j9/4 4. > Patant Case No. C-43-21-4024-A. Additive package for Pent oil solution color lightening, and Fenta oil solution sludge inhibitor. Additives ara a mixture of p-Chlorophenol, trlphenylphosphite, dlamylphenylphosphctI acid and dl(dlamylphenyl) phosphoric acid. Although not specifically stated la this ceee, the phosphoric acid partial esters (RT-53A) have been shown to be a very effective molten'Fenta stabilizer against dioxin bulld-up. The RT-53A is specifically used for this purpose in the pro posed distillation refining proceas for Fanta (FED Report No. P-1729). The use of RT-53A for this purpose is stated in another disclosure (No. 4133). Although the subject case was allowed, It was never issued be cause of fear of disclosure of too much technology to competitors. (S.H.Vogel, D.R.D111). 5. Disclosure-No. 4133. Commercial Fenta refining process by distillation. (J. E. White, D. R. Cova). Status - Inactive July, 1973. During the period of this work, a Dow Chemical Co. patent "Stabilized Distil lation of Fentachlorophenol", Patent No. 3,816,268 (1974) was issued. 3 They claim decomposition during distillation Is Inhibited by presence of one or more amines, stable at the distillation temperature. More recently, a patent (US 3909-349} by Vulcan Materials Co. vas issued and also relates to Penta purification.^ The patent describes purification by distillation In the presence of zinc dust or ethylene thiourea vhich results in very light colored product. FINAL REPORT NO. 3996 3. REFERENCES 1. Schultz, K. H. - Clinical and Experimental Studies on the Etiology of Chloracne. ARCH. Klin. Exp. DERM. 206:569-596 (1957). 2. Kimmig, J. & Schultz, K.H. Occupational Acne (eo-celled Chloracne} due to Chlorinated Aromatic Cyclic Ethers. Dermatologia 115 (A): 540-546 (1957) O'er. '* 3. Kimmig, J, & Schultz, K.H. Chlorinated Aromatic Cyclic Ethers as a Cause of the so-called Chloracne. NATURWTSS. 44:337-338 (1957) (Ger.). 4. Sandeman, W. et al, Studies of Wood-Protection Chemistry. IX invest: gations Concerning Oily Blue Stain Protection Agents. HOLZFORSCH. HOLSVERfc'EJvT. 10 (4); 57-&6 (19:8;. 5. Sanger, VL, Scott,!., Handy A, et al; Alimentary Toxemia in Chickens. J. AMER. VET. MED. ASSOC*. 133:i72-176, 938. 6. Schmittle, S.C., Edwards, H.M., Morris, D.A. A Disorder of Chickens Probably due to a Toxic Feed - Preliminary Report. J. AMER. VET. MED* ASSOC. 132:216-219, 1958. 7. Edgar, S.A., Bond, D.S., Melius, F. et al: The Effect of a Toxic Substance In Fat on Poultry. POULTRY SCI. 37:1200-1201, 1958. 8. Flick, D.F., et al: Studies of Chick Edema Disease. POULTRY SCI. 44:1214-1222, 1965. 9. Flick, D.F., et al: Studies of Chick Edema Disease, POULTRY SCI! 46: 186-191, 1967. 10. Cantrell, J.S., et al: Search for Chick Edema Factor. CHEM. ENG. NEWS 45:10, 1967. 11. Cantrell, J.S., et al: Acta. Crystallogr., B25, 150 (1969). 12. Higginbotham, G.R., et al: Chemical and Toxicological Evaluations cl Isolated and Synthetic Chloro derivatives of dlbenzo-p-dloxin. NATURE 220:702-703, 1968. 13. Kimbrough, R.D.: Toxicity of Chlorinated Hydrocarbons and Kvidtcd Compounds. ARCH ENVIRON. HEALTH, Vol. 25:125-131, Aug. 1972. 14. Shea, K.P. t* Lindler, B., "Pandora and Lhe Sruragv Tank". ENVIRONMENT, Vol. 17, No. 6, 12-15. (1975) 15. Preliminary Dioxin Implimentation Plan - U. S., EPA, Office of Pesticide Programs Criteria & Evaluating Division, Wash. D.C. 11974). FINAL REPORT NO. 3996 Huff J.E., Wassom, J.S.: Health Hazards from Chemical Impurities Chlorinated Dibenzodloxlns and Chlorinated Dibenzofurans: Intern. J. ENVIRON. STUDIES, VOL. 6:13-17, 1974. King, M.E., et al, Carcinogenesis BioAasay of Chlorinated Diber.zodioxins and Related Chemicals. ENVIRON. HEALTH PERSPECTIVES. (No. 5):163 (1973). Progress Report No. 1, Organic Chem DIv. Job No. 2-02-760.018903309, July 1970, J. P. Mieure. WGK, TSD Report No. 4242-2, Pentachlorophfcnol/hntobnte Dioxin Study. 8-15-70, S . H . Vogel (See Appendix A). Memo communication J. P. Mieure to D. R. Dill (1970). No longer in file. MICC R&D Progress Report No. 1, 11-15-71, Job Ko. R&D 090171 -Pente Quality Improvement Studies. S. H. Vogel. (Fente Additives) MICC R&D Progress Report No. 4, Pentachlorophenol/Sar.tcbrice Dioxin Studies, Job No. 090171, 11-15-71, S. H. Vogel (See Appendix B). Dow Patent No. 3,816,268 - Stabilized Distillation ot Pentachlorophenol. June 11, 1974. MICC R&D Progress Report No. 1, Job No. 7309103 - Pentachlorophenol/ Santobrlte Quality Improvement & Maintenance. S. H. Vogel 12-15-72 (See Appendix C). Organic R&D Progress Reports No. 1, 2, 3, & 4 - Distillation of Pentachloronhenol - Job No. 0930205 and 0930305. J, E. White,' 10-25-71, 4-28-72, 7-20-72, 1-8-73. PED - Distillation of Pentachlorophenol to Remove Polychloro-dibenzop-dioxins - Report No. P-1677, Job No. 093035 - Aug. 24, 1972. J* E* White. Industrial Bio-Teat Labs, Northbrook, 111. Rtport. No. BTL-72-49 (1972). On file at Monsanto Medical Dept. Re: Dr. P. L. Wright. W.C.K. Plant - Standard Analytical Method No. 13.107, Pent A h way by CLC. Addendum to Spectroscopy Method 70-20 "Hexa-and-Hepta-Chlorodloxinti in Pentachlorophenol". March 2, 1971, J. P. M U u r e - Applied Sc K ik-o . C.0. Applied Science Spectroscopy Method No. 70-70, ''Dioxins In Fantuchlorophenol" J. P. Mieure, July 1970. Vulcan Materials Co. - Patent No. US-3909-349, "Separating Feniachlorophenol from Colored Impurities. FINAL REPORT NO. 3996 7. -32. Applied Science Spectroscopy Method No. 70-15 "Analysis of Pentachlorophenol for 2, 3, 7, 8-Tetrachlorodibenzo-p-dioxin". July 1970. J. P. Ml cure. 331 MICC Progress Report No. 2 "Fentschlorophenol - Manufacturing Support" Job No. 7309103, S. E. Vogel 3-31-73 .(See Appendix D) 34. MICC R&D Pinal Report No. 3882 Job No. 7309103 end 7311103. "Process Development for Recovery end Treatment of Aqueous Phenolic Wastes Associated vlth Chlorophanol Operations at WGK Plant. S. H. Vogel 6-28-74. 35. WGK AR&D Report No. 2715, June 9, 1975. Dept. 236 - FentachlorophenolPenta Sample Screening by Internal Marker GC. P.Riley, J.Francis. 36. Memo. "Pentachlorophenol/Chloracne/Chick Edema Disease". P. L. Wright to D. R. Dill, May 14, 1973. 37. Medical Dept. Reports CMU-73-27 and BTL-73-28A thru 287, 1973. 38. PED Report No. P-1729 "The Distillation of Fentachlorophenol to Remove Polychlorodlbenso-p-dloxlns: J.E.White, May 5, 1973. 39. MICC R&D Progress Report No. 4, Job No. 0930305 "Fentachlorophenol Purification", 1-8-73, J. E. White. 40. MICC R&D Progress Report No. 5, Job No. 1430305- "Fentachlorophenol Purification" 3-20-73, J. E. White. 41. TSD Progress Report No. 16, Job No. 91370:9000 Pilot Plant Prepara tion of Penta Solution, R. W. Grlscom, July, 1974. 42. Applied Science, Spectroscopy Method No. 73-1, Analysis of Pentachlorophenol for 2, 3, 7, 8-tetrachlorodibenzo-p-dioxin and other chlorodioxins. J. P. Mieure 3-14-73. 43 Monsanto Newport R&D Report No. 2, Job No. 759.72, "Dev. of Chrom. . Method for Quant. Det. of Dioxins & Predioxlns in Penta and Santobrite. Comparison of Methods Currently in Use by European Penta Producers." K. M. Baker, A . J . McLoughlln, R. A. Lldgett, May 20, 1975. 44. WGK AR&D Report No. 2448, "Dept. 239 - Santophen-1 - Analysis for Presence of Chlorodioxins in Product", 4-18-74 H. L. Campbell. 45. Amendment "F" to Dept. 236 Standard Manufacturing Process Description for Molten Penta, Prilled Penta, Block Penta, & Penta 60 dated Jan. 30, 1972, Amendment date Aug. 24, 1973. S. H. Vogel. 46. Amendment "H" to Dept. 236 SMPD for Penta Mfg. dated Jan 20, 1972. Amendment date April 4, 1974. S. H. Vogel. 47. WGK Plant Process Research Progress Report No. 2, "Improve Pentachlorophenol Quality", Job No. R&D 090171, 1-13-72, N.E.S. Thompson. 8. FINAL REPORT NO. 3996 48. Applied Science - Special Study AC-75-SS-3 April 7, 1975, J. P. Mieure, G. W. Mappes. 49. Cutis Vol. 13, (1974) pp 585-591, "Chloracne - A Continuing Problem", J. S. Taylov, MD, Cleveland, Ohio. 50. Monsanto Newport England R&D Statue Report No. 2, May 20, 1975 Job No. 759.72 R. A. Lidgett et al. 51. Further Studies on Impurities in Chlorophenols", J. CHROM. 89 (1974) 325 - 333. C. A. Nillson et al. Dept, of Organic Chem., U. of Umea, Sweden. 52. "Thermal Chemistry of Chlorinated Phenols" Advances in Chemistry Series No. 120 (1973) H. G. Langer, et al. Dow Chem. Co. Eastern Research La., Wayland, Mass. 53. "An Artifact in the Gas Chromatographic Determination of Impurities in Fenta Chlorophenol", J. CHROM. 67(1972) 247-253, C. A. Nillson et al. 54. Chromatographic Evidence for the Formation of Chlorodioxlns from Chloro-2-Fhenoxyphenols", J. CHROM. 96(1974) 137-147, C. A. Nillson, et al. 55. Applied Sciences - Special Study S-75-SS-17, June 1975 "Chlorodibenzodloxlna In Fentachlorophenol Treated with Morpholine". G W. Kappes. 56. J,. Chrom. 107(1975) 295-310 "Analysis o Polychlorinated dibenzodloxlns and Furans in Chlorophenols by Mass Fragmantography." H. R. Buser. 57. Molecular Pharmacology, 9, 736-749 (1973) Alan Foland & Edward Glover "Chlorinated dibenzo-p-dioxins: Potent Inducers of g- Amino-levulinic Acid Synthetase and Aryl Hydrocarbon Hydroxylase." 58. Applied Science - Special Study AC-75-SS-1, "Determination of Fenta chlorophenol and Chiorodlbenzodioxins in Wood Borings", 0. Hicks, J. F. Mieure, March 27, 1975. 59. Microbial Decomposition of Fentachlorophenol. J. Water Poll. _2 pp 359-64. Kirsch, Edwin J, Etzel, J. E. (1973). 60. Oil Paint & Drug - Chem. Market Reprot Nov-17-1975, Page 26. "Dow Antimicrobials Safer - Say Purdue Researchers." FINAL REPORT NO. 3996 9. NOMENCLATURE 6 ABBREVIATIONS Alkali Insolubles BTL CDD's , CDF's CEF CMU 2, 4-D DBF DiCDD Dioxins DPE ECGC EPA Furans GC G.O. H3 2-HDFE 4-HDPE hepta CDD hexa-CDD Isopredioxin - Gravimetric measurement of non-phenolic high boilers in technical pentachlorophenol. - Bio-test Laboratories - Northbrook, 111. - Chlorodibenzo-p-dioxins - Chlorodlbenzofurans - Chick Edema Factor - Carnegie Mellon University - Pittsburg, Pa. - 2, 4-dichlorophenoxy acetic acid - Chlorinated dlbenzofurans - Dlchlorodlbenzo-p-dloxln - Chlorodibenzo-p-dioxins - Chlorinated dlphenylethers - Electron Capture Gas Chromatography - Environmental Protection Agency - Chlorinated Dlbenzofurans - Gas Chromatography - Monsanto General Office - Creve Coeur, Mo. - High boiling Pentachlorophenol impurities; chlorination by-products - Dioxin precursors; predioxins; Folychloro2-hydroxy dlphenylethers; Polychloro-2-phenoxy phenol; hindered phenols. - iBopredioxins; polychloro-4-hydroxy diphenyl ether; polychloro-4-phenoxy-phenol; - hepta Chlorodibenzo-p-dioxins - hexa Chlorodibenzo-p-dioxins - Folychloro-4-hydroxy diphenyl ether (4-HDPE); polychloro-4-phenoxy phenol. *> 10. M&MT MDL MICC non-phenolics OCDD para PED Penta Penta CDD FPT Precureor Predloxln R&D rate of rise REC RT53A Santobrite Scrap Penta SMPD FINAL REPORT NO. 3996 - Mechanical & Materials Technology @ WGK - Minimum Detection Limit - Monsanto Industrial Chemicals Co. - High boiling chlorinated compounds found In technical Pentachlorophenol that contain no hydroxy group; alkali Insolubles; chlorobenzenes, chlorodlphenylethers, chlorodibenzo-furans, chlorodibenzo-p-dioxins - Octachlorodibenzo-p-dioxin - p-Chlorophenol - Process for Engineering Design - Pentachlorophenol - Pentachloro-dibenzo-p-dioxin - WGK Plant Process Technology - Predioxin; polychloro-2-hydroxy diphenylether (2-HDPE); polychloro-2-phenoxyphenol - dioxin precursor; polychloro-2-hydroxy-diphenylether; polychloro-2-phenoxy phenol; hindered phenols (2-HDPE) - Research & Development - Final segment of phenol chlorination to Penta chlorophenol when batch temperature parallels crystallizing point from 125C. up to end of batch at''w 185C. - Rabbit Ear Chloracne; chloracne test by Carnegie Mellon University, Pittsburg, Pa. - 50:50 mixture of di-Amylpheny) phosphoric acid and di-diamylphenyl phosphoric acid; high mole cular weight partial acid ester of phosphoric acid MOI additive in Penta & Molten Penta stabilizer. Sodium salt of technical Pentachlorophenol. - Chunks of Penta from Prill Tower. - Standard Manufacturing Process Description FINAL REPORT NO. 3996 Spread SPS-30 SPS-21 2, 4, 5- T TCDD TCP trl-CDD Venturi Penca 11. Degree C difference between chlorination tempera ture and crystallizing point during rate of rise in chlorination of phenol to Penta. 30X (vt) sodium Pentachlorophenate (Tech.) in water* 20X (wt) eodium Pentachlorophenate in water represents typical feed to Penta oil solution in Dept. 716. 2, 4, 5-Trichlorophenoxy acetic acid Tetrachlorodlbenzo-p-dloxln 2, 4, 6-Trichlorophenol Trichlorodibenzo-p-dioxin Sublimed Penta in exit air during prilling which is collected in a Venturi Caustic Scrubber. FINAL REPORT NO. 3996 13. EXPERIMENTAL RESULTS AND DISCUSSION A. THE DIOXIN STORY Dioxins first became a concern In late 1937. K. H. Schultz1, In his study of chloracne cases at a 2, 4, 5-trichlorophenoxy acetic acid (2, 4, 5-T) factory, traced the cause to an Impurity in the 2, 4, 5-T. The impurity was identified-as 2, 3, 7, 8-Tetrachlorodibenzo-p-dioxin (TCDD). About the same period, Schultz also studied and found TCDD to be responsible for 31 cases of chloracne in a German 2, 4, 5-Trlchloro- phenol factory.2 The cases had originated in 1933, when vorIters vere exposed to decomposition products generated in a run-avay reaction of caustic and 1, 2, 4, 5-tetrachlorobenzene. Ensuing studies 34 revealed that TCDD was not only chloroacnegenic, but also extremely toxic. This prompted intense concern for TCDD, and dioxins in general, in other chlorinated hydrocarbons. The concern extended to Pentachlorophenol (Fenta) in the late 1960's when Peata was implicated as a source of "Chick Edema Disease". Millions of commercially raised chicks had died mysteriously in' 1937.3 The disease was labeled "chick edema disease" and had the common feature of increased amounts of pericardial fluid (hydro-pericardium). .At first glance, an epidemic seemed apparent. It was soon discovered that residues of certain'distilled animal fats in the chick's diet would induce the chick edema s y n d r o m e . 7 intense studies on the "toxic fat" promptly ensued.* It was not until 1967-69 that chlorodlbenzo-p-dioxlns (CDD's) were identified in the toxic fat.^0,11 to have come from commercial Fenta contamination in the leather and tanning industry where the fats originated as by-products. There have been other instances of dioxin poisoning in the environment. Among these wete E-disease In cattle,13 and most recently, horse poison ings at three riding stables in central and eastern Missouri.^ Recent literature cites numerous toxicological studies on the subject of dioxins and more studies are currently in progress.13 To date, CDD's have been shown, or suspected to be, responsible for occurrence of chlor acne in humans, liver disease in several species, X-disease in cattle, chick edema disease, teratogenic effects on mammals, and mutagenic effects in bacteria.1 Carcinogenicity has also been suggested.1? Monsanto is continuing to study the dioxin problem - particularly as it relates to the safety and environmental aspects of Penta. Our findings to date are presented in the following sections of this report. B. DIOXIN STUDIES - PENTA/SANTOBRITE - 1970 1- Product/Process Stream & Competitive Penta Dioxin Survey Work first began on the dioxin problem in the early part of 1970. After analytical method development by Applied Science,1 a study was carried out with the following objectives: FINAL REPORT NO. 3996 (Cont'd.) a. Measure the dioxin concentrations In all Penta/Santobrlte process and product streams. b. Measure the dioxin concentrations in competitive Penta and Santobrite. c. Define possible ways to reduce or eliminate the dioxins. The details of this study are presented In WGX TSD Report No. 4242-2 "Fentachlorophenol/Santobrlte Dioxin Study", 8-15-70, S.H. Vogel (Reference No..19). The report is found In Appendix A. In summary, the major findings from the study were as follows: a. Heptachlorodlbenzo-p-dloxln (hepta CDD) and Octachlorodibenso-p-dioxln (OCDD) were found In finished Penta and in Santo brite* OCDD was found in concentrations up to 5000 ppm. Hepta levels were In the 500 ppm range. b. Somewhat lesser amounts of OCDD and hepta CDD were found in Santobrite and Santobrite aqueous solutions* Some of these dioxins are removed during processing as alkali insolubles at the cone settler. These insolubles are still routinely sewered (1976)* Qantobrlta production was discontinued in 1971 Santobrite solution is still being manufactured how ever and is a feed stream to Penta oil solution (Dept. 716).J c. Dioxins are primarily formed during the later stages of chlorination -..They also fora, but at'slower rates, during the molten hold period prior to prilling. There was also evidence of further dioxin formation in Fenta/oll solutions when the solutions were cooked for an extended period of time at wood treating temperatures. d. Lover chlorinated dioxins ( < hepta CDD) were not detected In any product or process streams. Levels of about 1 ppm tetra CDD (TCDD) were found In the chlorination mass at the trl-tetrachlorophenol chlorination stage.- e. Competitive Penta and Santobrite were found to be comparable In dioxin content to Monsanto*b products. f. Laboratory chlorination study suggests dioxin formation can be moderated by milder chlorination conditions. Discussion - Analytical Method Inadequacy - Cyclization of Dioxin Precursors Dioxin method of analysis utilized In these studies envolved an Initial caustic clean-up of Penta/benzene solution and quantitation by flame ionization GC (MDL 50 ppm) against an OCDD standard solu tion. 18 FINAL REPORT NO. 3996 15. 3. 2. (Cont'd.) We have just recently learned however that the Initial caustic (NaOH) clean-up procedure causes cycllzation of some dioxin precur sors (polychloro-2-hydroxy diphenyl ethers) to dioxins. Thus the reported values are Inflated. In addition since the precursors are hindered phenols they do not have good solubility In the caus tic. Hence they are,not totally extracted by the caustic. This leads to further inflation of the dioxin values due to thermal cycllzation of the precursors to dioxins In the GC. ^ The caustic catalyzed cycllzation phenomena also tells us we are probably generating dioxins In the plant In all aqueous Santobrlte solutions. A small excess of caustic Is normally present In these solutions. Host of the generated dioxins are likely being removed as alkali insolubles at the cone settler and are currently severed. These Insolubles should be handled with extreme caution due to sus pected high concentrations of hexachlorodlbenzo-p-dloxln (as well as hepta CDD & OCDD) The solution Itself should also be handled with extreme care due to remaining dissolved dioxins. Toxicological data on this solution Is lacking. C. DIOXIN STUDIES - PENTA/SANTOBRITE - 1971 1. Survey - Dioxins in Penta & Fenta/011 Solutions Research efforts during 1971 were primarily directed at development and Implementation of our MOI/MOT additive package for the Penta p r o c e s s S o m e dioxin work.In the areas of problem definition, chlorination studies, stabilization, and purification approaches was carried out however. These studies are reported In detail In MIC R&D Report No. A, "Pentachlorophenol/Santobrlte Dioxin Studies" Job No. R&D 090171 (11-15-71), S. H. Vogel (Reference No. 22). The report Is found In Appendix B. Some of the major findings are summarized here and In the following immediate sections. Further assessment of dioxins In Penta/Santobrite and Fenta/011 solution revealed: a. Santobrlte and pentachlorophenol (Penta) each slowly Increase In dioxin content when held at elevated temperatures. In each case, the formation rate Increases with temperature. Typical temperatures however, in end-use processes are quite moderate (80 - 220F.). Formation rates at these tempera tures are very meager. For this reason, dioxin build-up in treating plant oils or their sludges is negligible. b. Finished Penta and Santobrlte contain at least three differ ent dioxins - hexa, hepta, and octa chlorodibenzodioxin. Octachlorodibenzodioxin (OCDD) is the major dioxin present. Other lower chlorinated dioxins have not been detected in finished Penta or Santobrlte. Trace amounts ( ^ 1 ppm) of 2, 3, 7, 8-tetrachlorodibenzodioxin (TCDD) however, have been detected at the "dichlor" stage of Penta chlorination. FINAL REPORT NO. 3996 b. (Cont'd.) Ultimately further chlorination transforms TCDD to the high er chlorinated dioxins (hexa-octe). TCDD if present in_____ renta, i s < 0 . 5 ppm which is the detection limit of our test. "Block" Pent! contained comparable levels of dioxins to that f.ound in Prills. Chlorination Studies w Temp. Effect on Dioxin Formation Limited studies once again show dioxin formation during chlorination can be suppressed through lower temperature chlorination. 22 This is particularly true during the final stages of chlorination when dioxin formation rates were shown to be most rapid (Repta CDD and OCDD). There was also suggestion that maintaining a higher hydro- gen ion concentration in the mass during this period would further .suppress dioxin formation rates. Chlorine - Initiator In Dioxin Formation Chlorine was shown to be the initiator for dioxin formation in molten Penta.22 When free chlorine in molten Fenta is eliminated, dioxin formation is stopped. Introduction of small amounts of chlorine scavengers to molten Penta just after chlorination, ef fectively prevents further dioxin formation* Phenol and parachlorophenol are examples of such scavengers. Fara-chlorophenol Is currently being used in the plant in conjunction with MO I/MOT additives. Consequently, all normal Penta production at this time is dioxin stabilized during the molten hold period. Methods of Refining - Feasibility Study Various'purification techniques were shown to be feasible wayB of refining Penta*22 Distillation, solvent extraction from sodium pentachlorophenate solution, and recrystallization were each ef fective in removing dioxins and other impurities from crude Penta. Distillation appeared the most economical means. Development of a distillation process for refining by J. E. White of Organic Re-, search began in 1971 23 and continued into 1972 & 73. Discussion - Dioxins Method of Analyses As in our first study (1970), method of analysis for dioxins utilized an initial caustic clean-up step.3 Because of this, hepta & OCDD dioxin values are inflated (See Section B, 2. for explanation). TCDD analyses were carried out by J. Mieure of Appli ed Science at G.O. by GC/mass spectroscopy (a modification of the then newly developed ECGC method for dioxin analyses - Method No. 70-20). 3 0 32 FINAL REPORT NO. 3996 17. D. DIOXIN STUDIES - FENTACHLOROPHENOL - 1972 1. Trlchlorophenol Stability Studies Research on the dioxin problem In 1972 covered a number of areas. Included were studies on molten Fenta stabilization, trlchlorophenol tability/reactivityr chlorination studies, and competitive Fenta monitoring. These studies are presented in detail In M1CC R&D Progress Report No. 1, Job No. 7309103- Fentachlorophenol/Santobrlte Quality Improvement and Maintenance - S. B. Vogel, 12-15-72 (Refer ence No. 24), The report Is found In Appendix C. Major findings in other areas are described In following Immediate sections of this report. We had observed In chlorination profiles of Phenol - Pentachlorophenol that a surge of high boiler (HB) formation normally occurred when trlchlorophenol vaa near maximum concentration. A HB hump would occur in the chlorination profile. We also observer terminal EB content In Fenta was lover when the hump was lover. Chlorination and stability studies were carried out on 2, 4, -trl chlorophenol to determine what conditions favored or suppressed HB formation In this stage of the chlorination. Temperature was shown to be the most critical factor effecting EB formation. Acidity of the mass was also found to be Important. HjSO^ and HC1 were each effective stabilizers against EB formation in molten trlchlorophen ol (In absence of CI2 ). Bases accelerated HB formation. No alter nate catalyst system shoved measureable benefits. 2. Effect of Various Chlorination Conditions on Dioxin Formation Extensive chlorination studies ^4 aimed at establishing more opti mum chlorination conditions to minimize HB levels In Penta were carried out In 1972. the major findings were as follows: a. Heat history, or minimal temperatures throughout the entire chlorination Is most essential. It Is most critical during the later stage of chlorination. b. Catalyst concentration, catalyst addition point, Cl2 rate, mass mobility, CI2 bubble size, and system pressure (10 lbs.) had little or no effect on terminal HB content in Penta. c. No alternate catalyst systems (besides aluminum) shoved a benefit. d. Chlorination in the presence of solvents at lover temperatures resulted in significant HB reduction. These studies are presented in detail in Reference No. 2b. 18. FINAL REPORT NO. 3996 D. 3. Distillation Process Development & Stabilization Studies Studies on the development of distillation process for Penta refining began in Feb. 1971 by J. White of Organic Research. This work continued In earnest up to August 1972 at vhich time a Process for Engineering Design document was issued. : : -* +L '1* One of the major problems encountered in the development of the dis tillation process was low yieldsTdue to thermal degradation of the Penta. The use of p-chlorophenol was tried as a stabiliser but would not stand up under the'harsh distillation conditions.' .(p-Chlorophenol had earlier been shown to be an effective stabiliser* for molten Penta In the plant 22). A search was made for a mors effective stabiliser. Several compounds were screened. In summary the studies revealed the following; a. Bast performers vara those compounds which readily react-with -chlorine-or-chlorlne-Tadical, and-were- thermally stable-them-. .' "-T-i* selves^*t>t h e ^ t e s t ^ s a p s r a t u r e s i ? soa^b^obv^-oiessf'= * 8 * < ^ b. Resorcinol B-naphthol, parafln wax and RT53A2* were the most effective compounds found. RT53A was recosnended for use as stabiliser In Penta distillation. Zt was chosen over the other compounds primarily due to Its known compatibility In end*use Panta - (was currently an Ingredient In cur additive package Dioxin analyses In support of the distillation process development In 1972 were also carried out. The leachablllty of dioxins from soil by water was also tested and found to be nil. These and other process development work are presented'in detail In Reference No. 24. 4. BioAssay - CEP on Typical WCK Penta - Lot KH-261 During 1972, Penta Lot KH-261 (typical product) was submitted to In dustrial BIO-TEST Labs, Northbrook, 111. for Chick Edema Factor (CEF) studies. The results of the studies are found in Report BTL-72-49 which is on file in the Kedlcal Department at G.O. (Reference No. 27). The sample elicited a positive CEF response at 150 ppm dietary concen tration, but not at 50 ppm. This was our first instance of confirma tion that commercial Penta could produce the Chick Edema disease. 5. W.G.K. Penta - Dioxin Analyses Monitoring of dioxin levels In WGK manufactured Penta began on a fre quency basis in 1972. Values obtained on 1972 material, and on Penta produced in subsequent years are presented in AppendixqE.1 Applied Science, Method No. 70-20 was used for all analyses. * J See following discussion section for comments on results. \ FINAL REPORT NO. 3996 19. D. 6. Discussion - High Boiler & Dioxin Methods of Analyses 28 Highboller (HB) method of analysis used in the trlchlorophenol (TCP) and Penta chlorination studies vas by a normalized GLC pro cedure. The same HB response factor used for Penta analysis vas used for HB determination at all stages of chlorination. This may have caused slight Inflation of HB values at the lover stages of chlorination. Regardless of this, the values are valid for com parative purposes. Aa In our earlier dioxin studies (1970 - 71), the method of analy sis for dioxins, utilized an initial caustic extraction of a Penta/ benzene solution. 29,30 All hepta-CDD and OCDD values are there fore erroneously high. (See Section B.2 for explanation). Again hovever, the values are valid for comparative purposes. In addi tion, all hexa CDD values reported Include co-elutlng chlorodlphenylethera. 29 Aside from comparative purposes* the values should be disregarded.* E. DIOXIN STUDIES - PENTACHI0R0PHEN0L - 1973 1. Introduction Feasibility studies concerning by-product reduction during chlor ination, purification procedures, and molten Penta stabilization vere carried out In early 1973. These studies are presented In detail in KICC Progress Report No*.2, Job No. 7309103, S. H. Vogel, 5-31-73 (Reference No. 33 - See Appendix D). Research efforts during the later portion of 1973 vere primarily directed tovard process development for chlorophenol vaste treatment, and recovery of Penta from prill fume scrubber liquor. ^ Some minor dioxin studies, covering a number of areas vere carried out in latter 1973. This work, along vlth the major findings In Ref erence 33, are summarized in the folloving Immediate section of this report. 2. Chlorination Studies - Alternate Catalysts & Pressure Chlorination Again chlorination temperature vas found to be the most Important variable affecting by-product formation during chlorination. Chlorine assay vas also shovn to be. Important. Lov assay chlorine (98% and lover) gives higher HB as veil as lover chemi cal assflv. Chlorinating under pressure (up to 38 lbs.) had no measurable effect on suppression of dioxin or HB formation. No alternate catalyst systems vere found that had any measurable benefit over aluminum. A large unexplained variation in HB con tent vas observed in replicate batches made at standard as veil as amended conditions. A normalized GLC procedure 28 was use(j for Penta HB determination on all samples. We have recently-learned this method vill occassionally give iov HB values on Penta that has been chlorinated FINAL REPORT 3996 (Cont'd.) to higher then normal X Pt. (>178.5C.). Polymeric HB'a do not elute from the GLC. This phenomena was discovered when ve vent to an Internal maker GLC method to measure ylelds/composition in Fenta distillation studies. ^ Since that time ve-have also observed HB values reported by the analytical section on routine Fenta product are occasionally low. 35 (The normalised GG procedure is still in use.) Because of the method inaccuracy all HB values reported in reference No. 33 should be viewed vlth some skepticism.) Dioxin values reported in Reference No. 33 are also questionable (except for comparison purposes) due to the method of analysis used. (See Section B, 2 for explanation.) BioAsaay CEF & Chloracne Response - Lab Low Temp. & Distilled Penta About mid-1973f Monsanto Medical Dept.- carried out a Toxicon data base search to secure Information on chlorodloxine as related to CEF and chloracne diseases.^" The search indicated the most potent in ducers were i trichlorodibenso-p-dloxin tetrachlorodlbenzo-p-dloxln hftxachlorodibenro-p-dioxln tetrachlorodlbenzofuran pentachlorodibenzofuran Isomers cot specified' The search also revealed data shoving OCDD and hexa ODD*were not readily absorbed by the chick in CEF studies, and-were therefore not nearly as toxic as the lover chlorinated homologues. It vas suggest ed that TCDD vas probably the major causative factor in CEF disease due to its most severe toxicity. Following this search, a series of six pentachlorophenol samples were submitted to BTL and CMU for CEF and chloracne testing vlth the hope of developing a more precise relationship of dioxin concentra tion to biotesting responses. Three of the Fenta samples selected were lab preparations from chlorination studies " made earlier in 1973. These samples had shown reduced amounts of dioxins and HB's. For the most part, they exemplified low temperature chlorination. Samples of Monsanto Lab refined (distillation), and Dow's EC-7 refined material were Included in the tests. Our standard KM-261 lot vas also resubmitted. LOT NO. OR 210598 OR 211535 OR 211525 OR 211544 OR 211544 OR 214234 IDENTIFICATION Lab prepared - in Perchlorethylene solvent " M - Low Temperature " " -" " - high level Cl2 Monsanto Commercial KM-261 Dow EC-7 (distilled) Monsanto - Lab distilled FINAL REPORT NO. 3996 21. 3. (Conc'd.) 37 ^ The results of the Chloracne teats Indicate that .staples A, B, 1 and C vere essentially equivalent and produced the "least detectable! change". Sample D, the commercial material, produced a marked effect and the distilled samples vere essentially vithout effect.in these tests. *7 With respect to CEF toxicity, neither the solvent chlorination or distilled samples shoved a response at the highest level (1500 ppm) tested. Samples B & C vere negative at the 500 ppm level, vhlle KM- 261 vas positive at this level (lovest dietary level tested). KM-261 vaa positive at 150 ppm In an earlier test (Reference No. 27). The encouraging results from these tests prompted a trial of lov tem perature chlorination In tha plant vhich vas carried out In early 1974. Results of this trial are presented in a later section^ of this report. (See Section F, 4.) 4. Distilled Penta a. Process Development i Some major modifications to the 1972 PED for Fenta distillation, P-1677 (Reference No. 26) vere made in early 1973. The major j changes vere: 1) elimination of the need for dichlorobenzene i condensing solvent, 2) an increase in tetrachlorophenol content of the crude and refined Fenta, and 3) a reduction In the typical dioxin content of the product. Thle vork vas carried out by J: E. White of MIC R&D. The amended PED document for Fenta distil lation - Report No. P-1729 vas issued In May, 1973. (Ref. No. >38) b. Distilled Penta - Comparability in Santobrlte Process i There vas some concern that dioxins would build up again when distilled Penta vas used In the preparation of Santobrlte or San tobrlte aqueous solutions (by Monsanto or a Penta user). To ans wer these questions, the following experiments vere run (OR 211701). Distilled Penta, Lots OR 214233 and OR 214234, vere composited. A 30Z aqueous Santobrlte solution was prepared. A portion of the solution vas oven dried at 110C. for 2 hours to simulate Santobrite manufacture. Another portion vas neutralized with aaueous HC1. The released Penta vas collected by filtration and air dried. Analysis of this material vas used to determine if dioxins were generated during the 30Z solution preparation (a simulation of t SFS-30 manufacture). Two more portions of the 30Z solution vere cooked at 80C. with and vithout contact to steel coupon. Samples vere taken from each at 1, 5, 12, and 30 days for dioxin analyses. This test vas design ed to detect any detrimental effects during SPS-30 storage. 22. E. 4. b. (Cont'd.) FINAL REPORT NO. 3996 i i i SAMPLE Hepta CDD OCDD Composited distilled Penta Oven dried Santobrlte (2' @ 110C) Penta Released & Air Dried N.D. N.D. N.D. 0.5 ppm 0.5 ppm 0.5 ppm WITH STEEL CONTACT WITHOUT STEEL CONTACT : Hepta CDD OCDD Hepta CDD OCDD SPS-30, 1 day @ 80C. SPS-30, 5 days @ 80C. SPS-30, 12 days @ 80C. SPS-30, 30 days @ 80C. ^ 0 . 1 ppm ^0.1 0.3 * 0.5 ppm 0.5 1.3 2.4 ^.0.1 ppm .60.1 0.5 * 0.5 ppm 0.5 1.2 1.5 i * Unable to determine due to GO co-eluting Interference!. Dioxins by. Method 70-20 i The first set of results above indlcete.no change In dioxin j content due to Santobrlte or SFS-30 preparation conditions. The second set of results (SPS-30 cooking) suggest a alight or ' meager build-up of dioxins vlth time. Contact with steel may ! accelerate the build-up slightly. There was also a noticeable I build-up of low boiling unidentified non-phenollcs in each of j the SPS-30 samples. Thera was also a slow build-up of-one high- I boiling non-phenollc which eluted 3 - 4 minutes beyond OCDD on i the GC. No attempt was made to identify these unknowns. I i c. Crude Penta X-Pt. VS. X Distillation Residue (OR 211708) Crude Penta feed to distillation,^ is expected to have a crystallising point of about 170C. Blending distillation residue with the crude feed should lover the X-Pt. This would permit reduced molten storage temperature prior to refining, and thus less thermal degradation. In addition and more impor- tantly, this residue might be blended with Penta chlorination mass, and thus allow a significant reduction in chlorination temperature due to lower X-Pt. Hence leas dioxins and H 3 Ta would be formed, and thus improve the overall yields of the chlorination/refining process. . Distillation reboiler residue, Lot OR 214240, and crude Penta, Lot OR 214240, were blended at various proportions and X-Pt determined. (See Graph, Page 22 a.) j ' | | j i | ii | X-Pt is indeed suppressed by addition of the residue. Since the residue is already highly chlorinated, it is not likely to chlorinate further. Finishing the Penta chlorination at tempera-1 tures no higher than 145C., are expected to result in much less ^ by-product formation, as was the case when PCE solvent was used, j The feasibility of using the residue as a "chlorination solvent" ! OR 211708 S.H.Vogel 9/4/73 FIUAL REPORT NO. 3996 23. E. 4. c. (Cont'd.) has not been demonstrated yet, mainly because of Insuffi cient quantities of residue available to try It. If Mon santo does decide to pursue refining by distillation, ve should consider utilizing the residue In this manner as a means of Improving yields/economlcs. d. Distillation Stabilizers/Corrosion Inhibitors Although RT 53-A had been demonstrated to be a useable and effective decomposition inhibitor In Penta distil lation, it did possess some disadvantages. It's cost vas high, availability somewhat uncertain, and lastly, showed some thermal instability Itself in the distilla tion process. 33 (Low boiling degradation products were found In the vent scrubber system.) ;Ie had shown in earlier stabilization studies that Penta decomposition could be retarded or stopped when raw chlorine In molten Penta was eliminated. ^2 High corrosion rates were also of concern In the reboiler,and preheater In the proposed distillation system. ^5 was therefore desirable to find an RT 53-A alternate which would: 1) provide some corrosion inhibition and, 2) provide stabilization by consumption of raw chlorine. A number of commercially available corrosion Inhibitors as well as various.organic residue streams available throughout Monsanto were screened for corrosion inhibition on A-235 steel In molten Penta. All tests were carried out at 220C. for periods of 10 days or more. The inhi bitor (0.32 by wt.) was first added to Penta in a 100 ml stoppered vial. The vials were then Immersed in a heated oil bath. After the Penta came to temperature, A-235 steel coupons were added to each. Three groups of inhibi tors were screened in this manner. Our original plan was to select the samples which shoved good corrosion inhibi tion. The selected samples would then be analyzed for dioxins to determine stabilization effectiveness. Our plan vas sidelined (other priorities) and the dioxin analyses were never run. Some indication of stabilization effect iveness is apparent however, from the color of the Penta sublimate which collected on the underside of the vial stopper in each test. White sublimate Is indicative of stabilization; without stabilization, the.sublimate Is yellow. 24. FINAL REPORT NO. 3996 E. 4. d, (Cont'd.) MOLTEN PENTA (220C.) 258 HRS. CORROSION INHIBITOR SCREENING TESTS (A 285 STEEL) SERIES NO. 1 (OR 211704) INHIBITOR (0.3Z Wt.)'Control - No Additive Edwin Cooper's RT 53-A Stripped RT $3-A (211587) Trlphenyl Phosphate Residue Trlcresyl Phosphate Residue Cresylic Acid Residue S-140 Residue (W.G.K.) N2 CP Residue (Queeny) Triphenyl Phosphite MPY"- COMMENTS 2 Moderate ccc under deposit, light pits and craters. 2 Light ccc under deposit, light pits and craters. 3 Moderate ccc under deposit, moderate pits and craters*. "1 Same as above. 4 Light ccc under deposit, moderate pit ting and cratering. 3 Moderate ccc under deposit, heavy pits and craters. 3 Light-ccc7under deposit, light pitting and cratering. 4 Light to moderate ccc under deposit, light to moderate pits and craters. 2 Light ccc under deposit, light crater ing and pitting. The coupona vere examined by WGK M&KT. Ve were surpriaed at the low corroalon rate# in all aeaplea, especially the control sample. We had expected the rates to be comparable to those experienced in the distillation studies (50 mpy & higher) 25 Fresh molten Fanta from the plant was used in these tests. In tha earlier tests, Prill Fanta had been remelted. We there fore surmised the difference in corrosion rates might be due to different moisture levels In tha Penta. Remelted prills would be wetter. In the next two sets of inhibitor screening tests, we used re melted Penta (Lot KM-509). FINAL REPORT NO. 3996 25 MOLTEN PENTA - CORROSION INHIBITION SCREENING TESTS-ON A-285 STEEL, *~220C.t 2 WK5. SERIES NO. 2 (OR 211711) (0.30Z) INHIBITOR _ SUPPLIER MPY Control - Reaelted KM-509 n i AA m --------" pc a v i u w u p c r 2 9 Polyrod-0515-- -- Hercules 1 BH6rCUXCB Da I 11 Unainin& 0-- -- Lonza Unamine OZ----------- --- Lonza 3 9 9* 3 3 1275-1 Nirus AOL.-- Morton vvirlpA* -- -- m 1 Sarkoeyl-------------- -- --------CAB Geigy 3 9 3 A H u G 3 R E - 4 1 0 ------------------------ ------- GAF 2 SUBLIMATE COLOR @ 3 DAYS DISCUSSION OF INSPECTION Yellow Medium all over cratering, light pitting. White Moderate to heavy all over cratering, light pitting. White Medium all over cratering, light pitting. - White Light to moderate all over cratering, light pitting. White Moderate all over craterln light pitting. . White Moderate to :heavy craterin light pitting. White Heavy all over cratering, light pitting. White Light to medium all over cratering, light pitting. White Heavy all over cratering, light pitting. White Medium to light all over cratering, light pitting. White Medium all over cratering with some areas of heavy cratering. General light pitting, Several areas of medium depth pits. White Light to moderate all over cratering, light pitting. 26 FINAL REPORT NO. 3996 MOLTEN PENTA -- CORROSION INHIBITOR SCREENING TESTS ON - A 285 STEEL* 200C. - 2..UES. SERIES NO. 3 (OR 2 U 7 2 1 ) INHIBITOR (0.30Z) SUPPLIER SUBLIMATE MPY .COLOR 6 3 DAYS COMMENTS * Control - Remelted KM-509 2 Yellow Medium to moderate all over cratering and light pitting. SB-10----- Air Products Chem. 2 Virco Pet 20-- --Mobil 5 Tyfo Amine-D -- Hawthorn 5 Nopco Chex. RA-- Diamond Shamrock 5 Corexit 7802 *-- -- Exxon 3 Armohib 28 -- -- Armak 4 Wric'o-514 --- -- ^Wright Chea. 3 WayFos-MM40-- -- Wayland 3 WayFo S-M60------ Way land 3 WayFos-A----- *-- Wayland 2 RM-510--------- G A F 2 PE-510--------- G A P 2 White White White "* White White White White White White White White White Same as above. i ti. ti i tt....... h *. '*V * it it If it n if ti h it it ti h . -h h . ti it it ti n ti h h it h if M m * CURSORY EXAMINATION BY M&MT FINAL REPORT NO. 3996 27. E. 4. d. (Cont'd.) Again, all corrosion rates observed (Series 2 & 3), were lover than anticipated. There was actually no meaningful differences In corrosion rates observed in any of the tests. All inhibited samples and control samples in the three sets of tests had near-equivalent corrosion characteristics on A-285 mild steel. There vas also no difference in corro sion rates betveen fresh molten Penta and remelted Penta. None of the corrosion rates observed vere even severe enough to be considered unacceptable - thus mild steel could not be ruled out as a material of construction for the distilla tion process. Nearly every inhibitor tested gave evidence of "dioxin" stabilization as indicated by the vhlte sublimate generat ed in each test. The sublimate vas yellov in all control samples. Actual*stabilization effectiveness on all inhibi tors is unknown however, since dioxin analyses vere not run. A number of the inhibitors could probably be used in place of RT 53- in the distillation process. The thermal stabil ity and tarrlng/charring characteristics of the Inhibitors should be tested first however. In the course of our chlorination studies, the following compounds vere also tested for effectiveness as molten Penta stabilizers. All tests vere carried out at 200C, for 7 hre.- (METHOD 70-20) No. Test Hepta CDD OCDD Chem. Assay 1 Control - not cooked 0.029X 0.28X 97. IX 2 Control - Cooked 0.052X 0.64X 94.9X 3 N02 - (Sparge) 0.068X 0.69X -- 4 H 2S - (Sparge) 0.035X 5 o-Chloronitrobenzene(1,OX)0.064X 6 p-Chloronitrobenzene(1OX)0.046X 0.32X 0.70X 0.54X 96.8X -- ----- . 7 Nitrobenzene (1.0X) 0.052X 0.60X H2S appeared to be very effective.. This compound should be considered for use as a stabilizer in Penta distillation since it would also have the advantage of stabilizing the distillate as veil as feed, reboller, and column material. There was no detectable H2S odor in the solidified sample. Of the other compounds tested, only p-Chloronitrobenzene shoved some stabilization effectiveness. 28. FINAL REPORT NO. 3996 E. 4. e. Distilled Penta - Stability In Woodtreating Solutions One question raised about pure Penta was whether or not dioxins would build up again In Woodtreating solutions. If dioxins were formed under the treating conditions, It would hardly make sense to remove dioxins In the first place. To answer this question, 52 Penta solutions In two different woodtreeting oils were prepared - then cooked to simulate treating conditions. Additive and non-additive current commercial Penta solutions were also prepared In one of the oils and similarly cooked. All samples were cooked while in contact with a rusty steel coupon. Cook temperature was 190F (somewhat higher than maximum anticipated In woodtreating processes). (OR 211583). ___________OCDD (Method 70-20) IDENTIFICATION 0 Hrs. 120 Hrs. 336 Hrs. Pure Penta In Shell-Med.Arom. Pure Penta in Hobll-Certrex Additive Penta In Shell Med. Arom. Non-additive....... . " " Additive Fenta-as ls-not cooked Pure Penta-as ls-not cooked 25 ppm 25 ppm 0.461 0.56Z 0.53Z 1.2 ppm 25 ppm < 2 5 ppm 0.42Z 0.561 2 5 ppm 2 5 ppm 0.452 0.532 Applied Science spectroscopy Method No. 70^20 (Ref. No. 29), with the modification of no benzene during caustic extraction, was used to analyze the Fenta/oll solutions. Due to Interfering oil components, the MDL for OCDD was 25 ppm. Lower chlorinated diox ins were not resolved well enough to allow quantitation. OCDD did not build up In any of the solutions during the two week cook period. The tests Indicate Pure Penta is not likely to increase In dioxin content In woodtreating processes. 5. Recrystalllzation Studies Late in 1972, there were Indications that distillation of crude Penta might not remove TCDD completely. Secondly, there was also fear TCDD might build up again during storage of molten distillate. Thirdly, however, ve had just learned that by Improved recrystalllzation tech niques, we could prepare pure Penta with less than 100 ppm total diox ins with just one recrystalllzation. Past work had Indicated two or more recrystallizations would be required to get below the 100 ppm level." The need for only one recrystalllzation made the economics of refining by this approach much more attractive. Additional recrystalllzatlon studies were then carried out and are presented in Reference No. 33 (See Appendix D). The key to efficient dioxin removal in recrystalllzation o-f Penta is adequate removal of the mother liquor from the crystals. This is ac complished by repeated washing with fresh solvent. FINAL REPORT NO. 3996 .29 E. 5. (Cont'd.) This revelation led to expanded end extensive recrystallization studies by MICC R&D. These studies are presented In detail In the following two MICC, R&D Progress Reports by J. E..White: Report No. 4, Job No. 0930303, 1-8-73 and Report No. 5, Job No. 1430305, 3-20-73 (References 39 & 40). .t__ In summary, a number of recrystalllzatlon solvents were evalua ted. Penta yield losses, due to unpreclpltated Penta from the solvent were high In all cases. Thus, "economically", refining by recrystalllzatlon didn't match refining by distillation, and work on the recrystalllzatlon approach was terminated in the Spring of 1973. 6 . Wood Treating Penta/011 Solutions - Dioxin Survey Our first survey of dioxins In Penta/oll solutions is presented in Reference No. 22. At that time (1971) ve concluded that build up of dioxins in wood treating solutions was meager, to negligible. In addition, a survey of the wood treating Industry failed to re veal any past or current problems of toxicity to workers or users attributable to dioxins. (There were some hearsay Instances of chloracne however when materials were handled In a non-hyglenlc manner.) Aside from our finding In 1971, another .survey of diox ins in treating solutions was .made In 1973.," Treating solution samples were, obtained from two treatment plants designated as follows: Sample No. Identification Source 1 Fresh solution in Mobil Certrex 50-C 2 Solution from Boulton measuring tank - over 200 hours treatlng/heatlng - Shell oil Wyckoff* Wyckoff 3 Sludge from cylinder (Retort) Wyckoff 4 New Mix Baxter** 5 6 7 Storage 07 Storage - sampled during low level Baxter Baxter * Wyckoff Co., Seattle, Washington ** J. H. Baxter Co., Eugene, Oregon Penta contents of the solutions were first determined by simple hydroxyl titration (OR 211567). Applied Science, spectroscopy Method No. 70-20 with the following modifications was used to determine hepta CDD and OCDD levels; 1) Oil sample weight varied such that each contained 1.0 gram of Penta and, 2) Ben zene omitted during caustic extraction steps. Control Penta solutions in Certrex 50-C and in Provalent 4A oil were also 30. FINAL REPORT NO. 3996 E. 6, (Cont'd.) prepared and analyzed to test method validity. (ASSAY) ppm Dioxins Wt. X (Penta Basis) SAMPLE IDENTIFICATION PENTA hepta CDD OCDD Wyckoff 01 Wyckoff #2 Wyckoff #3 (Sludge Free) Wyckoff #3 (Homogenous) 5.49 6.34 6.33 5.65 400. 300. 300. 300. 6500 2600 3300 4300 Baxter 04 Baxter #5 Baxter 06 11.00 6.06 6.07 320. 330. 400. 2900 2800 3000, 07, Penta In P4A Oil 08, Penta in Certrex 50-C #,,, Penta, used In 7 & .8 . 5.06 10.16 -- (Used for assay control) 500. 1800, 400. 1800, Except for sample 1, all dioxin values are comparable to typical levels found-In prill Penta. Replicate analyses of sample no. 1 varied as much as 3400 ppm (OCDD). Two modified methods were tried In efforts to improve precision and reproducibility. One envolved benzene dilution and filtration of the oil sample prior to caustic clean-up. -The other envolved azeotropic removal of the oil from saponified sample. Neither approach was any better than the Initial method. In spite of method problems, the tests Indicate there is probably no gross build-up of dioxins In treating processes. The reason for the inconsistent results In the tests is now .thought to have been due to exposure of the samples to caustic. Caustic causes c y d l z a t l o n of dioxin precursors to dioxins (See Section B,2, for explanation) 7. Penta Esters - Dioxin Analyses_____ Early In 1973, Catomance of England (Monsanto customer) requested dioxin analyses on a series of Penta esters prepared by them from recrystalllzed Penta. The samples were analysed by the following pro cedure. (OR' 210571-4).1 (1) 1.0 grams ester Is dissolved in 10 ml nanograde benzene and 5 ml of morpholine (Fisher Cat. No. M-263) in a 4 or. capped bottle. (2) Mix for 30 minutes with magnetic stir bar. (3) Transfer to 250 ml separatory funnel. (4) Complete the analyses as described In Applied Science, Spectro scopy Method No. 70 - 20 (Reference No.,, 29 & 30). The morpho line Is removed In the normal washing procedure. FINAL REPORT NO. 3996 31. E. 7. (Cont'd.) The testa are summarized'in the following table. A number of con trol samples using our standard technical Penta (Lot KM-261) are Included to Illustrate the effect of morpholine on dioxin values. A 1.0 gram sample was used In all tests. SAMPLE IDENTIFICATION hepta CDD OCDD 1 ) KM-2 61-Normal work-up (Method 70-20) 2) KM-261-Normal work-up (Method 70-20) 3) KM-261-25 ml Morp.-not cooked-immediate normal work-up. 4) KM-261-25 ml Morp.-l hr. Boll-Normal work-up. 0.030Z 0.0311 0.026X 0.057Z 0.3B3Z 0.3B0Z 0.305Z 0.8B5Z 5) KM-261-5 ml Morp.-24 hr. @ 25C. Normal work-up. 6) KM-261-5 ml Morp., 10 ml benzene, 30 min. Stir @25C.f Normal work-up. 8) 9) 10) 11) 12) Catomance Recrystallized Fenta Tech. Fentachlorophenyl Laurate (Catomance) Tech. Fentachlorophenyl Laurate (Ventrn) Pentachlorophenyl Laurate (Catomance) Fentachlorophenyl Acetate (Caatoaance) 0.02LZ 0.029Z 4.3 ppm 679. ppm 159. ppm 0.7 ppm 0 .2 ppm 0.275Z 0.2450Z 18.2 ppm 6869. ppm 7160. ppm 16.6 ppm 2.3 ppm This was our first indication that time/temp. exposure of techni cal Fenta to morpholine would cause dioxins to Increase. For this reason, analysis of the eaters was carried out under condi tions which would minimize this exposure. We did not recognize at the time that aqueous caustic would also cause dioxins to In crease. We surmised that dioxin precursors were Implicated, but did not Investigate further. No comment Is made about dioxin values obtained on the ester sample since nothing was known about preparation conditions etc. We now know however that dioxin values on all samples which originally contained dioxin precursors are probably Inflated (See Section B, 2 for explanation). 6 . Dioxin Analyses a. W.C.K. Penta Dioxin analyses were run on every 10th lot of W.G.K. Fenta produced during 1973. These, along with values on every 10th lot from other years are presented In Appendix E. 1. Method No. 70-20 was used. 30 (See comment regarding validity of values Section B, 2.) b. Prill Fumes (Venturi Penta) Exit air from Penta prilling contains Penta fumes. This "sublimed" Penta is collected by passing the exit air FINAL REPORT NO. 3996 (Cont *d.) through a Venturi Bcrubber containing caustic solution - thus the name Venturi Penta. Penta collected in the solution was analyzed for dioxins. (Penta released from solution by acidi fication with H 2SO4 , then filtered, washed, & air dried.) Chlorophenol analyses were also run on the released/dried Penta. Some lover chlorophenols were probably lost due to solubility in water. (OR 211760) (Dioxins by Method 70-20) Phenol p-Chlorophenol trl-Chlorophenol tetra-chlorophenol Penta Chlorophenol High Boilers Hepta CDD OCDD 0.402 0.302 0.102 15.962 83.252 0.102 24 ppm 55 ppm Phenol comes about through degradation of our tri-phenyl phos phite additive, p-Chlorophenol is also an additive. As ex pected, dioxin values are very low since the Penta is essen tially sublimed material. Bloassay tests for CEF on a 482 Penta/oll solution prepared from the. Venturi* liquor in Pilot batching (Reference No. 41) came out negative at highest dietary level tested (1500 ppm). Chloracne response on this solution was not determined. These tests are discussed In more detail later in this report (See Section F. 5.). Aqueous Santobrite - Alkali Insolubles Alkali Insolubles were collected by filtration from aqueous Santobrite solution sampled at the dissolver in Dept. 236. The dried Insolubles were analyzed for dioxins by Method No. 70-20 procedure. Hepta CDD OCDD 3.92 10.22 These results paralleled earlier values on this material.^2 The Santobrite solution represented a mixture of Penta from prill fumes and scrap Penta. Competitive & Refined Penta (211706 & 211520) Penta produced by domestic competitors was collected and analyzed for dioxins. Applied Science Method No. 70-20 was used for analyses. Our standard lot KM-261 is in cluded in the analyses summary table below for comparison purposes. FINAL REPORT NO. 3996 33. E. 1) 2) 3) 4) 5) 6) 7) 8) 9) 10) 8 . d. (Cont*d.) SAMPLE IDENTIFICATION 1 '' Monsanto KM-261 Relchhold Relchhold Relchhold Relchhold Lot 312173 'Lot 6-6-73* Lot'342273 Lot 352473 Vulcan QC 086 Dow 0825223711D Dow 0819223120A Dow EC-7 (Lab Sample) Monsanto Distilled, OR 194436 Hepta CDD 0.030Z 0.063Z 0.065Z Typical 0.052Z 0.090Z 0.065Z 0.079Z 0.9 ppm 1.7 ppm~-;: OCDD 0.38Z 0.24Z 0.22Z 0.30Z 0.21Z 0.44Z 0.37Z 0.31Z 34 ppm 4.6 pp There Is no gross difference In dioxin content between any of ..the technical grade samples. The Relchhold Penta dioxin GC chromatograms are a little different from the . other'-samples;- Peaks normally eluting just prior to hepta CDD and* just prior to OCDD are absent'or extremely small. (See Section B, 2 concerning accuracy of dioxin v a l u e s .) During 1973 we received our first sample of Dow's ex perimentally1"refined Penta. The sample was analyzed for dioxin content by our routine method and; Is compared to Monsanto's experimentally refined Penta*In the preceedlng table of analyses. The sample had the physical .form of prills and was somewhat "off" white In color-. .Examination of crushed prills under the microscope Indicated an amor phous crystal. This type of crystal Is formed when molten Penta Is rapidly cooled. This Indicated the material had been prilled from the molten state rather than through compaction of crystals - say from recrystalllzatlon. This served as a clue to their mode of refining and distillation seemed most likely. Monsanto refined material is somewhat lower in dioxins and Is also white Instead of "off" white. 9. Improved GC Method - Dioxin Analyses During 1973, Applied Science developed an Improved method of analyses for TCDD in Penta (Reference No. 42). The Improvement was in the manner of sample clean-up. Instead of caustic ex traction for phenollcs removal, a benzene solution of the sample is chromatographed over alumina. Although the method was develop ed primarily for TCDD analyses, it was suggested it might also be 34. FINA! REPORT NO. 3996 E. 9. (Cont'd.) *S.\ . used for quantitation of higher chlorodioxina in Pent. Initial evsluscloa of the method for OCDD analyaea indicated low recoveries. In s Ister evaluation (256755) the recovery, problem Is Improved by better AI2O3 chromatographing conditions. Ve have just recently (1975) gone to routine use of this method for hepta CDD and OCDD analyses In Fenta since It avoids the precursor cycliration problem (See Section B. 2 for explanation). F. DIOXIN STUDIES - CHLOROPHENOLS - 1974/1975 1. Introduction Work described in the following lssnedlate section of this"report'was carried out Intermittently during 1974 & 1975. Subject natter, for the most part Is presented In chronological order. Areas"covered Include Plant tests. Lab chlorination studies, Biotest data, Dioxin analyses on various Chlorophenols, Dioxin precursor study, hcxechloro benzene analyses, and dioxin mechanism of. formation. The work was carried out at W.G.K. Plant Process Research. :*Some supportive analy tical work was also done by Applied. Science Group of G.O. * *, 1 t ' * 1.1 2. Plant Demo - Dual "Circulation - Pump" Test 1 cl2 Assay In early 1.974, the plant conducted a test in which two circulation pumps were.used to circulate chlorination mass through the external heat exchanger during chlorination. Circulation .Is usually done with only one pump.', Our hope was to Increase cooling efficiency through Increased circulation rate. Thus, X-Ft/batch temperature spread might be reduced which would result in a lower dioxln/HB product be- - cause of reduced "heat history". Kaos transfer would also be Increas ed which might further suppress dioxln/HB formation. Data from the tests are presented In the following table. Normal circulation rate (1 pump) Is approximately 430 gpm through the exchanger. Dual pumps gave about 530.gpm as determined by pump amperage change. (OR 212175) ** r.r.'. X Type Clr. 130C. Final r.r. Heat Chea. Z . Z XZ Z Phenol Pumps Spread Spread Hours History Assay Tetra Fenta HB MOI MOT 96X ' 94 96 94 94 VP VP F VP F 1 2.8C. 6 .6C. 1.75 1 2.6 6.6 1.92 1 -3.0* 6.4 2.33 1 3.0* 6.4 2.00 1 3.0* 6.9 1.67 12.4 15.7 12.0 15.9 13.2 97.5 5.9 86.4 7.7 .4:0 65.0 98.2 6.0 87.0 . 7.0 4.0 55.0 97.5 5.2 87.5 7.3 4.0 68.0 97.7 5.4 B'7.0 7.6 2.0 53.0 97.5 5.3 87.4 7.3 2.0 62.0 92 VP 2 3.8 6.4 2.33 16.4 96.8 4.9 85.8 9.3 6.0 39.0 94 F 2 2.8 6.4 2.00 17.4 97.9 4.7 87.6 7.7 4.0 37.0 94 F 2 2.8 5.6 2.25 19.4 96.8 4;6 66.0 9.4 4.0 55.0 94 F 2 2.6 6.2 1.75 14.4 97.8 5.8 86.9 7.3 2.0 63.0 * Estimated VP - Virgin'Phenol *-F> Fractions Phenol' `r.rl1Ratsi.of Rise ** (ma1) area between X"Pt. & batch temp. - from 130C. batch temp, to end of batch (from strip chart). O ' Ln *> H O OO U) FINAL REPORT NO. 3996 35. F. 2. (Cont'd.) High boiler (HB) contents vary widely from batch to batch. There was no apparent HB reduction when using two pumps. There was also no dlscemable effect on other quality features. On the basis of these findings, and the. cost and time Involved, dioxin analyses on the samples were not run. (Chlorine feed rates may have also varied from batch .to batch, which would. In turn, affect rate of heat generation and thus the "spread".) GLC was by internal marker method (OR 211759). IhlB Is the same method as Reference No. 28, but with dlphenylether Internal mark er added. 3. Dioxin Analysis a. W.G.K. Penta Dioxin analysis was run on every 10th lot of WGK Penta pro duced In 197' 4',W and on a por~ t--i--o---n- ---o--f----t--h- -e lots prod--u---c--e--d- ----in 1975. These, along with dioxin analyses from previous yeara are pre;sented In Appendix E. 1. Again, Kethod No. 70-20 29,30 was used for analyses. For this reason, values are Inflated (See Section B. 2. for explanation). b. Monsanto Newport Penta (OR 256718) A survey of dioxins in Monsanto Newport Penta was made in February 1975. Sample identification and analyses are pre< sented in the following table. BATCH X hepta CDD hepta CDD Pre--OCDD* OCDD 332-End of Chlorination " 1/2 Flaked 11 Final Flakes 818 ppm 775 749 1251 ppm 1225 1403 397 ppm 459 389 4612 4980 5542 441-End of Chlorination " 1/2 Flaked " Final Flakes 897 852 920 1448 253 4974 1325 219 5128 1322 267 5253 447-End of Chlorination " 1/2 Flaked " Final Flakes 738 988 632 1248 351 5329 1672 389 5737 1690 393 7151 * This compound is probably Octochlorodibenzofuran.^ Method of analysis was 70-20.^ ^0 The OCDD response fac tor was used for lower chlorodioxin quantitation. (Relative response factors are near unity - OR 211536). The ^ h e p t a CDD values represent all eluting non-phenolics prior to hepta CDD elution. OCDD values on the average are about 2000 ppm higher than typical W.G.K. Penta. OCDD also Increases slightly during