Document mp39Z0O8VoKmYNO55xQwG4RkO

St. Louis ftesooreh Report lo. 1640 October 22, 1956 - ^e t* ft-* < v f4 /-'*- ^w. * J ^ H *-3 FiML nrota OR AftOCLOR 1242 - STUD! OP UK3TABLX FRODOCT Job Ro. 2-02-750.01-3269 Monsanto Cbenlcsl Conpony Organic Cfcealcals Division St, Louis Reaoarcb Departasnt Vorl: done bj: A. K. Ellenburg H. Merten Writtoo by: B. Merton DSW 167198 STLCOPCB4041356 DISTRIBUTION OP REPORT HO. 1640 1. Pile 2. A. M. Ellenburg 3. J. H. Lun - P. B. Zlenty - file 4. Duplicate file 5. Dire. Assoc. end Assist. Rea. Dir. - 3. D. 6. H. Merten 7. WGK Plant 8. VGX Plant 9. VGK Plant 10. Anniaton 11. Anniaton 12. MCL (England) 12. MCL (England) 14. Sstra 15. Extra This report contains confidential Information which la the property of the Monsanto Chemical Company and which shall he disclosed only to duly authorised persons. The recipient is held accountable for the filing and safe custody of the report, which must be returned on demand. DSW 167199 STLCOPCB4041357 tabis op contents Page Ho. INTRODUCTION.................................................................................... SUMMARY........................................................................................... conclusions.................................................................. RECOMMENDATIONS ........................................................................... REFERENCES ...................................................................................... EXPERIMENTAL WORK....................................................................... DISCUSSION...................................................................................... DESCRIPTION OP RECOMMENCED PROCESS...................................... APPENDIX. ......... .................................................. 1 1 1 I 2 2 5 5 7 DSW 167200 STLCOPCB4041358 2. xmssspxm During July and August 1956 tbs Anniston plant experienced that the corrosion chloride stability of Aroclor 1242 vat periodically out of specification. At the sane tine it vat also reported that the plant had difficulty vlth free chlorine in the hydrogen chloride off-gas delivered to the nurlatlc acid tovers. This lead to the conclusion of a catal/st deficiency in the chlorination process. In the plant the chlorination of biphenyl is carried out in the presence of iron turnings which, vhen subjected to chlorine, are converted to ferric chloride. It vas the purpose of this investigation to study the effect of varying amounts of ferric chloride catalyst on the stability of the chlorination product, and to date rains the minimum catalyst require ments for the production of a stable Aroclor 1242. 30SMK Several biphenyl charges were chlorinated vlth ferric chloride as a catalyst in concentrations froa 0.01 to 0.03)1. The chlorination products were treated vlth lias and attapulgus earth slallar to the present plant practice. In several cases the aaount of line, or of attapidgus earth vas increased in order to study the effect of such treatasnt. The yield and the corrosion chloride value (deter mined by Dr* Hunch's group) indicated the effect of the varying catalyst concentrations on the stability of the chlorination product. ^QNgmtpg It vas found that a minima aaount of 0.0># of ferric chloride is required for the production of a stable Aroclor 1242. tower catalyst ratios resulted in increased addition chlorination, indicated by higher corrosion chloride values and lower yields. Ho improvement vas found vhen the amount of lime or of attapulgus earth vas increased, except in those cases vfcer* the amount of catalyst used vas belov the minimum charge required. REC OMMEHDATI0H3 It is recemended that a ferric chloride concentration of 0.0? - 0.05# should be practiced for the chlorination of biphenyl at a temperature betveen 120-140*C in order to guarantee the production of a stable Aroclor 1242. The present plant practice, using 0.25# of lime for the distillation and 0.2# of attapulgus earth for re sistivity improvement, should not be changed. DSW 167201 STLCOPCB4041359 a rarareacbs 1. Process Description for Aroclor, Department 246, by C. H. Schvarting tad 0. B. Schwarts, Bov. 12, 193?, revised* lurch 15, 195^. 2. Special Report, Job Bo. 117-1907, A Method for Determining the Stability of Trlchlorobensene, by 0. V. Ashworth, April 25, 1948. affBflffmAkypr& In e 2-1.ACE-reactor, equipped vlth gas-inlet, etlrrer, thermo meter, air-cooled condenser, and heating mantle, charges of 1000-1200 g. ox biphenyl were chlorinated vlth 0.05, 0.03, 0.02, and 0.01J? or ferric chloride as catalyst. The chlorinatlons vere carried out between 120-140'C, and at such a rate that a specific gravity between l.?80-l.390/25#C* was reached within 10 or 12 hours. Then the addi tion of chlorine was discontinued, and the reaction product was blown vlth air for about four hours at 00-90*0, in order to remove absorbed hydrogen chloride gas and chlorine. The crude chlorination product was velghed, 1.0 or 0.25J1 of line vere added and the material vas distilled at 20 am. Bg pressure. The amount of the distilled product determined the yield. This material vas agitated vlth 1.0 or 0,2ft of ettapulgue earth at 0O*C. for three hours and then filtered. The corrosion chlorine analyses vere carried out by Dr. Munch's group. The values listed in table I represent hydrogen chloride In ppm evolved et 210*C. after the first 8 and second 8 hour periods. * Eo. Catalyst TABLE I Spec. ** * Lime Earth yield Orav. 25*c. Corrosion Chloride ppm Is- 6 hr 2nd B hr 332-A 332-B 332-C 333-3 0.05 0.01 0.01 0.02 333-3 0.02 333-A 33*-A 0.03 0.05 334-A 449 512 0.05 1.0 1.0 1.0 1.0 81.0 0.25 1.0 63.0 1.0 0.2 93-0 0.25 0.2 90.0 0.25 0.2 98.6 0.25 0.2 99.0 1.0 0.2 98.5 1.391 1.257 1.25* 1-388 1.388 1.389 1.393 1.392 0.049 205. 224. 28.7 31.1 49.4 46.4 0.10 0.12 0.16 0.22 0.23 0.11 8:i! 0.055 201. 184. 17.7 22.6 17.1 37.8 0.02 0.06 0.10 . 0.25 0.C9 0.13 0.06 0.05 3.50 3.&6 DSW 167202 STLCOPCB4041360 3- For the purpose of coBparlson the corrosion chloride values or t satisfactory plant lot. Ho. 449, and a rejected plant lot. Ho.512, were added in Table 1. TABLE 11 Resistivity of Aroclor 1242 Sasole Ho. 3 Catalyst f Peed Resistivity oha -cm x K 332-A 332-B 333-B J33-B 333-A 334-A 334-A 0.05 0.01 0.01 0.02 0.02 0.03 0.05 0.05 1.0 1.0 0.25 1.0 0.25 0.25 0.25 1.0 14.400 1,700 (390 after hectirv 16 hrs.) 870 3,800 1,700 12.400 10,200 11,300 (14,700 after heating 16 hrs.) Measureaents were Bade on General Radio Resistivity Bridge Type 544 - B34 at 500 VDC and 100*C. Readings were Bade after 30 seccnds charge tine and 30 seconds on operate setting. In Table 1X1 are given the comparative results of numerous plant samples aade before and during the reported plant difficulties. DSW 167203 STLCOPCB4041361 table iii Analyses of Plant Saaples of Aroelor 1242 Saaplo Lot Ro. Chlorides, ppn (210*1 1st 8 ara. 2nd 8 hj Remarks S-414 512 0.31 0.36 0.30 1:5* 3.09 515 1.62 1.81 510 1.66 2.09 516 0,71 0.82 502 1.87 2.49 449 0.15 O.13 426 0.23 0.26 423 0.26 5-415 0.25 0.21 5-414 plus 0.26 0.4l photo Aroelor 92.8 3-414 plus 2.03 Photo Aroelor 587 5-414 1.29 512 8.54 5.11 0.24 0.27 0.24 5:11 f:?5 1.55 1.36 2.00 0.70 0.84 1.73 2.44 0.06 0.05 0,23 0.26 0.33 0.26 0.23 0.23 32.4 187 1.04 1.32 8.05 4.77 MORA 802 (Plant E) OATX 2988 (Anniston) Rejected by 0. E. 0,25 ppm Cl Anniston-Rejected ij O.E. 0,25 ppa Cl * Rejected by G.E. O.J ppa Cl " OK by O.E. 0.1 ppa Cl Rejected by 0. 0.3 ppn Cl " OK by 0. E. 0.1 p| Cl " OK by Q. E. 0.1 ppa Cl OK by 0. E. 0.1 ppa Cl Blent B Photo catalyzed Aroclor added to regular Aroelor Tests aade at 2>C*C to note effect of teaperature If to All tests on chlorides were aede by 0. W. Ashworth. Saaples cf C ^raa slse were used In the stability tests. Titration of the lorides evolved was aade using 0.005 H Ag]fOs. DSW 167204 STLCOPCB4041362 ft ft 5. DISCUSSION The result! of the experiments listed In Table I indicate a minimum catalyst requirement of 0.02# of ferric chloride for the production Of stable Aroclor 12*2. When a catalyst concentration of 0.01# of ferric chloride was used, the chlorination stopped after about three hours. By supplying additional heat the chlorine consumption continued for another hour, then the reaction had to be dlecootlnued, since considerable foaming occurred. The speolflc gravity vat only 1.237 at 25 *C. Those experiments Indicated In general that the stability of the chlorination and product depended on the degree of the addition chlorination. The less addition chlorination had taken place tho more stable vas the chlorination end product. It vaa noted that the degree of addition chlorination vas not only a function of a minimum amount of catalyst but also a very definite function of a minimtan reaction temperature. At reaction temperatures belov 120*C. the addition chlorination Increased considerably. The addition of lime after the bloving of the chlorinated produc with air before the distillation accomplished a dual purpose: convex lng the catalyst into a none volatile product, and eliminating small amounts of addition chlorination products. Tho present plant practice, using 0.23# of lime seemed justified and should not be lovered. DESCRIPTION OF THE RBCOWENIED PROCESS Biphenyl should be charged vlth 0.03# of ferric chloride and chlorine should be introduced not before the reaction mixture has reached a tenperature of 120'C. The rate of added chlorine should be regulated in such a vay that this reaction temperature la main tained for the first period of the chlorination. The first period of the chlorination represents the introduction of nt least one chlorine Into the biphenyl molecule. This chlorina tion product does not sublime and the rate of chlorination can be increased, which means that the reaction temperature can be raised to 160*C. When a sample of the crude chlorination product Indicate e specific gravity of 1.380-1.390/25*C. the addition of chlorine should be discontinued and the reaction mixture should be allowed.to cool to about 80*C. and should then be blown vlth air for four hours at this temperature. DSW 167205 STLCOPCB4041363 6. This crude air-blown chlorination product should bs distilled over 0.25^ of line at reduced pressure. The distilled mterlal should be agitated with 0.2g of activated attapulgus earth at 80*C. for at least tvo hours and should then be filtered. The resistivity aeasureaents In Table IX show a correlation be tween the sore stable Aroclor and high resistivity values. The "Munch' test for chlorides Indicates that a total of lass than 0.5 ppa chlorides after heating for 16 houra at 210*C signifies a stable ccopound that trill pass the O.E. corrosion test for 07l ppn chlorides. The use of a large excess of line in the distillation of unstable saaples bad sane beneficial affect. The extent of this effect in the plant Is not known, but it nay explain why som lota were aore unstable than others during the period when unstable Aroclor were nade. Resistivity of tvo sanples was checked after beating for 16 hours at 100*C. A decrease In the resistivity of the eaaple mde with 0.0i FeCl is a further Indication of Instability, whereas the stable Aroclor frca 0.05JK FsCl Increased in resistivity. Balabaugh cells were used In the resistivity aaaaureaents because only saall sanples were available in torn eases. The resistivities with these cells are invariably higher than the values obtained with the O.E. cells. Therefore these data are only comparative and do not show exactly vbst my be found in plant practice. The data ahovn In Tsblt III clearly points up the ability of the Kunch test to pick up the corrosion chlorides In the Aroclor that was rejected by Oenersl Electric* Anniston had been wable to detect these unstable lota by their test aethod. With aodlflcatlons of their test aethod Anniston has now been able to duplicate the O.E. results. This is the first case of unstable chlorides in tbs Anniston production and shows that It Is possible to produce the unstable aaterlsl. DSW 167206 STLCOPCB4041364 7. Teats vert aade using saall amounts of unstable Aroclor added tc stable material. The chlorides evolved shov that only trace quanti ties In the order of 0*002 of unstable aaterial vill give corrosion chlorides In excess of 0.1 ppa by the O.K. test. These results indicate the extreme importance of sufficient ferric chloride catalyst at all tlaes in the chlorination step. APfEJfBIX The detailed experimental data and procedures sre described In the notebook 1936, - A-70}>5 and A 01878. ds 1/17/57 A. X. Ellenburg B. Merten DSW 167207 STLCOPCB4041365