Document evbnKbrMveNazvjeMjp17BE4

*? *.. \V. *,-'0 if 1 3 9 a *"/ * I- " ' ? ` _ V' < V ^ The Analytical Approach Edited by Claude A. Lucchesi .r , , - . , ^' Industrial Analytical Chemists and OSHA Regulations for Vinyl Chloride - r $ * S. P. Levine, K. G. Hebei, J. Bolton, Jr., end R. E. Kugel 5 Stauffer Chemical Co., Eastern Research Labs, Dobbs Ferry, N.Y. 10522 -3 -i In 1971 the newly created Occupa October 1974 (Figure 1) (2,3). How VCM down to the 0.25-ppni level with tional Safety and Health Administra ever, the emergency temporary stan a precision at the 95% confidence limit tion (OSHA) with the advice of its dard remained :n effect until April 1, of 50%. technical arm, the National Institute 1975. The final standard sets a maxi for Occupational Safety and Health (NIOSH), adopted a 500 parts-per- mum permissible level of 1 ppm for an Analytical Approach 8-hr time-weighted average exposure. million (ppm) by volume permissible A ceiling limit of not more than 5-ppm To aid industry in monitoring pro level for worker exposure to vinyl VCM over a 15-min period has also grams designed to comply with this chloride monomer gas (VCM). In 1974 been set. In addition, an action level of standard, NIOSH published a prelimi a review of this standard was prompt 0.5 ppm was set up; exposures above nary procedure for VCM sampling and ed by reports of several deaths from a the action level require periodic moni analysis that was classified as "opera rare form of liver cancer called angio toring, medical examinations, and tional, but not thoroughly character sarcoma among polyvinyl chloride training (4). ized" (5). This method calls for collec (PVC) plant employees. This suggest The emergency and the final stan tion of VCM in glass adsorption tubes ed a possible relationship to PVC pro dards called for vastly different ap containing one of the specific NIOSH- duction. Animal studies and epidemi proaches from an industrial hygiene approved lots of activated charcoal. ological surveys indicated that expo point of view. Under the temporary Air from the breathing zone of the sure to VCM might be a causative standard, a survey was made to deter worker is drawn through the adsorp agent involved in the development of mine areas of emission, and only tion tube with the aid of a small low- angiosarcoma in humans (7). These "grab" sampling was performed. It flow battery-operated pump. After facts led OSHA to issue in April 1974 was only necessary to ensure that sample collection is completed, the $ a temporary emergency standard of a work areas did not exceed 50 ppm of tube is capped and sent to the labora 50-ppm ceiling exposure. This stan VCM in the air. This sampling ap tory for analysis. VCM is desorbed dard also provided for regular moni proach was changed with the advent from the charcoal with CS?. and the < toring of the work space by personnel of the permanent standard. Now areas resulting solution is injected into a gas monitoring systems able to assay must be regulated by both the ceiling chromatograph (GC) for analysis. Sep 5-ppm VCM with a relative precision (maximum) value and by the time- aration of VCM from other compo of 20% (average for a 10-min air weighted average exposures of workers nents is performed with an SB-30 col sample). This was to have a profound in those areas. These requirements umn. Since NIOSH realized that this effect on the vinyl chloride and the call for classifying the areas and types procedure had not yet been thorough s polyvinyl chloride industries which employ approximately 360,000 work of jobs in plants and for monitoring the actual exposures of workers over a ly characterized, the final standard al lowed this method or any equivalent ers in over 7,500 plants. typical workday. The combined sam method to be used. Following public hearings, OSHA pling and analytical methods used Preliminaiy testing by our laborato published a final standard for VCM in have to be capable of determining ry, as well as by others, indicated that - RSV 0017285 /% OSH& Regulalions .. j" .r-Vinyi'Chloride Regulations, isfactory, their precision, reliability, and delivery date were all unknown at the time that the VCM personnel monitoring surveys were started by Stauffer Chemical Co. In an industrial environment, meth od development frequently involves more method adaptation than actual invention. The development and ad aptation problems for thia project in volved two categories, the sampling system and the analysis system. Sampling System We have investigated the utility of two types of personnel sampling sys tems for organic gases. The first in volves concentrating the sample in an adsorbent tube, such as that used in ;Oema*r1S74. OSKAImum 4ml ttandsnl 'A*-. Raduciranofespowmfo .. . : ^-ppffl/B-hc tin* wetgfilsd wwpe with .- Vittun pinAt. *ta*&aao &- Vrk Vppnt mu and 0.6*ppm action level the NIOSH procedure. Although cer tain drawbacks have been noted in the NIOSH procedure, variations of ad sorbent tube design, adsorbent, and/or VCM desorption techniques have I rti Rgur# 1. Genesis of OSHA regulations ACOH " American Ccnltmocm of Oovfrwrmn HdueMel Bygiantsta. MOSH " National tnstttuta Of Oc been applied successfully by several groups. These variations involve the cupational Safety and Health. OSHA " Occupationaf Safety and Health Administration use of modified reusable charcoal tubes, heat desorption devices, head space analyzers, and desorption with CS2 at Dry Ice temperatures. Advan tages of the adsorbent tube approach there were some disadvantages with the number of personnel monitoring are the small size of the sampling ap the recommended method. These samples could be minimized. This re paratus and the fact that large vol were: poor storage stability of VCM on quirement, plus a well-designed sam umes of air can be drawn through the the charcoal tubes, lot-to-lot varia pling program, was needed to ensure tube, thereby concentrating the VCM tions in charcoal, low and variable de the validity of the resulting VCM ex by several orders of magnitude. A sorption efficiency of VCM with CS2, posure data, because the variations drawback in the heat desorption and the inadequacy of the SE-30 column due to personnel, work shift, process, head space analysis procedures (which to resolve VCM from other compo and even day of the week are not al are applied to the adsorbent tubes) is nents (of the plant air) and/or CS2 im ways controllable. that gas chromatographic analysis can purities, and the toxicity and flamma Due to the effective date of the be performed only once; repetitions bility of CS2. In addition, the volatility standard (January 1,1975, delayed to are not possible since the sample is ei of CS2 made it difficult to prepare sta April 1,1975, by court order) and the ther totally consumed in a single de ble standard solutions of VCM in CS2. time required to train personnel, strict termination or its concentration has Because of these problems, our lab time limits were imposed on the ana been substantially changed. The use oratory, the Analytical Section of Stauffer Chemical Co.'s Eastern Re lytical method development stage of of Dry Ice baths to minimize losses of this project. This timing precluded the VCM and/or CS2 during desorption search Center, sought to develop an use of semiautomated VCM analysis from charcoal tubes was developed by improved method capable of VCM systems that have since appeared on Dow Chemical Co. (6). This procedure personnel monitoring for Stauffer the market. Although many of these is a variation of the NIOSH-developed Chemical Co.'s PVC resin and fabri commercial systems are perfectly sat method and has been tested by our cating plants. In addition to the re quirements set forth by OSHA, we Inlet Tube . '"* had several other considerations to in clude when deciding which analytical approach to use: The sampling device had to be ca pable of storing VCM with no losses for periods of up to one week to per mit shipping of samples from several plant locations to a central laboratory for analysis. The GC column must cleanly re solve VCM from interfering sub stances that might be found in plants employing VCM or VCM-containing materials used in a wide variety of synthetic and/or fabricating formula tions. The analytical procedure should exceed in both accuracy and precision the stated OSHA requirements so that Figure 3. Loss of vinyl chloride gas from Teflon gas sam pling bags laboratory. Although somewhat timeconsuming, it is more reliable than the original NIOSH procedure. A second ba*>c sampling procedure involves the collection and storage of the sample gases without concentra tion. The method of choice for person nel monitoring involves the use of gas sampling bags. A battery-operated pump is used to draw air from around the worker's breathing zone and ex haust it into the bag. The contents of the bag are then analyzed directly by gas chromatography or any other suit able analytical technique. The pump and the bag are placed in a small dayhike backpack which is then worn by the worker for a complete workahift (Figure 2). Gas sampling bags are commercially available and are usually fitted with a metal twist-lock valve, although some are also equipped with a permanent or replaceable septum or with a filling snout. Although the storage stabilities of a wide variety of volatile materials in these bags have been summarized in the literature (7-77), none of these reports has dealt with 0.2-1.0 ppm comentrations of VCM in air. There fore, the storage stability, memory ef fect (from previous samples), and loss es of VCM in two commercially avail able gas sampling bags were studied. In addition, the precision and accura cy of the total sampling system (pump, bag, and tubing) were defined. Chosen for this study were a ^pHon _ bag equipped with a replaceable sep tum and a twist-lock valve and an alu minized Scotchpak three-layer bag equipped with a valve. Figure 3 shows the loss of VCM from Teflon bags to be in the range of 20% per Hay. It was not determined whether this loss resulted from the permeability of Teflon or from me Flgura 4. Stability of vinyl chloride gas in aluminized Scotchpak gas sampling bags 1078 A ANA! YTICAL" CHEMISTRY, VOL. 47. NO. 12. OCTOBER 1975 chanical problems. There is really lit tle need for a septum on a gas sam pling bag since maximum GC preci sion can more easily be achieved by using gas sampling valve injection rather than gas syringe injection tech niques. Figure 4 illustrates the storage sta bility of VCM in aluminized Scotch pak bags. There is no detectable loss of VCM for a period of one week over the concentration range of 0.1-1.1 ppm VCM in air. Because of the possi bility of leaks in gas sampling bags, it is recommended that they should be leak tested with clean compressed air for a period of several hours before use or reuse. In actual Held use, we find about a 10% "mortality" rate for alu minized Scotchpak bags when they are used repeatedly. All further studies were carried out on only the aluminized Scotchpak hags. Bags experimentally filled with between 1.0 and 10 ppm VCM had no detectable amount (<0.03 ppm) of VCM remaining after two repetitions of vacuum pumping of the bags and refilling with compressed air. It is, therefore, our practice to perform three pump-and-fill cleaning cycles before each reuse of a sampling bag. The performance of the entire sam pling and analytical system was checked by pumping 1.0 ppm of VCM in air from a full gas sampling bag through connecting tubing and a sam pling pump into a second bag which had been evacuated prior, to the exper iment The lengths of Teflon-lined neoprene tubing used and the pump were the same as would be used in the field. This was a simulation of the RSV 0017287 9832.1-00 aaVD 3DIA83S 830X311 NO 61 313*0 Z09M 'AN 'Jajsaqooy *;aajts ueujpoo> ijjjon ZZZOl `uofsiAia sionpojy leondo dmiuaps 'quid 8 qosnee ASti uoiiBJisuoujsp pue 6o|jbo aajj joj uioozoajais qtuoi e qosneg 'sjoui pue japu;; m3|a piaij aptM xoi Maiauj ajnsodxa ue snjd `aejijl die ajam-sax isibiujoj ejaujeo sjoieuiiunit! 9 pue spuejs g `sasuaj Ajeiuaui -aiddns g 'sasaidaAd luajajjtp g uiojj maias ubd noA pue (xOZ 1 luui/saun OOG iseai ;e) n6iq Ajauiaji -xa si uotwiosatj asueisip Suwjom am 6ut6ueip mo -MIm uo;stoajd pue Aipndujis m;M pausnduioooe si XZ OJ XI UiOJj 6uiuj002 'X08Z Ol XS'Z ujojj 6u;MdjA a|iqM snaoj lueisuoo pue Aja6eiu! pjaij ieu 'djeqs ia6 nox sasA|eue H0Ui|O3ds ssoj6 fiugpueuiap }soui am joj. jomocI 6uiA|osau adoosoioiui g woozosiotg S> aiAioi -s Hasnva oq ^9 P*)B1(JVU1 JOU 81 uibisAn Siqj, UOtJBOtUnUlUiOO |9U08J8d 'UOJ -*5uuy` O30uwi3-aj^MM`op92: ,:{ (51) `1161 >4*M)9*OH PH1W *uy * 03 iKiiuianQ naqs 'u^Ud '3 *H <11) `1161 `*p8u3 'mo *8pu318UJ -uiaqQ `*P3 *a|pvW V H ,,`*301d 'duiAg <)uwuiiub}uu3 auaqavoui -IV jo suiiduisg aainog,, ui /sjod*^ pus sag ioj 8ut{duiBH `*4f -tuiXaiH a 'j (oi) 2161 *`D`d ,`*`niui mm 'eei *e-t <19 "v miraH uaauauiv `aamuiuioQ aauaiawa^ui t/n Aptiy pua iuji'diUBg jty jo'spoqqaj^,/ (g) *(1961) SIS I "uQJMVz 'wtuty `auanips f i (8) *0161 'SZS-891-OV SUN m'"<1A 1u*a|S joj apoifjaw 8ui[auiB Jty sooub^ jo *!*HM`nir,0'Pu!W^W:X`H <1) '9161 '> ""f 'uuiw `Btiodtauuij^ 'aauajajuoQ auata ^H lui*npuj uBauatuy ^uaoiuojtAug iBixjfrtpui ub u; 8iodB/v JuaAjog pa)Buuopi3 jatfjo pu* apuom5 jAm^ o pae'dig \auuou>d SuuovuoWt* `Aio^S THPW 'u*S 'M T `au*"UV *V *V <9> >161 *0*811-1 *811 dd `131*91 M3H `S*fl -**P PW lwiAluyjo [BnuBW msOINk (9) (H6I) m * P*J If) (9161) * *t 'Vi/WH 7S W ,JP*!0 *0 (G) mi *mo 'puupu!3 `BiviuatlAH {Bu^mpui {bjoboi uiwoo jo jpunoQ uBsuMay m'HIOOV Aq pajdopy j;y 0000(10^ ut saMiBjaqng |B3joiato joj aaniiA Jiuin PRMIL. (Z) (>161) 69 *0CZ -aofBy *pa#v nty r`*m w w p *uomof n w "*f M 0 "if *MWD T r H iX) tMUOJiJAy *GB[dmB8 auutnniotzi jauuot -jad Suipuodsaiioo jo i)nsaj am BuopBjjuoouop ^oa J?B luaiqoza a]qu -|jba aurjejajjoo jo X^qiqedeo aq^ aAeq lsnm umiuoin wqa *uoi(uonz snonui^uoa jo uo^ezqpn am 'iMMoq 'aq }anoi ^3A unsodxa laqjOM 3uiu;uuaiap joj jtoS aSuu-Suo| au.T. 'Sap^aa [au^snp ui ue joj qaeojdde pnt^X[vuv a|qatA a s^uaaajdaj aAoq* pauipno poipaa aq.y, luatudo(aAap pue uot)BjdBpv.pomaai SAisuajxa autjB-jissaoau anm 'uzvjS ojd ono jo tfeoi am ajvnbapa ^ou aiaj* HSOIN ^9 panddna aauqapinS BtaAfeuv pue SuqdtuB^ -asm Voqa e tn uia^aXa jmiuooi lauuowad j^ja e}d(d moo e 8u!i[B)8tn ptn `JuiAOid p(ay `8utdO(aAap m^ paaaj um tpnpojd (Xuia paqtfuy jo uo^eauqej joj 3Ad pue uo^anpojd uiaaj joj WOA 9811 qo}MM aauqanpui wp `aotvnpuoo uj *(;rj paen j aApA^iod-oi pmetn mnv uy *MuU|od sruuod jo tramioa papApoe poip e pue jaucA[od soojod miM (Rvejoj oo auepo-u) qedejn^j jo uexnjoaaid t Mieiodioaoitaum** oqe| Aueui Aq avn at Apuaaajd tuapAa qBnyqaeq-pae^na y *p(OA "*9^ "!1 oopueiaj jaftioi Suueq a(dum an am coojj ntoRiptooo jo uuni|09 39 am eeap <n Amaaoau agp ^ 61 rann aoAjeA aaaqj, *aA{iA paiemome a(8u 9 ut uopaafu] pue qm^qaeq^pue )to pue `uotpafut pue qsnuqoeq uuxn -(oo jo uoipunj am eiejodioaut miHM sauo)ejoqe( jaqio Aq pado(aA9p uaaq aAeq saorjemJijuoo SutApA {*23495 jib u; pyOA J9 vuopemaaauoa tndd-^o'O ^1 pe^ap ueo mat sAa stqj, *sauzp ajom jo om) ajdaiea qoaa plafut Afautjnoj aM ^auopaafu; jo jaqomu am J ?ooj ajenbe/x ** aaaaaaa 111 uoieioaid am aauig *{M{ tadd-gj'o ' am %U pue'iaAa[ mdd-Q*x am 1* %? jo (uojpafui a[8uia `jaAaj aouap -ijuoo %9g) auoiapajd 9At)e[aj m^ a^piaaj pue uopn[oaaj apnbape uem ajoui pjatA 01 aim g pue g uaaMtaq jo aamp aiaApue jeuimou punoj aoeq a^x *uottnjoaaj jo aao[ luvoyraSiatno ipui paan aq ueo pa g-g oj dnjo sdoo[ a|doi95 *M[ia uopaafai'aeS am mH8 uoipafui a{dmea Suunp uondnutp auqaesq jouiux A[uo ui qpuw ipp(A ajnaaajd ipn ao| e aajmbaj qedajnQ jo ean aqx'Soiqoed ({tvejoj uo 00t XBMOqje^) qedunQ e eao oj oaaoqa aaeq *uuxn{oo aiqdejSojeaxxnp ni 06-3S papuarnmoow-HSOIN *m jo Aavnbapeut >mJ0 "nesag uaaiais ptoftiieuy *WOA*n Jo%Z6-96JO^e*09itzeavepa' ^ ejep aqx '*uoi)^Mxoa aan japan papadxa aq ppm mp aaaaof CONFIDENTIAL TENTATIVE PROCESS FOR VYSET RE (VINYL CHLORIDE, ETHYL-*HYDROYYETHYL FUMARATE. ISORUTY LSNE TERPOLYMER) VOLUME ! - TENTATIVE PROCESS TENTA FIVE PROCESS No 1595 Mnroh. i*o-4 rl; K * S .il Ki'.imu'lts IS. i-ti in c> ) ; ' . r. J;.nkt*r n . ,.t :* * r v L .J . !.i !ui<* r n O ror nlcv 1 ! . P. Hotl.niov R .1 . Sli'i I'liiht' S O V.ir'.i: L .1 i^rfikl.indcr n C. 1 r *. 0 r K 1; . D\ I ! . j'.o'.kor K 'A Si i: 41* :* K i : l. i i O N !) \ .-i.or No. 1 I G AA ' iu'i. r Ev " O 11 D.\ . ^ ; on o >v No > M. 1'1 r* j r Mr i _r S: Low;s * D.V S L.V. 0 'V No. L' '\ N,,. 0 ov No S.I 'V No. 0 ? No. 7 0 _v,- No v'1 j-. NN ; No a O n\. :! 0. No ' Z . N> . : . N> : -V - . No. R E CI'.aT'n^e r 1 1: :: J R . C::0rii^u R I . Dur.Tio J R . Ei k M F. X. G:C;o::: R - E. G r c; c :*. E :krc, . J . Hi-.r. D . Mo v. - v N . * , ,, " / M " S - R. u 0 r K i\ J . St'.- r. = confidential information * r: a r. . . :r :r. - - . .-t=s a.*.* - a - v .. ? w: +r. *+l . ct?**"** 'J* zr : r.#*S `: tr-m ar.* f^c-;p4*r. il : h t " ar.a fr? ;i t i~szr 31 mp; >34. T* * r* .Mr. buiiiir-. s **c:r* ._:6: S:j r : r. c 1 ' 01 ci S:>r ; :\l ic ici S11 r : C * i e I d S:. Lou;? S:. Lou: 5 S'j r ;:. ^U: S*.. Lou:s Sur : S*. Loul 5 So r .t: j: ;ciU So r * *\ N Lou.> Sit . 0:. v. RSV 0017289 :r "is.. "* p' n.i* Jcr .-* j: * ; 1 sr=t*~ * -v HI I A 1 a* *1 Shawiniga n Resina Corporation Copy No. lb H. Dabagian Central Research Department Copv No. 17 L. J_. Breuklandcr Copv No. 18 J . D. Calfee Copy No. lo D . Carter Copv No. 20 R. S . Gordon Copv No. 21 H. P. Holladav Copv No. 22 13 . T . Junker Copv No. 23 H. W. Mohrman Copv No. ZA R . W. Schuler Copv No. 23 R. J . Sioccmbe vCopv No. 26 J . D. Upham :*^fcopv No. 27 Technical Reports Li Copv No. 2S Technical Reports Li Copv No. 2 Technical Reports Li Copv No. 30 Extra Copv No. 31 Extra Copy No, 32 Extra Springfield St. Louis St. Louis St. Louis St. Loui s St. Louis St. Loui s St. Louis St. Loui s St. Louis St. Louis St. Loui s St. Louis St. Lot: i s St. Lou i s St. Louis St. Louis RSV 0017290 i i i I 8 --gB | m I 1 - 1 I IB m| 1 a J______ TABLE OF CONTENTS II. III. rv. V. VI. VII. VIII. IX. X. XI. XII. INTRODUCTION CHEMISTRY SYNOPSIS OF PROCESS BILL OF MATERIALS RAW MATERIALS MATERIAL BALANCE FLOW SHEET AND EQUIPMENT DESCRIPTIONS OPERATING PROCEDURE A. Preliminaries to Reactor Operation B. Reactor Start-Up and Forward Feed Operations c. Reactor Recvcle Operation D. Reactor Shutdown E. Precipitation F. Filtration and Cake Washing G. Drying H. Acetone Distillation DISCUSSION A. Poly me rizatior. B. Monomer Stringing C. Precipitation D. Filtration and Cake Washing E. Drving F. Acetone Recovery G. Materials o: Construction H Packaging and Storing VYSE7 RE Powder I. Product Properties J. Raw Materials K. Waste Disoosal L. Product Stabilization COST ESTIMATE PATE NT STATUS TOXICITY AND HAZARDS A. Hazardous Operations B. Hazardous Compounds Page No. 1 4 Q ^4 54 !> !> FT ^8 o0 t>! tI t>2 h2 73 7^ 7n 81 S3 83 84 84 -I 1 RSV 0017291 I XIII. * XIV. XV. XVI. XVII. ANALYTICAL METHODS A. Nonvolatiles in the Reactor Charce B. Reactor Charce Assay c. Recycle Stream Assay D. Volatile Residue in Polymer Solution E. Residual EHF in the Reactor Charce F. Residual EHF in Polvmer Solution C. Acidity and Hvdroxvl Content c: VYSET RE H. Physical Properties Tests for VYSET RE I . Bulk Density J . Water in Drv Polvmer K. Water in Acetone L. Acetone in WatcjM. Chlorine in Drv Polvmer N. Trace HC1 in VC1 Monomer and Recycle Stream PROD YCT SPECIFICATIONS A C K NOW L E DO M E N T RE EE REXCES APPE NDICES TO VOL"ME ONE Pace No. 6 QC CO IOC 100 101 103 10s 1C* 1CI c1010- i 0- I O' 10- i I0 RSV 0017292 LIST OF FIGURES_____________________________________ I Pace No, 1. Material 3alance VYSET RE Process S b Flow Shcc t-Polvmcrization and Monomer Stripping 10 3. Flow Shcc t - Precipitation, Filtration, Drying and Acetone Ret ovc ry t. Monomer ,Stripping Column 6. Pilot Plan : Acetone-Water Separation 11 16 21 Pilot Plan t Acetone-Wato r Separation - Low Coix emraticn Ranee -* > Free Monomer in Reactor Charge and Composition of Poly me r n Simplified Material I3alar.ce tor Stripping System ^. V apor - Lie .::d Equilibrium Curve and Composition - Tempera*ure Diagram for System VCl-Acctore 10. Acet one -"V'titer Separation 1 1 . At e tone - **V ater Sena rat:ot'., Low Com er.t ra`: or. Range -> Roi i i ng- Pc .:t* Composition. A. etone - Wu: e r System * Sample :3c mb SB- I and SB-2 1 4. Re il *or 5-i mpling System r4 TO 71 T{? 7 a7 a* RSV 0017293 LIST OF TABLES i Pace No. i I. Equipment List Polymerization Pilot Plant II. Estimated Production Costs, Capital Requirements, and Selling Prices lor the Manufacture of VC 1 / EHF/I13 i Tcrpolymer III. Cost Estimate for VC1/EHF/IB Terpolvmer i (Revised 7/31/63) ST 66 i a i i i RSV 0017294 -'1 c. n^ OX < n-- !il,' r* no -- nn. r, na! o-o O1 !>i' :i: . r; ;i ^ -- COSI; H I HC-C-O Cs H i OH RSV 0017295 |c <?' J> I. INTRODUCTION VYSET RE is a. low molecular weighs terpolymer of vinyl chloride {VCD, cthyl-s-hydroxycthyl fumaratc (EHF) and isobutylene (IB). The end use for this polymer is thermosetting metal surface coatings. The resin was in vented in the Central Research Department Polymer Research Section (1,2). V'YSET RE films on metal have substantially better impact resistance and adhesion than existing surface coating products such as acrylics and melaminealkvds. Wilson (3) has concluded as a result of a market survey that at 40-45^/lb VYSET RE can compete successfully with these products :n sprav and dip coatings and in the rapidly expanding coil coating** field. I: has been shown bv Dybdal (-4) that VYSET RE can he sold :n the *i0-*4?^/lb range and earn 20^ return after taxes. A tentative process for VYSET RE synthesis \\a. developed during at the Central Research Department. The tem-vve process is described :n this report. The report is published in two volumes. Volume I presents the tentative process proper, and contains the information normally repaired bv design and manufacturing functions. Volume II is an experimental supplement which gtves the details of all experimental work performed in the course of developing the process. A tentative process tor the synthesis o separately (5). !:F monomer has been published I r.e term, applied to coating long strips of metal sheet prior to fabrication. * : s s i -le nt 1 v :.ihra Jtffi : r.t o "u >r ] * :;.i ::> * does, nut :re RSV 0017296 I I CH-=CHC1 * HC-C-O-CgRiOH C3 He -C -CH I tl O I EHF I I 80'C Acetone Solution AZBN I V C l / E H V /IB ter oolv mo r I I I I 1 I RSV 0017297 i I III. SYNOPSIS OF PROCESS i i VC1, EHF, IB, and AZBN'-acetone solution arc continuously pumped to a con tinuous stirred tank polymerization reactor operated at i'O'C ana 120 ps;g pressure. Heat of polvmeriration ;s removed from the reactor bv transter to the jacket. Polymer solution consisting of about 30-T VCl/EHF/IB tcrpolvmrr, i 30V acetone, 30!To monomeric VC1 and IB and I^p monomeric EHF is contin uously withdrawn from the reactor. i The monomeric VC1 and IB are distilled from the reactor effluent \n a sieve tray stripping column, condensed, and rcwclen to the reactor. Open super heated acetone vapor is used to operate the stripping column. Paraplex G-e2 stabilizer is mixed with the polvmcr solution effluent from the bottom of i tlie stripping column. The polymer is precipitated from the stripped polvmer solution bv mixing with i water in a tank with specially designed high ^mensitv agitation. The polymer slurry from the precipitation operation is filtered and washed i with water on a rotarv vacuum filter. The wash nitrate is used in the pre cipitation operation. The filtrate is distilled to recover acetone, wkiv h is recvclcd to the polymerization reactor and to tin* stripping column. i The wet filter cake produced on the rotarv vacuum filter is dried :n an air convection oven and packaged in polvcthvlene-lined Leverpaks for shipment. RSV 0017298 IV BILL OF MATERIALS The following raw materials are required to make 100 lbs of finished goods Vinyl chloride Ethyl-?-hydroxyethyl i'umarato Isobutylene Azo-bis - isobutyr onit rile Acetone Paraplex G-62 CUv water " -i. 3 lb 1 *1 > lb t1 . --1 lb I. 0 lb 5. 3 lb 0. - lb 2120.0 Ih RSV 0017299 i i i i ji i i i i i i T I V. RAW MATERIALS A. ACETONE Source: Enjay Chemical Co. , 60 West 4th St. , New ork 20, N. \ . Grade : Commercial grade. Comment: 1. Enjay commercial grade acetone is made by isopropanol dehydrogenation. Acetone made by other processes should not be used without laboratory investigation. 2. Baker and Adamson A. C. S. grade* acetone was used in the early pilot plant work. The performance ol the Ernav acetone was in all respects equivalent to the A.C-.S. grade acetone. B. AZO- BIS - ISO BUT V R O NIT RILE Source: E. I. du Pont dc Nemours A: Co. , Industrial and Bioche mic al* Dept., Wilmington 2S, Del. Grade: Commercial Alternate Source: Eastman Organic Chemicals. Distillation Products Industries, Roche ste r 3, N. V. Comment: 1. The du Pont Company are the assignors of U. S. patents 2,47l,5? and 2, 500,023 wine:', relate to the use ot a;*.obis-isobutvronitrile in polymer manufu*. ture. Their current practice is to grant a license to use these patents to all applicants for a royalty fixed at 10c 'lb catalyst. It is their current practice to require cis'cmcrs to exe*. ute a license agreement before they will sell ar.o-b.sisobutyronitrile in greater than development quantities. C. E7HY L-5-HYDROXYETHYL FUMARATE Source: Monsanto Chemical Co. , Plastics Division, Sprir.utielc, Mass. Grade: Ethvl"~-hvdroxvethvl fumarate monomer prepared according to the methods of CRD Tentative Process Report 1577 ( 5 ). RSV 0017300 Specifications: Ethyl--hydroxyethyl fumarate Hydroxyl content Acidity Glycol difumarate Ethyl-*-hydroxyethoxyethyl fumarate Color > "5% > " <0.01 meq/ml < 5% < 11% lipht amber D. ISOBUTYLENE Source: Petro-Tex Chemical Corp. * P. O. Box Houston, Texas. Grade: 9% isoburylcnc. Specifications : Isobutylene Butanes Butylenes (other than IB) Propane and propylene Sulfur Water QO*% min. 0. f"f> max. 0. S"' max. 0. 0?% max . 10 ppm max. 100 ppm max. Alternate Sources: Phillips Petroleum Co., Special Products Division, Bar tie sville, Oklahoma. Enjay Chemical Co. , 60 West 4th S:. , New York 20, N. Y. Comment: 1. All research work was done with Phillips material distrib uted by the Mathcson Co. The spev. locations tor the products of the other manutac ture r s are vompur-ihlo to Phillips. Z. Monsanto Chemical Co. Hytlroi .irbor.s Divrsion have a process stream at the Chocolate H.ivou plum trom which isobutylene may be extracted. Is*. :> lone ruv become available from this source ;n the r.ture. PARAPLEX G-62 Source Rohm and Haas, Inc. Phiiadelp WATER Sou re e: Springfield, Mass. , municipal water supple. C omuu? nt 1. I he uiTtorrnaikc ot Sprmciield :r. . .pul water was equivalent to that c: deionized waiter :r. the pilot plur.t. RSV 0017301 e. 1 I VI. material balance Fipure 1 is a material balance for the VYSET RE process. RSV 0017302 vn FLOW SHEET AND EQUIPMENT DESCRIPTIONS i i Figures 2 and 3 are process flow sheets. Table I is an equipment list for the polymerization pilot plant. s Table I Equipment List - Polymerization Pilot Plant i CV-1 CV-2 Control valve - reactor effluent flow Control valve - reactor inerts bleed CV-3 Control valve - reactor jacket water temperature CV-4 Control valve - stripping acetone flow i CV-5 CV-6 Control valve - recycle system nitrogen bleed Control valve - stripper preheat water temperature CV-7 Control valve - polymer solution takeoff i CV-8 Control valve - acetone column bottom pressure D-l Monomer stripper i D-2 Acetone column E-l Reactor reflux condenser E-2 Stripper feed preheater i E-3 E-4 Recycle condenser and collector Recycle subcooler E-5 Product cooler i E-6 E-7 Stripping acetone vaporizer Stripping acetone superheater E-8 Acetone column feed heater i E-? E-10 Reflux head Acetone column bottoms cooler F-l Reactor effluent filter i F-2 F-3 VC1 monomer filter Stripping acetone filter F-4 Water filter i F-5 F-6 Rotary vacuum filter Acetone column primary filter F-7 Acetone column secondary filter i F-8 FI-1 Recycle acetone filter Nitrogen bleed rotameter (reactor) FI-2 Nitrogen bleed rotameter (recycle system) FI-3 Precipitant rotameter FI-4 Wash water feed rotameter FI-5 Acetone column feed rotameter RSV 0017304 Q N >u r i i INITIATOR SOLN |WATR RSV 0017306 J-l LC-1 L.C-2 LR-l LRC-1 LT-I LT-2 LT-3 LT-4 L.T-5 P-1 P-2 P-3 P-4 P-5 P-6 P-7 P-8 P-? P-10 P-11 P-12 P-13 PI-1 PI-2 PI-3 PI-4 PI-5 PI-6 PI-7 PI-S PI-9 PI-10 PI-11 PI-12 PRC-1 PRC-2 PRC-3 PT-1 PT-2 PT-3 R-l S-l T-l T-2 T-3 Feed system ejector Monomer stripper liquid level control Recycle collector liquid level control Feed system level recorder Reactor level recorder controller VC1 level transmitter IB level transmitter EHF level transmitter Initiator solution level transmitter Reactor level transmitter VC1 feed pump IB feed pump EHF feed pump_ Initiator solution feed pump Recycle pump Polymer solution pump Precipitant pump Polymer slurry pump Wash filtrate pump Filtrate pump Vacuum pump Blowe r Acetone column feed pump VC1 standpipe pressure indicator IB standpipe pressure indicator EHF standpipe vacuum indicator Initiator solution standpipe vacuum indicator Stripper steam trace pressure indicator Steam pressure indicator stripping acetone system Flow control pressure indicator Pressure indicator stripping acetone tank, Item T-5 Pressure indicator stripping acetone tank, Item T-6 Pressure indicator VC1 storage system Pressure indicator steam Pressure indicator acetone column feed Reactor pressure recorder controller ^ Recycle system pressure recorder controller Acetone column bottom pressure recorder controller Reactor pressure transmitter Recycle system pressure transmitter Acetone column bottom pressure transmitter Polymerization reactor Discharge snubber VC1 metering standpipe. IB metering standpipe EHF metering standpipe RSV 0017307 12 T-4 T-5 T-6 T-7 T-8 T-9 T-10 T-l 1 T-12 T-13 T-14 T-15 T-16 T-l 7 T-l S T -1 9 T -20 TI-1 TI-2 TI-3 TI-4 TR-1 TRC-1 TRC-2 TRC-3 TT-1 TT-2 U-l Initiator solution metering standpipe Stripping acetone storage tank Stripping acetone storage tank VC1 monomer storage tank IB monomer storage tank EHF monomer storage tank Initiator solution tank Polymer solution tank Precipitant feed tank SRC disintegrator Slurry hold tank Wash filtrate receiver Filtrate receiver Filtrate storage tank Distillate receiver Bottoms receiver Initiator injection tube Temperature indicator* acetone vaporizer steam supplv Precipitant feed temperature indicator Disintegrator temperature indicator Acetone column feed temperature indicator Temperature recorder recycle system Reactor temperature recorder controller Reactor jacket temperature recorder controller Feed preheater jacket temperature recorder controller Temperature transmitter reactor jacket Temperature transmitter feed preheater Tray dryer ITEM CV-1 CONTROL VALVE - REACTOR EFFLUENT FLOW Function: To control reactor effluent flow to stripper (D-l). Description: Research Control minim flow valve, Trim M Mat'l of Const: B_>dy and trim, stainless steel Comment: 1. See comment,Item LRC-1. ITEM CV-2 CONTROL VALVE - REACTOR INERTS BLEED Function: To control nitrogen bleed inerts from polymerization reactor (R-l). De sc ription: Research Control minim flow valve. Trim P-1 Mat'l of Const: Body and trim, stainless steel Comment: 1. See comment, Item PRC-1. RSV 0017308 a ITEM CV-3 CONTROL VALVE - REACTOR JACKET WATER TEMPERATURE Function: To control steam addition to polymerization reactor (R-l) jacket tempered water system. Description: Research Control minim flow valve, Trim D . Mat'l of Const Body and trim, stainless steel. Comment: 1. See comment, Item TRC-2. ITEM CV-4 CONTROL VALVE - STRIPPING ACETONE FLOW Function: To control stripping acetone flow to acetone vaporizer (E-61. Description: Research Control minim flow valve, Trim F-l. Mat'l of Const: Body and trim, stainless steel. Comment: 1. See comment, Item FC-I- ITEM CV-5 CONTROL VALVE - RECYCLE SYSTEM NITROGEN BLEED Function: To control inerts bleed from recycle system. Description: Research Control minim flow valve, Trim L. Mat'l of Const: Body and trim, stainless steel. Comment: 1. See comment, Item PRC-2. ITEM CV-6 CONTROL VALVE - STRIPPER PREHEAT WATER TEMPERATURE Function: To control steam addition to the hot water system lor the stripper feed preheater (E-2). Description: Research Control minim flow valve, Trim N. Mat'! of Const: Body and trim, stainless steel. Comment: 1. See comment, Item TRC-3. ITEM CV-7 CONTROL VALVE - POLYMER SOLUTION TAKEOFF Func tion: To control flow of polymer solution from the stripper (D-l) bottom. Description: Research Control minim flow valve, Trim O. RSV 0017309 14 I Mat'l of Const: Comment: Body and trim, stainless steel. 1. See comment, Item LC-1. ITEM CV-S CONTROL VALVE - ACETONE COLUMN BOTTOM PRESSURE Function: To control flow of steam to the acetone column (D-2) bottom. Dcsc r iption: Research Controls minim flow valve, Trim O. Mar '1 of Const: Body and trim, stainless steel. Comment: See comment. Item PRC-3. ITEM D-l MONOMER STRIPPER Func ti on: To remove unreacted IB and VCl iroin the reactor (R-l) effluent stream. De sc ript ion: Precision Distillation Apparatus Co. . 2 in. dia. Olaershaw column with 5 sieve trays. Mat* 1 of Const: Stainless steel. Services: Steam. Comment : 1. Details of the monomer stripper arc yivcn in Fipure 4. 2. Material balance and typical operating cona-.t.ons: reeus (c ..r i omp. Pressure NonvoJa?; ie s \'C1-IB Acetone {"O tpsig) fa rs> fsj f j J r 44, 8 '>4. ^ 67 5. 8 53 L - 48*. 0 i18 M 744. S o. 0 7 Ci 8. 6 85 mI N s<7e. r 365, 2 55 3. Heat Balance: me -s. nr:ot be j;:ve , .or.s o: v a o o r a r.d !1 'e ,,.d the monomer stripper feed stream J are unknown. The heat balance eiven below is based on pilot pln: Operations, usir.c the proport.or, v: vapor and i.c:u:d commuted bv the methods described in Section IX-3-2. RSV 0017310 2 I This heat balance should be used in design with fulI awareness of its limitations. I Basis: One hour's feed. Reference: Liquids at feed temperature (55"CJ zero enthalpy. I Thermodynamic properties (at 35 psia); I a. VC1 (6) Latent heat of vapor:z.ation AHv = 70. 1 cal / ct I Specific heat of vapor Specific heat of liquid Cp - 0. 227 cal^ c e C CP1 = 0. 27 :aI/c*C b. Acetone (7, S) I Latent heat of vaporization AI iv = 6. S t a 1 / g Specific heat of vapor {50*C) = 0. 36 ral':0C 1 ( I 1 Se C) = 0. 40 c al / c *C Specific heat of liquid c01 - 0. 57 .al'VC Polymer (?) Specific hea of liquid Cpi = 0. 25 cal / c *C I JL MN Flow Er.thalpv Flow Er.thalpv Flow Em rialDv Flow Er.thalpv I c/hr c al/hr VCl - IB i.vapor) 84 8. o 5 c', 4 87 C /r. r . c a 1! h r C / r. r i ai. h r c ''hr ^ a!,'hr VCI - 13 Liquid) 1 3 6. 2 0 -- 4 7^ . I Acetone ( vaporl 27 1. 1 26, 242 Acetone ' La ula) 404. 7 0 488, 0 5 5,5 18 -- -7 0 ft. 6 14, 5 no 565.2 '5,n 4 7 . Pol-, mer i liquid) 744. S -- 7 4 4 . r 5 . 5 ft - I Total :*.! * h r B*u ' h r o5,7 2 340. I 5 3, - : r 81 . 7 20, 5",J 4 j 4 . Ki i 04, 212. 3 1 Radiat- ion losses i n : h e c o 1 u m r. J - L - M - N : - 5 . ^ 3tu ' h r 4. The vapor "liquid eq ull.br. a netessarv for mak'r.c rigorous theoretical stage iahu'.a- ons are r.o; available. I Approximate cal:-l&tions have* beer, mlide .ors.during ?he binary system \ Cletore, ir.ese taL jlat or.s are LT-'er. in detail .r. Sv. non IX-B. I 4 I u e '.'isi om'v el "0 a . ; l: !' t pi.: r p i ' \ : m r -. . i .1; 5C*C is about 1 30 >.p. A plu.n: at r .up' :.>u x o:.-nr. should be cesicr.ed to handle u I:cu;c phasv of this *-: s. o: *v. RSV 0017312 6. An entrainment disengaging space is required at the top of the column. Vapor velocity in this space should not exceed 0. 1 ft/sec. The use of a wire mesh demister at the top of the column is not recommended because of possible plugging problems. There was little entrain ment in the pilot plant column. 7. Control of the stripping column operation is discussed :n detail in Section IX-B-3. * RSV 0017313 ITEM D-2 ACETONE COLUMN Function: To recover acetone from rotary vacuum rlter (F-5) filtrate accumulated in filtrate storage tank (T-17). De sc ription: Rectifying section - 1 in. ID > 10-trav Oldershaw sieve tray column. Stripping section - 1-7 / S ir.. ID , x 2*4 in. i oiumn packed with Goodloe wire mesh packing. Mat'l of Const: Comment: Rectifying section - Pyrex glass,. Stripping section - Pvrex glass with stainless steel packing. 1. The acetone column was operated with open steam to avoid a rcboilcr fouling problem. Z. Calculation of the number of theoretical stages reqmrcd for the separation is properly done by the modified latent heat of vaporization method (10). 3. Latent heats of vaoorization at 36. 1 eC (acetone bc:''nc point). Water : IS, 327 3tu / lb-m. Acetone: 12, 492 Btu/lb-m. Modified molecular weieht acetone Material balance lor modified mole c i;la r we Basis: 100 lb-m feed actual Reflux ratio 2 F' = 99. 493 lb-m. Xj- 0. 0108S D* - 1. 132 II xD- 0. oil-. S^ Note : 203 lb-m.- X\v - 0. 000302 Primed ssvymbols refer to modified mo'.*, -..l-s quantities. The X's are modified mole of acetone. o RSV 0017314 5. Figure 5 is a McCabe-Thicle diagram for the actual pilot plant column. Figure 6 is a plot for the low concentration region. The material balance in Figure 5 is for actual pilot plant operation and uses a more dilute feed shown in Figure 1. The plots show five stripping and seven rectifying theoretical stages. 6. The designofa distillation column for recovering acetone from dilute aqueous solution is considered in detail in Section IX-F. Provision must be made for the very high liquid rates in the stripping section. ITEM E-I REACTOR REFLUX CONDENSER Func t ion: To condense VC1 and IB vapors from reactor R-i. I De sc r iption: 3 ft length of jacketed 1 in. schedule ^0 pipe. Mat'l of Const: Stainless steel. Service s: Cooling water. Commcr 1. This condenser was not actually used in any of the pilot plant polymerization runs. The reactor (R-l) temper ature was controlled bv adjusting the temperature of the water in the reactor jacket. 2. The condenser was used when the re.ii. tor was . leaned with refluxing solvent. ITEM S-2 STRIPPER FEED PREHEATER Func tion: Description: To preheat monomer stripper (D-l) feed. 2 ft length of jacketed ? ; e 11 e 10 *j i u e . Mat1! of Const: Stainless steel. Se rvices: Comment: Hot water. 1. The inlet temperature o e stripper :oeci o re nea ;j r wets measured at 30cC. This temperature rotloi.ts the radiation losses from the reactor effluent flowing through the effluent filter (F-l) and the effluent line loading to the flow control valve (CV-1). 2. A heat balance for the stripper feed preheater is deportee on the inlet temperature whith : r. plan: operation, would be expected to be eonsidc rablv r.igher than iOc C be-, ause o; smaller radiation, losses. The slant hear balaru e should be based on consideration of expected radiation. losses. RSV 0017315 ITEM E-3 RECYCLE CONDENSER AND COLLECTOR Function: Desc ription: To condense and collect the recycle vapors from the stripper (D-l). Young Radiator Co. heat exchanger Model 202Y tollowed by a 3 ft length of jacketed Zi in. schedule *40 pipe. Mat'l of Const: Stainless steel. Services: Cooling water. Comment: 1. The condenser sloped downward from the stripper (D-l) to the vertical collector. The feed line extended 24 inches into the collector. Top of the collector served as an inerts separator. In a plant installation the condenser should be operated vertically with the vapor inlet at the top to give a subcooled feed for the rccvelc pump (P-5). 2. The sensing probe of the liquid level controller (LC-2) extended 6 inches through the bottom of the collector. 3. A drain port was provided for draining the collector to waste in the case of entrainment of polvmer from the stripping column (D-l). 4. The maximum temperature at which the recycle vapors will condense at 35 psia is 22. 5eC. 5. Heat balance: (Basis liquid condensate at r.cro enthalpy. Sec Item D-l, Stream X. ) Enthalpy condensate Enthalpv feed Heat duty 0 414. Btu'hr - 4 4. u Rtu/hr - 14 0.4 Btu 'lb \ aoor ITEM 5-4 RECYCLE SUBCOOLER Func ti on: To subcool condensed recycle vapors from the rm v*. 1c condenser and collector (E->). De sc r ip ti or.: 1? ft length of jacketed k in. schedule 40 pipe :ns.de a 24 x 6 x 7if in. sheet metal brine buth. Mut 1 ui Cunst: Stair.le s s steel. Se rvices : Refrigerated brine. Comment: 1. The purpose of E-4 was to subcool trie -emu le s'ream to prevent flashing across the suit tor. v-ilves o: the recycle pump (P-5). RSV 0017318 2 . The bath was designed to accommodate and cool the remote pump head of pump (P-5)uti?;* ing available refrigeration. 3. In the plant use of an adequate head between Ac recycle collector (E-3) and recycle pump (P-5) will prevent flashing across the check valves. For a mixture of 70% VC1 and 30% acetone at 60*F the head should be calculated on the basis of 2. 6 ft of head per ps: pressure drop across the suction valve. With this arrangement no recycle subcooler will be required. 4. In pilot plant operations, a small amount of ice which fouled the check valves ol the recycle pump (P-5) formed when the brine in the recycle subcooler was operated below -5"C. ITEM E-3 PRODUCT COOLER Func ti on: To cool the polvmer solution effluent from the bottom ci the monomer stripper (D-l). Dc sc r iption: i-in.OD 20 -ft coiled tube in 2-in.schedule 40 pipe. Mat `1 of Const: Stainless steel. Se rvic c s : Cooling water. Comment 1. The product cooler was used to cool the poivmer solution below its normal boiling point (bh'. SeC) to prevent flashing of acetone during letdown to atmospheric oressure. The heat capacities arc (8. ) Polymer CD - 0.25 cal/g'C Acetone Cl, - 0. 357 cal/gfC at 50cC Cl, - 0. 37 8 cal'g^C a; > 5 * C To cool 47. 3% polymer solution from &5 to 58*C, assuming negligible heat of mixing (basis 1.0 g) 0.H = -0. 47 3 (0. 25}(85-5 0) - 0. 52 3 (0. ^~i(85-50) uH = -10. S5 cal/'c C.K - "20. 1 Btu lb oolvmer solution ITEM Z-5 STRIPPING ACETONE VAPORIZER *;on: Des: notion: To vaporize the acetone fed to the bottom of the stripper (D-l). Jacketed coiled tube 3.- 5 in.OD. Mean coil diameter 2* ,n. 23 ft long. Total length of heat exchanger 2t ft. RSV 0017319 Mat'l of Const: Service s: Comment: Stainless steel. 120 psig steam. 1. Heat duty (basis: 1.0 g). See Item D-I for thermodynamic propertie s. Sensible heat duty liquid acetone (85-30) 0. 57 = 31.4 cal/g Heat of vaporization = 96. 8 cal/g = 128. 2 cal/g = 231 Btu/lb. 2. Steam pressure in the jacket was regulated. ITEM E-7 STRIPPING ACETONE SUPERHEATER Func :ion: Description: To superheat the acetone feed to the bottom of the monomer stripper (D-l). 3 --ft length of jacketed t-in.OD tubing. Mat'l of Const: Stainless steel. Se rvicc s : 1 20 psig steam. Comment: 1. The steam supply to heat exchangers (E-7) and (E-6) is connected in series with (E-7) leading (E-6). Temper ature control is effected through steam pressure regulation. 2. Heat Duty. Sensible heat dutv acetone vr.por AHS = (1 18-85) 0. 39 - 12. Q cal/g ^ 23. 2 Btu/lb. ITEM E-8 ACETONE COLUMN FEED HEATER unction: To heat feed to the acetone column (D-2). De sc riptior.: 12 mm dia x 30-in.glass tube wound with r.ichromc heating wire. Mat'l o: Ccr.s:: Py rex g la s s. Se rv-.ee s : I 10 v electricity. C cm me r.t: 1. The temperature o: the heater effluent was controlled by manual adjustment o: a variable transformer in the heater power supply. 2. Heat Duty : Inlet temperature 20*C. Outlet temperature 36eC. Soecific heat o: mixture = 0. S 3tu/lb-eF. Heat dure - (0. 96)(36-20){1. S) = 2S. 2 Btu/lb. RSV 0017320 ) c r i i i i i i i i i i i i i T I i i 3. This heater would not be required in a plant, since use of cold feed to the acetone column . u 2) is desirable to minimize the number of theoretical stages required. ITEM E-9 REFLUX HEAD r unction: To condense vapor from acetone column (D-2) and split distillate and reflux. Description: Swinging funnel type glass reflux head. Mat'l of Const: Services: Comment: Pyrex glass. Cooling water 110 v a-c electricity. 1. The reflux ratio to the acetone column was controlled by a Flexopulse timer in the power supply to a solenoidoperated valve in the reflux head. 2. Heat Duty: aKy. - 225 Btu/lb vapor condensed at reflux ratio 2. Heat duty = (3)(ZZ5) 675 Btu/lb distillate. 3. For plant application, a cooler for the distillate might be required to keep the distillate receiver (T-18) from running hot. Hot acetone may not be used to dissolve initiator because of the explosion hazard. ITEM -10 ACETONE COLUMN BOTTOMS COOLER r ur.k. ::on: To cool acetone column (D-2) bottoms. Description: 20 mm dia x 10-in,glass tube with water jacket. Mi: `1 of Const: Pyrex glass. Services'*. Cooling wate r. le n: 1. This stream need be cooled only if value can be recovered from the sensible heat,or if discharge of ho: water into the plant sewers is prohibited. ITEM F-l REACTOR EFFLUENT FILTER r cr.cnon: To filter reactor (R-1-) effluent to remove polymer "grit". Dr sc riotior.: Cuno "Micro-Klean" fiber cartridge filter Model 1H1 w:th filter cartridge 2278-33 (5 u.). RSV 0017321 2b. i Comment: i This cartridge filter is in general use in the paint and varnish industry where filtration is required after formulation. A similar filter cartridge made of wool fibers (Cuno 2278-Cl) should not be used because it discolors polymer solution. ITEM F-2 V Cl-MONOMER FILTER i Function: To filter VCl monomer charged to VC1 metering standpipe (T-l). i Description: Hoke 540 series 50 micron filter. Mat'l of Const: Body - stainless steel. I Gasket - Teflon. t ITEM F-3 STRIPPING ACETONE FILTER Function: To filter stripping acetone upstream of stripping acetone flow controller (FC-I). i De sc r iption: Hoke 540 series 50 micron filter. i Mat'l of Const: Body - stainless steel. Gasket - Teflon. 1 ITEM F-4 WATER FILTER Function: To filter water feed to the disintegrator (T-l 3) and the i rotary vacuum filter (F-5). De sc ription: Cuno "Mic r o-KIeanH fiber cartridge filter Model Cl AG with i 5 p. cartridge. Mat' i of Const: Filter housing - stainless steel. l Filter cartridge - cellulose. C omme nt: 1. Used to filter rust oarticles from water suddIv. i i i RSV 0017322 ii 1 i Hu ,|i|||ii 'i i"t i "i ........................................j win ' i "i iiii i|1 1 i ITEM F-5 ROTARY VACUUM FILTER Function: To filter and wash precipitated polymer accu iulated in slurry hold tank (T-14). Description: Eimco Corp. drum type laboratory filter station 18 in. dia x 12 in. face. Mat'l of Const: All wetted parts stainless steel. Service s: 220 V, 3-phase electricity. Seal water for vacuum pump Item P-11. Comment: 1. The laboratory filter station consisted of the following components: a. Filter drum: 18 in. dia x 12 in. wide. b. Filter medium: Cotton filter cloth on 4 x 4 wire mesh support. c. Drum drive: t hp variable speed (0. 1-1 rpm). d. Agitator: Two steel arcs for swing type agitation. During operation it was found that filter cake formation on the drum was improved with a more vigorously agitated polymer slurry. A small airdriven laboratory mixer was employed for agitation of the slurry in the filter tank. e. Vacuum pump, Item P-11. f. Blower, Item P-12. g. Filtrate pumps, Items P- and P-10. h. Filtrate receivers, Items T- 1 5 and T-16. ITEM F-b ACETONE COLUMN PRIMARY FILTER function: To remove polymer solids from acetone olumn (D-2) feed. Descrip'.on: Enzinger (Duriron Co., Dayton, Ohio) Laboratory vertical lea: pressure filter. The filter had 3 leaves, one with two 4-1/8" x S-l/2" and two with two 2-i / 6" x 8-1/2" filtering surfaces. The leaves were dressed with cloth bags sewn over them. RSV 0017323 8. Mat'l of Const: Comment: Filter - type 316 stainless steel. Gaskets - Buna-N. Cloth - polypropylene style B-3407 (Nationa Corporation). . ilter Media 1. The sire of filter required for the plant will depend on the amount of solids passed by the rotary vacuum filter (F-5). An undetected cloth failure on the rotary vacuum filter could produce a very large surge of solids into the acetone. This did not occur in the pilot plan:, but in extended plant operation an occasional mishap of this type would be inevitable. Plant designs should be developed to deal with this situation. Z. In the pilot plant an average cake of about 1/64" polymer was developed upon filtering 10 kg of acetone column feed. _ The material was mostlv, but not entirely soluble in acetone. 3. A small amount of polymer was found on the leaves, inside the filter bags. Tms was mostly, but not entirely soluble in cold acetone. A Sparkler-tvpc filter, which has easily cleaned internals, is recommended over the pressure leaf type filter for this service. ITEM F-7 ACETONE COLUMN SECONDARY FILTER Func tion: To remove fines passing the acetone column, primarv filter (F-6) from the acetone column (D-2) feed. Description: Cuno model 1C1AG Micro-Klean filter wit}'. (Cuno Engineering Co., Meridan, Conn. ). element Mat'l of Const: Body - die-cast aluminum. Element - cellulose. Comment: 1. The very small amount of material collected in this filter from 10 kg of acetone column feed appeared roughly equal to that deposited m the acetone column (D-2). For this reason, inclusion of the secondary filter in plant design appears justified. Another advantage is that it will prevent a large amount of polymer from entering the column in the ever.: of a failure in the acetone column primarv fiber iF-5). ITEM F-8 RECYCLE ACETONE FILTER rur.ction: To :ilter recvcle acetone downstream of the distillate ri-i'i'ivcr ( T - 1 8). RSV 0017324 " Description: Cuno Model iClAG ^cro'-^eanVilter with^ ^ element*^^^ Mat'l of Const: Body - die-cast aluminum. Element - cellulose. ITEM FC-1 STRIPPING ACETONE FLOW CONTROLLER Function: To sense and control flow of liquid stripping acetone to the acetone vaporizer (E-6). Description: Foxboro model M /59 flow controller mounted on Foxboro model 1 3A d/p cell with 0.020-inch integral orifice. Mat'l of Const: Body - stainless steel. Gaskets - Teflon. Sc rvic es Instrument air. Comment: 1. Controller set point signal was indicated by PI-7. Flow was recorded on FR-1. The output signal was applied to CV-4. ITEM FI-1 NITROGEN FEED ROTAMETER (REACTOR) Function: To indicate Na flow to the reactor (R -- 1). This nitrogen provided a Nj pad over the reactor charge. Description: Fischer and Porter 1700 series rotameter with 02F-1/8-165/70 tube and 1 / 8-in.sapphire ball float. Mat'l of Const: Tube - glass. Body - stainless steel. Gaskets - neoprene. Comment: 1. When the reactor (R-l) was operated at autogeneous pressure, this rotameter was needed only lor purging during start-up. ITEM FI-2 NITROGEN FEED ROTAMETER (RECYCLE SYSTEM) Function: To indicate flow to the recycle condenser (E-3). Description: Fischer and Porter 2700 * ;ries rotameter with FP-1/16-20G-5/81 tube and 1 / 16-in.sapphire ball float. Mat'l of Const: Tube - glass. Body - stainless steel. O-Ring - Buna-N. RSV 0017325 30. ITEM FI-3 PRECIPITANT ROTAMETER Function: To indicate flow of precipitant to first stage of SRC dis integrator (T-13). Description: Fischer and Porter 700 series rotameter with B-4-21-10/27 tube and special float. Mat'l of Const: Tube - glass. Body - stainless steel. Gaskets - neoprene. ITEM FI-4 FILTER CAKE WASH ROTAMETER Function: To indicate flow of wash water to rotary vacuum filter (F-5). Description: Fischer and Porter 700 series rotameter with B-4-21-10/27 tube and special float. MatM of Const: Tube - glass. Body - stainless steel. Gaskets - neoprene. ITEM FI-5 ACETONE COLUMN FEED ROTAMETER Function: Description: To indicate flow of acetone filtrate feed to acetone column (D-2). Fischer and Porter 2700 series rotameter with 2F tube and X-in.ball float. Mat'l of Const: Body - stainless steel. Tube - glass. Float - stainless steel. Gaskets - Viton. ITEM FR-I STRIPPING ACETONE FLOW RECORDER Function: To record stripping acetone flow to acetone vaporizer (E-6). Description: Foxboro universal case pneumatic recording receiver. Comment: 1. See comment, Item FC-1, for details of application. RSV 0017326 31. Function: Description: Mat'l of Const: Service s To evacuate all feed system components requiring charging by vacuum (T-3, T-4, T-5, and T-6). Worthington 2^ cpa steam ejector with 9-in.barometric condenser used at atmospheric pressure. Porcelain. 100 psig steam, water to condense plume. ITEM LC-1 MONOMER STRIPPER LIQUID LEVEL CONTROL Function: Description: Mat '1 of Const: To control product withdrawal rate through liquid level control in the bottom of the monomer stripper. Robcrtshaw Fulton model 305-A3-N1 Level-Tek liquid level control operating a Skinner X-5 three-wav solenoid valve through an Agastat 22QT time delay relay. Sensing probe - Teflon covered, in stainless steel mounting gland. Se rvicc s: 110 V, 1-phase electricity. Instrument air. Cornmen': 1. The 110 V output of the liquid level control relav was converted to a 3-15 psi off-on air signal utilizing the three-way solenoid valve. 2. The Agastat time delay relay was included in the svstem to prevent too rapid cycling of the solenoid valve. ITEM LC-2 RECYCLE COLLECTOR LIQUID LEVEL CONTROL r unc tion: To control the operation of the recycle pump (P-5) through liquid level control in the recycle collector (E-5). De sc ription: Robertshaw Fulton model 305-A3-N1 Level-Tek liquid level control operating the recycle pump (P-5) through an Agastat 22QT time delay relay. Mat'l of Const: Sensing probe - Teflon covered, in stainless steel mounting gland. Services: 110 V, 1-phase electricity. Comment: 1. The Agastat time delay relay was included in the svstem to prevent too rapid cycling of the recycle pump. RSV 0017327 32. ITEM LR-1 FEED SYSTEM LEVEL RECORDER Function: To record the signals from the feed metering standpipe level transmitters (LT-1, LT-2, LT-3 and LT-4). Description: Minneapolis-Honeywell Brown 6-point strip-chart recorder; 0-10 mv range. Services : 110 V electricity. Comment: 1. The signals recorded indicate the liquid level in the standpipes (LT-1, LT-2, LT-3 and LT-4). 2. The instrument included a device for marking the chart at half-hour intervals. ITEM LRC-1 REACTOR LEVEL RECORDER CONTROLLER Function: To record and control the reactor (R-l) liquid level by setting the control valve reactor effluent flow (CV-1). Dc sc ription: Bristol series 532 recording pneumatic controller with 5-8 psig input span. Services: Instrument air. Comment: 1. For proper operation of the stripping column (D-l), the flow of reactor effluent to the stripper must be controlled by a flow controller cascaded with the output of the reactor level controller (LRC-1). See Discussion Section IX.-A-3-C. 2. See comment, Item LT-5. :TEM LT-1 VC1 LEVEL TRANSMITTER Function: To transmit VCI level in YC1 metering standpipe (T-l) to feed system level recorder (LR-1). 1 De sc ript: or : Foxboro model 613 electronic d/p cell transmitter. Mat 'I of Cons ?: Body - stainless steel. Servic e s: 110 V electricity. ITEM LT-2 : ur.L t: on : IB LEVEL TRANSMITTER To transmit IB level in IB metering standpipe (T-2) to feed system level recorder (LR-1). RSV 0017328 33. Ma.t'1 of Const: Body - stainless steel. I Services: 12 0 V electricity. I ITEM LT-3 EHF LEVEL TRANSMITTER I Function: To transmit EHF level in EHF metering standpipe (T-3) to feed system level recorder (LR-1). Description: Foxboro model 613 electronic d/p cell transmitter. ! Mat'l of Const: Body - stainless steel. i So rvices: 110 V electricity. ITEM LT-4 INITIATOR SOLUTION LEVEL TRANSMITTER r Function: To transmit initiator solution level in initiator solution metering standpipe (T-4) to feed svstem level recorder i (LR-1). Description: Foxboro model 613 electronic a/p cell transmitter. i Mat'l of Const: Body - stainless steel. Se rv-.i. e s : 110 V electricity. i ITEM LT-5 REACTOR LEVEL TRANSMITTER function: To sense and transmit reactor (R-l) level to reactor level recorder controller (LRC-1). Des: riprlon: A device for measuring the force exerted on a paddle attached to the end of a lever by the swirling motion of lia4uid in the reactor and transmitting a pneumatic signal i proportional to this force. Ma* 1 oi-Const: Paddle and lever assembly - stainless steel. Sen-. ;<:es: Instrument air. Comment: 1. Details o: the instrument are given in rhe Central Research Department Analytical Devices Laboratory Drawing G-6272. The device consisted essentially of a lever extending into the reactor (R-l) through a top nozzle, a paddle at the end of the lever which was RSV 0017329 34. a;. ?***T*tV: VSv.Vsl _............................................................ partially immersed in the process fluid, a bellows seal on the lever at the nozzle, a fulcrum just above the nozzle and a Tate Emery DA-60 pneumatic load cell above the fulcrum for measuring the force exerted on the paddle by the swirling liquid. 2. In operation the force exerted on the paddle was proportional to the depth of immersion of the paddle in the reactor charge. As the level rose, the signal transmitted by the load cell increased; as it fell, the signal decreased. 3. A new design for a level transmitter employing the principle of measuring the force exerted on a partially immersed paddle by a swirling liquid has been prepared. In this design the measuring head of a Foxboro Type 18A target flow transmitter is used to sense the force on a lever extending into a stirred reactor. The Foxboro transmitter contains sensing device, fulcrum and seal. Its use offers the advantage of substantial mechanical simplicity. 4. The calibration of this device was checked periodically by opening the outlet of an inverted dip tube which extended into the reactor from the top and terminated at the desired reactor operating level. ITEM P-1 VC1 MONOMER FEED FUMP Fur.c ::on: To pump and meter VC1 monomer feed to the reactor (R-l). Des: riot ion; a) For flow rates above 800 g/hr - Lapp Puisaieeder type i CPS-1 positive displacement diaphragm pump. b) For flow rates below 800 g/hr - Lapp Pu'sateeder type i LS-20 positive displacement diaphragm pump. Me; 1 o: Const: Reagent head - stainless steel. Diaphragm, CPS-1 - stainless steel. LS-20 - Kel-F. Ser . i;es: 110 V, 1-phase electricity. 1. The check valves on the larger pump typ CPS-1 were unsatisfactory for VC1 service and were supplemented with Hoke series 570 check valves at both suction and discharge locations. RSV 0017330 35. The feed rates of>U Lapp pumps were 'monitored m continuously during runs and the pump settings were adjusted at half-hour intervals to maintain desired feed rates. This was done because the volumetric efficiencies of the diaphragm pumps, particularly when operated below 50% of rated capacity, were poor and exhibited pronounced sensitivity to fluctuations in suction pressure* A complete charge of a standpipe would increase the suction pressure 2-3 psi with a con-, current increase of flow rate of about 10%. In the case of the volatile feed streams fluctuations in the static pressure controllers caused considerable drift in pump feed rates (as much as 20%). Through continued monitoring the feed rates were generally maintained within x 2% of the total feed. ITEM P-2 IB MONOMER FEED PUMP Function: To pump and meter IB monomer feed to the reactor (R-l). Dcsc ription: Mat ' 1 of Const: Lapp Pulsafeedcr type LS-10 positive displacement diaphragm pump with 4:1 gear reduction in the motor drive. Reagent head - stainless steel,Diaphragm -Kel-F. Services: 110 V, 1-phase electricity. Comment: 1. See comment, Item P-1. TEM P-3 EHF MONOMER FEED PUMP r uncnon: To pump and meter EHF monomer feed to the reactor (R-l). Desc rip:lor.: Lapp Pulsafeeder type LS-20 positive displacement diaphragm pump. Ma-'l o: Ccr.st Reagent head - stainless steel. Diaphragm - Kel-F. 110 V, 1-phase electricity. 1. See comment, Item P-1, ITEM P-4 INITIATOR SOLUTION FEED PUMP ur.c nor.: 10 pump and meter initiator solution feed to the reactor (R-l). De s, riotior.: Same as Item P-3. RSV 0017331 36. Comment: 1. See comment, Item P-1. ITEM P-5 RECYCLE PUMP Function: To pump condensate from recycle collector (E-3) to reactor (R-l). De sc ription: Lapp Pulsafeeder type LS-3 positive displacement diaphragm i pump with .remote head. Ma ('1 of Const: Reagent head - stainless steel. Diaphragm - stainless steel. Se rviccs: 110 V, 1-phase electricity. Comment : 1. See comment, Item E-4. ITEM P-6 POLYMER SOLUTION PUMP Function: To pump and meter polymer solution feed to SRC Dis integrator (T-13). De sc ription: Viking model C54G pump with Vickers variable speed drive. Mat '1 of Const: All iron. Sc rvR es : 220 V, 3-phase electricity. ITEM P-7 PRECIPITANT PUMP Function: Des<_ rip' 'or: To pump precipitant (water) from precipitant feed tank (T-12) to SRC Disintegrator (T-13). Eastern model D-ll centrifugal pump. Ma*'! of Const: Body and impeller - stainless steel. Seal gaskets - Teflon. Se re ices: 110 V, 1-phase electricity. ITEM P-8 POLYMER SLURRY PUMP Fur.c *ion: To transfer polymer slurry from slurry hold tank (T-14) to rotary vacuum filter (F-5). De sc rip:. or.: Eastern model D-ll pump with mechanical seal. RSV 0017332 37. i Mar'l of Const: Body and impeller - 316 stainless steel. Seal gaskets - Teflon. i Sc rvices: 110 V, 1-phase electricity. t ITEM P-o WASH FILTRATE PUMP Function: To Transfer wash filtrate from wash filtrate receiver (T-15) i Description: to precipitant feed tank (T-12). Worthington Worthite model l-CNG-64 centrifugal pump. Mat'l of Const: Stainless steel. Comment: 1. A check valve is required in the pump discharge line to preserve vacuum in the filter system. Z, This pump is part of the rotary vacuum filter (F-5). ITEM P-10 FILTRATE PUMP Function: To transfer filtrate from filtrate receiver (T-16) to filtrate hold tank (T -17 ). i Description: Worthington Worthite model l-CNG-64 centrifugal pump. Ma*'t of Const: Stainless steel. Comment: 1. See emment, Item P-9. TTEM P-11 VACUUM PUMP Fun- `.or.: To maintain vacuum in the rotary vacuum filter (F-5). Dot', rip'ion: Nash rotary wet pump model MD-573. Mat 1 of Const: Unknown. Comment: 1. The vacuum Dump is oart of the rotarv vacuum filter (F-5). ITEM P-12 3LOWER run, :;on: To supply air for blowing filter cake from filter cloth on rotary vacuum filter (F-5). I Dr: s v r.Dt .or.: Roots-C onner sville model 22-AF low-pressure blower. i RSV 0017333 3b. Mat'l of Const: Comment: Unknown. 1. The blower is part of the rotary vacuum filter (F-5). ITEM P-13 ACETONE COLUMN FEED PUMP Function: To pump feed to acetone column (D-2) from filtrate storage tank (T-17). Description: Eastern Industries model D-ll centrifugal pump. Mai '1 of Const: Body and impeller - stainless steel. Seal gaskets - Teflon. Sc rvic e s: 110 V, 1-phase electricity. C omme nt: 1. A steel pump may be used in this service. ITEM PI-1 VC1 STANDPIPE PRESSURE INDICATOR Function: To indicate pressure in VC1 metering standpipe (T-l). Description: Ashcroft 30-0-150 compound pressure gage. Mat'l of Const: Steel. ITEM PI-Z IB STANDPIPE PRESSURE INDICATOR Function: To indicate pressure in IB metering standpipe (T-2). Description: Ashcroft 30-0-150compound pressure gage. Ma: '1 o: Const: Steel. ITEM PI-3 EHF STANDPIPE VACUUM INDICATOR Function: To indicate vacuum in EHF metering standpipe (T-3) during charging operation. Description: Ashcroft 0-30-invacuum gage. Mat'l o: Const: Steel. RSV 0017334 ITEM PI-4 INITIATOR SOLUTION STANDPIPE VACUUM INDICATOR Func Lion: To indicate vacuum in initiator solution standpipe (T-4) during charging operation. Description: Ashcrolt 0-30 in vacuum gage. I it Mat'l of Const: Steel. ITEM PI-5 STRIPPER TRACE STEAM PRESSURE INDICATOR Function: To indicate steam pressure on trace line on the exterior surface of the bottom section of the monomer stripper (D-l). Description: Ashcroft 0-100 psi pressure gage. Ma*`I of Const: Brass. ITEM PI-6 STEAM PRESSURE INDICATOR STRIPPING ACETONE SYSTEM Function: To indicate steam pressure of steam supply to stripping acetone vaporizer (E-6) and superheater (E-7). Dcs> rjptton. Ashcroft 0-200 psi pressure gage. Ma:'l of Const: Brass. ITEM Pr-7 FLOW CONTROL PRESSURE INDICATOR Fujivtiort: To indicate set point signal to stripping acetone flow controller (FC-I). Des- r p* on: Foxboro small case indicating gage model B5084, 3-13 psi. Mat 1 o: Const; Steel. i IE MS PI-s AND PI-o STRIPPING ACETONE TANK PRESSURE INDICATOR rar.c'.or.: 10 indicate pressure in the stripping acetone storage tanks (T -5 ) and (T-6). Oes-. r.pt.on: Ashcroft 0-100 psi pressure gage. Mu: 1 u: Cons*.: Brass. RSV 0017335 40. i i ITEM PI-10 VC1 STORAGE TANK PRESSURE INDICATOR Function: To indicate transfer pressure in the VC1 monomer storage i tank (T-7). Description: Ashcroft 0-300 psi pressure gage. i MatT of Const: Steel. ITEM PT-11 STEAM PRESSURE INDICATOR Function: To indicate steam pressure in steam line to steam-water i mixer used to heat precipitant feed to SRC Disintegrator (T- 13). i Description: Ashcroft gage"D-200 psi. Mat'! of Const: Stainless steel tube. i TEM PI- 1Z PRESSURE INDICATOR ACETONE COLUMN FEED Func lion: To indicate pressure in acetone column (D-2) feed system. Dost ription: Ashcrolt 0-15 psig gage. MatM of Const. Steel. Coni men* : 1. The pressure in the feed system was controlled by recycline part of the output of the acetone column feed pump (P-13) to the filtrate storage tank (T-17) on manual control. ITEM PRC-1 REACTOR PRESSURE'RECORDER CONTROLLER Furu. F on: To record and control pressure in reactor (R-l). Dus- r:p': or.: Foxboro model M40 recorder controller. Ser\ ':e s: Instrument air. C ommer.t: 1. Pressure control was effected through ir.erts bleed through control valve (CV-2). ITEM PRC -2 RECYCLE SYSTEM PRESSURE RECORDER CONTROLLER r i To record and control pressure in the recycle system. 1 RSV 0017336 41. I  Description: Foxboro model M54Consotrol receiver recorder with integrally mounted M5 8 controller. | Services: 110 V electricity. Instrument air. j Comment: 1. Pressure control was effected through inerts bleed through control valve (CV-5). ITEM PRC - 3 ACETONE COLUMN BOTTOM PRESSURE CONTROLLER Func tion: To control pressure drop across acetone column (D-2) as indicated by acetone column bottom pressure.transmitter (PT-3) by controlling flow of open steam to the column bottom. De sc riprion: Foxboro M-40 pneumatic recording controller with proportional and reset control actions. Sc rviccs: Instrument air. C omme n: : 1. Acetone column bottom pressure control valve (CV-8) was set bv this controller. ITEM PT-1 REACTOR PRESSURE TRANSMITTER F`.:uc von: To transmit reactor (R-l) pressure to reactor pressure recorder controller (PRC-1). Dose ription: Foxboro model 45 indicating pressure transmitter, 0-350 ps Mat 1 o: Const: Sensing element - stainless steel. Sv.-vl. es: Instrument air. TEN! PT-2 RiECYCLE SYSTEM PRESSURE TRANSMITTER De sc r ip:; on: To transmit recycle collector (E-3) pressure to recycle system pressure recorder controller (PRC-2). Foxboro model 45 indicating pressure transmitter, 0-75 psis: range. Ma:`l o: Const: 2'j rv: t,* s : Sensing element, stainless steel. Instrumer.t air. RSV 0017337 ITEM PT-3 ACETONE COLUMN BOTTOM PRESSURE TRANSMITTER Function: To sense pressure in steam line leading to bottom of acetone column (D-2). Description: Taylor pneumatic differential pressure transmitter 0-40 in Ha O range. Mat'l of Const: Comment: Stainless steel. 1.' The high-pressure tap of the transmitter was connected to the steam line leading to the column bottom. The lowpressure tap was open to the atmosphere. 2. See comment, Item PRC-3. ITEM R-l POLYMERIZATION REACTOR Kune rion: Reactor for solution polymerization of VC1, EHF and IB. Dc sc riplion: Glascote type LR-2 2-gallon jacketed glass-lined laboratory reactor with agitator. Mat'l of Const: Vessel - glass-lined steel. Flanges - stainless steel. Gaskets - Teflon envelope. Safety head - nickel. Service s: Jacket - hot water. Drive - 220 V, 3-phase electricity. Commen:: 1. Reactor was rated for 500 psig internal working pressure The safety head was rated to burst at 305 psig. 2. Reactor agitator: Two-bladed anchor *vpo, glass .oated, to fit 1? in.rotary mechanical seal. Agitator drive: Single speed belt driver. 345 rpm. Rotary mechanical seal: Durametallic type consisting of lj in.rotary double dura-seal assembly and unitized seal oil circulating and cooling sy stern Dura-pres sure unit G-5-I00. The seal oil pressure was set at 30 psi above reactor operating pressure. Seal oil SAE *10. 3. Reactor nozzles: Head : Agitator shaft - IX in. Vent assembly - 1 in. Feed inlet - 1 in. Bottom: Product removal - 2 in. RSV 0017338 53. Vent Assembly: One in. stainless steel flange containing a 1 in. schedule 40 vent line which was connected to the reflux condenser and from which a 1 in. schedule 40 pipe was branched to the safety head and the manual vent valve. Contained within the vent line were a i in. stainless steel tube recycle line which extended 6 in. below the flange face of the nozzle and a i in. stainless steel dip tube, the immersion length of which provided a reference for the reactor volume. Feed Inlet: The mounting plate for the reactor level transmitter (LT-5) acted as a flange containing four 1/8 in. stainless steel feed inlet tubes which extended 2* in. below the mounting plate face without interfering with the motion of the liquid level sensing lever (LT-5). Bottom: Two in. stainless steel flange containing a * in. product drain flush with flange face, a * in. tube thermowell extending 3 in. above the flange face and a i in. sample port extending 2 in. above the flange face. 4. Heat Release: The standard heat of polyme rization for vinvl chloride e stimated by Roberts (11) is -IT kcal/g-mole (liquidliquid). This estimate is taken as the heat of reaction for VC1/EHF/I3 terpolviner at sOeC in the absence cf better data. For reaction at SC'C with the feed and recycle input streams at 20eC, the enthalpy of the feed relative to liquid at 80C per hour is: VCI t IB Ace: one H- O AZB X (1 45 c. 2) <C. 27) (20--80) = -23o 5*; .ul ' hr ( 165. 1) (0. 56) (20.-80) - - 5 4 r I cal f hr ( 100. o) (0. t>2) (20--SO) = - 3742 cal/ ( 675. S) (0. 57) (20--80) = -231:2 cal' hr ( 2) c. 00) (20--80) = - 4 32 cal' hr ( l . 3) (C. 50) (20--SO) = - 21 z .a! hr -5er25 t al h r Mole 5 poly me hr - 10. 0 5 r g. mole h r. RSV 0017339 Heat release = (17, 000) (10. 058) 56625 = (170986 - 56625 = 114361 cal/hr = 114361/730.7 - 156. 5 cal/g polymer ~ 282 Btu/lb polymer ITEM S-l DISCHARGE SNUBBER Function: To separate seal water from exhaust products of blower (P-12). Description: Burgess STC-1 snubber. Mat'1 of Const: Unknown. Comment: 1. The discharge snubber was part of rotary vacuum filter (F-5).- ITEM T-I VC1 METERING STANDPIPE Function: To store VC1 under helium pressure for continuous feed to reactor (R-l). De scription: 7 ft length 3 in. schedule 40 seamless pipe mounted vertically. Mat'l of Const: Stainless steel. Se rvices: 100 psig helium, vacuum. Comment: 1. Helium pressure was adjusted to 15 psi above the vapor pressure of VC1 at ambient temperature. ITEM T-2 IB METERING STANDPIPE r unction: To store IB under helium pressure for continuous feed to reactor (R -- I). De sc riptior.: Mat'l of Const: 7 :: length 3/4 in. schedule 40 seamless pipe mounted verticailv. Stainless steel. Se rvice s: Comment: 100 psie helium, vacuum. 1. Helium pressure was adjusted to 15 ps: above :he vaocr pressure of IB at ambient temperature. RSV 0017340 45. ITEM T-3 EHF METERING STANDPIPE Function: To store EHF for continuous feed to reactor (R-l). Description:' Mat 'I of Const: Sc rvice s: Comment: 7 ft length 3/4 in. schedule 40 seamless pipe mounted vertically. Stainless steel. Vacuum. _ Standpipe was charged by applying 10 in. Hg vacuum. ITEM T-4 INITIATOR SOLUTION METERING STORAGE STANDPIPE Function: Desc ription: To store initiator solution for continuous feed to reactor (R-l). 7 ft length 2 in. schedule 40 seamless pipe. Mat'l of Const: Service s: Stainless steel. Vacuum. Comment: 1. Standpipe was charged by applying 10 in. Hg vacuum. ITEMS T-5 AND T-6 STRIPPING ACETONE STORAGE TANK Func tion: To store under nitrogen pressure stripping acetone lor continuous feed to monomer stripper (D-l ). Desc ription: 2000 cu. in. Air Force surplus oxygen tank mounted vertically on platform scales. Mat '1 of Const: Stainless steel. Se rvi^ es: 50 psic nitrogen, vacuum. Comment: 1. Nitrogen supply line was equipped with 50 psig pressure relief valve. 2. Tanks were charged by applying vacuum. ITEM T-T___ VC! MONOMER STOOGE TANK Func tion: To store under helium pressure VC1 monomer for transfer to VC1 monomer metering standpipe (7-1). RSV 0017341 46. .*. /I Description: Mat'l of Const: Services: Comment: Size 1A cylinder with full-length dip tube. Carbon steel. 100 psig helium. 1. The tank was equipped with a 200 psig pressure relief valve. ITEM T-8 IB MONOMER STORAGE TANK Function: To store under helium pressure IB monomer for *ransfer to IB monomer metering standpipe {T-2). Description: Size 2 cylinder with full-length dip tube. Mat'l of Const: Carbon steel. Service s: 100 psig helium. C omment: The cylinder shut-off valve was equipped with a 1400 psi bursting disc. ITEM T-9 EHF MONOMER STORAGE TANK Function: To store EHF monomer for transfer to EHF monomer metering standpipe (T-3). Description: 2 liter burette. Mat'l of Const: Glass. ITEM T-10 INITIATOR SOLUTION TANK Function: To prepare and store initiator solution fed to the in:*iator solution metering standpipe (T-4). Description: 1 gallon jug. Ma: '1 o: Const: Glass. Comment: 1. The preparation of large quantities of initiator solution should be carried out observing the following precautions a. The concentration of azo-bis-isobutyronitriie in acetone should never exceed 4%, i. e. , the azo initiator should be added to the bulk of acetone in an agitated vessel. RSV 0017342 * l . b. Avoid mixing and storage facilities which would permit accumulation of nondissolved initiator. c. Avoid exposure of initiator solution to temperatures in excess of 35 *C. d. Details of hazards involved in handling initiator are given in Section XII-A-5. ITEM T-l 1 POLYMER SOLUTION TANK Function: To collect polymer solution effluent from monomer strippe: (D-l), stabilize with Paraplex G-62, and store for feed to disintegrator (T-13). De sc ription: 1 gallon jug. Mat'l of Const: Glass. Comment: 1. In the plant the stabilize should be metered to the polymer solution in an agitated vessel. Paraplex G-o2 can be handled in a metering pump without dilution. ITEM T-l 2 PRECIPITANT FEED TANK Func tion: To store precipitant for feed to SRC disintegrator (T-13). Desc ription: 55 gallon drum. Mat'l of Const: .Phenolic lined steel drum. ITEM T - 1 3 SRC DISINTEGRATOR Func tion: To precipitate ace tone-polv me r solution stored in polymer solution tank (T-l I) by mixing with water from prec.pitant feed tank ( T- 1 2}. De sc ription: Mat'l of Const: Service s: Comment: Four inch, ID x 3 inch cylindrical tank with special high intensity agitators. Tanks and agitators - stainless steel. Packing - Teflon impregnated asbestos. 220 V, 3-phase electricity. 1. The design of this equipment :s closelv held proprietary information of Shawrnigan Resins Corp. Details o! the pilot plan: SRC di s inte c ra tc r are elver. :r. Shawir.icar, Resins Corp. drawing RD 4?. RSV 0017343 4 b. 1 ITEU T-14 SLURRY HOLD TANK 11 Function: To collect and store slurry from the SRC disintegrator (T-13) for feeding to the rotary vacuum filter (F-5). Description: 20 gallon polyethylene garbage can. 1 Mat'l of Const: Polyethylene. Comment: 1; In a plant operation this tank will pose a substantial fire 1 and explosion hazard because of the small amounts of VC1 and IB which may be evolved there. Two solutions are possible -- blanket the tank with inert gas, or I maintain a he aw stream of air through it to dilute the VC1 and IB. 1 ITEM T- 15 WASH FILTRATE RECEIVER 1 Function: To collect and store wash filtrate for feeding to the precipitant feed tank (T-*2>. De sc ription: 10 in. dia x 20 in. receiver. ] Mat'l of Const: Stainless steel. 1 Comment: 1. This receiver was part of the rotary vacuum filter (F-5). I ITEM T-16 FILTRATE RECEIVER 1 Function: To collect and store filtrate for feeding to the acetone recovery system. Description: 10 in. dia x 20 in. receiver. Mat'l of Const: Stainless steel. C omment: 1. This receiver was part of the rotary vacuum filter (F-5). ITEM T-17 FILTRATE STORAGE TANK F unc tion: To accumulate filtrate from the rotary vacuum filter (F-5) for feeding to the acetone column (D-2). De sc ription: 55 gallon drum. Mat T >of Cons t: Polyethylene-lined steel. RSV 0017344 t?. Comment: 1. This tank should be equipped with a suitable flame arrestor in the vent. 2. This tank was located on platform scales. ITEM T-18 DISTILLATE RECEIVER Function: To collect and weigh distillate from acetone column (D-2). Description: 1 gallon jug. Mat'l of Const: Glass. Comment: 1. Steel or stainless steel can be used in plant operations. Stainless is recommended to prevent contamination of acetone by rust. 2. In a plant application this tank should be equipped with a suitable flame arrestor in the vent. 3. This tank was located on platform scales. 4. For plant application, a day tank equipped with gage glass is recommended. 5. For plant application, a product cooler between the reflux head (E") and the distillate receiver may be required. Hot acetone may not be used to prepare initiator solution in initiator solution rank (T-10) because of the explosion hazard. ITEM T-1Q BOTTOMS RECEIVER Fum tio: : To collect and weigh bottoms from acetone column (D-2). Dost, ription: 5 gallon can. Mat'l of Const: Steel. Co?nmeT'*' 1. This tank was located on platform stales. 2. This tank will not be needed in a plant. The bottoms raav be cooled bv mixing with cold water ana se..: to the waste disposal or sewer system. ITEM T-20 r ur.c tior. INITIATOR INJECTION TUBE To store the initial charge of A Z 3N-solut: on for transfer to the reactor (R-l) under helium pressure. RSV 0017345 50. ! Description: Mat'l of Const: Services: Comment: 2 ft length of in. schedule 40 pipe. Stainless steel. 300 psig helium. 1. The contents of the initiator injection tube were transferred to the reactor through the sampling manifold (Figure 14). ITEM TI-1 TEMPERATURE INDICATOR ACETONE VAPORIZER STEAM SUPPLY Function: To indicate steam temperature in the jacket of the acetone vaporizer (E-6). Desc ription: Ashcroft dial thermometer 0-300*C. Mat'l of Const: Stainless steel. ITEM TI-2 PRECIPITANT FEED TEMPERATURE INDICATOR Func tion: To monitor precipitant feed temperature to SRC Disintecrator (T-13). Desc ription: Ashcroft dial thermometer 0-100eC. Mat'l of Const: Stainless steel stem. ITEM TI-3 DISINTEGRATOR TEMPERATURE INDICATOR Function: To monitor disintegrator (T-13) temperature. De sc ription: The r mo mete r. Mat' 1 of Const: Glass. C ommer.t: 1. Used in exit line to indicate slurry temperature. Preferred location for sensor would be inside dis integrator near slurry exit. Should be a temperature controller to regulate steam flow to precipitant feed. See Section IX-C. ITEM TI-4 ACETONE COLUMN FEED TEMPERATURE INDICATOR r unc tion: De sc rip: ion: To indicate temperature of acetone column (D-2) feed. - "i Minneapolis-Honeywell copper-constantan poter.tiometric temperature indicator, range 0C-200CC. RSV 0017346 51. ITEM TR-1 TEMPERATURE RECORDER RECYCLE SYSTEM Function: To indicate and record temperatures at various points in the recycle system. Desc ription: Minneapolis-Honeywell 12-point copper-constantan 0-200*C strip-chart recorder. Service s: 110 V electricitv. ITEM TRC" 1 REACTOR TEMPERAT URE RECORDER CONTROLLER Func tion: To record and control temperature in reactor (R-I). Description: Foxboro Dynalog recorder controller M/40 integral plus derivative control mode. Service s: Instrument air, 110 V electricity. Comment: 1. The controller output war used as set point signal tor reactor jacket temperature recorder controller (TRC-2). ITEM TRC -2 REACTOR JACKET TEMPERATURE RECORDER CONTROLLER Function: To record and control temperature in the reactor (R-l) jacket Dose ription: Foxboro tvpe M/40 Stabilog recorder controller. Service s: Instrument air. Comment: 1. See comment, Item TRC-1- ITEM TRC-3 FEED PREHEATER JACKET TEMPERATURE RECORDER CONTROLLER Function: To record and control water jacket temperature in the teed preheater (E-2). Desc ription: Foxboro type M/40 Stabilog recorder controller. Se rvice s: Instrument- air. ITEM TT-1 TEMPERATURE TRANSMITTER, REACTOR JACKET Func tion: To sense and transmit reactor jacket water temperature to reactor jacket temperature recorder controller {TRC-2). RSV 0017347 52 Description: Service s: Foxboro type I2A pneumatic temperature transmitter. Instrument air. ITEM TT-2 TEMPERATURE TRANSMITTER FEED PREHEATER Function: To sense and transmit feed preheater (E-2) water temper ature to feed preheater jacket temperature recorder controller (TRC-3). Description: Foxboro type 1 2A pneumatic temperature transmitter. Service s: Instrument air. ITEM U-l TRAY DRYER Function: To dry filter cake from rotary vacuum filter (F-5). Description: Proctor and Schwartz 10-tra\ dryer. Mat'l of Const: Scrvices : Trays - stainless steel. 220 V, 3-phase electricity, 100 psi steam. Comment: 1. The drying tray arrangement in the oven was changed to hold 20 trays 20 x 29. 75 in. Z. Filter cake was deposited 3/4 in. thick on drying trays. 3. Operating temperature 50C. 4. Average drying time 15 to 20 hours. 5. Charge of oven 32 lb dry polymer. 6. To prevent dusting the fan speed of the dryer was reduced from 1169 rpm to 681 rpm. RSV 0017348 5 I 1 I I I s Vin. OPERATING PROCEDURE A. PRELIMINARIES TO REACTOR OPERATION 1. Reactor a. Turn on cooling water to reactor (R-l) seal oil circulating system. Turn on seal oil circulating pump and reactor agitator. b. Bring the reactor pressure to 160 psig by feeding nitrogen through nitrogen feed rotameter (FI-1). Pressure check reactor for 15 minutes and find and repair leaks if necessary. While the reactor is at 160 psig pressure, zero the reactor level transmitter (LT-5) and check the reactor thermocouple for continuity. c. Vent the re actor (R-l) to 2 psig through the reactor vent valve (V - 3). Close the reactor vent valve. a. Open the reactor drain valve (V-3) and start nitrogen purge to re actor at the rate of 60 std cu ft/hr. After 30 std cu ft of nitrogen have been fed, close the reactor drain valve without interrupting the nitrogen feed and bring the reactor to operating pressure. When the reactor is at operating pressure, reduce the nitrogen feed to 1.5 std cu ft/hr and put the reactor pressure controller (PRC -1) in operation. e. Turn on water to the reactor reflux condenser (E-l). f. Fill the reactor effluent filter (F-l) with acetone and close the re actor effluent valves (V-l and V-2). Feed Svstern a. Energize the electronic feed system level transmitters (L.T-1, LT-2, LT-3 and LT-4) and the feed system level recorder (LR-1). Che^k the zero and span of each instrument. Span is checked bv applving gas pressure to the high pressure side of the standpipe. b. Purify VC1 monomer if VC1 of satisfactory purity is no; available. * Clean, uninhibited VC1 will presumablv be available in a plant. Puriiu av.on is discussed in Section IX-J. RSV 0017349 54.  ,.-.C ;#.<<-.j c. Prepare the initial charge and the feed solution of AZBN in acetone. The initial charge solution consists of 6. 0 g AZBN in 150 g acetone. The feed solution consists of 25. 6 g AZBN in 2474. 4 g acetone. CAUTION: ADD AZBN TO SOLVENT. DO NOT EXCEED 4% AZBN CONCENTRATION. d. Evacuate all standpipes (T-l, T-2, T-3 and T-4) and charge with VC1, IB, EHF and pure acetone, respectively. The volatile monomers VC1 and IB are pressured in from the storage tanks (T-7 and T-8) by helium pressure. The EHF and acetone are drawn into the standpipes from the storage tanks (T-9 and.T-10) by vacuum. e. Pressure the VC1 and IB standpipes to 20 psi above the monomer vapor pressures with helium. Vent the EHF and acetone standpipes to atmospheric pressure. : . Fill the initiator injection tube (T-20) with the initial AZ E5N-acctonc solution (6. 0 g AZBN in 1 50 g acetone) and connect the tube to the bottom of the reactor by wav of the bottom sample connection. 3. Rccvcle Svstem a. Pressure check the recycle system at 50 psig. b. Energize the monomer stripper liquid level control (LC-1) and the recycle collector liquid level control (LC-2). De-cnergizc the re cycle pump relay in LC-2 until recycle operation commences. c- Prepare stabilizer solution (7.4 weight percent Paraplex G-62 in acetone}. d. Charge the stripping acetone storage tanks (T-5 and T-6) and pressurize with 50 psig nitrogen. e. Bleed acetone through the stripping acetone flow controller (FC-0 and check zero adjustment. : . Set recycle system pressure controller (PRC-2) at 21 psig. 3. REACTOR START-UP AND FORWARD FEED OPERATION During this operation the reactor (R-l) is operated without recycle. The un reacted VC1 and IB monomer in the reactor effluent is stripped from the polymer solution in the monomer stripper (D-l) and wasted. 1. Pump the following initial charge as indicated by the standpipe levels to the reactor. VC1 IB Acetone 2040 g (Standpipe T-l) 13 g (Standpipe T-2) 57 g (Standpipe T-3) 3330 g (Standpipe 7-4) RSV 0017350 0 During this time, adjust the pumps to the rates required for forward feed conditions. VC1 IB EHF Acetone 1430 g/hr (Pump P-I) 94 g/hr (Pump P-2) 165. 1 g/hr (Pump P-3) 713 g/hr (Pump P-4) The pumping rate checks should be made with feed and suction pressures leveled out at operating conditions. Drain remaining pure acetone from initiator solution metering standpipe (T-4) and fill with feed solutio^(25. 6 g AZBN in 2474. 4 g acetone). Refill all other standpipes. VC1 and IB standpipes are refilled a: their operat:nc pressures (20 psig above vapor pressures of the monomers). Heat the reactor (R-l) charge to operating temperature (S0*C) by applvmc hot water to the reactor jacket. The temp: rature of the water is controlled by the reactor temperature control system (TRC-1, TRC-2). To avoid excessive overshoot oi the charge temperature, limit the output of the re actor jacket temperature controller to 5"C during the heat-up period. When the reactor (R-I) temperature has leveled out at 50*C, pressurize the initiator injection tube (T-20) to 300 psig with helium and blow the AZBN'-acetone solution into the reactor- The recorded run time is started at the instant the AZBN-acetone solution is in. This is indicated by a rise in the reactor pressure. Start all reactor feed pumps (P-1, P-2, P-3, and P-4) operating a: the rates indicated in Section VII1-B-1- Set the control point of the reactor level controller (LRC-1) to the operating level. Open the manual reactor effluent line valve (V-l). As soon as reactor effluent flow starts, turn on acetone feed to bottom c: the monomer stripper (D-l). Turn on steam to stripping acetone vaporizer (E-6) and stripping acetone superheater (E-7). Turn on he: water to stripper feed preheater (E-2). Adjust the monomer stripper to operate under the following conditions: Stripper feed preheater(E-2) Stripper bottom (D-l) Stripper head (D-l) Acetone vapor to stripper (E-7) Stripping acetone flow (FR-i) 53CC 55 eC 55CC llc*C 500 g/hr During this feed :orward period the vapor from the monomer stripper is vented to waste by the recycle system pressure controller (PRC-2). The coding water is turned off the recycle collector and condenser (E-3) during this period. Some liquid may accumulate in the collector during this period which must be drained to waste. RSV 0017351 56. t the monomer stripper liquid level controller (LC-I) to withdraw ripped polymer solution from the bottom of the monomer stripper >-l). Mix stabilizer solution {7.4 weight percent Paraplex G-62 in etone) with the stripped polymer solution at the rate of 50 ml/hr. jmove the initiator injection tube (T-20) from the sampling manifold and 'nnect the conversion (nonvolatiles) sampling system (Section XIII-A). Mermine conversion (nonvolatiles) by the method described in ction XIII-A at one-hour intervals starting one hour after the run was arted. icord all operating data at hourly intervals starting 30 minutes after the art of the reaction. leek feed pump rates and adjust pump settings as required at naif-hour tervals. eady state is reached when the three successive hourly determinations nonvolatilcs differ by no more than * 1%. This normally requires 6-S urs. If the steady-state condition, when reached, differs from 30To non- datiles by more than 1 unit,adjust the AZBN concentration in the AZBNitiator feed solution according to the formula r New AZBN x _ 'Present AZBN N ______________ 30v, ^Concentration./ Concentration S x \ observed % nonvolatilesV onrir.ue operation until a new steady state at 30 1% is reached. IEAC TOR RECYCLE OPERATION he reactor (R-l) reaches steady-state operation with nonvolatilcs content 7>, recycle operation is started. repare a solution of 5$. 7 g AZBN in 24-41. 3 g acetone. This initiator .'lution is to be used during recycle operation. Li a chance has been made AZ3N during the forward feed operation to attain 30 ^ l^o solids in the fluent, a proportional change should be made in the AZBN content of this .'lution. urn on cooling water to the recycle condenser and collector (E-3). Turn " refrigerated brine to recycle subcooler (E-4). * nergize the recycle collector liquid level control (L.C-2) and turn or. the jcycle pump (P-5). If the level control and recycle pump operate ecycle subcooler will not be required in plant operation if sufficient head is provided for the recycle pump (P-5). RSV 0017352 3-. roperly, stop the pump. * Drain the initiator solution metering standpipe f-4) and charge with the solution of 58. 7 g AZBN in 2441. 3 g acetone. hange the reactor (R-l) feeds to following rates: VC1 IB EHF Initiator solution 511.9 g/hr 31. 1 g/hr 165. 1 g/hr 316. 8 g/hr (Pump P-1) (Pump P-2) (Pump P-3) (Pump P-4) nmediately start the recycle pump (P-5). 'perate the recycle system under the following conditions: Stripper feed preheater _ Stripper bottom Stripper head Vapor to stripper Stripping: acetone flow Recycle condenser water Recycle subcooler brine Stripper head pre ssure (E-2) (D-l) (D-l) (E-7) {FR- 1) (E-3) (E-4) (PRC-2) 53*C 85 "C 55 "C 11SC 500 g/hr < 22. 5C 100-20*C 21 psig t hourly intervals determine nonvolatiles in reactor charge (Section .III-A), VC1, IB and acetone in reactor charge (Section XIII-B) and VC1 ontent of stripped polymer solution (Section XIII-D). Adjust AZBN oncentratior. in initiator solution feed as required to maintain nonolatiles at 30 * 1%. Adjust IB feed as required to maintain VC1/IB eight ratio atT3. Adjust stripper feed preheater temperature and cetone vapor temperature as required to maintain acetone stripper ottoms flow at 157 1 + 30 g/hr and VC1 content at less than 1% (wt). adjust VC1 feed to maintain 40. 8% free VC1 * IB in reactor. REACTOR SHUTDOWN hut off all feeds and apply cold water to the reactor (R-l) jacket. >hui off recycle pump (P-5) and operate monomer stripper (D-l) a. ..t orward feed operation (Section VIII-B-7). )uring the last few minutes of recycle operation note the average output : the reactor level controller (L.RC-1). When the feeds and recycle ump are shut off, put the reactor level controller on manual and adjust e pilot plant the recycle stream was. pumped to a tared cylinder for a , hour period to see that all equipment was operating properly and that e were no leaks in the recycle system. This can probably be dispensed in a plant if a good pressure test of the recycle system is maue before : -ud. RSV 0017353 58. ' vv. _5i`v*r - ^ _i_-T ' ?' inr" v-^vl .'5 'T. .'" ^'v' .`ii'.- the output to this average reading. - Continue operation of the stripping column until all material in the reactor has been stripped. 4. Shut down and drain the stripping column. Vent the reactor. 5. Fill the reactor (R-l) with two gallons acetone and reflux for 45 minutes with the stirrer running. Drain the reactor hot through the monomer stripper (D-l) and mix with a large amount of water in the SRC dis integrator (T-13). Drain the acetone-water solution to waste. 6. Shut off reactor agitator and seal oil system pump and cooling water supply. Shut off water to reactor reflux condenser (E-l). t, Remove filter cartridge from reactor effluent filter (F-l), clean filter and replace cartridge. 8. Drain and vent feed system standpipe (T-l, T-Z, T-3 and T-4). Shut manual valves in the reactor feed lines. Rinse EHF metering standpipe (T-3) and initiator metering standpipe (T-4) with acetone. 9. Shut off brine system and nitrogen, helium, vinyl chloride and acetone supply lines. 10. Shut off steam and cooling water supplies to the recycle system (D-l, E-5, E-6, E-7, E-S and E-9). 11. Drain the stripping column (D-l). Fill column with acetone, allow to soak 30 minutes and drain acetone to waste. PRECIPITATION 1. Turn on water sprays to cool SRC disintegrator (T-13) packing glands, Z. Turn on the precipitant feed pump (P-7) and adjust the precipitant rate to 460 lb/hr as indicated by precipitant rotameter (FI-3). 3. Expel the air trapped above the outlet nozzle of the SRC disintegrator (T-13) by placing a hand over the slurry discharge to force air out through the agitator packing glands. 4. Adjust the precipitant temperature to ZS^C as indicated by precipitant feed temperature indicator (TI-Z). * 5. Turn on agitator in slurry hold tank (T-14). If lower bulk density product is desired, the precipitant temperature should be lower. See Section IX-C. RSV 0017354 59 6. Turn on SRC disintegrator (T-I3) agitator drive. Adjust tip speed of beaters (agitators) to 1600 ft/min. 7. Start polymer solution feed from polymer solution tank (T-ll) to SRC dis integrator (T-13). Adjust speed of polymer solution pump (P-6) to feed polymer solution at the rate of 46 Ib/hr. 8. Start filtration (Section VIIl-F)as soon as polymer slurry reaches slurry hold tank (T-14), 9. When polymer, solution feed is exhausted, flush polymer solution feed pump (P-6) with acetone and turn off. 10. Allow SRC disintegrator (T-13) to run until effluent is clear. Then turn off agitator and precipitant feed. 11. Filter all slurry in the slunrv hold tank (T-14) and shut off agitator. 12. Shut down filtration. F. FILTRATION AND CAKE WASHING 1. Start feed of polymer slurry by polymer slurry pump (P-8) from slurry hold tank (T-14) to rotary vacuum filter (F-5). 2. Start agitator in slurry tank of rotary vacuum filter (F-5). 3. When the level of polvmer slurrv has reached the overflow level, start rotation of rotary vacuum filter (F-5). Turn on vacuum to filter drum and adjust wash water to 6 lb water/lb dry polymer as indicated by the wash water feed rotameter (FI-4). Simultaneously start the vacuum pump (P- 1 I ) and the blower (P-1 2). The doctor knife should be 1/8 inch from th surface of the filter. 4. Start the wash filtrate pump (P-a) and the filtrate pump (P-11). 5. Adjust the rotation of the rotarv vacuum filter (F-5) to a rate which matches the rate of slurry production by the SRC disintegrator (T-13). 6. Add water to the precipitant feed tank (T-12) as necessary to supplement the supply of wash filtrate used as precipitant. 7. To shut down the rotary vacuum filter (F-5), turn off the polymer slurry pump (P-8) and continue operation of the filter until cake formation ceases Shut down the wash filtrate pump (P-9), the filtrate pump \P-10), the vacuum pump (P-11) ana the blower (P-12). Turn off the wash water feed. S. Drain the residual slurry into the slurry hold tank (T-14). v. Wash the rotary vacuum filter (F-5) down with a hose. Turn off the drum rotation. RSV 0017355 oO G. DRYING 1. Clean all dust and foreign particles from the tray dryer (U-l) tray compartment, trays and plenum chamber with a vacuum cleaner. 2. Charge drver trays with filter cake to a depth of 3/4 inch. Break up any filter cake lumps. Place trays in dryer. 3. Start up tray drver and operate at 50*C. Check trays periodically for lumps and break up any lumps which may form. 4. After 15 hours determine moisture content of polymer (Section XIII-J). If moisture content is below 1%, terminate drying and discharge product to polyethylene-lined L-cverpaks. If moisture content is above I7o, continue drying until an analysis shows it to be below 1%. H. ACETONE DISTILLATION 1. Set reflux head (E-) for total reflux and turn on cooling water. 2. Turn on cooling water to acetone column bottoms cooler (E-10). 3. Turn on acetone column feed pump (P-13) and adjust by-pass so that feed pressure is 3 psig. 4. Start filtrate feed to acetone column (D-2) at the desired rate, 12.3 kg/hr. 5. Turn on open steam to bottom of acetone column and set the acetone column bottom pressure controller to maintain 3 in. Hs O g pressure at the column bottom. 6. When the acetone column head temperature is at 56*0 or less and the feed trav temperature is 72*0, start distillate takeoff to distillate receiver at reflux ratio 2. 7. Adjust acetone column feed as required to maintain feed tray temperature at 7le-7 3cC. Increasing the feed rate lowers the feed tray temperature. 5. When feed is exhausted, immediately set reflux head lor total reflux and turn of: acetone column feed pump. Turn off open steam feed and allow column to drain. Turn off cooling water. * G. Examine feed filters (F-6 and F-7) and discharge cake i: necessary. I: .i tresh supplv of filtrate for distillation is ant i c ipa t ed, the column may be left on total reflux ir.defir.iteiv. RSV 0017356 IX. DISCUSSION [ A. POLYMERIZATION The polymerization reactor (R-l) is an overflow, or continuous stirred tank reactor. VC1, EHF, IB and AZ BN-initiator solution are continuously pumped I to the reactor by metering pumps. In normal (recycle) operation the condensed vapors from the monomer stripper (D-l) are also continuously pumped to the reactor. A solution of polymer, unreacted monomer and acetone is continuously I withdrawn from the reactor. 1 Reactor Operating Conditions I An extensive synthesis and evaluation program has shown that VC1/EHF/IB terpolvmer containing 42.5% w: chlorine and 1.65% wt hydroxyl and having a Gardner viscosity* of Z-Z2 (specific viscosity 0. 240***) is required for I surface-coating applications. This corresponds to a monomer content of '5% VC1, 20.6% EHF*** and by difference 4.4% IB. (2). I To produce polymer of this type in the continuous reactor it has been found experimentally that the following physical limitations applv: I a. The total concentration of free monomer must not be less than 3S% wt in order to achieve reasonable polvmerization rates. Under these conditions the polymerization rate is approximately 6.0 lb/hr ft of reactor. The AZBN initiator consumption is about 0.01 lb/lb polymer I made. b. The solvent content of the reactor effluent must be no less than I 0.75 lb/lb polymer and preferably should be 1.0 lb/lb polymer in order to have fluid solutions. This establishes the known feasible upper operating limit at about 35% nonvolatiles in the reactor. The operating I level for design and research purposes is taken at 30%. c. Reactor holdup must be no less than 2. 10 hrs. Below this value the initiator consumption increases sharply (in excess of 1 00%) \ ielding [ polymer with undesirably low molecular weight. I 1 ".he viscosity of a solution o: 25% M1BK, 2r>-o xylene and rO1': polvmer measured at 25*C. I **i% polymer in cyclohexanone at 25eC. *"*Fcr EHF monomer containing a. 0% OH. I RSV 0017357 I 62. 1 d. EHF monomer is by far the most reactive of the monomers in the system. Figure 7 shows the relative weight proportions of the free monomers to produce polymers having the desired 75.0% VC1, 20, 6% EHF and 4. 4% IB composition. The low concentration of EHF monomer is important in that no recycle of EHF monomer from the reactor effluent to the reactor is required. e. Molecular weight of the product is established primarily by chain transfer. This is clear from the fact that about 10 moles of polymer are produced per equivalent of initiator decomposed. Isobutylene is an active (but not. the only) chain transfer agent present. Fine control of the molecular weight of the polymer can be maintained by adjusting the free isobutylene concentration in the reactor. In the pilot plant isobutylene was fed to maintain the ratio of VCl/IB at 13. 1. f. The operating temperature in the reactor must be maintained at S0*C. The operating pressure oi the reactor must not be less than the autogeneous pressure of the reaction mixture (about 120 psig). 2. Control of Polvmerization Reaction In the pilot plant the reactor temperature, pressure and liquid level were automatically controlled. The makeup teed streams were metered by manually reset diaphragm metering pumps. Analytical methods were available for monitoring the nonvolatiles content and the VC1, EHF, IB and acetone contents of the reactor charge. In all operations the temperature, pressure and liquid level were set to predetermined levels and maintained constant. The feed streams were manipulated to obtain the desired nonvolatiles, product composition and product viscosity. Two modes of operation must be considered -- the feed forward mode ;n which all unreacted monomer is wasted, and the ret vcie mode in which the vapor from the monomer stripper (D-l) is condensed and immediatelv re cycled to the reactor. In the feed forward mode the monomer feeds are calculated to give, for a set conversion, the proper amount and proportion of free monomers ;n the reactor to produce the desired product. The concentration of non volatiles in the reactor effluent (conversion - residual EHF) is controlled by adjustment of the AZBN (initiator) concentration in the AZBX-acetonc solution feed. It is not necessary to monitor the reactor monomer ar.d acetone concentrations during feed forward operation. In the normal operation mode, the recycle mode, the reactor feed streams are augmented by the recycle stream, of condensed vapor from the monomer stripper. The composition of this stream, is somewhat variable, subject to the losses ir. the stripper bottoms and to the deviations from, target nonvolatiles ir. the reactor effluent. A a practical matter, it is necessary to monitor the relative proportions oi VCi, IB and acetone :r. RSV 0017358 tS o FIGURE 7 FREE MONOMER IN REACTOR CHARGE AND COMPOSITION OF POLYMER RSV 0017359 I - _I the reactor effluent and make small adjustments in the monomer feeds as necessary to maintain the proper proportions of these materials in the reactor. The VC1, IB and acetone determinations are made by gas I chromatography. As in the feed forward mode, the proportion of nonvolatiles in the reactor I effluent during recycle opeation is adjusted by adjusting the concentration of AZBN in the AZBN-acetone solution feed. In making these corrections, the AZBN concentration is adjusted in direct proportion to the deviation of observed nonvolatile from the desired level of nonvolatiles. Experi mentation showed that this linear adjustment was to be preferred to making the change in initiator feed proportional to a power or root of the deviation. The water content of the acetone feed has some effect on monomer con version. In general, about 5%-10% more initiator was required to obtain 30% nonvolatiles in the reactor effluent when acetone containing 2% water was used than when acetone containing 0.5% water was used. The hydroxyl content of EHF monomer may vary from about 7. 8% to 8. 2% (5 ). The polymer material balance assumes use of EHF monomer con taining S. 0% hydroxyl. If the EHF monomer does not contain 8. 0% hydroxyl, the EHF feed rate is adjusted to give the same hydroxyl feed rate as if the S. 0% monomer were being used. Details of the development of the feed plan for a typical pilot plant run are given in Appendix A. 3. Instrumentation and Control Analyses for the Polymerisation Process a. Reactor Pressure Control In most of the pilot plan: work, the reactor pressure was controlled well above the autogenous pressure by feeding a small stream of nitrogen to the reactor and setting the reactor pressure controller (PRC-1) to bleed nitrogen as required to control the se* pressure. In other cases the reactor was operated at autogenous pressure. Operation at the higher controlled pressure was preferable because it gave a constant head for the feed metering pumps i*h* outputs of which were quite sensitive to discharge pressure) to work agamy. Ir. the plant it will be desirable to dispense with the nitrogen stream, and operate the reactor a: or near to autogenous pressure.* If the reactor is sufficientlv leak-free, there may be enough ir.erts in the teed streams to permit operation above autogenous pressure without adding nitrogen. Conditions should be adjusted so that there is no flashing ;r. the reactor etfluent filter (F-l). See Section IX-A-7. RSV 0017360 b. Reactor Temperature Control In the pilot plant the reactor temperature was controlled by a temper ature controller (TRC-1) cascaded with a temperature controller (TRC-2) in the jacket water supply. This was necessary to eliminate upsets due to water supply pressure and temperature fluctuations. In the plant a cascaded system may or may not be necessary depending on local conditions. Reactor temperature must be controlled to SO * A 1 *C. c. Reactor Level Control Reactor liquid level must be controlled very precisely in the VYSET RE process both because fluctuations in level produce severe upsets in the monomer stripper {D-1) and because of upsets in conversion. No commercially available level transducer was suitable for use in the small and rather cramped space of the pilot plant reactor. A transducer (LT-5) which measures level by measuring the tangential force on a paddle inserted into the surface of the swirling liquid in the reactor was developed. This transducer performed exceilentiv. It was able to detect the withdrawal of 20 cc of fluid from the twogallon reactor charge and was not subject to troublesome zero drift. A plan: version of this device has beer, built and tested in a 30-gallon baffled reactor. It is recommended that this device be used on the interim plant reactor. In the plant reactor it will be r.eccssarv to cascade the reactor level controller with a flow controller in the line leading from the reactor to the monomer stripper. This will ensure a steady flow of teed to the stripper. Without it much difficulty with flooding in the stripping column would be expected. Reactor Feed Metering Successful operation of the VYSET RE pilot plant required accurate metering of the various liquid streams fed to the reactor. Until this was done through installation of the metering standpipes (T-I, T-2, T-3 and 7-4), control of product composition, conversion and product viscosity was impossible. The metering system permitted control of the feed rates to 2ro of the desired feed rate. This specification should be met by any plant feed metering system. With respect to a plant installation :: is noted that 1) metering pumps will no: meet this specific atior. for accuracy unless they are continuously calibrated ar.d reset; 2) continuous recording of feed rates should be provided, ar.d 3) independent calibrating equipment (volumetric tanks or weigh tanks) should be provided tor each feed s t r e a m. RSV 0017361 e. Nonvolatiles Determination The nonvolatiles are determined by evaporating a weighed sample of reactor effluent to dryness and weighing the residue on a moisture balance. In a plant operation the nonvolatiles measurement might possibly be replaced by a measurement of the heat evolution in the reactor followirg the methods of Wstring (12). This could not be investigated in the pilot plant because of the relatively large radiation losses from the small reactor. Installation of equipment to measure heat trans ferred to the cooling water in the plant is recommended. f. VC1, IB and Acetone in Charge For control analyses in the pilot plant, the proportions of VCl, IB and acetone in the reactor effluent were determined using a standard laboratory gas chromatograph. This method, if good technique in sampling and sample injection is used, is accurate. However, it requires about 20 minutes to take a sample and run and interpret the chromatograms. The use of a process eas chromatograph in a plant would be desirable; however, a suitable sampler and sample vaporizer would have to be developed. Polymerization Reactor Fouling The reactor was opened frequently to check for fouling. Each time a thin film (less than 0. 01 in. ) was found deposited on the reactor wall, the agitator and the sensing lever of the reactor liquid level transmitter. In addition, a slight accumulation of deposits was observed on unwetted stain less steel parts of the sensing lever and the feed nozzles in the proximity of the liquid interface in the reactor. The amount of deposits did not appear to increase with operating time o: the reactor. Analyses cf these polymer deposits indicated approximately l^o nitrogen which suggests the possible presence of toxic decomposition products of AZBN. (Micro analysis of precipitated VYSET RE polymer showed OTi N The deposits were soluble in hot cyclohexanone and partially soluble in acetone. .) It is considered probable that much of the slight deposits found in the reactor were polymer which remained on the walls when the reactor was drained. There was never enough polymer deposited to interfere with* reactor operation. On present evidence the possibility that a plant reactor would have to be given an occasional cleaning with hot cyclohexanone cannot be ruled out, although it is considered unlikely. Reactor Agitation The reactor was stirred by ar. anchor type agitator operated at i*rpm. Since the polymerization is a solution process, the only requirements on the RSV 0017362 agitation system are that it provide good mixing of the feed streams into the charge, minimize temperature gradients in the charge and support adequate heat transfer to the reactor walls. Design on the basis of the correlations given by Perry, Chilton and Kirkpatrick ( 8 ) should be satisfactory. 6. Use of a Reactor Reflux Condenser In very large reactors it would probably be desirable to cool the reactor by using a reflux condenser. This was not investigated extensively in the pilot plant; however, there would appear to be no serious impediment to using a reflux condenser when needed. 7. Reactor Effluent Filtration {Grit Removal) Midway in the process development work it was reported by the Springfield evaluation group that the pilot plant product contained small but objection able amounts of insoluble polymer called "grit" (13). This "grit" caused roughness in coatings applied to metal. The problem was solved by installing a reactor effluent filter (F-l). The filter used is made especially for filtering paints and resin solutions used in paint manufacture. The use of a filter of this type in plant operation is required. The pilot plant filter was operated upward of 100 hours without evidence of much solids accumulation. For plant applications, it is recommended that two filter stations be used in parallel so that they can be changed without interrupting operation. Gauges should be provided upstream and downstream of the filters to indicate when the cartridges should be changed. The operating reactor pressure and the static head on the filter combined should be high enough that there is no vaporization in the filter cartridge due to the pressure drop across the filter. 3. MONOMER STRIPPING In the monomer stripper (D-l), the unreacted VC1 and IB monomers are stripped from the reactor effluent for recycle to the polymerization reactor. The distillate contains acetone, VC1 and IB; the bottoms 5 0T} polymer in acetone solution containing not more than 1 To VC1 and IB combined. The column is operated by open superheated acetone vapor. 1. General Considerations Vapor-liquid equilibrium data for the system V Cl - IB-a ce tone-polvme r are not available, so an exact piate-to.-plate calculation for the stripper is not possible. Inferences about how the system behaves can, however, be drawn from available physical data. These inferences are: RSV 0017363 Ip a. The boiling points of VCl (-13. 8*C) and IB (-6. 9*C) are so much lower than acetone (56.5*C) that there is little likelihood of VC1acetone or IB-acetone azeotropes--and there is no chance at all of azeotropes containing large proportions of acetone, which is the region of concentration of interest in stripper design. No inter ference by azeotropes can logically be expected. b. Polvmer in solution apparently has only moderate effect on the volatility of its solvents. This is inferred from the fact that a 50% solution of polymer in acetone boils only about 2*C higher than the pure solvent. It is therefore concluded from these general considerations that a separation giving VCl-IB-acetone as distillate and polymer-acetone as bottoms is technically feasible. 'This was demonstrated in the pilot plant to be fact. Consideration of VCl-Acctone System To show what must be considered in designing and selecting operating con ditions for a stripping column, the binary system VCl-acetonc is now conside red. Figure 8 is an idealized material balance for the stripping system. In this material balance the moles of VCl correspond to the actual moles of VCl and IB combined as taken from Figure 1. The moles of acetone correspond to the acetone flows of Figure 1. The other components-polymer, EHF, water and initiator --are ignored. Figure 9 is an x-v diagram and a composition-temperature diagram for the system VCl-acetone estimated from vapor pressure data using Raoult's La w. In the idealized material balance, Figure 8 , the internal column liquid and vapor rates are considered fair approximations of the minimum liquid and vapor rates which actually existed in the pilot plant column. Part of the feed to the pilot plant column was vapor which merely passed through the top of the column without contacting the liquid on the top plate. This vapor stream is the stream J' on Figure 8 . A large part of the open acetone vapor was condensed in the column to supply sensible heat and to make good radiation losses. This is the stream L' in Figure 8 . The operating line for the stripping section with open acetone vapor is W V ~V ' W Xw V y = 2. 52 X - 0. 0363 This operatir.c line is plot:tec on the x-v diagram , Figure , and the theoretical stages are stepped off, starting from the specified bottoms RSV 0017364 VCL ACETONE (3.73 g-m 4.74 g*m 18.47 g-m Jr VCL 15.75 g-m ACETONE 6.30 g-m 22.05 g-m VCL 2.15 g-m ACETONE 6.88 Q-m 9.03 g-m X * 0.238 VCL 15.88 g-m ACETONE 11.62 g-m 27.50 g-m Am VCL ACETONE 2.02 g-m 1.56 g-m 3.58 g-m X--0.565 ACETONE 3.58 g-m VCL 0.13 g-m V W ACETONE 6.90 g-m 9.03 g-m Xg - 0.0144 M ACETONE 8.40 g-m C ACETONE 4.82 g-m VCL ACETONE 0.13 g-m 13.72 g-m 13.85 DUWN CHECKED OATE SOLE FIGURE RSV 0017365 Monsanto Chemical Company 100 N. UNDIEKH UVD. ST. LOUIS 66. 1*0. SIMPLIFIED MATERIAL BALANCE F0 STOIPOIVG SYSTEM " rQJ. NO. LOCATION DIVISION PW4 wo. aOdVA Nl IDA NOliDVdd DOW 3M concentration, x s 0.0144. It is seen that about three theoretical stapes are required. T^is is reasonable in view of the fact that the pilot plant column had five trays plus the sump at the column bottom. The criterion of operability of the stripping process is that the operating line be under the equilibrium line up to the feed composition, the liquid coming to the top tray. If this is satisfied, it is only a question of providing an adequate number of theoretical stages to attain the desired separation. If an adequate number of theoretical stages is provided for an operable set of flow rates, the operation of a stripping column is virtually self regulating, i. e. , an excess of theoretical stages will only result in a slight lowering of the low boiler con-tent of the bottoms below specification. Changes in the slope of the operating line can be made either by changing the temperature of the feed, which changes the split of vapor and liquid in the feed, or by changing the vapor feed rate. 3. Consideration of the Real Column Returning to c on side ration of the real pilot plant column, the following differences between the simplified case and the real case should be noted. a. . The vapor-liquid equilibrium relations for the real case are undoubtedly different from the simplified case. Probably the relative volatilities for the real case are less than for the simplified case. b. The assumption of constant molal overflow cannot possibly hold exactly in the real case because a large part of the liquid stream is nonvolatile polymer which must be heated as it descends the column. c. The pilot plant column had, despite tracing, rather large radiation losses. All these factors would tend to make the pilot plant theoretical stage requirement greater than found in the analysis of Section IX-B-2. For this reason, it is recommended that the plant be designed on the basis of 1? times the number of theoretical stages indicated in Figure and that the slope of the operating line in r.o case be greater than that used in Figure 9 . The independent operating variables for a real column are 1) the temper ature of the feed, which controls the ratr and composition of the liquid teed to the top of the column and 2) the superheat in the open acetone vapor feed. The problem in column control is to adjust these independent variables so that the stripper bottoms will contain the correct amount of acetone and be at or below spe cificati on- i n VC1 content. The instrumentation of a plant stripping column should include pressure control at the vent, measurement ana control of feed temperature, control RSV 0017367 -1 of the open acetone feed rate, control of the temperature at seme selected tray in the column by control of the temperature of the superheated acetone vapor feed, measurement of temperature on all trays, and measurement of either the bottoms flow rate or the bottoms density. The column operating pressure should be as high as is acceptable to permit the use of the highest cooling water temperature in the recycle condenser (E-3). Pilot plant experimentation showed this to be 21 psig: above this pressure the polymer solution from the bottoms was overheated (above SS'C) and was discolored. The split between vapor and liquid in the column feed should not result in a liquid phase containing much more than 50% polymer because the solution would be too viscous to flow properly in the column. The theoretical case (Section IX-B-2) indicated a 5S% solution. For the reasons indicated above, the actual concentration in the pilot plant was probablv somewhat lower. 4. Addition of Stabilizer There is reason to suspect that it would be desirable to add the Paraplex G-62 stabilizer to the column feed rather than to the bottoms. Advantages might include improvement of polymer stability and alleviation of a very slight corrosion problem in the column.. This was not done in the pilot plant: however, provision of nozzles so that it can be tried in the plant is recommended. C. PRECIPITATION Solid VYSET RE polymer is precipitated from the monomer stripper (D-l) bottoms solution by mixing with water in a special high intensity mixing vessel. The use of the high intensity mixer is critical: attempts in the pilot plant to use stirred tanks and pipeline mixers resulted uniformly in unmanageable stickv, gooey messes. The mixing device used in the precipitation is called the SRC disinteg ratc*r. This mixer was developed by Shawinigan Resins Corp. for use in precipitation of polymers from solution. Readers of this report are cautioned that the existence of this device Is a closely held Monsanto trade secret. The SRC disintegrator is a vertical cylindrical tank with two intermeshing beater type agitators. The action of these beaters is identical with the ordinary kitchen egg beate r. Ir. operation polymer solution and water sire fed to the top of the vvlinder through separate nozzles. Slurry is withdrawn from the bottom nozzle through a leg with syphon break so that the mixer always runs full. The experimental work on precipitation was done using a 4 ir.. ID x t> :r.. high pilot plant SRC disintegrator loaned bv Shawinigan Resins Corp. The standard production SRC disintegrator is 16-1/4 in. ID x 24-3/Id in. high. RSV 0017368 7 3. Shawinigan Resins Corporation have scaled up numerous polymer precipitation processes using this equipment. In their aggregate experience, as outlined by Dabagian and Crozier (14), it has been possible to run in the production units all products which were successfully run in the pilot plant unit and, additionally, to run some products in the production unit which could not be run in the pilot unit. Since VYSET RE was successfully run in the pilot plant under a variety of conditions (cf. Volume II, Sections XVIH and XIX) there is little doubt that it can be run in a production unit satisfactorily. The independent variables in the precipitation operation are operating temper ature, water/polymer solution ratio, feed rate and beater (rotor) tip speed. The dependent variables arc polvmer particle size distribution and bulk density. The Shawinigan experience does not provide a method for predicting the exact operating conditions for a plant SRC disintegrator to obtain particle size distribution and bulk density which duplicates a pilot plant material. Indeed, some experimentation with plant rotor speed and operating temperature arc almost alwavs neccssarv. The pilot plan: experimental work was designed to find the range of feasible operation and to show, within this range, in " ha* direction changes should be made to change bulk density and particle size to match specifications. Tile results of the experimental work are given in detail in Section XVIII. lollowing general observations were made. The 1) Bulk density increased from 17-20 lb/cu.ft. to 21-25 Ib/cu. ft. upon increasing the slurrv outlet temperature from 10-20C to 30`-35*C. 2) Beater tip speed had to be 13^0 :i, sec or above for satisfactory operatic r. 3) In the range 1350-1740 ft/sec beater tip speed liar: little effect on particle size distribution or hulk densitv. 4) Operation with volumetric total feed rates from 0. 4 1 to 1. 4h SRC disintegrator volume s ' minute gave satisfactory precipitate having essentially the same particle size distribution and bulk density. ^ ) The maximum operating temperature limit for routine operations sr.ould be 30*0 tor acetone in the water phase o: the slurry. '-sing the maximum feed rates listed above, a 10,000. 000 lb.'vr VYSET RE plat:*. v- calc require at least two standard SRC disintegrators operating in parallel. :*. might be desirable to design ar.d develop a larger SRC disintegrator for a large plan:.' z ollowmg precipitation in the SRC disintegrator, the precipitate must be r.arueneci bv learning residual acetone from the solid particles. Shawinigan Resins Corporation does tms :n a s ev_ one -s :a ge SRC disintegrator. The pilot resuT~> me;- that an be.it be ucne ::: the VYSET RE prov css ; r. at', agitated r.ciu t a rex ( T -1 4) r.aving volume equal to about one hour s slurrv production. RSV 0017369 ye tv?.. yew- In the pilot plant VYSET RE was also successfully precipitated from solution in a one-gallon Waring blendor operated either as a batch or a continuous over flow operation. Scale-up to plant scale of the Waring blendor type operation would be difficult because no plant-scale analog of a Waring blendor exists. * D. FILTRATION AND CAKE WASHING i he hardened polymer slurry which overflows the slurry hold tank is filtered and washed on the filter to remove residual EKF and acetone. Two types of filters -- rotarv vacuum filter and perforate basket centrifugal -were tested in the plant. Both types of filter were used successfully. The rotary vacuum filter (F-5) was chosen because of a probable difficulty with dazing of the filter cake heel in a perforate basket centrifugal in a plan: operation. This common difficulty encountered in polymer filtration in pertorate basket centrifugals is caused by the effect of the discharge plow on the surface ol the heel of polymer which remains in the basket to produce an impervious glazed surface. This could not be evaluated in the pilot plan; because the pilot plant did not produce enough material to permit a test in an automatic discharge machine. ** The rotary vacuum filter operated well in the pilot plant and seems an ideal choice for the VYSET RE process. Dry cake production was 50 lb/hr-sq.ft. Sonic pilot plant tests were also run on dewatering (but not washing) VYSET RE tr. a batch solid bowl centrifugal. It was found that all polymer in the slurry feed settled to the periphery of the bowl. This suggests, but does not prove, that a continuous solid bowl centrifuge could be used for dewatering VYSET RE. DRYING VYSr-T RE powder discolors in air above 65 CC and fuses to a stickv mass at "0e`-60f`C. Any drying process must necessarily avoid these limiting ond ' * tons. -A. hign-shear bottom driven mixer known as the Papenmeier mixer (Welding Engineers, Ir.c. , Norristown, Pa. ) was tested. It failed because uses a round tar-< v. ;th batfle rather than the four-lea: clover type tank -see bv the Varing blenuor. The baffle cuicklv became fouled with oolvmer deposits. The Decatur Nuke r - Pe rk: r, and washing, with glazing o :r:lan polymer plant uses both rotarv vacuum filters and er Mccr type perforate basket centrifugals for polymer filtratio r.hev prefer the rotarv vacuum filters because o: the difiiceltv t.u* r.eel ir. the basket centrifugals. RSV 0017370 The VYSET RE produced in the pilot plant was routinely dried in a mechanical air convection tray dryer. The long (18-24 hour) drying cycle and low capacity of tray dryers make their use in a plant completely impractical. Several types of drying equipment which were potentially suitable for a plant operation have been given preliminary consideration. These were continuous through-circulation, direct rotary, fiuidized bed, flash and vacuum tumble drvers. The vacuum tumble drver was ruled out because of low capacity and high cost. The flash dryer works, but offers serious control problems. The continuous through-circulation, direct rotary and fluidized bed types probably would be satisfactory; however, more work must be done in order to make a proper choice and obtain information for sizing. This was not done in the pilot plant because the experimental work can be done much more economically and reliably using larger quantities of material which will become available from the projected Spr ingficla interim pla:nt. The Springfield faicilitv will use a ditect rotarv drver which is al:eadv installed Er.o ugh work has been done in the Sp-ringfield pilot plant to shov. : hat th:is dryer can handle the ou1tout of the proposed inte rim plant. Te s t s of the c onti r.uous through-circulation and ;luid bed drvers c an be carried out ir. established test facilities. Th c Decatur (Chemstr and) pilot plant have an exet;Uent through test set up including: the nece ssarv pell eting eouipme r.t. General Ameri can Transportation Corpora::ion maintains a fluid bed test ir. st&llation. Pohmer product quality can deterior ate severely during drying. u-valuiation t-f d r y: re tests must definitely include p roduct property ova lua tions. Anv drving S', st cm chosen m\:s: be shown to be c apable of producing a produ: : havi ng good appl ieation prope:*tie s. Pci-.-mer dried ir. the oilot olant trav drver was somewha t luniov. The s e lumps were generally abrogates of smaller particles which persisted from the filler cake. These lumps were considered objectionable by the product ^valuation groups. They were broken up either by screening or by running the dry product through a Model M Fitzpatrick Commutating Machine (The Fitzpatrick Company, Chicago, Illinois} without screens. Ir. the plant if polymer is dried in a rotary drver, a fluid-bed dryer or a flash drver, it may no: be necessurv to install equipment to break up lumps. If the pclvmcr is dried in a through-circulation urver (which requires pelleting prior to drying) some grindir.g apparatus to break up the dry pellets will probably be required. F. ACETONE RECOVERY Acetone is distilled from the filtrate from the rotary vacuum filter (F-5). This predominately aqueous stream contains about c,3re acetone, a trace of EHF and ar.v solid polymer wr.ich may, through a cloth defect or otherwise, have passed rotary vacuum, filter. The design and operation o: the recover-.- system are considered belc.>.. RSV 0017371 Filtration of Column Feed Any solid particles in the acetone column (D-2) feed will inevitably become deposited on the feed tray or on the trays immediately below. Such deposits would inevitably be difficult and expensive to remove. For this reason it is important that an efficient means for removing solids from the column feed be provided. In the pilot plant this was' accomplished by two filters in series. The primary filter (F-6) was a pressure-leaf filter in which the bulk of the solids was removed. The secondary filter (F-7) was a cartridge filter which removed any traces of very small particles which had passed the primary filter. This system worked satisfactorily in the pilot plant. For plant applications, it is recommended that filters having internals which can be easilv cleaned, c.g* Sparkler filters, be used. The possibility that difficultly removable solids might accumulate inside the leaves of pressure filters cannot be ruled out on present knowledge. Column Stage Calculations Extensive vapor-liquid equilibrium data are available for the systems ace tone-wate r (15). This makes it possible to calculate the theoretical stages required to effect anv given separation with precision. Figures 10 and 11 show the calculation for the number ol stages required for the separation given in the material balance, Section VI, for reflux ratio two. The modified latent heat of vaporization method was used to calculate theoretical stages. This method was chosen after a few test calculations showed that sensible heat effects had little effect on the location of critical sections of the operating lines. * The calculation shows that for feed at 25 *C, four theoretical stages are required above the feed tray and four belov.. The minimum reflux ratio for the acetone-water separation is substantially influenced by the temperature of the feed. In general, the slope of the operating line for the stripping section is very steep and increases with increasing temperature of the feed. This increases the minimum reflux ratio required to obtain an intersection of the stripping and rectifying operating lines below the equilibrium line . Heat ol mixing and acetone vapor heat capacilv data are available (If*, 17 ), so plate-to-plate enthalpy balances can be made if it is ever considered necessary. RSV 0017372 RSV 0017373 In the pilot plant the acetone recovery column was successful1'/ operated at reflux ratio 2 with feed temperatures up to 45*C. Operation was better with feed at 35eC-45*C than at 25*C because the higher vapor rate in the stripping section caused much difficulty with flooding. In a plant column with properly sized downcomers, there would seem no good reason other than possible heat economy by using waste heat for pre heating the feed. In the pilot plant distillation column, Figures 5 and 6, the 1 in. ID 10-sieve tray Oldershaw rectifying section achieved a separation equal to about five theoretical stages. The 24 in. Goodloe wire mesh packed stripping section achieved about seven theoretical stages or 3. 4 in. height equivalent to a theoretical stage. This suggests that stage efficiencies in the acetonewater system tend toward thc.low side and that a conservative approach to tray rating in ? plant column is desirable. 3. Mechanical Desien of Acetone Column The unusual feature of the acetone column operation is the high, liquid rates in the stripping section. In the pilot plant, glass Oldershaw columns could not be used in the stripping section because the downcomers were too small to handle the liquid flow at vapor rates high enough to operate the column. In a plant column, unusually large downcomers will certainly be required in the stripping section. It is further recommended that in a plant column, the oversi/.e downcorners be continued for two trays above the feed tray. Da situ r (Is) has reported that this is needed for acetone-water separations and pilot plant observations tend to confirm this. The presumed reason for the difficulty is boiling in the downcomers in the region where temperature chances rapidly. Either a sieve tray or a bubble tray column could be used in the plant. A sieve tray column is recommended because it would be easier to clean in the event that mechanical cleaning is ever necessary. The pilot plant acetone recovery column was heated by open steam rather than by a reboiler. Open steam is clearly preferred because a reboiler would undoubtedly be subject to some fouling. 4. Control of the Acetone Column ihe pilot plan* acetone column was controlled by setting the reflux ratio a: two, setting the open steam feed to the bottom o: the col'., mr. to maintain - at. H; O g pressure at the bottom, and adjusting the feed to maintain the teed plate temperature at 71c-72cC. This is the proper procedure to obtain maximum throughput at the established reflux ratio. In a plant the column feed will be established bv the rate o! p r ii m t: iu: el nitrate by the rotary vacuum filter. In this case the proper urexedure will be to set the column feed to equal the filtrate production rate ur.d to RSV 0017375 adjust the open steam feed to maintain a selected temperature at a control point in the column. The control point of the pilot plant acetone column was the temperature of the vapor above the feed tray. Figure 12, the boiling point-composition curve for the acetone-water system, shows this to be a region of substantial sensitivity. In a plant column thermowells on the feed tray and on several travs near the feed tray should be provided for possible use as control point. Although it was not tested in the pilot plant, a continuous differential refractometer would appear to be a desirable instrument for monitoring the distillate water content. This may also be done in the field by specific gravity measurement; however, extremely careful observations must be taken to obtain accurate results. Water Content of Rccvcle Acetone The water content of recycle acetone affects conversion in the polymerization reactor. For this reason it is important that the water content of the recycle acetone be maintained constant in the i.T'7o-2. 3% range. Any fresh acetone added to the recycle system should also be diluted to 2% water content. Column Fouling The acetone recoverv pilot plant was operated a total of 7S hours. During this time a very small amount of yellow solid accumulated or. the packing in the section just underneath the feed inlet. This material appeared to be partially but not entirely soluble in ho: acetone. The amount of material accumulated was so small that it could not be properly characterized. In the design of a plant column it is recommended that provision be made for washing the column with hot solvent and for easy access to the trays for cleaning. There is at least a remote possibility that the solid deposits were catalyst residue and possible cuite poisonous. This point is discussed in Section XII-A-3. MATERIALS Or CONSTRUCTION Reactor and Recvcle System Pilot plant operation of the polymerization reactor and the recycle system indicated that glass-lined steel and stainless steel are suitable materials of construction. Stainless steel coupons placed in the vapor space above the reactor, in the reaction mixture and.in the top of the stripping column showed no evidence o: corrosion. The stainless steel flanges on the reactor and the stainless steel recvcle condenser uere not corroded. One RSV 0017376 1. RSV 0017377 incident of corrosion was observed when polymer solution in the monomer stripper was accidentally overheated to decompose the polymer. Stain less steel parts exposed to the decomposition products showed spots of rust. It is possible that rusting due to accidental decomposition of polymer solution can be minimized by supplying the epoxy resin stabilizer feed to the stripping column inlet* although this was not tried. 2. Precipitation and Filtration Materials of construction used in precipitation and filtration equipment should be stainless steel with exception of the slurry hold tank {7-1*0 which requires glass or resin-lined steel. In the early stages oi pilot plant work an open stainless steel slurry hold tank was replaced with a polyethylene tank when traces of corrosion were observed on the stainless steel vessel wall in the proximity of the liquid-air interface. This change was made before epoxy resin stabilizers were added routinely to the polymer solution, subsequently no evidence of corrosion in the pilot plant system was observed. 3. Acetone Recovery Stainless steel is an acceptable material of construction for the acetone recovery system. Possibly carbon steel could be used except in the condenser and reflux splitter. H. PACKAGING AND STORING VYSET RE POWDER Dried VYSET RE powder was routinely packaged in polyethylene-lined Leverpacks and stored at room temperature. There was no evidence of de:e rioration during storage. I. PRODUCT PROPERTIES ihe definitive test of the quality o: a VVSli RE product is how- it pe rform s in applications. Product made by the methods of the Tentative Pr oce s s and meeting the specifications listed in Section XIV has been judged sat; siactorv by Central Research and Plastics evaluation groups for metal c oav.r.c appl: cations. The impact specification > 5Gro pass was the most important speci::cat:or.. Damage to the polymer during drying was the most common cause of failure to meet this specification, provided the viscosity and chemical composition was wi: h i n specification. :`Ov currence oi accidental decomposition of polvmer solution appears unlikelv m tr.e plant v/nere heat addition to the monomer stripper is effected bv addition of open superheated acetone vapor only. RSV 0017378 The impact resistance of pilot plant polymer was substantially better at the upper end of the viscosity range than at the lower end. J. RAW MATERIALS The suppliers for raw materials listed in Section V are those found by Dr. A. D. Gott of the Monsanto Central Purchasing Department to be the most likely sources for a manufacturing operation located in New England. All these materials were tested in the pilot plant. The vinyl chloride used in the pilot plant was obtained from Texas City in 200 lb steel cylinders, which frequently contained rust, polymer and presumably ferric chloride. It was necessary to purify this VC1 monomer according to the procedure given in Section XX, Appendix A. The bulk monomer delivered by tank car to a plant will presumably be transported under clean conditions and therefore not require purification. This was tested in the pilot plant by obtaining a shipment of VC1 in new cylinders from Texas City and using it successfully without purification. The acetone used in the process was obtained fiwm a plant which makes acetone by isopropanol dehydrogenation. Laboratory experimentation should precede any attempt to use acetone made by the cumene-phenol process, since traces of phenol would be expected to inhibit polymerization. EHF monomer is not an ar. le of commerce and must be made by Monsanto. A tentative process for EHF manufacture is available ( 5 ). K. WASTE DISPOSAL The principal waste stream from the plant is the bottoms from the acetone column, which is mostly water with a trace of EHF and possibly minute amounts of catalyst residues. Disposal of this stream would presumably no: be difficult; however, it must be evaluated with respect to the existing local conditions. During polymerization reactor start-up and shutdown operations, substantial amounts of VC1, 13 and acetone mixture may have to be disposed of. The magnitude c: this disposal problem will depend on the scale of operation and local conditions. It is noted that in some cases where VC1 is flared, it is necessary to scrub the combustion products with caustic to remove acid prior to discharge to the atmosphere (1). Small amounts of acetone used in reactor and. column cleaning will have to be disposed o: oeriodicallv. RSV 0017379 S4. L>. PRODUCT STABILIZATION The practice of adding an epoxy-type stabilizer to the polymer solution effluent from the monomer stripper (D-l) was adopted midway through the pilot plant work. This was done because the unstabilized dry powder showed a tendency to rust mild steel when stored in contact with it. Evaluation groups have agreed that addition of stabilizer has eliminated this problem. Addition of stabilizer has had ancillary advantages. It virtually eliminated the problem of corrosion in the precipitation and led to a slightly lighter colored product. Two epoxy stabilizers -- Epor. 828 and Paraplex G-62 -- were used in the pilot plant. Both were effective. Paraplex G-62 was preferred by Plastics Division Research (13). RSV 0017380 65. X. COST ESTIMATE (by E. C. Pybdal) Estimated manufacturing costs and capital requirements for the production of 5, 10, and 20 million lb VYSET RE (designated as VCl/EHF/lB terpolymer in the memorandum to Dr. R. W. Schuler) were prepared by Dybdal (4 ). Summaries of these estimates are given in Tables II and III. Detailed estimates were prepared for the base case; 10,000,000 lb a year, and conversion costs and capital requirements for 5,000,000 and 20,000,000 lb a year were factored ("0.6" factor for capital) from the base case. For 10, 000, 000 lb a year (Table II) estimated raw material cost is 17. 846 a lb. Estimated raw material requirements and costs as well as details of the estimated conversion cost (7. 26 a lb) are shown in Table III. An over-all yield of 86% was assumed which included an estimated 5% loss for scrap and a yield of approximately 90% on monomers. The estimated bulk manufacturing cost is 25. 06 a lb. Estimated M5cE is $1, 35 0, 00C and fixed capital is $2, 175,000 based largely on the estimate that nonmanufacturing capital will be about 50% of MkE. Working capital ($1, 300,000) was assumed to be 28% of r.et annual sales. Selling price for a 20% return after Federal income taxes is 466 a lb (based on a tax rate of 52% of net profit). At 5, 000, 000 lb a year the estimated raw material cost becomes 20. 36 a lb (approximately 1, 100, 000 lb EHF at 556 a lb are required). Conversion cost is 10. 16 a lb and the bulk manufacturing cost is 30.46 a lb. Estimated M&cE is 3SQ1,000 and fixed capital, $1,437,000. The selling price for a 20% return after Federal income taxes is 5a lb. A: 20, 000, 000 lb a year, estimated raw material costis 16. 36 a lb (approxi rr.atelv 4, 430,000 lb EHF at 37a lb). The conversion cost is 5. It and the bulk manufacturing cost is 21. 41 a lb. Estimated M&E is $2, 047, 000 and fixed capital: $3, 321,000. Selling price for a 20% return after Federal income taxes is 3 96 a lb. Prices for EHF at the three production rates are estimated prices and were presented ir. a memorandum to Dr. R. W. Schuler, Cost Estimate for the Manufacture of Ethvl, 5 Hydroxyethyl Fumarate, Revised July 30, 65. The following is a brief description of the projected process of manufacture cr. which these estimates were based: 1. VYSET RE is produced in an overflow reactor by continuous addition of four streams: (a) vinyl chloride and EHF dissolved in a portion of the mquirrid acetone, (b) isobutylene fed separately, (c) a*o catalyst dissolved in acetone, and (d) a recycle stream of unreacted vinyl chloride dissolved in the remaining acetone. The effluent from the reactor is a 30% solution of terpolymer in acetone and unreacted monomers. Reaction temperature RSV 0017381 86. TABLE II ESTIMATED PRODUCTION COSTS, CAPITAL REQUIREMENTS, AND SELLING PRICES FOR THE MANUFACTURE OF VCl/EHF/IB TERPOLYMER Composition - 74. 5% vinyl chloride, Plant location - Springfield Yield on monomers - 86% 19.6% EHF. and 5. 9% isobutylene Production rate, lb/vr Raw material cost, d/lb Conversion cost, d/lb Bulk manufacturing cost, d/lb M&E, S Fixed capital, $ Working capital, $* Total fixed plus working capital Selling price, 20% return after taxes, d/lb** 5,000,000 20. 3 10. 1 30. 4 891.000 1.437.000 809.000 2.246.000 58 10,000,000 17. 84 7. 16 25. 00 1.350.000 2.175.000 1, 300,000 3.475.000 46 20,000,000 16. 3 5. 1 21. 4 2.047,000 3, 321 , 000 2, 15, 000 5, 480, 000 39 "23% of net sales :"`SARE expenses 15% of net sales, Federal income taxes 52% of net profit RSV 0017382 87. o COST ESTIMATE PRODUCTION 10, 000.000 lb LOCATION Sprinefield M&E - $1, 350, 000 Mfc. Bide. - $150,000 RAW MATERIALS vinvl chloride EHF isobutylene azo catalyst acetone TABLE m H flvCl/EHF/lB Terpolymer (Revised 7/31/63) Ml VIIS as product in bulk storage OPERATION | TIELO , ON 8000 mrs./'yr.I 86 iIMonomers i UNIT QUANTITY S PER UNIT T JOATI ! S PER YEAR MEET MO. or joo numocn OH*D* iOAT i % PER too UBS, lb lb lb lh lb -- 18. 755,000 0. 0706 ! 2. 216. 700 i 688. 600 ! 60.000 i0.44 10. 0796 ! 1. 70 961.900 10. 0355 - :i ______________ t| --... j 1 1 1 i .l 61 8. 1 00 975.400 54.800 102.000 34.200 ................ 6 1A 9. 75 0,55___ _______LJL2____ ; 0.34 ..... >-------- ------ -- ! i j_______________ L I To*ol Gro R.M, CREDITS 1,784,500 : 17.84 TOTAL NET R.M. DIRECT EXPENSE labor SUPERVISION PAYROLL CHARGES STEAM ELECTRICITY COMPRESSED AIR REPAIRS WATER-COOLING WATER-PROCESS-CITY WATER FUEL-GAS FACTORY SUPPLIES L AOOR AT ORv CLOTHING 4 L A U N DR v i ManHr M lb CKWH J/ gal >*{ pal 26,670 78, 000 19, 560 75,000 16, 260 3. 60_____ 0. 88 1.131 0. 126 0. 126 96,000 12.000 17,000 6. 340 22. 120 183,600 Q, 450 2. 050 6,750 56.000 840 ; 0. 96 0. 12 0. 17 0. 6Q 0. 22 < 1. 84 0. 10 0. 02 0. 07 0. 56 0. 01 TOTAL D.E. INDIRECT EXPENSE DEPRECIATION - M & E DEppEC! AT ION - 9LDG. vTAXES AND INSURANCE . -> - ^ other indirect J *" TOTAL 1 .D.E. BULK MANUFACTURING COST PACKAGING & shipping TSYal P i"s COST TOTAL UAHbFACTURlHC COST f.o.b. 475,150 4. 76 112.500 7.500 1 18. 800 1. 13 - 0.08 1. 19 - 2 3 H, H00 2. 40 498, 450 25. 00 RSV 0017383 is SO*C and the pressure# 200-225 psig. The reactor is a one-thousand gallon jacketed, stirred vessel of 304 stainless steel for a capacity' of 10,000,000 lb terpolymer a year. 2. The product stream from the reactor passes through a letdown valve and the pressure is reduced to 20-60 psig depending on the cooling water temperature. A heat exchanger is provided to supply the heat of vaporisation of VC1, IB, and acetone and compensate for the cooling on expansion. Vapor and-liquid are separated in the letdown tank. The liquid phase containing the dissolved terpolymer is pumped to the feed tank for polymer precipitation. 3. Polymer is precipitated by pumping the product stream through a nozzle into a high-velocity stream of water in a pipe loop formed by withdrawing slurry from the bottom of the precipitation tank through a 500 gpm pump and discharging the slurry into the top of the tank. The capacitv of the precipitation tank is 300 gallons. Slurry from the precipitation tank is riltcred by centrifugation to recover the polymer.* *. Two 4S-inch suspended-ba sket programed batch centrifugals are assumed. The cake is washed with water before it is discharged into the cake hopper for the dryer. Filtrate and wash water are kept separate. Tne filtrate contains about 7. 5% acetone which is recovered bv distillation. :i The centrifuge cake contains 60To water and is dried in a Proctor and Schwartz continuous through-circulation drver to about 0.5% water in the product. The cake is preformed in a Bonnot pelletizer before it is fed to the dryer. The polymer is then transferred to product storage. 7 ne tentative process calls for precipitation in an SRC disintegrator and nitration on a rotary vacuum filter. The numbered equipment costs for this equipment have been compared with the costs for the same steps in the estimate. It was :ound that the costs would be insignificantly less than those usee i r. tr.e e s timate :he nee the estimate may be considered valid lor the te r.ta tive crocess. RSV 0017384 69. XI. PATENT STATUS (Prepared by J. D. Kennedy) The following patent applications are on file at the Patent Office. They have not yet been on file long enough to receive any official action by the Patent Office. RE 2744 - Hydroxyalkyl alkyl fumarate copolymers with vinyl chloride; high solids solutions of such copolymers in aromatic solvents: process of preparing such copolymers, particularly from monomer of low- acid content. RE 231? - The use of organic amines in the interpolymerizatio** of vinyl chloride with hydroxyalkyl fumarates, with or w`'',out isobutylene, specifically the use of trialkyl amines in amo . stoichiometric to the monoalkyl fumarate in the monomer to dc-ctivatc same. RE 2S20 - Terpolvmers of vinyl chloride, fumarate esters (particularly alkyl hydroxyalkyl fumarates) and isobutylene; such polymers containing particular ranges of monomers; methods of conducting the polymerization. RE 2824 - Interpolymers of vinyl chloride with bis (hvdroxyhydr ocarbv 1) fumarates, optionally including dialkyl fumarates as additional monomer, and polymerization processes. The filing of additional applications covering interpolvmers from related monomers, process details, and curing and stabilization systems is contem I plated. The prospects for sound patent protection, such as composition of matter claims, are better for the terpolvmers containing isobutylene as third monomer than for the two component polymers. I No infringement problems, i.e., adverse patents, of sufficient significance to prevent commercialization of these polymers have been found. Information concerning studies which have been made can be obtained from the Patent 1 Department. I I I I RSV 0017385 Q0. XII. TOXICITY AND HAZARDS* l--------------------- A, 1. HAZARDOUS OPERATIONS Polymerization 1 The polymerization operation, carried out according to the operating procedure of Section VIII, is straightforward. It has only the hazards normally associated with high-pressure exothermic reactions of flammable materials. 1 Particular attention should be'paid to protection of personnel from the consequences of leaks or errors during reactor sampling. While taking 1 samples operators should wear face shields, chemical goggles and gloves. In the event of cooling water, power or mechanical failure, the reaction in 1 a large plant reactor will have to be stopped to prevent overheating and overpressuring the reactor. The most effective wav to do this is to vent the volatile vinyl chloride and isobutylene monomer to a safe place. In cases where the reactor stirrer is still operating, a free radical acceptor 1 such as hydroquinone might be pumped into the charge to stop the reaction. 2. Precipitation 1 The acetone content of the liquid phase of slurry produced by precipitation of stripped polvmer solution is about 57o (wt. ) acetone. At 90" F a B concentration of about 7. 37o acetone in water will produce explosive vapors. Under the operating conditions outlined in Section VIII, the acetone con centration in the vapor over the precipitated polymer slurry would be below the lower explosive limit. However, an operating upset which resulted in 1 too much acetone in the stripper bottoms or one in which filtrate rather than wash filtrate were recycled to the precipitation could lead to acetone vapor concentration ir. the explosive range. 1 A small amount of ur.reacted VC1 and IB monomer is present in stripped polymer solution. Some of this monomer may be present in the vapor over 1 the precipitated polymer slurry, thereby adding to the fire and explosion hazard. Designs should provide for proper venting of unreacted monomer vapor from the slurrv hold tank (T-14). *The lower explosive limit for acetone in air is 2. 55 % (vol. ). The vaporliquid equilibrium data giver, by Chu (13) indicate that this vapor acetone con centration corresponds to about 7. 37a (wt. ) acetone in the water. RSV 0017386 91. 3. Filtration The problem with potentially explosive vapors is the same as that discussed in Section XII-A-2 above. Proper precautions against fire and explosion should be taken. 4. EHF Handling EHF monomer used in this process causes severe skin irritation bv contact even in dilute form as in the reactor effluent. The only way to avoid skin irritation when handling EHF is to avoid all contact with the skin. In achieving this, the following principles should be kept in mind: a. The fumarates are essentially nonvolatile: hence they remain where they are spilled. Spills may be washed up with isopropanol, ethanol or copious amounts of water. Tools may be cleaned with hot water. b. Temporary protection is obtained with rubber or vinyl cloves; however, the fumarates are soluble in rubber ^ nri vinyl and soak through in time. Gloves which have been exposed to fumarates should be dis carded immediately. c. Leather also absorbs fumarates. Any shoes which have come in contact with fumarate material should be discarded. d. Janitors and others handling contaminated waste may be affected. Special provisions l'or handling waste should be provided. e. Clothing which has been contaminated with fumarate should be changed immediately. A system should be worked out so that laundry workers will r.o: be exposed to the contaminated clothing. At the Research Center, contaminated clothing is sent to the laundry in polyethylene bags and laundry workers have been warned to dump the clothing into the washers directly from the bags. Details c: EHF toxicity studies are given by Carter ( 5 ). 5 . Handling At o-bi s-Isobutv ronitr ile (AZ3X) (20) Packaged AZ3N should not be exposed to temperatures in excess of 50eC as vigorous, but not explosive, decomposition is liable to occur with the risk of fire. Since some of the potential decomposition products are highly toxic, it is strongly recommended that all operations be carried out with adequate ventilation. Avoid contact with eves, skin and doming. Use of dust masks, rubber gloves and protective clothing is recommended. AZBN can be decomposed by mechanical impact without explosion. An ex plosive hazard appears to exist only for solutions of AZ3X in acetone (20). The maximum safe concentration of AZ3N in acetone is unknown: in the pilot plant solutions up to -1% wt c one v nt r a! 2 on w err: handled withon* difficulty. Trc temperature of the solution should not exceed 35 6C. RSV 0017387 92. I 1 During preparation of initiator solution, the initiator should be added to ^ the bulk of solvent in an agitated vessel. Configurations of equipment leading to entrapment of partially dissolved initiator should be avoided. In some experimental work AZBN which had been recrystallized from ethyl acetate was used. The recrystallization procedure is given in Section XX-B. Do not attempt to recrystallize AZBN from acetone. B. HAZARDOUS COMPOUNDS 1. Acetone (21) Description: Colorless liquid fragrant mint-like odor Boiling point: 56.48*C Flash point: 0F (closed cup) Explosive range in air: 2.55-12.8% Autoignition temperature: 1000eF Fire extinguishers: COs, dry chemical Toxic hazard rating:* Acute local: Irritant 1, Ingestion Z. Inhalation 2 Acute systemic: Ingestion^, Inhalation 2, Skin Absorption 2 Chronic local: Irritant 1 Chronic systemic: Ingestion 1, Inhalation 1, Skin Absorption 1 Maximum allowable concentration in air: 1000 ppm I. C.C. classification: Flammable liquid, red label 2. Azo-bis -Isobutyronitrile (20) Description: White crystalline solid: stable at room temperature Flash point: Can be ignited readily with an open flame. Will decompose vigorously, but not explosively, at temperatures above 50C Fire extinguishers: COs, dry chemical Toxic hazard: Decomposition products highly toxic. Toxic effects I produced through inhalation of dust or vapors and absorption through skin or eyes. Explosive hazard in acetone solution: Maximum allowable. Concentration unknown. Do not heat solutions of AZBN in acetone B above 35*0. Do not attemul to recrv stallize in acetone. I I *TOXIC RATING CODE: 0 - None, 1 Sli g 2 - Moderate, 3 igh. I U = Unknown RSV 0017388 G 3. B 3. E thy Ihydroxy ethvl Fumarate ( 5 ) Reference ( 5 ) gives details of a toxicological study of EHF. Description: Straw colored liquid Boiling point: > 250*C Flash point: Unknown, probably high Explosive range in air: Unknown, probably not explosive .. Autoignition temperature: Unknown Fire extinguishers: CO*, dry chemical Toxic hazard: Powerful skin irritant, very dangerous to eyes, toxicity when ingested. I. C. C. classification: Not established moderate 4. Epoxy Resin (Epon S2S, Paraplex G-62) (22) Toxic hazard: Primary skin irritant. Sensitizing agent giving rise to some forms of hypersensitivity 5. Isobutylene (21) Description: Volatile liquid or easily liquefied gas Boiling point: - 6. ?CC Flash point: < 20CC Explosive range in air: 1. S-8. 8% Autoignition temperature: S69*F Fire extinguishers: CO; , dry chemical Toxic hazard rating: Details unknown. May have asphyxiant or narcotizing action Maximum allowable concentration in air: Unknown I. C.C. classification: Flammable compressed gas, red label 6. Methvl Isobutvl Ketone (21) Description: Colorless liquid Boiling point: 115. 1*C Flash point: 75eF (open cup) Autoignition temperature: 860F Fire extinguishers: Foam, COs , dry chemical Toxic hazard rating: Acute local: Irritant 2, Ingestion 2, Inhalation 2 Acute systemic: Ingestion 2, Inhalation 3, Skin Absorption 2 Chronic local: U Chronic systemic: U Maximum allowable concentration in air: 100 ppm I. C.C. classification: Flammable liquid, red label RSV 0017389 94. Vinyl Chloride (21) Description: Colorless liquid or gas; faintly sweet Boiling point: - 13.4*C Flash point: - 108"F (Cleveland open cup) Explosive range in air: 4-22% Fire extinguishers: COa, dry chemical Toxic hazard rating; Acute local: Irritant 1 Acute systemic: Ingestion 2, Inhalation 2 Chronic local: TJ Chronic systemic: Ingestion 1, Inhalation 1 Very dangerous when heated to decomposition, will emit phosgene. Maximum allowable concentration in air: 500 ppm 1. C.C. classification: Flammable gas; red label RSV 0017390 >& I I 1 I I k 8 I LI I I I I k M 3 Xm. ANALYTICAL METHODS A. NONVOLATILES IN THE REACTOR CHARGE Connect sample bomb (SB-1) (Figure 13) to the reactor sampling manifold as shown in Figure 14. Purge the sampling manifold of residues of previous samples by sweeping a small amount of reactor charge through valves SV-1 and SV-2 into the purge volume. Valve SV-3 is opened to permit withdrawal of a reactor sample into the sample bomb (SB-1). Subsequently valve SV-3 and SV-1 are shut off and the amount of reactor charge trapped in the purge volume is flushed through valve SV-4 into the purge receiver. The sample bomb is then removed from the sampling manifold, any spill over from the reactor is washed off with acetone, and the net weight of the reactor sample determined. The determination of nonvolatiles is carried out on an Ohaus moisture balance (Model 6000). Adjust zero balance with an aluminum liner in place on the pan of the moisture balance. Discharge the contents of the sample bomb on the aluminum liner. The residue in the bomb is rinsed onto the liner with acetone to insure that the entire reactor sample was transferred. Commence to drive of: the volatile components of the sample at a heater setting of 35 (approximately 65C) on the Ohaus balance for 40 minutes and weigh the dried sample. A nitrogen purge of the interior of the Ohaus balance must be provided to prevent explosion of the acetone vapors in the balance. From the net reactor sample weight and the dry weight calculate the nonvolatiles of the reactor charge: tr , weiehr drv sample net weignt reactor sample . The sample bomb SB-1 is cleaned thoroughly with acetone, dried and tared prior to the next sampling. B. REACTOR CHARGE ASSAY For reactor control purposes, the volatile monomer and solvent charge in the reactor (VC1, IB and acetone) can be determined by a moderately fast (10 min. ) isothermal gas chromatographic method without internal standards. Gas chromatograph: Column: Packing ; Helium flow: Oven temperature : Injection, block power: Samole size: Perkin Elmer Model 154 t in. OD x 1 0 ft copper 5% 2, 2'b:s(cyano ethoxy) dipropyl ether on Teflon 6 75 ml/min 55 C 65 (setting) 1 pi RSV 0017391 96.  NPT X SWAGELOK TUBE SS CONN. 4-SS HOKE NEEDLE VALVE NPT X " TUBE SS SWAGELOK CONN. j I.D. X 6^* LG. SS TUBWGl J SS KONCENTRIC UN!N SAMPLING SEPTUM SB-2 WITH SS CONAX THERMOCOUPLE GLAND WELD SB-1 WITH KONCENTRIC UNION WELDED SHUT DUWN CHtCXEO DATE SCAL1 FIGURE RSV 0017392 13 Monsanto Chemical Company 800 N. UHPBEI9N ILYD. ST. LOUIS 66. MO. SAMPLE BOMB SB-1 B SB-2 ntoj. mo. LOCATION DIVISION DW6. MO. OMWN CHECK* OAT VlAli RSV 0017393 Monsanto Chemical Company too N. UNDtIKH UVD. FIGURE 14 REACTOR SAMPLING ST. LOUIS 66. MO. SYSTEM oj. mo. LOCATION orvisiON owe. MO. t I Sample the reactor using sample bomb SB-2 (see Figure 13) which has been charged with 20 ml methyl isobutyl ketone. The reactor sampling procedure is the same as outlined in Section XIII-A. The sample weight need not be I determined. Shake the bomb and withdraw a 1 fil sample with a syringe directlv from the sample bomb through the sampling septum. Measure the area response of IB, VCI and acetone on the chromatograph. The area I response factors for IB, VC1 and acetone are determined from a standard consisting of approximately 90% wt MIBK, 0. 5% wt IB, 5% wt VC1 and 4. 5*To wt acetone. Calculate for the standard: F_ % wt (i) i % area response (i) where i - IB, VC1 or acetone (Note 'd i rea response {i) area response (i) x 100) total area response I From the area response of the reactor sample calculate: wt IB = ro area response (IB) x Fib wt VC1 = To area response (VC1) x FvCl B To wt acetone = T? area response (acetone) x F^cejone I In order to express the above distribution of volatile constituents in terms of the reactor charge they must be multiplied by the following correction: 100 - % nonvolatile v 1 v Too ,, The peak for the sample diluent MIBK will appear on the chromatogram B approximately 10 minutes after injection of the sample. 2, 2'bis(cyar.o ethoxy) dipropyl ether is not available commercially but .has been synthesized in laboratory quantities at the Research Center (23). E C. RECYCLE STREAM ASSAY Assays for IB, VCi and acetone in the recycle stream (G) are carried out identically to Section XIII-B. D. VOLATILE RESIDUE IN POLYMER SOLUTION B Assays tor 13, VCI and acetone in the polymer solution (M) from the stripper bottom tre carried out identically to Sec tion .XIII-3. Since the polymer solution is .it p r s ^ > j r , a ^ ml .sample can he take n di r cc t ly into glassware B containmc approximately 30 ml MI3K. During normal operation o: the polymerization system, the area response of 13 is negligible. OQ (f RSV 0017394 E. RESIDUAL EHF IN THE REACTOR CHARGE For control purposes, up to 2% EHF may be determined in the reactor charge by a quick isothermal gas chromatographic method employing dibutyl phthalate as an internal standard. Gas chromatograph: Column: Packing: Helium flow: Oven temperature: Detector temperature: Injection block temperature : Sample size'. Aerograph A-90-P (Wiikens Instrument and Research Co. ) tM OD x 32" stainless steel 1% Carbowax 6000 on acid washed CHromsorb W 60/S0 mesh 60 ml/min 23*C 273*0 26S'C 1 ul Prepare a standard by charging about 25 ml MIBK to a 50 ml volumetric flask and weigh in about 5 g dibutvl phthalate. Make to the mark with more MIBK and shake well. Sample the reactor, proceeding in the same manner as outlined for Section XIII-A. Discharge contents of the sample bomb into about 30 ml MIBK. (If necessary, rinse sample bomb with MIBK to insure complete transfer of sample. ) To make a determination, charge Z ml of standard solution to the MIBK solution containing the reactor sample. Run this mixture on the chromatograph and measure the area response of the EHF and DBP. Simultaneously prepare a MIBK-EHF reference solution in the ratio of approximated 150:1. Add 2 ml standard solution to this mixture and again determine the area response of EHF and DBP. Calculate for the reference solution i area response icHF !_________ weight EHF ( area response DBP ^ weight DBP stanctara From the area response of the reactor sample ro wt EHF = area response EHF s- weight DBP N area response DBP x F 100 RESIDUAL EHF IN POLYMER SOLUTION Assavs foor reessidual EH: in polymer solutions are carried out ideennttiiccaally to Section XIII-E. Since the polymer solution-is at atmospheric pressure sampling can be hanadled directly in glassware. RSV 0017395 100. This method is applicable to polymer solution from the bottom of the stripper as well as dry polymer redissolved in either acetone or MIBK. G. ACIDITY AND HYDROXYL CONTENT OF VYSET RE (by R. J. Slocombc) 1. Hot Acidity Test a. Weigh 0. 5 to 1.0 g sample into 250 ml round-bottom flask. b. Add 10. 0 ml Analytical Reagent grade pyridine with moisture adjusted to 0. 3-0. 5% based on Karl Fischer analysis. c. Reflux 40 minutes under an air condenser (ground glass joint greased with "Lubriseal") in a constant temperature oil bath adjusted to 126 t 0. I'C. The 40-minute reflux time is checked with a stop watch to the nearest second after reflux drops begin falling from the condenser tip. d. Cool immediately in ice bath. e. Remove condenser and wipe grease from the ground glass joint. f. Pour contents of flask into 250 ml beaker, rinse the joint of the flask with a tiny amount of acetone, put the condenser back into the flask and rinse down the condenser with 15 ml acetone. Swirl the flask to completely rinse the sides with acetone, and pour it into the beaker. g. Pipette 25. 0 ml cyclohexanone (best quality available) into the flask, swirl to completely rinse the sides and add it to the beaker. h. Add 10. 0 ml distilled water to the flask, swirl to completely rinse the sides and add it to the beaker. : . Put the beaker in a plastic or glass ice bath, mount on a magnetic stirrer and allow the temperature to drop to about 5*C. j . Titrate pote ntiometrically using 0. 1 XaOH. Approach the pH of 8. 0 carefully, then reduce the additions of caustic to 0. 10 ml between pH reading s. k. The largest break is usually observed between pH values of ?, 5 and 10.5. The end point is determined by interpolating from the largest breaks to give the value of the titre to the hundredth of a milliliter. 1 . Run a blank using everything exactly the same except omit the sample of polymer. m. Calculate as follows: 101. RSV 0017396 (titre for sample) - (titre for blank) x N of base _ meq. acidity per weight of sample gram of polymer 2. Cold Aciditv Test a. Weigh a one-gram sample (accurately) of polymer into a 250 ml beaker. b. Add 25.0 ml cyclohexanone and 15.0 ml acetone to the beaker. c. Add a magnetic stirring bar, mount the beaker on a magnetic stirrer and allow the sample to dissolve. d. Cool the solution to about~5'C by setting the beaker in a glass or plastic ice bath on the magnetic stirrer. e. Add 10. 0 ml distilled water. f . Titrate potentiometricallv (while stirring in the ice bath) with 0. 1 normal NaOH. Approach the pH of 8.0 carefully, then reduce the additions of caustic to 0. 10 ml between pH readings. g. The largest break is usually observed between pH values of 9. 5 and 10.5. The end point is determined by interpolating from the largest breaks to give the value of the titre to the hundredth of a milliliter. h. Run a blank using evervthing exactly the same except omit the sample of polymer. i . Calculate as follows: (titre for samp~ le--) --- (v:-t-i--t-r--e------f-o----r----b--l--a--n---k--)----x-----N-----o--f----b--a---s--e------------ -------------------. - = meq. weight of sample gram of polvrrier acidity per 3. Hvdroxvl Determination a. Weigh 0. 5 to 1. 0 g sample into 250 ml round-bottom flask. b. Add 10.0 ml hvdroxvl reagent. (Hydroxyl reagent prepared daily as follows: Add 30.0 ml acetic anhydride, AR grade, into a 250 ml volumetric flask. Fill to graduation mark with AR grade pyridine containing 0. 3 to 0. 5% water based on Karl Fischer determination. ) c. Reflux 40 minutes under an air condenser (ground glass joint greased with "Lubriseal") in a constant temperature oil bath adjusted to 126 - 0. I'C. The 40-minute reflux time is checked with a stop watch to the nearest second after reflux drops begin falling from the con denser tip. d. Cool immediately in ice bath. 102 RSV 0017397 i I n i i i e. Remove condenser and wipe grease from the ground glass joint. f. Pour contents of flask into 250 ml beaker, rinse the joint of the flask with a tiny amount of acetone, put the condenser back into the flask and rinse down the condenser with 15 ml acetone. Swirl the flask to completely rinse the sides with acetone, and pour it into the beaker. g. Pipette 250 ml cyclohexanone (best quality available) into the flask, swirl to completely rinse the. sides and add it to the beaker. h. Add 10. 0 ml distilled water to the flask, swirl to completely rinse the sides and add it to the beaker. i . Put the beaker in a plastic or glass icc bath, mount on a magnetic stirrer and allow the temperature tc drop to about 5*C. j. Titrate potentiometrically using 0.5 N NaOH. Approach the pH of 8.0 carefully, then reduce addition of caustic to 0. 10 ml between readings. k. The largest break is usually observed between pH values o:' . 5 and 10.5. The end point is determined by interpolating from the largest breaks to give the value of the litre to the hundredth o: a milliliter. l. Run a blank using everything exactly the same except omit the sample of poly me r. m. Calculation: v. v? V, V\ N- Na wt: Wt} A - vol. base used in hot - vol. base used in ho: - vol. base used in hyd - vol. base used ir. hyd - normality of 1/10 X ! - normality 0. 5 X XaO - weight of sample for - weight of sample for - acidity correction test (V; - V- ) x N; x Wtg ' Xj X w t; = Acidirv Correction - A <V4 - A) - (V3) x o.o:~oi x 100 x N; Wt; OH H. PHYSICAL PROPERTY .TESTS TOR VVSET RE (by H. P. Holladay) Early tr. the VVSET RE program a routine procedure was adopted which gave a good comparative evaluation of research polvmers. The earlv polvrr.ers varied considerably from lot to lot in `er.ner.i. y to release HC1 to cause premature RSV 0017398 103. reaction of Resimene. The following solution procedure essentially removes that variable. Solutions are prepared according to the following formula: Polymer Xvlene mibk 50 pts 25 pts 25 pts If the polymer does not contain stabilizer, 1 pt epoxy stabilizer is added. Complete dissolution is normally obtained by rotational mixing 60 minutes at 60*C. The solution is then equilibrated at 75*F overnight. Resimene U Q20 (6.25 pts) diluted with MI3K {12.5 pts) and xylene (12.5 pts) is added with stirring, followed by 30 minutes of mixing on a 2 rpm solution wheel. The resulting solution is allowed to stand five minutes before castings arc made. Two solutions are prepared from each polymer lot. Hastings are made on SAE 1010 cold rolled steel (Gardner PG-1 300B) to give i. 5 ml thick c oating s when err. Three panelc are pr e pa red from each s olution. Castings are dried 15 minutes in 75eF, SO^o RH air before baking 15 minutes at 1 50 * C. Baked c oating s ar e c onditioned at 7 5 * F and 5 0% RH ove might before te sting. Testing procedure emphasizes impact resistance. Twenty-four reverse impacts (50" lbs) are run for each polymer and rated according to decree of failure and number of failures. Ratio of "no failures" tc number of tests run is reported. A Gardner Impact Tester is used. Bend tests are run by quickly deforming the panel, coating side up, over a j" rod. A test strip wide is cut in a transverse direction to coating. Test is rur. parallel to coating direction. Decree and type of failure is reported. Adhesion is observed as par: of impact and bend tests. Aaditionallv a 1/16" grid :s cut in. coating with sharp razor blade and tested by a fast pull, at 0E to plate, with ''Scotch-Tape" firmly attached to coating. Percent lift of grid squares is reported. Pencil hardness is measured using graded engineering drawing pencils with lead ilattened by sanding at a right angle to pencil. The sharp edge is used to gouge completely through the coating. The hard leaa pencils are used progressively to softer leads until a softness is reached which will not cut through coating but next harder lead will cut through. The hardness of the pencil just too soft to penetrate coating is resorted. Color of baked coating is reported in cual at;ve terms. "Viscosity is measured in Gardr.er-Holdt 3ubble Viscometers at -dcF. Polvmer 104. RSV 0017399