Document 3Nq1dD1Nxkjx2zvJ72Gpr7EJa

B.F.GOODRICH CHEMICAL COMPANY DEVELOPMENT CENTER V0 To: *L. A. Bennett *A. R. Berens S. M. Davis *F. E. Krause *R. M. Kresger P. H. Lawrence R. N. Rylands *G. D. Schaaf E. G. Schwaegerle R. J. Wolf H. L. Wuerth Date: From: December 29, 1967 H. H. Marty KHM-U4-67 cc: Development Center V. N. Arnold H. L. Brandt E. A. Collins *A Jf.Fairlie - *H.S.Haller *P. J. Horvath H. W. Long G. M. O'Dell *W. A. Reed *R. F. Reinhart Central Technical File Cleveland OT Bixby B. C. Martinelli l.D.Bcott - J.L.Belaoo V. M. anith Avon Lake General Chemical Plant --^.ATBsSy------- :------- :----------- I*.' K. Hnalright, Jr. % A, Ivkanec 0. J. McDonald *R. V. McKay *P. P. Reber *R. M. Sandfry J. M. Whitney Thosa Attending Akron P.T.Whitmire - R.S.Reynolds Calvert City C.L.Woods - S.B.Farbstein Henry C.B.Cooper - J.P.Piers *N. L. Kuchenoeister J. J. Palkovic Long Beach A.R.Webber - W.D.Robb R. K. Miller Louisville t.O.Crunkleton - 8.3.Michels H. I. Barrett G. D. Boldrick M. J. Carmack A, J. Oeborae V. Tesowitch Niagara Falla J.R.Linak"- J.A.Kluper \ Breckaville "I. W._Soyr M. R. Frederick > WC teak Force Technical Meeting T5TT357 These ninnies are divided into two sections. The first is a series of brief stetemente of stoAies, a^erinants or changes suggested by/for one or sore of the plants. The saoood is nore detailed report of the nsetiag acUon. A. Brdrofull I 1. Make 60,000 lba. of 103BP-PTS In a hydrofull aystam to evaluate roi of flahayeo neaaured in bloua files. .2 put Vo. 6 poly on hjdxofull all the tins - as soon as possible, reaming more polys hydrofoil to get e faster evaluation of this approach in our 3300-gallon polys. 20618001 BFGQ4241 o >- 2 3. Louisville. On the hydrofull polys, install third blade and screw coupling, sterna valves on Jacket, batch controller with alarm, filter in injection water line, and a recycle system in the initial fill piping. Most of this work was planned originally on the twelve hydro full polys, but it was deferred. Louisville. Go to hot water charging in the hydrofull polys to speed up production. 5. Avon Lake. On the four hydrofull polys, bring than up to the operating capabilities of our other sites. Proceed toward automated operation as rapidly as possible. 6. All Plants. Accumulate data on interior wall and dome conditions depending upon: whether polys are charged normal or hydrofull, the VCI/H2O ratio being used, the types of agitator configuration and speed employed. 7. All Plants. Accumulate accurate, in-the-bag yield data in hydrofull operations. How does it compare with non-hydrofull operationsf This is important information for evaluating the value of the hydrofull approach. .8 All Plants. Determine the meet effective route to increased productivity. What balance vlll you make between faster reaction times and increased loadlngf A common unit should be used - will this be lbs ./gal./dayT 9- Avon Lake, Henry, Louisville. By January 15, 1968, send sm your hjrdro- full procedures, descriptions of your polymerisation, shortstop and recovery equipment, end Instruments. Know and report an the reliability of your operations. What are your bottleneckst We shall set a meeting date for a discussion on what methods are bast and should be used in the future. At this meeting we 1 set time and achievement goals. 10. Louisville. Measure the timinratiira in the dome of a hydrofull poly, ftoes it differ from regions where the poly is Jacketed! 11. All Plants. Broaden the porosity, conversion, bulk density, powder mix time"and fusion time relationships far more specific resins. Establish any porosity-conversion shifts due to hydrofull operations. B. Agitation 1. SSSSL* OpanXu at slower agitator speed. Does this improve fieheye characteristics! 2. Louisville. Bvalnete the difference in poly cleenlinees and particle site vfaea'a third blade is sddsd to the hydrofoil polys. Z008T90Z -aty'v.; -1 r i 4fe:;.sk-jafesa^S'2riL BFG04242 3. All Plants. Consider how extensive a study should be made on agitatloo characteristics (speed, nuaber and types of blades and baffles), and emulsifier level with hydrofull and non-hydrofull systems. C. Conversion Control 1. Long Beach. A T instrument will be installed in the jacket of a poly with a constant recirculation rate. The system is expected to serve as a modified BTU route to measure conversion. 2. Long Beach. Begin an extended study of early vinyl chloride removal for blowdown at a known conversion point. Determining vhat quality resin results will be an important objective. D. Shortstop 1. Henry. Run an M.A. to produce 30,000 lbs. of resin shortstopped with TCA (thlo carbanilide). 1 saluate product at Louisville and Development Center. Compare with like product shortstopped with BPA and vlth no shortstop. Bow effective is TCA as a shortstopT What effect does TCA have an resin quality, especially on heat stability, and in compounding! 2. Development Center. Vhat effect does addition of caustic to the acetone solvent have on WA solubility in larger scale operation? Does the caustln altar resin properties? R. P. Reinhart is responsible for following this. 3. Develnyent Center. R. P. Reinhart to work with Lee Traynor in evaluating more shortstops. One prime candidate is Dillydap (dilauryl thlo dipropionate) and BRA. h. Louisville. Design a shortstop system which will accomodate TCA as well as BRA. E. Oacygen Content 19 67 8 1. Avon Inks. Saaple and determine the heat stablX^F or the suae resin(s) aada in tbit A sod Bldg. fc6l. Is there any difference? 2. Avon Lake. Calculate the oxygen level present in the vapor space in polys la bolt A and Bldg. U6l. 3. LoulevUle. Colonists tbs oxygen level present in the vapor space in polya la iLdg. 1 and 111. h. All Rants. Verify oxygen content of the vapor space in the polys by analysis, We hope the Greenbrier Oxygen Analyser will help us here. 20618003 BFG04243 L F. Programmed (007) Charges) 1. Development Center. Complete the evaluation of 007-5 by Ins-.ron C7K.' (stability) and Prodex extrusion. 2. Avon Lake. Run acre programmed charges. We may want to look in iepth at resin with a Lever inherent viscosity next. 3. Avon Lake. f s totter or mixed catalysts. U. Avon Lake, or H^ry, or Louisville. Seriously consider use of cat control with can speed correlated with conversion. G. Heat Transfer 1. Avon Lake. On Poly 113, get data on overall and heat transfer coefficients on. the inside and outside surfaces. 2. Avon Lake. Hold a meeting with L. K. Himelright, M. A. Ivkanec, P. P. Reber, V. A. Reed and me to discuss the X-poly condenser testing. Svmluate use of the condenser on X* *oly. Determine properties of resin produced. 3. All Plants. Cooling water, quantity and temperature, greatly restricts our ability to increase productivity. What steps do you plan to take to remove this bottleneck? H. Catalysts 1. W. A. Reed will look at mixed catalysts to determine how effective a mixture can b* developed. This will include scam of the new, fast catalysts. I. Vinyl Chloride Racortry 1. Ability to remove unreacted vinyl chloride rapidly and without tine daisy seams to be a problem at both Avon Lake (Bldg, U6l), and Louisville. Older certain conditions, removal is no problem. This is not true all the tisw. How can this bottleneck be reeuvedf d Attachmt. (l) H. H. Marty 20618004 jTl: BFG04244 x__fa. 1 SUMMARY OF MEETING I. CONVERSION CONTROL A. Henry. (N. L. Kuchenmeister) Henry made one formal and several informal runs evaluating the hydrofull technique. The formal run compared hydro full, vs. non-hydrofull at air free and normal levels of air. See the December 13, 1967 report by N. L. Kuchenmeister, Authorization to Manufacture, No. 3031*. In the summer, twenty charges of 103EP-F72 were made hydrofull. The general results checked the 102EP-F5 run. The conversion was 70% based on porosity. There was no Increase in cooling demand at the end of the reaction. The reaction time was one hour shorter than with non-hydrofull charges. The powder mix time was faster than for non-hydrofull resin. When the operating pressure was varied from l4o to 170 pslg, no change in reaction time was observed. A 60,000 lb. K.A. is written for 103EP-P72 to see if blown film fisheyes are reduced in a hydrofull system. Henry wants to look at 109EP-F2 made hydrofull to see if control is better. No difference In dame buildup was noted when operating hycLrofull. Poly 6 was super-cleaned before every charge so differences would have been seen if there. The agitation speed will be lowered on 103EP-FT2 to determine whether flsheye level is improved. Kuchenmeister felt strongly a study should be undertaken to quantify speed of agitation, emulsifier level, and resin properties. Be also has shown that changes in gelatin level affect poroaity and conversion. B. Louisville. (V, Tesowitch) Eleven polys were converted for hydrofull operation in Bldg. 111. By Decanter 12, 224 charges of 10ZEP-F1 were made and there were no losses by venting. The reaction time was reduced from 10 hours to 8.5*9 hours baaed on powder mix time. Properties of the dried resin were consistent and within spec. Poroaity was slightly low. Powdar mix time was less than target but within speo. The first compounding run was satisfactory. Conversion la to be determined by weight collected off the drier. The No. 6 poly was the only poly initially to have a screw coupling and tbrae Inmet 111 blades on tbs agitator shaft and an automatic steam valve in the control system. Lowering the speed from 36$ to 310 RFM reduced the drive loed from 8-9 to 5-$.$ amps. Loading was Increased in Poly #6 from 3600 to 4600 lbs. vinyl without requiring full flow of cooling water. Twenty percent more Elvanol was used in the recipe for this poly to gat the aame particle else. The drive load Increased from 5-5 to 20618005 7 7 ampa at the 5-6 hour period and then dropped off to ;W. A. Heed thought thia might occur with the agglomeration phenomena. A discussion of cleanliness of the poly in hydrofull operations brought out these points. With the two bladed polys, rupture discs were charged every 3 to 4 charges. More r*. ratings were taken from iydrofull polys (2-3 buckets va. 1^ buckets for non hydrofoils). This may be due to more thorough cleaning of the dame. A question to be resolved is whether there is a change in poly (wall and dame) cleanliness when the agitat ion change la made from 2 blades at 365 MM to 3 blades at 310-320 RPM. The tem perature at the top of No. 6 poly will be measured to see if it differs from the control point. Louisville believes gelatin in the emulsifier increases the powder mix time. Recovery time varies from 30 minutes to 3-1* hours. This points up a definite need for a shortstop so qusllty will not vary. W. Yesowitch noted the changes still to be made for the hydrofull polys: * Convert the initial fill water system to recycle so polys can be filled with less trouble. * Install automatic steam valves on all the polys. Get ball valves installed and go to hot water charging. Install screw couplings, a third blade and slow the speed to 310*320 RJtt on the rest of the polys. ym la the volume in the agitator nostie so buildup will not take place there. Install filters in the water injection lines so the DP cells won't plug from resin backing up the line. * Install batch controllers to signal when water injection into a poly is almost caplets. * Design a shortstop system. C. Lang Beach. (Rune report from R. K, Miller) On thm AT method, part of tbs instrumentation still is to be shipped. Sams work was done sines the last report with the early VC1 removal method. TWo polys are piped with a 13/64 inch orifice In the individual recovery lines. This should^restrict VC1 removal to `00-600 lbs./hr. A back pressure regulator in'the line Is set at TO psig. The normal recovery rats is 6-10,000 lbs ./hr. of VC1, so if several polys ware on early TCI removal, the removal rats in ths poly* shouldn't be hurt. So Tint arAmi 9008T90Z r far, 15-20 charges were made with this method. However, water purity problems developed and this raised questions about the resin properties. A new series is underway. To date, based on laboratory tests and the TOPS #37 stability test, we saw no differences. The early recovered sample looks no different than normal Long Beach production on blown film fisheyes. The point at which monomer may be removed without hurting properties (porosity, particle size) may be closer to Uot than 30J& as proposed earlier. This will be verified later. D. Avon Lake. (M. A. Ivkanec) There have been many problems with both process and product control. An early change was to separate and to Increase the size of the bypass water system for initially filling the poly. After this, it was shown that the demineralized water pressure fluctuated badly. This was layroved. Temperature control of the water was erratic. A selfcontained D.M. water system for the injection water is in the works. Porosity results were erratic. More careful attention to injection weter DP cell operation improved this. Addition of shortstop at the target time was again shown to be an important factor in making specification resin. When the recovery system couldn't accomodate immediate blowdown of a charge, the need far an effective shortstop wee highlighted. Avon Lake (Coctd.). (A. N. Pairlie) Before the concentrated efforts to Improve properties, the porosity range was from 0.17 to 0.380. After the push for improvement, the porosities ran from 0.270 to 0.31$. Considered against the expected range as a blended and dried resin (about 0.009 porosity units) this is an acceptable Improvement. There still was e difference between results from the two polys where one consistently produced resin with lower porosity. Control temperature and DP cell calibration are suspect. Avon Lake (Coctd.). (H. H. Marty) A series of changes are planned to the four hydrofull polys to improve operations, product uniformity and process controls. These fall into five categories . a series of agitation changes; an improved, selfcontained injection water system; ljproved, automated reaction control methods; increased coding capabilities; and an isgrovad recovery operation. Sea reports by I. H. Marty on this equipment, "Minutes FK task Poores Neetli*", Ootobar 20, 1967. Appendix 1, Utt-36-67. "State* leport Z on Poly 109-112 (Udg. `61)". HW-Ul-67. I. Overall Ve have three plants with hydrofoil systems at dlfl development end with different equipment sod diffej -20618007 u procedure#. Each plant will write out its charging procedure, -hurt-to;, blowdown and recovery procedure, describe its equipment (including , instrumentation) and define its reliability of production and product I control. This information is to be sent to me by January 15, 1968. Then we shall set up a meeting of the men from each plant who have worked with the hydrofull systems to try and adopt, as much as possible, tue best features of each plant. The test property relationships for resin made using the hydrofull methods should be determined. These properties are: porosity, bulk density, powder mix fusion time and flow time as related to con version. The relationships for the various resins should be found. A systematic determination of the yield and the conversion obtainei with the hydrofull method as compared to the aon-hydrofull method is required. This data la important, in making a deci-ion for, or against, the hydrofull approach. All plants should collect data on the amount and type of buildup on the walla, dame, baffles and blades. Hydrofull and non-hydrofull should be compared. The type of emulsifier' and the VCl/water ratio should be taken into account as well as surface 'material of construction and type of agitation. II. FROQBAIigD CHABGIS A. 007-5 Charges (attempt to repeat 007-1 *ade in September, 1966). Three chargee were made in October, 1967, to try to reproduce 007-1. The charges ran slower than 007-1. It was found that the catalyst contained 89* effective peroxide rather than 95-100*. The inherent viscosity of the third charge waa similar to that of 007-1. Ihe three chargee were blended as slurry end dried together in the small ALGCP Venturi drier. The properties of the individual charges of the blended, dried resin ere listed in Appendix 3 of the FVC Task Force minutes of the October 20th meeting dated Rovember 10, 1967 (HW-36-67). Results still to be obtained are the Ins cron CVHS and the bottle extru sion work. B. 007-6. (an attenpt to lower the inherent viscosity). He proposed to nako a progressed resin with an inherent viscosity of 0.fli*-0.68. Our first attenpt waa a runaway (though contained) because e second shot of catalyst wee added too early. The next attenpt was kept in control satisfactorily. However, the inherent viscosity was tha same as 007-1. (0.9lk). The tamparaturs-conversion profile snployed was 150*F. for UO* conversion, 130*F. for 10*, 120*F. for 10*, end 115* F. for 10*. Ia order to get lower viscosity resin, we propose to do two 20618008 ..JkWWtt.. 'vwe things: Add more catalyst and use a different temperature-conversion profile. This latter will be: 150*F. for bo* conversion, lUo F :'or 10*, 130*F. for 10*, and 120* F. for 10*. To make lower inherent viscosity resins, we should consider mixei or hotter catalysts. We shall depend upon recommendations from A. R. Eerer.s and W. A. Reed on how to proceed along these lines. Norb Kuchenmeister and Bill Yesowitch pointed out that their [ lart.- l,^.e cam controllers which can be used to make programmed charges . A.: rill suggested, if the speed of the cam motor can be regulated, it would ce possible to adjust for conversion. III. SHORTSTOP(a) A. Plant Test (M.A. #3115) - M. A. Ivkanec Two plant comparisons were run in the 1100-gallon hydrofull polys. There were sampling mixups in the first series ao it was necessary to repeat the series. The objective was to run two charges which would be shortstopped with BPA. Three charges were to be shortstopped with TCA (tfcio carbeollide) - one at a aolar concentration equal to the IPP concentration, one at one-half the IPP aolar concentration, and one at one-fourth the IPP aolar level. Bach of these five charges was to be sopled right after the shortstop was added and six hours after the addition of shortstop. During this fci<Hng period, water was to be added to keep the charge hydrofull and nsasnre progress of conversion. Finally, a sixth charge was to have dimethyl sulfoxide (EKS0) added at shortstop tine, saapled and blown down. was the TCA solvent in these experiments. In the BPA charges (0.04 parts BPA), sapling after shortstopping was not on schedule. One charge was scaled one-half hour late. The socalled shortstopped saaple showed 7b* conversion by porosity. The six hour saqtla had 8b* conversion (hy porosity). By the amount of water added during the six hours, the conversion went to US* - an obvious error. Par the second charge the shortstop saaple was taken one hour late. This one showed 75i* conversion. The six hour sople was at 82* conversion. Both the aolar and 1/2 aolar charges vent frost 73* conversion at short stop tins to T9* conversion after six hours of holding. The molar charge went fraa 72* conversion to 77j* conversion. In order to get fairly rapid solution in 2KSO, it was necessary to reduce TCA concentration fraa 15* to 7*. B. Stability Tests - 1. P. Reinhart Based on rigid tin canpounding, the BPA samples looked slightly better than the TCA ssaplss. These conclusions verified the Research results. 20618009 BFG04249 As A1 Berena pointed out, BPA is a good material for improving heat stability while TCA is a better shortstc,, when IPP Is used as the catalyst. C. Statistical Analysis of ALGCP Results - A. M. F&irlie Analysis of the data reported by Ivkanec and Reinhart was made by Fairlie and Haller (report Issued December 12, 1967 as HSH-5-67). The analysis confirmed that TCA was a better shortstop than BPA. Heat stability by all the different tests, showed very little difference between BPA and TCA. Andy strongly recommended that the heat stability cosiparison of Henry resin be run in duplicate. By this he means the samples should be individual from the weighing on. D. Manufacturing Authorisation to Evaluate TCA at Henry - H. L. Kuchenaeister and R. F. Reinhart An M.A. is approved to make 30,000 lbs. each of resin shortstopped with BPA and TCA. Production will be made in January. The charges will be run non-hydrofull. Acetone will be the TCA solvent rather than EMSO. TCA will be charged at 0.01 parts which la slightly higher than one-half molar (0.006). Objectives, again, are to show how effective TCA is as shortstop snd to find out if TCA affects product properties. 8plee of each type are to be sent to the Davelafment Center, along with oon-sbortstopped resin, to compare boat stability characteristics. ^lantitles of each resin and a non-shortstopped control are to be sent to George Boldrich at Louisville. He should compare the TCA-shortstopped resin to the other two resins In several compounds. For one thing, how does TCA-shortstopped material react in a cosg>ouod with 3-27 stabilizer? TCA shows a cost seringa over BPA. At the one-half molar concentration (of IPP), Ita coat is ons-fourth that of BPA. E. Wort at lesearch - A. R. Berens Addition of caustic solution to scstone (up to 20 ot the acetone weight) increases the solubility of TCA. Kuchsomslster asked whether esetone would gradually buildup la VC1 over a period to time. Work by Lee Traynor has shown that dibutyl sulflds is an effective short stop. However, the sulfide laqparts a characteristic -dcr to the resin. Plllydsp (dUauryl thio dlproploeate) decopposes IPP but not as fast as dlbutyl sulflds. It has not been tried alone as a shortstop. However, it was cnOlnsil with BPA so charging of each was eqpal molar with IPP. As a shortatop the combination was as good as TCA. A tin stabilised recipe produced e product that was amre stable then that where TCA waa the shortstop. Dlllydsp is PDA approved. A satisfactory solvent for this material will have to be found. K. P. Reinhart will work with 20618010 7 Lee Trsynor to evaluate these and other shortstops - including one a lower molecular weight than Dlllydap. Louisville expressed a need for an effective shortstop. In order to get uniform properties, it will be necessary to stop each reaction at the same point. They will proceed with a design of a shortstop system. At present no difference in design can be seen between a system for B?A and TCA. IV. KERRY ACT-FREE PEARL EXPERIMENT A. Experimental Work - IT. L. Kuchenmeister A series of charges were made in April and May to look at hydrofull and non-hydrofull charges at normal (for Henry) and so-called "air-free" conditions. Six charges were made hydrofull. Three of these had normal air-level and three were air-free. The non-hydrofull charges were made alternately normal air-level and air-free - five of each. All charges were ran in Poly Ho. 6. See Status Report on M.A. 303^, December 13, 1967 by I. L. Kuchenmeister for tbs story on plant operations. Arena for further work are: l) Verify oxygen level using the Greenbrier Gbqrgaa Anslyzer. 2) Establish the porosity-conversion relationship; watch tor any shifts. 3) Convert more polys to hydrofull. B. Statistical Analysis of Run Data - A. M. fairlie The data is reported In detail in a report by Kuchenmeister and Tairlie, Hydrofoil and Air-Free Geon 103P-F5 Resin Sun at Henry". AMT-36-67. C. Other Rents Louisville has dona nothing special to lower oxygen content except use of the degasslfier on water for addition to the hydrofull polys. Hot watar shsrglng will improve the oxygen level In the poly end this will be paahed. BUI Tesowltch will calculate the amount of oxygen present la thf'joly using Knchsfslrter1 s matbod. At Base Ltkr, the D.M. water contains 1 ppm oxygen after tbe degasslfier. ty the time the weter reaches either Bldg. b6l or Unit A, the oxygen level, fta up to at least 7 ppm. Peal Baber win-see that the Kuchenmeister is made for ccqrgem level in both buildings. Oicn Schaaf aglets 1 that ths beat stability be determined when the same resin was made Be both Bldg. b6l sad Obit A. Bare the asms watar and vinyl is being charged so that variable is removed. The water system at Avon Lake ehoelg be studied to see vhat can be done to reduce resaturation of the wat1er with oxygen. % mM[jr- 20618011 V. HEAT TRANSFER A. Poly 113 (ALQCP) - H. H. Marty The work on this poly is aimed at determining which file (inside-slurry side - or jacket aide) controls heat transfer. In August and September We thought we would wrap this work up in a hurry. However, a couple of checks on our existing (so-called) BTU meter showed deficiencies. When 36-lnch manometer was installed on the existing orifice, the orifice measured 210 gpa flow while the magnetic flow meter read l4o gpm. The Foxboro H-69S flow conversion unit was sent back to the factory. Cn the temperature aide, we ordered calibrating resistors to check our resistance bulbs. The Foxboro temperature conversion unit (M-69U) did not respond properly. The "Span" did not work. This unit also was sent back to the factory. We plan to use an 11-inch chart to record A T rather than a U-inch unit so we can Improve analysis. At this point there was considerable discussion on productivity. Glen Schaaf again brouglout the two major routes to increased productivity. One route is increased charge size. The second is shortening the polymerisation time. Four to five hours reaction time is the point where stability starts to ba hurt - becausa of catalyst loading. This data is for 10SP-F5. Dick McKay pointed out that 103KP-F7 was made earlier in a six hour reaction time - and using caprylyl peroxide as catalyst. This is one indicator that the reaction time can be cut. One of the reasons faster cycles are difficult to obtain quickly is the quantity of cooling water attainable at the poly buildings. Avon Laka, Bldg. *6l, finds it difficult now to raise loading because the water pressure la so low. Louisville has low anough cooling water pressure that it has to be careful in Bldg. 111. They have discussed refrigerating part of their water to the cooling jacket in order to get a greater AT* Harold Long has promoted this idea for sene tins. B. Condenser Kvaluatlon - Long Beach (Phone report from B. K. Miller) Long Beach Obtained a special condenser which fit the top flange and had 22 aq. ft. cooling area. This added about 10-154 additional cooling surface. The condenser worked satisfactorily if no inerta were present. With Vase catalyst, the condenser had to be vented every half hour or continuously vented at 1-2 S.C.F.M. With lauryl peroxide catalyst, the condenser was vented two or three tines. When the condenser was blind ing with inerts, the (1 would drop froa 100 to 2 in about 20 minutes. Xn all cases, foan cans up into the tubes in the last hour. After every third chargs, tbs tubes were cleaned. Tbs buildup was due to the foam ing material. Id Martlnelll said that a $ ft. per second vapor velocity was the design rats to keep liquid or foan from getting Into a condenner. BFG04252 20618012 C. Develop--nt Center - W. A. Reed 9 In 15-gallon work, no significant product chsages were seen when a condenser was used in conjunction with the Jacket for cooling. See report by K. L. Christensen - C. A. Brunet, K1C-3-67, "Production of EP Resins With a Modified Cooling System", for statistical analysis of the data. D. Pechlncy-St. Gobain (Bulk PTC) - E, C. Martinelli Ed vent over the work P9G did on heat transfer. He shoved the plots of Lftt.per at conversion ead conversion per time vs. time. Through the installation of a condenser on the autoclave enough additional heat vas reserved that it vas possible to reduce the polymerization time froa 10 to $ hours. I. I-Poly Work Proposal - . I. Arne d On X-Paly in Bldg. b51 xt ALQCP, ve have a vertical condenser. Vince had planned to carry out n series of experiments looking at the practicality of the use of a condsnsar and the quality of resin produced therein. Since his transfer, this will he followed by the AL3CP Tech nical Department and the Task Ports. A meeting of P. P. Reber, M. A. Ivkanec, L. X. Hi--lright, V. A. Seed and H. H. Marty is to be arranged to discuss tbs lime of attack. PSQ says it hasn't been shin to use a condsnsar in a suspension system. The product contains too many fines to he acceptable. VI. CAXAtms - W. A. Reed Nazran aspects that we shell eventually use n different catalyst level or a mix of catalysts for each resin and/or operating tenperature. Another approach is to use a lower ta^iei still s and a chain transfar agent. He is investigating several hot catalysts. These include AC8P (acetyl cyclo hexane sulfo-yl peroxide) and Varcadox T-l6. H. H. Marty cl 20618013