Document 3QGQwYw2VNgDa7zdaaO4JBabE

366/ BFG TECHNICAL DOCUMENT The BFGoodrich Company Research and Development Division retention CODE 200-001 permanent record copy Corporate Research Quarterly Technical Summary Fourth Quarter 1980 DISTRIBUTION CORPORATE GROUPS R. j. Fawcett Corporate Information Center R&D Files Corporate Materials Evaluation C. F. Effler Corporate Planning M. K. Wolverton Corporate Research D. E. Ley E. D. Dickens, Jr. J. M. Marks Chemical Group List #1 EPG List tb Tire Group List #1 Corporate Technical Support C. H. Lufter/J. B. Pausch/ D. W. Askea/P. M. Zakriski Patent Department E. G. Fiorito J. R. Lindsay Technology Assessment S Planning C. E. Hilkes/K. C. Baranwal/ G. E. Stueber THE B.F. GOODRICH CO. JAM 27 1981 RESEARCH LIBRARY Information from this document may not be copied in whole or in part without the consent of T- 11 100^280 BFG07336 BFGoodrich nnd C^v^loom^nt Carter CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 TABLE OF CONTENTS Page MATERIALS RESEARCH.................................................................................................................................. 1 4003B - Direct Vinyl Chloride Monomer........................................................................ 1 4008F - Novel Low Combustibility Plastics.............................................................. 2 4013C - Guayule Rubber......................................................................................................... 4 4015D - Crystalline Polymers as Engineering Thermoplastics ................... 5 4016A - Low Melting Glasses as Smoke and Fire Retardants ........................ 8 4016H - Applications for Molybdates............................................................................. 9 4022G - Chemical Intermediates from VC1 and Polychloroethylenes ... 9 4025F - New Methods for Producing CPVCs ................................................................... 10 4025G - Reaction Zone Theory and Its Application in the Develop ment of Water Chlorination CPVC Processes................................................. 12 4027G - Organics and Plastics for Photovoltaic Systems ............................. 13 4028K - Polymers with Thermally Reversible Linkages......... 13 4028L - Bloom in Polymers...........................................................................................................14 4041B - Rubber-Plastics Composites ............................................................................ 14 PROCESSING RESEARCH...................................................................................... 4010B - Improved Rubber and Plastics Processing............................. 15 15 4018C - Alloys as the Basis for New CPVC Compounds ........................ ... 16 4045A - Process Innovation.............................................................. 16 PRODUCT PERFORMANCE RESEARCH............................................................................................................... 17 4002A - Diffusion in Polymers................................................................................................. 17 4002G - Vapor-Induced Relaxations in Glassy Polymers .................................. 17 4004A - Design of Composite Products..............................................................................18 4006G - Fiber Reinforced Plastics Composites .......................................... 19 4007G - Low Combustibility Technology.............................................................................. 20 4020B - High Performance Polymer Bonded FerriteMagnets..................................... 20 4029A - Optimizing the Performance of Rubber BlendProducts .... 21 4030A - Plasma Polymerization................................................................................................. 22 4032C - Conductive Composites................................................................................................. 22 4039A - Adhesion Science ..................................................................................................... 2 2 4040A - Highly Filled Materials and Composites .......................................... 23 4041C - Project LDF................................................................... 23 RESEARCH PLANNING ............................................................................. 24 4026A - Project Market Research.............................................................................................24 4026B - Market Research................................................................................................................24 ZO0LZQ0Z BFGoodrich Research end Oevelcoment Center BFG07337 CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 Through November Corporate Research expenditures were $2,771,585 which is 94.7% of the 11 month goal of $2,925,549. MATERIALS RESEARCH Direct Vinyl Chloride Monomer; (Project 4003B-80) R. T. Carroll -- The object of this program is a new vinyl chloride monomer process with reduced capital, conver sion, and utilities costs. Replacement of all or part of EDC cracking by direct processes can lead to such savings. Preliminary cost estimates by A. T. Schooley indicate savings may be sufficient to tolerate rather large recycle under favor able conditions and possibly minor sacrifice in yield. Our efforts to date concentrated on non-catalytic chlorination of ethylene em ploying low conversion and recycle to attain high selectivity. Ethylene and HC1 in the 1:2 ratio typical of oxychlorination feed was the recycle in the original concept. The assumption that HC1 would serve as an inert diluent in the chlor ination of ethylene proved false. It led to about 10% by-products, principally methyl chloroform and vinylidene chloride. This concept might have been attrac tive to BFG since these by-products sell at about twice the price of vinyl chloride. However, the required investment in additional columns to make neces sary separations cancelled savings in investment. (A. Schooley) This led us to examine the reaction of chlorine with ethylene without any dilu tion by HC1. We completed experimental designs to map the effect of temperature, chlorine concentration, and dilution by inert gas. Optimum selectivity for vinyl chloride requires a temperature in excess of 400 C. The combined selectivity of VC! + EDC is little influenced by ter^erature. Both VCl and the sum of VC1 + EDC are inversely related to chlorine/ethylene ratio which limits conversion of ethylene to 20-30%. However, combined VCl + EDC selectivity of about 95% (with about 90% VCl) were obtained. Simulated recycle of EDC produced 95% VCl based on converted ethylene + EDC. Much of our efforts during the early part of this quarter were in cooperation with Chemical Group technical personnel. We provided atmospheric pressure data for computer simulation and economic analysis of various process configurations. (A. Schooley and T. Kenat) We also worked in cooperation with John Lenczyk whose assignment was to investigate the effect of pressure. After early differences were resolvedwas able to confirm my results of about 95% VCl selectivity with simulatefLEDC recycle at atmospheric pressure. Increased pressure gave poor results due, I feel at least in part, to limitations associated with the 2-inch SS reactor he used. Our work to examine the chemistry and define possible potential of ethylene chlor ination directly to VCl is nearly complete. In most of our work we used sand as the fluidized reaction medium. This appears to have been a good choice, but questions remained about surface and/or catalytic effects. We have recently com pleted a short series of experiments in which the reactor was an open tube with minimal surface. There appears to be no advantages associated with this homo geneous system. Selectivity to VCl and EDC were lower than in fluidized sand. We are still concerned about some anomolous results which appear to be due to al tered surface or catalysts. Generally a temperature of greater than 400 C is 20627003 BFG07338 BFGoodrich ant Odntfjr CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -2- required to produce VCl as the principalQproduct. But we have observed instances of high VCl selectivity at as low as 350 C. Since lower temperatures might per mit higher selectivity, this merits some further investigation and is planned. Novel Low Combustibility Plastics; (Project 4008F-80) M. P. Dreyfuss -- Our ob jective is to study the products and chemistry of the addition chlorination of unsaturated polymers. The materials are to be evaluated for their apparently unique combustibility characteristics and for other key properties. We have prepared a series of addition chlorinated polybutadienes in which the extent of reaction was varied systematically from about 60% to 100%. This corresponds to chlorine contents of 44% to 56.7%. The partially chlorinated samples were found to be very sensitive to crosslinking under ambient conditions. The use of an antioxidant (CAO-1) during produce isolation was necessary and led to adequately protected products. The primary incentive for preparing this series was to establish the correlation between extent of reaction and smoke/ char values. Equally important, however, is the relationship between chlorine content and infrared spectrum, iodine number, Tg, density and thermal stability. This series should also give us useful insight into the conditions needed to obtain complete chlorination and the best analytical methods to use for follow ing the reaction. The smoke/char tests are currently being carried out by Bill Kroenke. Density has proved to offer considerable promise as a quick determina tion of chlorine level. The densities of our samples correlated linearly with percent chlorine over the above range. We are currently obtaining additional data points using older samples on our shelf as well as preparing other samples as needed. The density of a fully addition chlorinated polybutadiene appears to be about 1.456. This compares to a value of 1.406 determined for PVC. The feasibility of developing an infrared method for determining chlorine level in these polymers is currently being evaluated by the analytical group. Iodine numbers are also being determined in order to assess the validity of this proce dure for these chlorinated materials. Since the desirable properties of chlorinated polybutadiene appear to improve with chlorine content, it is of interest to determine if and how these proper ties change as chlorine content is raised above 56.7%. To do this, of course, requires post substitution chlorination. We are attempting to prepare these polymers by continuing to chlorinate a completed chlorine addition reaction, but now under illumination. In this way we hope to be able to achieve a reasonably random honvogenasup post chlorination. Such materials are expected to be more useful at thi^atage of our study than would a series of surface chlorinated powder particles. Our results to date have not been notably successful. In agreement with literature reports, it is rather difficult to chlorinate beyond 56.7% in methylene chloride solvent. Thus we are studying the use of chloroform and mixed methylene chloride/chloroform solvents. This work will receive in creased attention during the next quarter. We only carried out a few experiments in the area of interactive chlorination this past quarter. Addition chlorinations of polybutadiene were made in the presence of Butyl Carbitol (2-(2-butoxyethoxy)ethanol) and of e-caprolactone. Both materials appeared to be quite reactive and considerable amounts were in corporated into the products. The Tg's were 47 and 44 respectively. The chlorine content of the product made in the presence of the lactone indicated 0827004 BFG07339 BFQoodrich '""yrl P'avAlr'f*>mon* CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -3- that about 30% of the double bonds contain a lactone moiety, assuming no oligo merization of the lactone. Both polymers displayed limited solubility, but pressed films were prepared without difficulty. We hope to be able to devote more time to the study of interactive chlorination in the coming year. A research report covering our findings on and our current understanding of the addition chlorination of 1,4-polybutadiene was written this quarter. G. S. Huvard -- Solutions of CB221 are best chlorinated by using the heat of reaction to ramp the reaction temperature to reflux conditions. Concentrations as high as 2 wt.% CB221 in CH2CI2 have been chlorinated successfully. We find that injection of CH2CI2 liquid into the CI2 gas feed rather than bubble tower saturation of the feed Cl2 works best to prevent gel formation. We have noted a rather strange viscosity behavior during chlorinations in the lab. As chlorine addition proceeds, the viscosity of the solution increases to a maximum and then drops abruptly at about the endpoint of the chlorination. As yet, we cannot explain this behavior - we don't even have a good hypothesis 1 Yet the phenomenon is very reproducible and doesn't seem to have an adverse effect on the final polymer. The final cement is about 2.3 x the original wt.% concentration so a 2 wt.% original CB solution results in a very clear, 4.6 wt.% cement that has good flow properties (not overly viscous and should be pumpable). This reaction is being scaled to a 30-gallon reactor system. We plan sampling of the reactions for intermediate chlorine level polymers. Modifications of the reactor to be used are in progress. We also chlorinated an SBR polymer for Bob Beauregard. Bob plans to test this material as a "coupling agent" for CPVC and impart modifiers (such as Kraton) which have poor compatibility with CPVC. The SBR was particularly easy to chlorinate. We encountered no gel problem at the 2 wt.% level (could likely go much higher) and obtained a very non-viscous product cement. This reaction will be run at the 30-gallon level should Bob find good results in his tests and re quest lure material. w. J. Kroenke -- We are trying to quantify the influence of chlorine concentra tion on smoke reduction and char formation. The samples were made by Dreyfuss under very similar experimental conditions. we verified that the two post (substitution)-chlorinated H-H PVCs prepared by Schlegel (~65%'C1) have superior flammability characteristics. When 5 phr of MoO3 was added to each of these low smoke polymers, the smoke in the Smoke-Char Test was completely eliminated. The addition of the MoO3 did not change the very high char yield of >70% of the theoretical value. We need to find out how little post-chlorination is required to give superior flammability characteristics - with and without a smoke retarder additive. We also need to find out if lightly post-chlorinated H-H PVC has better flammability characteristics than the intermediate chlorine (post-chlorinated) PVC resins studied by Dickens. 20827005 BfG07340 BFGoodrich Parlor CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -4- t i Guayule Rubber; (Project 4013C-80) E. C. Gregg, Jr. -- A report on high-yielding guayule issued. In an expedition to Southwest Texas for wild guayule we noted considerable desert-plant growth as a result of recent rains. All the guayule had recovered remarkably from a severe drought of high temperature and lack of rain for 12 months. Both extensive flowering and seed-setting occurred on all the native guayule. Surprisingly, the guayule seedlings also survived the drought very well. Conversation with ranch managers local to the wild guayule revealed that Good year has harvested about 3000 pounds of wild guayule, presumably for operation of their pilot plant for extraction of rubber from guayule. The pilot plant is in Akron. Subsequently, a photograph of uprooted wild guayule appeared in an article about Goodyear's guayule activity in the New York Times, 4 November 1930. The photograph depicted a laboratory at Goodyear Research. We collected guayule samples from the wild and from an experimental plot of three-year-old cultivated guayule at the Texas ASM Experiment Station at Pecos. This collection of cultivated guayule was the systematic collection we alluded to earlier as a prospect. This collection is in response to the observed wide variation in rubber content from an earlier sampling of this cultivated guayule. Analysis of the rubber content of the wild guayule uncovered two bushes with 18.2% rubber. Analysis of the cultivated guayule showed one bush with 15% rubber. This bush is three years old from seeding. Only one guayule bush with 20-weight-percent rubber was transplanted from the wild. Severe rains nearly isolated us which cut short our work. Subsequently, the arroyos filled with water and prevented our return to obtain the remaining two bushes with 20-weight-percent-rubber. A guayule transplanter under development by the Bud Antle Co. in operation at Tempe, AZ planted 16,000 guayule seedlings per hour automatically. This is half of its capacity. Doubling the full capacity would be fairly easy and would match the field area covered by other crop machinery. The prospective commercial model could automatically transplant over 500,000 guayule seedlings per day, i.e. 40 acres per day. 1 3C-nmr andPMR examinations of a solid compound, C30H43O3, from guayule resin showed nearly 60 carbon atoms and more than 48 protons in disagreement with mass spectral data.'" This indicates that two compounds, CsoHujOs, were present. They were isolated'by gpc arranged for maximum resolution. In solution their molecular sizes were different by ca. 1/30 of a carbon atom. One of this pair of compounds has the following structural assignment: oto C3 to o C3 BFG07341 BFGoodrich CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -5- This is a derivative of a new parent ring hydrocarbon. We have named the ring parent, guayulane. The chemical name is 9,25-Cyclo,A-homo,7-keto,4a-oxaguayulan2-ol. A second solid compound, CjoHueOu, similarly shows two isomers. These isomers have been isolated and resubmitted for spectral examination. Crystalline Polymers as Engineering Thermoplastics; (Project 4015D-80) S. E. Home, Jr. -- In the last quarterly report, I reported that phase transfer catalysis could be used to effect condensation reactions. I have been exploring the scope of this reaction to sulphide polymers. x(CHz)nX + Na2S -HCH^S-fy or O" (ring compound) At 50C, using aqueous Na2S and either Carbowax 350 or benzyltriethylammonium chloride as catalyst, the following results were obtained: CH2X2 ______ X-Cl, 3r, or I CH2 /' \ ss cI h2 N s ' chi ring, crystalline CICH2CH2CI---------------------- 4CH2-CH2-S>n polymer, plastic CICH2CH2CH2CI BrCH2CH2CH2CH2Br H2C _ CHa i! ch2 - s ring CH2_ ch2 !! ch2 ch2 ring Br (CH2) sBr CH2 CH, CHZ ring CH2 CH2 S BFG07342 BFGoodrich Br (CH2) sBr Br(CH2)gBr CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -6- Ring + Polymer No Reaction BrCH2-/;)>-CH2Br Polymer ,C1 ch3ch2ch, \ Cl No Reaction / Cl / CH3-CH Cl No Reaction CH2-CH-CH2-------------------------------- No Reaction II Cl Cl I should note that the liquid ring compounds are very foul smelling. The reac tion is, however, an excellent method for preparing these compounds. There does seem to be a temperature effect. Below ~ 40C, no reaction occurs with C1CH2CH2C1, and at 70 C the reaction is very rapid. Holding the re action mixture at 70 overnight produces a very foul smelling mixture so there are secondary reactions to be considered. Perhaps the dihalides that showed no reaction at 50C will show a reaction at a higher temperature - this will be investigated. Copolymers can also be prepared from mixtures of CH2C12 and C1CH2CH2C1. These show a gradual change from the crystalline trimer to a gummy product back to the plastic polymer from CICH2CH2C1. Polymer characterization is made difficult by the insolubility of the poly mers in all ordinary Solvents. I am limited to characterizations that can be made on the solid materials. DSC results on a -S4CH2-CH2-S')jt polymer, 4015D-80-156B, shows some interesting results: First heat; 0C + 85C, no transitions Q Second heat; 0 start, Big endotherm @ 120 C another at 160 C Third heat; 0, endotherm @ 105 C, big endo at 160 C These indicate a crystalline phase change. However, the Ti/T2Qtest showed entirely different results, Ti below 0C, with T2 of about 100 C; these results were from two different polymers including the one used in the DSC test. BFG07343 BFGoodrich 2082700Q 1 CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -7- -<^>-012-3^ The polymer -S4CH2 (4015D-80-142B) shows on DSC: Second heat; start @ 0 - endotherms @ 120C and 155C with decomposi tion starting @ ~170 C. The Ti/T2test shows a T] of about 45C and a T2 of about 120C. The action of aqueous Na2S on THF solutions of various chlorine containing poly mers is also being studied. I feel the liable Cl's can be made to react with Na2S to give either ring containing polymers or crosslinked ones: Cl Cl 11 Cl Cl s or Cl Cl 1 11 ss 1I 1 Cl Cl Polymer PVC-2 grades CPVC-BFG CPVC-Nippon Carbide CPVC-Parker-lightly chlorinated " > 64% Cl " ," 70.5% Cl Geon 22S& 27% VCI2 Chlorinated CB, 54% Cl Halodenes - various levels of reaction Results No color change, soluble Dark brown, gelled rapidly N If If If Dark red-brown, gelled II If It II M It Dark purple, gelled rapidly No color change, gelled " "" , no gelling. Five runs were made with PVC, using two different phase transfer catalysts and levels of NaaS to 5 parts/100 PVC. The DSC results showed NO change in Tg (83-85C). The TGA results, in N2, likewise showed no differences in thermal stability. The dark brown and purple colors indicate unsaturation, and the gellation indicates crosslinking and/or unsaturation also. However, IR results on the dark polymers do not show indication of unsaturation. IR is not able to BFGoodrich r'. BFG07344 CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -8- detect the C-S-C linkage since it is a very weak IR band. Chlorine and sulfur analysis values are not available as yet. Exactly what the reaction is between the polymers and Na2S is now known. The re sults do clearly show the <^1 structure to be the reactive site. The stability C4C-C* dl of the Cl^ ^C1 grouping in the Halodenes, is in keeping with previous obser vations . Disclosure sheets have been written on the phase transfer condensation reactions. Low Melting Glasses as Smoke and Fire Retardants: (Project 4016A-80) R. E. Myers -- Last quarter I emphasized the role of char formation as it relates to flame/smoke retardance of organic polymers. I indicated that our study of char promoting, intumescent systems has developed from our previous work on low melt ing glasses. The goal of our work is to develop effective inorganic-based flame/smoke retardants for a variety of organic polymers. Our recent discovery of the ammonium pentaborate (APB) system is significant because it illustrates the potential flame/smoke retardant value of a simple, low cost, inorganic compound. In polyurethanes, such as Estane, we have shown that APB has a dramatic effect in reducing overall flame spread. A 1/8 inch thick sample of Estane containing APB will withstand at least 10 minutes exposure at 1000 C without sustaining backside damage to the polymer. A research report describing our preliminary work on APB will issue next quarter. This quarter we have investigated the performance of APB in phosphate plasti cized PVC. We have found that APB by itself does not have an appreciable effect in such formulations. However, during this study we did discover an effective flame/smoke retardant for phosphate plasticized PVC. By using a melamine molybdate/ammonium pentaborate system at 25 phr (total additive concentration) in a PVC recipe containing 50 phr of plasticizer, we observed increases in backbone char of up to 400% with a concurrent smoke reduction of 50%. The systems process well and prodace a flexible sheet. By replacing a portion of the melmo with the less expensive APB, we have extended the melmo to produce a more cost-effective material. Studies are in progress to determine the performance of melmo/APB in the more conventional DOP-plasticized PVC. This quarter we have investigated a novel phenolic resin-based system which, on the basis of preliminary data, serves to promote char formation and reduce smoke in phosphate plasticized PVC. The phenolic resin is a non-toxic material de rived from a natural, i.e., renewable resource. The resin shows an effect at the 10 phr level in a PVC formulation containing 50 phr of plasticizer. I became interested in this phenolic resin, because I felt that it had the potential to generate an in-situ chemistry in the presence of a phosphate plasticizer. During combustion the plasticizer and the phenolic resin may be involved in a trans esterification reaction, the net result of which is to provide a crosslinked phosphate plasticizer matrix. The physical manifestation of this process appears BFG07345 BFGoodrich -- -1 f'-.-al---*~ 20827010 t*or CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -9- in the increased char yields which are experimentally observed. Model experi ments (suggested by Paul Nicholas) are in progress to test the proposed mechan ism. We have also found that by combining the phenolic resin with our melmo/APB system, it is possible to attain smoke reductions of 55% with concurrent in creases in char of -500%. An invention record has been filed on this work. Future work will involve testing of the phenolic resin in DOP-plasticized PVC. OOP-plasticized PVC represents a difficult system with respect to flame/smoke retardation. Smoke is generated not only by the PVC component but also by the DOP. Previous BFG work on plasticized PVC has shown that smoke retardants act only on the PVC component. In analyzing our low melting glass/DOP plasticized PVC systems, I have now found what may be the first example of a smoke retardant which acts on both the PVC and the OOP. The implications of such a finding are significant in that we may be close to identifying a highly effective smoke re tardant for plasticized PVC. Future experiments will determine the reproduci bility of this result. Applications for Molybdates: (Project 4016H-80) W. J. Kroenke -- Our initial attempts to make Kc melamine molybdate more cost-effective by supporting it have been unsuccessful. Although we have been able to form a molybdenum com pound on the surface of mineral supports, it does not appear to be 1^. And, relative to Kc, the supported materials are poor smoke retarders for PVC. They also do not interact synergistically with copper oxalate. In order to see if our original idea has merit, we are switching to soluble amine molybdates. These molybdates can be directly applied to the surface of a support. They do not have to be prepared in place. Bob Lattimer and I have started to study the character of the chars formed from PVC containing SS-50 and related copper and molybdenum smoke retarders. We hope to gain additional information about the role of the smoke retarders. Our initial emphasis has been to prepare chars under combustion conditions and identify the fate of the metal additives in the chars, we have compared the weakly synergistic M0O3-CU2O system with the strongly synergistic Kc-CuOX (copper oxalate) system (SS-50). During combustion, MoOj decomposes differently from Kc, and CU2O differently from CuOX. However, in both cases the chars from the synergistic mixtures appear as simple admixtures of their respective component chars. That is, no new com pounds appear in the chars from the samples containing the synergistic mixtures, which are not present in the chars from the samples containing the individual components. Additional work is in progress to understand these results and re late them to understanding the nature of the synergism in SS-50. 20827011 Chemical Intermediates from VC1 and Polychloroethylenes; (Project 4022G) P. P. Nicholas -- This quarter, I completed two literature surveys. One covered the subject of this project, i.e., organo--transition metal catalysis that might j apply to the conversion of vcl and related chloroethylenes into useful chemical intermediates. The second dealt with organo-transition metal catalysis for oxy| genating olefins to epoxides with atmospheric oxygen-chemistry that mimics cer tain monooxygenase enzymes. I have recommended the former as my major 1981 pro gram. The main reasons for this choice include: (1) BFG's current and future bfg-8?93-a BFGoodrich Research and Development Center BFG07346 jd CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -10- leadership as a large volume, low cost VC1 producer and the importance of this business to our future, (2) Our ability to use by-product HC1, (3) the potential for wider application of successful technology in upgrading certain chlorinated by-products. In general, I believe that BFG is in the best position to exploit practical developments in this area. Moreover, the favorable technical promise for this study suggested by certain reports in the scientific and patent litera ture is still another plus. My program will begin with a study of amidation, i.e. the conversion of VC1 to acrylamides by substitution of chlorine with an amido group . This (-C-NR,R2) group will originate from carbon monoxide and an amine during the reaction. The raw materials cost for such a process appears favorable based on the current selling price of acrylamide, itself. The catalyst study will begin with bis(triphenlyphosphine )palladium II dichlorides of the general structure Cl2Pd||p (C6Hi,x) 3] 2 . Xt should be possible to assess the response of this reac tion to both electronic and stereochemical effects at Pd(II), depending on the nature and position of the substituent, X. This quarter, I began the synthesis of these complexes. New Methods for Producing CPVCs: (Project 4025F-80) Part 1 -- R. G. Parker -- Last quarter we produced high chlorine, high T,, CPVCs for study by B. L. Lee (see Part 2) and M. H. Lehr (see Part 3). During the Fourth Quarter, work was begun on the installation of the Hastelloy pressure reactor for use in the High Liquid Chlorine Process studies (with the Chemical Group). This reactor will be capable of producing pound quantities of highly chlorinated resins. We have begun to understand the lJC nrar spectra of CPVC much better, although the problem is far from solved (with R. A. Komoroski, Corporate Technical Sup port) . By studying a lightly chlorinated CPVC chlorinated in solution, we know that the first detectable structural sequence formed in CPVC is -CHC1-CHC1-CHC1-. Further, this sequence also appears to retain the tacticity information present in the starting PVC sequence. To substantiate these conclusions, we are cur rently studying the nmr spectra of two vinyl chloride-1,2-dichloroethylene copolymers which should contain the same sequences (with R. H. Backderf, Chemi cal Group). The first sequence which appears to contain a CC12 carbon is -CH2-CC12-CH2-. These nmr results have convinced us that it is no longer sufficient to describe the structure of CPVCs in terms of a polymer produced from the monomer units -CH2-CHCi-,-ClCH-CHCt,-CHj-CClr, and-CICH-CCIr. Rather, the sequence distribution must be described in terms of five (or seven) carbon sequences (including stereo chemistry) . For a "normal" CPVC, there are about 77 probable sequences. A re port was issued describing a new nomenclature system for use in describing the structural sequences present in CPVC (and other chlorinated polymers). Work continues on developing the syntheses of isotopically labelled vinyl chloride from 12C-carbon dioxide and 13C-acetaldehyde. The PVCs which will be obtained will be used in our chlorination studies, and in PVC pyrolysis studies (W. J. Kroenke, Project 4016H-80). 20827012 BFG-8793-8 BFGoodrich Research and Development Center BFG07347 CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -11- Part 2 -- B. L. Lee -- Most polymer processing operations involve moving and shaping a polymeric material in the molten state, and then cooling it "rapidly" to the solid state. Basically, this process involves material response under conditions of simultaneous deformation and temperature change. I have recently found that under appropriate conditions, significant changes in the mechanical properties of blends of CPVC/PVC can be achieved by rapidly quenching stressed material from the rubbery state. Specifically, the experiments were performed using the Instron Capillary Rheometer. The blends (75/25 with 3 phr of T-31 stabilizer) were extruded through the die slit at 205C. The extrudates were either subjected to on-line stretching, or no stretching. The Instron Tensile Test results indicated that the toughness and yield stress of the stretched material is drastically improved. These results suggest that it is possible to improve the toughness of CPVC using PVC, and no rubber modifier. Part 3 -- M. H. Lehr -- we reported earlier that lightly chlorinated CPVCs (LCPVC), prepared in liquid chlorine, showed markedly reduced recrystallization in fused samples containing 60-61% chlorine. We discovered later that resins prepared by the water slurry process also showed the same characteristic. We speculated that reduced crystallinity in LCPVCs might make possible the blowing of low-density vinyl foam, and the development of easier processing con^reunds. Laboratory rheological tests confirmed the expected benefits of reduced crystal linity in LCPVC. Tests were made on PVC (Geon 103EP) and LCPVCs (59-62% Cl), The LCPVCs were made from the same Geon 103EP by both thermal and photochemical initiation. The resins were made into compounds containing 2 phr T-31 stabilizer and 0.5 phr PE lubricant only so that the inherent resin properties could be measured. Instron rheometer results showed the apparent viscosity of PVC was more pressure sensitive than that of LCPVC. The pressure required to extrude PVC increased curvilinearly with die L/D ratio at 300 sec-^ at 170c and 200C. At an L/D of 40, the pressure required to extrude PVC was increasing steeply, and was signi ficantly greater than that needed to extrude the LCPVCs. In contrast to PVC, the LCPVCs showed a linear relationship between pressure and L/D ratio. PVC and LCPVC do not exhibit the same viscosity behavior. Apparent viscosity versus shear rate curves (Instron rheometer), and complex viscosity versus frequency curves, were steeper for LCPVC at 200C. This shows that LCPVC was becoming more fluid than pvc at the high shear rates. This result suggests that LCPVC would IXpgp more easily than PVC during injection molding. PVC and LCPVC Showed about the same die swell (>150 sec-1) at 170C where parti cle flow would be expected. However, at 200C, where melt flow occurs, PVC showed greater die swell than LCPVC. This work has been reported to the Chemical Group at a Project Review and an Ex ploratory New Products meeting. Effort is now underway to prepare large samples of the LCPVCs at the ALTC pilot plant. These resins will be tested by the Chem ical Group in foam, extrusion, and injection molding applications. 20827013 BFG-6783-8 BFGoodrich Research and Development Center BFG07348 CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -12- Reaction Zone Theory and Its Application in the Development of Water Chlorina tion CPVC Processes: (Project 4025G-80) G. S. Huvard -- During the fourth quarter, I've characterized rates of HC1 vapor diffusion in PVC, supported original CI2 vapor diffusion rate estimates, demonstrated that pore diffusion effects can alter rates of PVC photochlorination and shown that morphology of suspension and mass resins can have an enormous effect on chlorination rates and resin properties. We have also shown that CI2 uptake rates and HC1 loss by diffusion are much slower in H2O slurry systems than for simple vapor phase ex periments. Both of these observations require in-depth study as they affect rates of chlorination, resin properties and neutralization rates. To be more precise, HC1 diffusion coefficients from 30C to 60C are given by; sec 0.68 exp( -12200 RT ) The diffusivity at 30C is in agreement with Berens' correlation for numerous penetrants in PVC at 30C while the activation energy, 12.2 kcal/gmol, is in agreement with data given by Tihkomirov for activation energies for diffusion in PVC versus penetrant diameter. Chlorine diffusion times in vapor phase experiments sure in agreement with expec tations based on Berens' and Tikhomirov's correlations. These rates are much faster than times observed by Parker for CI2 uptake in PVC slurried in H2O. We believe this is a manifestation of pore diffusion limitations and not bulk phase to PVC surface mass transfer resistance. A dramatic effect of PVC particle morphology has been demonstrated with the help of Bob Schlegel (Chemical Group CPVC). We compared chlorination rates of standard 103EP suspension resin with reprecipitated 103EP having a drastically different morphology (much higher specific surface area). Reaction rates for these two were comparable yet the reprecipitated resin had better apparent heat stability. We repeated these experiments using 110x236, a resin which has very poor water chlorination rates and gives very poor CPVC. However, when reprecipitated to a high surface area morphology, 110x236 has rates faster than 103EP and apparently comparable heat stability. These two experiments dramatically demonstrate that evidence exists to support both intrinsic rate control as well as diffusion rate limitations to chlorination rates. Future work will be centered on separating and understanding the diffusion and intrinsic rates that occur during photochlorination and the interaction of these factors with Reaction Zone Theory. We are beginning an in-depth study of mor phology effect* on CI2 diffusion rates in PVC and CPVC water slurries. The Chemical Group has approved funding for a microplant system utilizing a con trolled photoillumination reaction zone. This system is expected to be on-line by middle or late summer of 1981. A research report covering the theory of photochlorination, the Reaction Zone Theory, diffusion related theory and initial Reaction Zone experiment has been completed and will be issued very shortly. I would urge all parties involved with PVC chlorination to read this report, formulate new and/or criticize old theory and report your ideas at Chlorpig meetings. frT 02,280; 3FG4783-S BFGoodrich Research and Development Center BFG07349 CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -13- Organics and Plastics for Photovoltaic Systems; (Project 4027G-80) L. Traynor -- We have continued to work on promising avenues in solar cell fabrication. I believe we can reproduce cells giving about 0.1 volt and 2 yamp/cm2 photo-out puts. These are cells made using a) electron donor additive, b) heat treatment and c) dichloroethane/2-methoxyethanol spin casting solvent. Approaches a) and b) were suggested by our work using narrow band pass filters. We need to opti mize electron donor structures and try even higher temperature heat treatments. An important discovery this quarter is the mixed solvent system. A continuing problem in this research has been the poor quality of the spin cast films. Using cellulose acetate polymer binder with the mixed solvent, we can obtain very good pinhole free films. There appears to be dye aggregation although macroscopically the films are uniform. I feel we have positive directions in which to proceed although progress is slow. We have continued to investigate polypyrrole as an inherently conducting poly mer. We have synthesized several N-substituted pyrrole derivatives. Only with the N-butyl derivative were we able to obtain a film suitable for conductivity measurement. Conductivity of this film was 4 orders of magnitude lower than polypyrrole itself. We have been able to make films using a comonomer mixture of pyrrole plus N-substituted pyrrole. Conductivity of these films is a little lower than polypyrrole. We have two of these films submitted for reflectance infra-red and Raman spectra to look for evidence of comonomer incorporation into the polymer film. The polypyrrole films are made by oxidative electrodeposition of pyrrole. We can easily measure the voltage applied to the electrolysis cell. The table shows the effect of applied voltage on the film conductivity. Clearly (and we have seen this in other series), the higher applied voltages give higher con ductivities. Attempts to measure actual electrode voltage are in progress. We Film No. 869-1 869-2 870 Applied Voltage 10 volt 5 2.5 Film Conductivity 25 ohm"lcm~l 18 3 will continue to explore variations in electrolysis conditions as well as other pyrrole derivatives and related compounds. Polymers with Thermally Reversible Linkages: (Project 4028K-80) F. L. Ramp and L. F. Stiberth -- This project remains in a highly exploratory phase. Thermally reversible bonds in the main chain and optionally in crosslinks should give a polymer with metal-like melt behavior, i.e. from rigid to very fluid. Two types of bonding linkages are under investigation: (1) the Diels-Alder reaction and (2) compounds which break reversibly to stable radicals. Diels-Alder ap proach: Furan compounds offer a wide variety of materials for the diene com ponent. However, a literature study of furan in the Diels-Alder reaction is not encouraging. This approach has been set aside at this time. Anthracene offers a stable but reactive diene system. However, studies to date have not un covered a satisfactory method of incorporating two or more anthracene structures in a compound. The latest approach, copolymerizing anthracene as a diene under free radical conditions, did not support earlier literature reports. This approach is still being studied. Bond cleaving to stable radicals: Compounds 20827015 3FG-9733.B BFGoodrich Research and Development Center BFG07350 CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -14- which have a carbon with electron donating and electron with drawing groups on the same carbon give stable radicals. Aminonitriles such as those derived from the Strecher synthesis possess the desired structure but lack hydrolytic stabil ity. 2,3-Diphenyl succinonitrile has been prepared with the object of closing a ring putting amino groups on the 2,3 position. The compound has proven to be very inert to attack at these benzylic positions. However, tetraphenyl succino nitrile breaks to oxygen stable radicals readily. The 2,3 diphenyl will be checked for radical production. A paramethoxy group should increase radical stability and can be easily incorporated. Bloom in Polymers; (Project 4028L-80) F. L. Ramp and L. F. Stiberth -- Work on this project by the authors has previously been reported under Project 4028K. The results to date are covered in detail in a status report. Briefly we have shown that sulfur bloom is due to supersaturation. DSC can be used to deter mine whether the rubber compound is taken from the last mastication above the critical bloom temperature. Stocks allowed to cool from above the critical bloom temperature, without mixing, will supersaturate and bloom. Bloom of zinc stearate has been shown to be caused by different factors. Zinc stearate bloom forms in two distinct manners with different driving force for the bloom. These two phenomena are conveniently categorized as: (a) thermal bloom and (b) high humidity bloom. Thermal bloom develops very rapidly on stocks above the melting point at the last mastication. This bloom is readily controlled by insuring that the stocks are cooled below the "zinc stearate" melting point in the last mixing. High humidity zinc stearate bloom develops over a period of hours or days at 50 c and ~90% humidity. Data to date indicate that this bloom is due to the conversion of zinc stearate - form II (see the status report) to the more stable zinc stearate - form I. The main focus of current efforts is to confirm these observations and to develop suitable preventative measures. Rubber-Plastics Composites: (Project 4041B-80) G. S. Huvard -- EPG's Les Fox was brought up-to-date on the composites reported last quarter. The properties of PE/scrap rubber composites and expected costs make these materials possible candidates for roofing membrane materials. I encouraged Les to consider this approach in his up-coming studies in this area. Further work has been suspended pending EPG's response. If favorable, I offered Les help in formulating and testing membranes made from such composites. 20827015 BFOa783-a BFGoodrich Research and Development Center BFG07351 r i_lLJ CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -15- PROCESSING RESEARCH Improved Rubber and Plastics Processing:(Project 4010B-80) V. L. Folt -- Our ob jective remains the determination of the optimum molecular weight and branching distributions required in our commercial polymers to optimize processing rates and product properties while minimizing energy consumption in processing opera tions. In the pursuit of these ends we are studying the viscoelastic and rheo logical properties of blends of linear and star-branched polybutadienes. In the last quarterly report, we discussed the melt viscosity behavior of blends of various branched polymers with a specific linear polymer. Although the in trinsic viscosity values for the blends increase linearly with an increase in the weight concentration of the branched component in the blend, the melt vis cosity increases in a very complex manner with increasing branch concentrations. And, the melt viscosity is also observed to increase linearly with increasing intrinsic viscosity but, on a log-log basis. There may be a critical branch concentration, depending on the branch length, at which the slope of the line changes abruptly. In all previously reported blend systems, the intrinsic viscosity of the branched component was higher than that of the linear component. We have now completed a study of a blend system in which the intrinsic viscosity of the branched polymer was lower than that of the linear component. In this case, the melt viscosity of the blend was observed to decrease linearly as a function of the intrinsic viscosity on a log-log basis, while the melt viscosity increased, again, in a complex manner, as the concentration of the branched component in creased. We also studied the behavior of a blend system in which the intrinsic viscosity of the branched component was virtually identical to that of the linear com ponent. Here we observe that the melt viscosity of the blends ranges from that of the branched polymer to that of the linear polymer, all at one specific value of the intrinsic viscosity. Simultaneously, there was a tenfold increase in melt viscosity observed as a function of increasing branch concentration. We have also prepared blends of linear and linear fractions. Preliminary data clearly demonstrate that linear-linear blends behave significantly different from branched-linear blends of comparable intrinsic viscosity values. We are now about to.extend the study to include blends of branched molecules with branched molecules. We hope, here, to clarify behavior differences between these due to branching from those due to variation in molecular weight distri butions. In view of the type of results we are observing, it has become imperative that we have accurate molecular weight data to properly interpret our results. Dale Harmon is applying laser light scattering techniques to attempt to measure the weight average molecular weights of our polymers used to prepare the blends. Data obtained on our linear polymers stand in agreement with literature values for polymers of comparable stereoregularity and intrinsic viscosity. Work is now in progress using our branched fractions. 20827017 9FG-8783-0 BFGoodrich Research and Development Center BFG07352 CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -16- Alloys as the Basis for New CPVC Compounds: (Project 4018C-80) M. H. Lehr -- The Hi-Temp group is planning a test program to evaluate phosphate, phosphite, and citrate salt co-stabilizers in Hi-Temp pipe compound. We discussed with them our citrate stabilization technology, which we developed initially for CPVC/ polyester alloys. We have also shown that Na+ and K+ citrates are effective co stabilizers in Hi-Temp compound. In addition, we showed how an anti-oxidant improves the stability of Hi-Temp compound. For example, .4 phr of Ethyl 702 or Goodrite 3125 afforded a 50-100% increase in Brabender stability time, depending on whether the CPE modifier was fused with the other ingredients or added later. The more stable compound was obtained when the CPE was added later after the other ingredients had banded on the mill. We also presented evidence that get ting good stability may depend on equipment factors. When we mixed our compounds in the Banbury at ALTC (with CPE present initially or added later) the benefit of AO was not realized. Process Innovation; (Project 4045A-80) J. Sidles and J. J. Blayne, Jr. -- We continued work on extruder blended curatives to achieve fast curing extrudates. Plans were finalized with Tire Development to build additional farm tires in the Miami, Oklahoma Plant, first quarter, 1981. This will yield additional tires for field testing. It will also enable an assessment of this system under factory conditions, using production size extruders. In other work, we showed that press-cured passenger retread tires using the "Fast Cure" approach could be cured in 40% less time than conventional compounds and methods. This work was undertaken to support Tire Group efforts aimed at licensure of the process to the retread industry. Work on ventless molding of passenger tires was continued. We evaluated a steam injection mode based on injection through small holes in the lower half of the mold. The steam was injected radially inward during the press pause just before total closure. This yielded the best looking tires to date, and will form the basis for a new, production size ventless mold now on order by Tire Development. Our plan is to try this mold at Brecksville, then place in pilot production at the Ft. Wayne Tire Plant. A Tire Group financial analysis shows a potential discounted cash flow return of 59.9% on an investment of $1.8 million, and an annual savings of $1.7 million. ETttZZSoZ aFG-8783-8 BFGoodrich Research and Development Center BFG07353 CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -17- PRODUCT PERFORMANCE RESEARCH Diffusion in Polymers; (Project 4002A-80) A. R. Berens -- Our previous vapor sorption studies of PVC indicated that an unusual "dilated" structure may result from exposure of the polymer to a swelling vapor, followed by rapid removal of the vapor under vacuum. We have renewed investigation of this process to learn whether it may be applicable to PVC films, as well as powders, to produce poten tially useful modifications of properties. Treatment of PVC resin with methyl chloride vapor, at room temperature and pressures above atmospheric, produces the "dilated" structure, as evidenced by increases in sorptive capacity and the appearance of a sub-T? endotherm in DSC curves. Unsuccessful preliminary attempts to produce this effect in a rigid PVC film have prompted a systematic experimen tal and theoretical study of the effects of vapor exposure and thermal history on the calorimetric (DSC) properties of PVC, now being carried out and reported under Project 4002G (see below). This study is helping to define conditions for producing the "dilated" structure and to predict its physical properties. In the continuing work on PVC particle-structure characterization through sorp tion kinetics analysis, the histographic distribution approach developed for gravimetric data is being applied to gas chromatographic sorption and desorption data. Gary Huvard has extended the computer program to calculate GC curves from known particle size distributions, and conversely. Experimental GC curves agree satisfactorily with the calculations for both sorption and desorption runs with MeCl on microsuspension PVC's. These results support our earlier suggestions that the "VCM desorption test" could be made more versatile by employing differ ent vapors for various particle size ranges, and less time-consuming by operating in the sorption mode, instead of desorption. Vapor-Induced Relaxations in Glassy Polymers: (Project 4002G-80) I. M. Hodge and A. R. Berens -- This new project covers the research being performed at Brecksville under National Science Foundation Grant No. CPE-7920740 to BFGoodrich and North Carolina State University, part of NSF's Industry/University Cooperative Program. The overall objectives of this basic research project are to apply vapor sorption procedures for producing and monitoring relaxation processes in rigid polymers, and to correlate results of sorption studies with thermal, mechanical, and dielectric techniques more conventionally used to study polymer relaxations. A major part of our proposed research involves the study of dielectric relaxa tions in polymers containing, or previously exposed to, penetrant vapors. Ex ploratory "dielectric creep" experiments, using equipment on hand, have demon strated that meaningful data can be obtained on porous discs formed by cold pressing PVC powder; DC discharge currents show the anticipated shift toward longer relaxation times during annealing after a thermal quench. We have sur veyed the requirements and capabilities of available equipment for both timedomain (dielectric creep) and frequency-domain (AC dielectric loss spectroscopy) measurements on polymer films and powders. An AR is being written for acquisi tion of equipment which will be of general value in BFG corporate and divisional R&D programs, beyond its immediate applications in this project. 9FG-8783-B BFGoodrich Research and Development Center BFG07354 GTUiZSoS <&: fi>nt''' n CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -18- Most of our attention this quarter has been directed to an experimental and theoretical study of the effects of vapor exposure and thermal history on the calorimetric properties of PVC (cf. Project 4002A) Experimentally, DSC curves have been obtained for PVC samples which were ex posed to varied pressures of MeCl vapor, or thermally quenched at varied rates, and subsequently aged at varied temperatures and times- Development of a pro nounced sub-Tg endotherm, growing with aging time and shifting to higher temper ature with increased aging temperature, is observed for the vapor treated sam ples and for the most rapidly thermally-quenched samples (plunged into liquid nitrogen from 120 C). The parallel theoretical study involves a mathematical relaxation model which contains parameters expressing the non-exponential rate of relaxation (due to a distribution of relaxation times) and its non-linearity (related to the depend ence of relaxation rate upon free volume.) A computer program based on this model can simulate a DSC experiment and predict the general effects of varied thermal quenching and annealing history. This program has successfully modelled all of the major effects observed experimentally. We may tentatively conclude that the development of a sub-Tg endotherm, like the more familiar "Tg overshoot" endotherm, is a consequence of enthalpic relaxation occurring during the prior aging period. The early and magnified appearance of this endotherm during aging, even near room temperature, suggests that vapor exposure, like a very rapid thermal quench, produces sufficient "frozen-in" free volume to allow unusually rapid and pronounced relaxation. Further application Of the model and computer simulations, coupled with experimental DSC data, seems to hold real promise for prediction of long-time aging effects on polymer physical properties from short-time laboratory data. Design of Composite Products; (Project 4004A-30) M. K. Chakko -- The objective of this project is to achieve better scientific capability for designing our cordrubber composite products. In our study of the fatigue behavior of tire belt composites, the main loading parameter has been the dynamic amplitude of the extensional strains applied to the belt samples. This quarter we investigated the effect on fatigue life of changing the mean strain level. The data obtained for 3% peak to peak dynamic strain amplitude with mean strain varying from 0 to 6% are shown in the figure. The fatigue life decreased as the level of mean strain was increased. As in the previous tests, the separations and failure occurred in the interply rubber, which in these samples was mostly natural rub ber. Yet, the decrease in fatigue life with increasing mean strain (or minimum strain) was observed. At the levels of mean strains below 1.5% in these tests, the samples were subjected to compression and buckling. The fatigue life, how ever, was higher. This may be due to the fact that the critical interlaminar shear strains at the edge of the samples remain near zero when the sample buckles, thus reducing the actual dynamic shear strain amplitude in the test. The fatigue testing conducted so far was mostly on steel belt composites. Cur rently, we are evaluating the fatigue characteristics of Rayon belt composites. Results to date show that the Rayon belt composites are superior to the steel belt composites in edge separation resistance. Using the new MTS Automated Dynamic Test System, we have measured the visco elastic properties of rubber under non-sinusoidal deformations. For deforma tion wave forms we used triangle. Sine and a rectified Sine similar to bfg-8783-8 BFGoodrich Research and Development Center BFG07355 S02 CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -19- those occurring in tires. Differences in energy loss per cycle were observed among the different waveforms. In the case of the rectified Sine waveform of deformation, the elastic modulus was up to 25% higher than the sinusoidal value. 20827021 Fiber Reinforced Plastics Composites; (Project 4006G-80) M. L. Dannis -- The in formation on creep characteristics (at room temperature) has been written up in a research report now almost ready for release. As the whole set of data was considered, several premature "conclusions" had to be revised. At the moment, the early part of the creep curve seems to fit the Andrade equation. E(t)/Ei ** t"n This is a curve-fitting expression, without important meaning related to creep itself as a molecular phenomenon. However, it does provide a useful comparison of initial modulus value Ei and of n (the log-log slope) in families of related materials. It must be emphasized that this Andrade relationship covers only 8FG-8783B BFGoodrich Research and Development Center BFG07356 CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -20- particular ranges of the much wider relationships discussed by John Ferry and co workers, i.e. the W-L-F shift concepts. The various relationships discussed in the report show the importance of extend ing the range of the creep data in both time and temperature. A six station creep assembly is being built to do just that. On the practiced side, a glass fiber reinforced Estane made at ALTC had a sur prisingly high initial modulus, attained without major damage to the fiber load ing. This has some possibilities as a sheet molding compound, but would be al most worthless for injection molding. The difference lies in the moderate re shaping of the SMC compared to major fiber destruction during injection molding. Further work is needed to advance this concept and to establish a preliminary cost-value relationship. Low Combustibility Technology: (Project 4007G-80) E. D. Dickens, Jr. -- I com pleted an initial evaluation of LDF foams being made by EPG (R. Srail). The foam tested was ~1.8 lb/ft3 density. Its smoke was very low and in line with values expected for the starting polymer. Efficiency of char formation was **60% for the foam as compounded versus -75% for the un-blown compound. The foam shrank drastically when exposed to a flame and lost most of its weight.(as HC1). A glow problem was observed and is believed to be due to the use of TiOj in the compound. Possible solutions, based on recent work by A. McRowe, Chemical Group, were suggested to EPG and are currently being evaluated. High Performance Polymer Bonded Ferrite Magnets: (Project 4020B-80) J. J. Blayne, Jr. -- The 400 lb. run of Kycar 4001 rubber from Avon Lake was received in early October. This material was tested and found to be equivalent to the lab samples that we had tested earlier. A factory trial at Marietta was planned for October 21. At the plant, a series of compounds from 55 to 70 volume percent ferrite were mill mixed, granulated, extruded, magnetized and tested. The results of this trial were very encour aging and the significant results were: 1. The Hycar 4001-AC540 binder system milled very well and accepted the BG-2 powder faster than the present Hypalon-Vistanex system. 2. The Hycar 4001-AC540 binder system extrudes at significantly lower head pressures than the production compound. 3. The highest loaded experimental compound ("69 volume percent BG-2) equals or exceeds most specifications for a gasket grade material. 4. The excellent thermal stability of the Hycar 4001 allows a wide range of extrusion temperatures ("280-400 F). This successful factory trial has encouraged the Marietta technical staff to proceed with investigation of this new binder system. Additional consultation and experimentation will be provided as needed. Z 0LZQ0Z BF&S783-B BFGoodrich Research and Development Center BFG07357 CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -21- Work has now resumed on the High Strength Magnetic compound based on Hycar 4001, AC540 and HM130. Initial experiments will determine optimum temperatures for extrusion using the existing four piece die. This will be followed by compound optimization for the highest magnetic strength with acceptable processability. Work on die design variables will proceed concurrently. A factory demonstration of the high strength magnet technology is planned for the near future. E. D. Dickens, Jr. -- In order to provide technical guidance to the extrusion trials and compounding program of Jerry Blayne, I have spent considerable effort in familiarizing myself with magnetic measurement techniques and repair of our hysteresisgraph. I feel confident in our ability to do and interpret these measurements. I have also made an effort (still ongoing) to pull together our past work on effects of volume loading and orientation. At the request of Marietta technical, I looked into the area of microwave ferrite gaskets. I have sorted out most of the basic physics used in this application. The ferrites used are not the type currently used by Marietta but are in fact specific "soft" ferrites tailored in composition to have an adsorption maxima at -2.5 GHz. I have received samples of potential powders from Ceramic Powders and they can be milled into current binders. At the request of Marietta techni cal and magnetic marketing, I talked to technical staff members at Hardwick Stove and Magic Chef to clarify the situation with respect to samples of our current magnetic gasket material submitted to them earlier this year. In all cases, our current material was found to be ineffective (an expected finding). The more technical sound customer (Magic Chef) is looking into patents a literature search found.that TDK (Japanese) has on this type of gasket and its use in ovens. No further action is planned until this legal issue is resolved since TDK is the sole supplier of this product at this time. (The patent situation has been re viewed with Bob Lindsay of the Legal Staff). Our successful extrusion trial of gasket grade magnets at Marietta (see above) pointed out a key difference between extrusion at Marietta and Brecksville. The extruders used in magnetic production operate in an almost adiabatic fashion. This means that the highest loaded compound generated a high stock temperature due to shear heating and this higher stock temperature offset a lot of the vis cosity increase due to the higher filler loading. The remarkable thermal stabil ity of Hycar 4001 allows us to take full advantage of this feature. I have worked out quite reasonable models for the extrusion processes at both Marietta and Brecksville- that take into account the effects of filler loading and the type of heat transfer involved (adiabatic vs. isothermal). The models clarify key issues involved in obtaining large increases in production output while giving at the same time a potentially large decrease in wear on existing equipment. The same type of strategy is being used in our work at Brecksville on the high strength magnet phase of this program. Optimizing the Performance of Rubber Blend Products: (Project 4029A-80) B. L. Lee __ Most rubber products are blends of two or more elastomers and several additives. The objective of this project is to develop new technologies that will enable our operating groups to optimize the performance and cost of rubber blend products. BFGoodrich Research and Development Center RFG07358 0S27023 CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -22- We have reported that in the SBR-CB tread compound, the hysteresis can be re duced by selecting the carbon black distribution in the individual SBR and CB phase. Most recently, we studied the hysteresis by selectively locating differ ent types of carbon black in the SBR and CB phase. The preliminary study is on 19955 compound, a low rolling resistance compound. The carbon black is mixture of N650 (coarse particle) and N234 (fine particle), with 36% of N650. The re sults indicate that the hysteresis can, indeed, be reduced. Adding the coarse particle to the CB and the fine particle to the SBR would reduce hysteresis. However, abrasion resistance is reduced too. Adding the coarse particle to the SBR and the fine particle to the CB phase would minimize the loss of abrasion re sistance due to the presence of coarse particle. The unvulcanized properties based on Dynamic Stress Relaxometer show marked difference in viscosity and stress relaxation behavior. This might affect die swell. We are planning to study the extrusion behavior. Plasma Polymerizations (Project 4030A-80) - No report this quarter. Conductive Composites: (Project 4032C-80) - No report this quarter. Adhesion Science: (Project 4039A-80) D. B. Rahrig -- The bonding of rubber to synthetic cord materials is achieved with resorcino1-formaldehyde-latex (RFL) adhesives. One of the primary goals of this project is to understand the inter relationships which exist among the chemistry, morphology and mechanical proper ties of RFL systems and how these interrelationships impact adhesion. In the last two quarters, we have reported on the use of the Dickie equations to predict the dynamic mechanical response of RFL films. Acknowledging the fact that RFLs are two phase systems, and assuming that the RF resin forms the continuous phase, one can predict the effect of a changing latex/resin ratio on the complex modulus and tan 6 behavior of RFLs using these equations. The experimental results suggest that for RFL films cast at 65 C and having ammonia in the recipe, the resin phase is continuous for latex/resin ratios of up to 20/1. For heat treated RFLs or those prepared without ammonia, however, the continuity of the resin phase appears to breakdown at lower latex/resin ratios. The results are summarized here: sample Latex/Resin Ratio Above Which Resin Continuity Breaks Down RFL withrammonia as cast pressed @ 175 C air cured @ 220 C >20/1 8/1 5/1 RFL without ammonia as cast air cured @ 220 C 10/1 10/1 Thus through dynamic mechanical experiments and the use of the Dickie equations, we have been able to learn details of RFL morphology which are beyond the re solving power of electron microscopy. 20827024 9FG-8783-8 BFGoodrich Research and Development Center BFG07359 CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1980 -23- This morphological information appears to be applicable to understanding adhe sive performance. Using two series of RFLs, one prepared with ammonia and the other without, and drying these adhesives on nylon at 220C; we found that ad hesion increased as the RFL latex/resin ratio increased. But for the RFLs con taining ammonia, this trend stopped when the latex/resin ratio went above 5/1. And for the RFLs without ammonia, the trend stopped when the latex/resin ratio went above 10/1. Thus in both cases the trend of increasing adhesion with in creasing latex/resin ratio stopped at the point where resin phase continuity is thought to breakdown. The RFLs prepared with ammonia also displayed a second adhesion maximum at the 12/1 latex/resin ratio, which is well above the level where resin phase continuity ends, however. The reason for this second maximum is not yet known. Highly Filled Materials and Composites: (Project 4040A-80) I. M. Hodge -- No report this quarter. Project LDF: (Project 4041C-80) G. S. Huvard -- During the fourth quarter, Dick Backderf (who performed the experiments) and I completed an assessmentof F-ll diffusion rates and mechanisms of F-ll diffusion in LDF compound resins. Our latest compounds sorb F-ll athennally and, depending on temperature and thick ness, will show both Fickian and Case II diffusion behavior. Based on our re sults, I successfully scaled a slurry sorption system to supply up to 150 lbm/wk. of F-ll saturated LDF compound cubes as a control feedstock for EPG's extruder and die design efforts. Ray Srail has developed a method for desorbing F-ll to give materials having the proper level of F-ll for extrusion. The technical interaction and progress on this project have been outstanding - Harry Ayres should have his extrusion designs optimized in the near future. He will con tinue to supply feedstock as long as is necessary. The slurry sorption system may be a viable alternative to an F-ll injection process. To this end, a process synthesis and Rapid Cost Estimate will be drawn up before the end 'of January 1981. A research report covering the funda mentals of high activity F-ll sorption in LDF resin will be written during December and issued in January or early February. 20827025 8FGa783-8 BFGoodrich Research and Development Center BFG07360 CORPORATE RESEARCH QUARTERLY TECHNICAL SUMMARY FOURTH QUARTER 1990 -24- RESEARCH PLANNING Project Market Research: (Project 4026A-80) J. M. Marks -- In support of Project 4025F, I provided Marv Lehr marketing information on lightly chlorinated PVC. This information plus.other technical information formed the basis for Marv Lehr's presentations at the Chemical Group's "Project Review" and here at Brecksville. Finally, in support of Project 4020B, I collected information on the 1979 net sales, market contribution (5 and %) and pound sales on the "Bonded Magnets" business segment of the Fabricated Polymers Division of the Engineered Products Group. Doug Dickens used this information when he reviewed the chronology of the High Strength Bonded Magnets Projects for the Group Technical Heads program "Studies of the Innovation Process within BFG". I have also included this information in a report "Magnetic Business 'Profitability'", which was issued during the quarter. Market Research; (Project 4026B-80) J. M. Marks -- During the quarter I completed and issued a report "Business Segment 'Profitability"*. This report gives the 1979 net sales, market contribution ($ and %), and division contribution for all of the business segments of Chemical, EPG, and Tire. I have also completed and issued a report "The Solar Space and Hot Water Heating Market". The purpose of this report was to determine if there are opportunities for BFG in this emerging market. Further, I held a brainstorming session for a product which had been made earlier at the Research Center but had not gone commercial. The purpose of this session was to determine possible applications for this product, known as Memolastic, which has a special property, memory. The next step will involve screening of these applications to determine if further investigation is warranted. Finally, I have begun collecting information on the synthetic fuels industry in preparation for a presentation which I am planning to give to the Management Committee sometime in the first quarter of 1981. BFG8783-8 N BFGoodrich Research and Development Center BFG07361 0827025 BFGoodrieh The BfGoodrich Company 3925 Embassy Parkway Akron, Ohio 44313-1799 rf/J `fut'i jjdsxnL. January 11, 1990 Mike Easterling Reneer Film Corporation Rt. 895 West Auburn, PA 17922 SUBJECT: GEONS 110X450, 110X500 and 110X911 - FDA STATUS Dear Mike: This is in response to our conversation regarding the FDA status of the above Geon PVC resins. Geons 110X450, 110X500 and 110X911 are homopolymer polyvinyl chloride (PVC) resins. Polyvinyl chloride is prior sanctioned for use in general food contact applications, both flexible and rigid. The prior sanction is based on an article by A. J. Lehman, from the FDA, published in the Journal of the Association of Food and Drug Officials. July, 1951. The current Good Manufacturing Practice (GMP) specifications for prior sanctioned PVC are a maximum volatility of 3.0% (1 hour at 105c) and an inherent viscosity of not less than 0.35 by ASTM D-1243-79. Geons 110X450, 110X500 and 110X911 meets the GMP specifications for prior sanctioned PVC, therefore meet the current requirements of the FDA for general food contact applications including food wraps. PVC is also listed as an acceptable ingredient of food contact articles under 21CFR: 175.105 175.300 176.180 177.1010 177.1200 177.2250 179.45 If you have any further questions, please call me at 216/374-3422. Sincerely, THE BFGOODRICH COMPANY W. C. Bachtel Toxicologist Health and Environmental Services 91024-3/jp cc: G. Schoff - CL-OTM-3 __ Jomzsoz BPG07362