Document MMa9v6B15p53OGabr3nGyk8Q9

Detention code soi-ooi' PROJECT REPORT PERMANENT RECORD COPY B F GOODRICH COMPANY TECHNICAL REPORT &S-6&10 |------------------------------------------- 1 | BPG TECHNICAL DOCUMENT | j Internal BFG Use Only j Volatiles Released During Lov Temperature Heating of Zepel RN. AUTHOR: R. P. Lattimer R. E. Harris Brecksville R&D Center D/8510 June 19, 1986 TO: M. B. Lattimer Brecksville R&D Center D/8526 6FD10D2 SUMMARY . I p..-v~ ' The volatiles released during lov temperature heating of Zepel RN are mostly residuals from polymerization (i.e., monomers, alcohols and associated fluorine-containing chemicals). TKe Zepel RN "monomer"-dsnoligomeric; it has been identified as lH,lH,2H,2H-perfluoroalkyl methacrylate. As you suspected, Zepel RN has excellent potential as an Early Warning Effect additive for use in PVC wallcoverings. ****** sample identification The BFGoodrich Co. 9-1. A ZepelR RN dried film, (Zepel RN is a DuPont fluorine polymer.) Brecksville PROBLEM DEFINITION R & D Files ft You are studying "early warning" fire protection systems for BFG Koroseal wallcoverings. This.work is related to information contained in a recent ' Westinghouse patent. Vestinghouse claims that materials containing DuPont's Zepel RN coating cause activation of an ionization type smoke/fire alarm during early stages of sample heating ("50-200C). This early detection can only be caused by the release of volatile chemicals from the sample that are capable of activating the alarm. The first report in this series shoved that Zepel RN releases numerous volatile chemicals at temperatures <200C, including numerous "fluoropolymer fragments." You requested that we examine these volatiles in more detail to obtain a better characterization of the evolved chemicals so that you could assess toxicity and subsequent manufacturing effects. EXPERIMENTAL APPROACH The Zepel RN material is quite complex chemically, so a number of mass spectral approaches were used. One technique used was GC-MS analysis of an acetone extract of the dried Zepel film. Both electron impact (El-MS, 70eV) and chemical ionization (CI-MS, isobutane reagent gas) were used at lov mass resolution. This approach has the advantage that the chemical components are separated prior to mass spectral analysis. BFG11125 21183001 2- - A second approach that vas used is high resolution (or atomic composition) mass spectrometry. In this, the mass of a particular ion is measured accurately to three or four decimal places. This enables one to determine, normally unambiguously, the atomic composition (i.e., the elemental formula) for the ion. This provides a very powerful tool for structure elucidation of unknown chemicals. -' ' A third approach that was used is the recently developed time/temperature profiling procedure.In'this, the sample is heated at a constant rate (5C/min) at atmospheric pressure under helium flow. This method allows one to determine the temperature at which a particular chemical component starts to evolve from the sample. RESULTS A GC-MS trace of the acetone extract of Zepel RN is shown In Figure 1 with the peaks identified. It is quite clear from Figure 1 that the volatiles in Zepel RN are a very complex mixture of oligomeric chemicals. The peaks were identified from their El and Cl mass spectral fragmentation patterns. High resolution accurate mass measurements (Table I) were very instrumental in assigning the structures. The low temperature volatiles are almost exclusively fluorine-containing chemicals which can be grouped into five oligomeric series: -- "A" series (monomers). CH2C(CH3)-C00-C2H4-C2nF4n+1 MW 132 + lOOn MW - 432 - 532 - 632 - 732 - 832 - 932 n-3-4-5-6-7-8. These are the Zepel RN residual monomers: 1H,1H,2H,2H-perfluoroalkyl methacrylates. Based on the patents you provided, DuPont suggests several possible monomers that can be used to make fluorinated acrylate polymers. " Methacrylate esters is one class of compound that is mentioned. One patent suggests, however, that the most common type of fluorinated alcohol may be lH,lH-perfluoroalkyl. This is in contrast to the actual alcohol that was found in Zepel RN, viz 1H,lH,2H,2H-perfluoro alkyl alcohol. -- "B" series (alcohols). H-C2H4-C2nF4n+l MW - 64 + lOOn MW - 364 - 464 - 564 - 664 - 764 - 864 n-3-4-5-6-7-8 These are the residual alcohols used in synthesis of the monomeric esters. - "C" series (ethers). C2mF4m+1-C2H4-0-C2H4-C2nF4n+1 MW 110 + 100(m+n) MW - 610 - 710 - 810 - 910 - 1010 - 1110 - 1210 m+n-5-6-7-8 - 9-10-11 These would be formed via self condensation of the alcohols. 21183002 BFG11126 -3- -- "D" series (iodides). I-C2H4-C2nF4n+l ^ 174 + 100n MW - 474 - 574 - 674 - 774 - 874 - 974 n=3-4-5-6-7-8 These must be formed from some side reaction of the alcohols with an active iodide species. -- "E" series (chlorides). Cl-C2H4-C2nF4n+1 MW 82 + lOOn (35C1 isotope) MW - 382 - 482 - 582 - 682 - 782 n-3-4-5-6-7 Like the iodides, these must be formed from some side reaction in the manufacturing process. The exact molecular weights of the chloride oligomers detected by GC-MS (Figure 1) are not knovn, since no molecular ions could be observed either by El or Cl analysis. Isobutane Cl time/temperature profiles are given in Figures 2-4 for the monomers, alcohols and ethers, respectively. (Profiles could not be obtained for the chlorides and iodides since these did not yield Cl MH+ ions.) The key fact to be noted from these profiles is the vide range of volatilities of the chemicals released from Zepel RN. While some components evolve at the beginning of the experiment (30C), others are not released until much higher temperatures (up to "120C) are attained. CONCLUSIONS 1. Zepel RN is a polymer derived from lH,lH,2H,2H-perfluoroalkyl methacrylate. The "monomer" is itself oligomeric, with molecular veights ranging from *300-1000 amu. 2. The volatiles released from Zepel RN in the range ~30-150C are almost exclusively residuals remaining from the polymerization process. These include fluorinated monomers, alcohols, ethers, chlorides and iodides. Minor amounts of other fluorine-containing residuals might also be present, but no other fluorine compounds were detected. The relative concentrations of the various oligomer series are not knovn. Fluorine compounds have widely different ionization efficiencies, so one cannot use the relative peak heights in Figure 1 to reliably estimate concentrations. 3. Our analysis confirms your hypothesis that Zepel RN.should be very attractive as an Early Warning Effect additive for PVC wallcoverings, for two reasons. First, since the Zepel RN volatiles all contain multiple numbers of fluorine and/or oxygen atoms, the chemicals should be excellent activators of ionization type smoke/fire detectors. Second, the Zepel RN residuals span a vide range of volatilities. Thus even if some components are lost through evaporation during processing, other components would still be available to cause an Early Warning Effect. BFG11127 eoofQTTS 4- - 4. It should be noted that the Zepel RN volatiles are released at much lover temperatures than the phthalate plasticizers vhich cause an Early Warning Effect in most commercial PVC vallcoverings. In our -earlier study, the ,, lowest temperature for evolution of any phthalate plasticizer was "117C, while some Zepel RN residuals are detectable at room temperature or slightly above. 2 5. In an earlier report, ve noted some other volatile residuals in Zepel RN. These include ethylene glycol (EG), poly(ethylene glycol), palmitic and stearic acids, and a trace of dioctyl phthalate. In our time/temperature profiling experiments, only ethylene glycol was detected below a temperature of "120 C. Ethylene glycol thus might contribute to an Early Warning Effect due to Zepel, but EG is so volatile that it would probably be lost from a wallcovering during processing, storage or end use. The other chemicals mentioned above are not volatile enough to contribute to an Early Warning Effect in the presence of the much more volatile fluorine-containing species. Robert P. Lattimer Robert E. karris Log Book Reference: Service Book 805, pp. 134-137; S. B. 817, pp. 1, 9-10; MS4957. /gg 5.2-86 cc: J. B. Pausch R. S. Varga E. A. Huettel (Marietta) _J^_D. Mason (Marietta) ~7RD'Files A H)0BTTZ BFG11128 5- - REFERENCES 1. L. N. Tannapoulos, R. D. Straw and D. C. Philips, U. S. Patent 4,520,157, May 28, 1985. 2. R. P. Lattimer. "Volatiles Released During Heating of a PVC Wallcovering and Coatings," Project 6IF10D2, December 18, 1985. . -- f I 3. R. P. Lattimer, "Time/Temperature Profiling of Volatiles Released from PVC Wallcoverings," Project 6FD10D2, June 18, 1986. 4. R. E. Johnson, Jr. and S. Raynolds, U. S. Patent 3,256,230, June 14, 1966. 5. R. E. Johnson, Jr. and S. Raynolds, U. S. Patent 3,256,231, June 14, 1966. SOOS8TTZ BFGlll 29 6- - Table I Accurate Mass Measurements and Atomic Compositions of Zepel RN Volatiles Measured Mass a Atomic Composition Calculated Mass*5 Ion Identity 69.0341 86.0368 c4h50 W>2 69.0340 86.0368 ch2-c(ch3)-co+ (monomer fragment) 1 Cfl2*C(CH,)-C00H+' (monomer'3 fragmen t) 113.0596 C6H92 113.0602 CH2-C(CH-)-C00-C2H4+ (monomerJ fragment)4 532.0306 632.0252 C14H92F17 C16H92F21 532.0331 632.0267 Monomer M+* (n - 4) Monomer M+* (n 5) 665.0028c 764.9925c C14H60F25 C16H60F29 665.0019 764.9955 Alcohol MH+ (n 6) Alcohol MH+ (n 7) 377.0194 477.0137 711.0234 811.0140 C9H60F13 C11H60F17 j C16H9OF26 I C18H90F30 377.0211 t 477.0147 i i 711.0238 ji 811.0174 j( C6F13'C2H4_0-CH2+ (etner fragmentj c8f17-c2h4-o-ch (etner fragmentj Ether ME* (a - 6) Ether MH+ (n - 7) 9!)t>S8TT2 i i ii BFG11130 7- - Table I (Con't) Accurate Mass Measurements and Atomic Compositions; of Zepel RN Volatiles Measured Mass3 Atomic Composition Calculated Mass*1 Ion Identity 327.0038 C8H3F12 327.0043 (M - HFI)+ (n =. 3) (iodide fragment) 426.9977 473.9154 C10fl3F16 C8H4F13I 426.9979 473.9152 (M - HFI)+ (n - 4) (iodide fragment) Iodide M+* (n - 3) 573.9079 C10H4F17I 573.9088 Iodide M+* (n - 4) 112.9962 c3b4p235ci 112.9970 35ci-c2h4-cf2+ (chloride fragment) 114.9936 - C-H,F37C1 J4l 114.9940 37ci-c2h,-cf2+ (chloride fragment) Measured by peak matching at a resolution M/4M 10000. Electron impact mode unless otherwise stated. ^Based on the given atomic composition. cIsobutane chemical ionization mode. 2J.183007 BFG11131 2118S00S RFni 11^ RIC HASS CHROHATGGRAHS li:06/ 06 14:41:00 SAllt' DATA: (IS4957C2 (11 CALI: TH U7 SCANS BFG11134 21183011 BFG11135