Document XOen77y3eZMq0Kw9zEXKVm0y4

S'DFG TECHNICAL DOCUMENT , B.F.GQOORICH CHEMICAI^CQMPANYJTECHNICAL REPORT ______ _= HCl Evolution From Geon 84295: Efficiency of the Calcium Carbonate Filler r* N*t'^ far <4*Qu**tco #v U7e a r ioit' 1/24/75 Peojic r so VMM* lev ~ * , --- ------ *j Geon 84295 is a commercial PVC cable compound that was designed to meet a low chloride flammability specification. This was accomplished by the use of high levels of calcium carbonate filler. Since calcium carbonate will react with hydrogen chloride under combustion conditions, we subjected Ge>*.. 04295 to NBS Smoke Chamber (flaming) conditions in order to determine the efficiency of the calcium carbonate in absorbing hydrogen chloride. I A compositional analysis of Geon 84295 indicates that the mole ratio of hydrogen chloride to calcium carbonate is 1.77:1 indicating that if the calcium covbonotf is 100% efficient it will absorb all of the h^liogen chloride. In verification of previous high temperature pyrolysis work, we found that calcium carbonate was only 50% effective in reducing hydrogen chloride evoltuion. Under NBS conditions, Geon 84295 gave 2654 ppm of HCl vs. theoretical value of 6082 ppm. Thus a significant improvement over and above simple filler dilution has boon attained. Future studies will involve upgrading the efficiency of calcium carbonate by substituting the current filler (Atomite) with a much finer (3X surface area) version. InitioI studies will be with Diamond Shamrock's Multiflex DX calcium carbonate. By doing this wo will attempt to produce a i . "no-HCl" PVC compound. introduction Geon 84295 is e new flsxfcls PVC wire and oobte compound that was issued in late 1973 to overcoeis a specific problem involving hydrogen chloride evolution during burning. The approach taken by marketing in designing this compound was to diluto tho PVC with an inert filler in order to meet a "maximum chloride allowable* specification. The diluent filler used to meet this objective was calcium coronate (Ca COe) which is not really inert. I previously hove shownd ) that Ca CO3 (i.e. AtomHe) sddbitis malar efficiency in reacting with hydrogen chloride at 600C, according to the following equation (a mere relevant version of the chemistry is presented in the Appendix): 2 HCl > Ca CO] L04 lOOi MPCMHCC 001-57-152 0 L. B. Crider J. K. Goots C. A. Clork G. A. Small J. E . Hartltx J. C. Heoly /T1 M. E. Woods J. Gallo ^e CaCI2 CO2 h2o 0) riCK C Q HO Brocksville E. b. dickens A. W. Mcftowo D. M. Kurts W. J. Kroenke Cleveland PTT.Tint E. G. Schwoegerle 8. A. DiUddo T. W. Williams 8. E. Wolch I.M.G.Zvdcke^ D . C . Kuharik J . W . brook F. Donat J. F. Mo lone el t to T 2. y 50% molar efficiency I simply moan that ona equivalent of Ca CO3 will react only with one equivalent of hydrogen chloride. Thus at these high temperatures, the efficiency of the Ca CO3 reaction with HCl is drastically reduced from the theoretical expectation. Recently^) we have demonstrated that a finer porticle-sized calcium carbonate may be much closer to the theoretical reoction stoichiometry than Atomite. This early worio^) is currently being re peated in an expanded study. The purpose of this current study was to determine the HCl absorption efficiency of Atomite in Geon 84295 under NBS Smoke Chamber (flaming) conditions. Composition of Gcon 84295 The following partial composition of Geon 84295 helps clarify the problem. Geon 84295 Nat 022, Geon 102 Atomite Others * issue 003 Wt.% "30" 34.2 27.8 * include stabiliser, plasticizer, processing aids and antimony oxide. Based on this composition, a 100 gram sample will contain 0.607 moles of HCl and 0.342 moles of calcium carbonate. If the stoichiometry of Equation 1 was obeyed, no hydrogen chloride would be transported to the atmosphere during burning. By meosuringfho actual hydrogen chloride in the atmosphere after combustion and correcting this for wall absorption losses, the amount absorbed by the calcium carbonate can be calculated. From this, the molar absorption efficiency of Atomite under actual combustion conditions can be determined. Experimsntol and Data Geon 84295 was obtained in 50 mil pressed sheets. The appropriate sample size was ejqmeed to NBS Flaming conditions and hydrogen chloride was analyzed by our "syringe extraction" method. From previous work, we estimated that the obtained HCl value accounted far only 70% of the smtual hydrogen chloride that was in the atmosphere and absorbed on the wells of the NK ehatabor. This correction was mode in the calculation (below). The hydbegM chloride evolution data is contained in Table 1. zooeesT z '7 T 3. TABLE 1 HCl Evolution from Geon 84295 NBS Conditions; flaming Sample mass: 21 .3 g Time (minutes) HCl, pf .5 10 1.0 103 2.0 469 3.0 978 4.0 1271 5.0 1858 7.5 1818 8.5 1790 Calculation and Discussion The calculation of the quantity of hydrogen chlorido absorbed by the calcium carbonate is based on the following premises; 1. 100 g of Goon 84295 yields 22.15 g HCl (14.04/ at STP corrected to 30C) . 2. chamber volume of 518,000 cc. 3. a 30% lots of hydrogen chloride due to incomplete sample combustion (minor) Therefore the following values were determined; theoretical HCl in chamfeii from 21.3 g sample: maximum HCl concentration, found (Table 1): HCl absorbed by Ca CO) in 21.3 g sample: moles Co CO) in 21.3 g sample: molar absorption efficiency; 6082 ppm 1858 ppm 2654 ppm 3428 ppm - 1775 cc z .0726 moles .0727 moles 50% 21383003 T 4. f The occurocy of this value which is dependent upon the accuracy of out corrected HCI con centration is probably i 12%. The corrected HCI value it based on previous work^) where the recovery of HCI was found to be a function of sample size. Thus the value used in this work is based on experimental data and, in our judgement, is a fairly good estimate. The molar absorption efficiency calculated in this study indicates that our previous wo*k^ ) which was based on inert pyrolysis was a fairly adequate model of combustion at least with respect to HCI absorption. Clearly Atomite is only about 50% effective in absorb ing hydrogen chloride. Future Work The significant point to be considered on this subject is the fact that Geon 84295 is a commercial compound that works and most importantly contains enough calcium carbonate to theoretically remove all of the chloride formed during combustion. It is now time to dis cover the method for increasing the efficiency of calcium carbonate absorption under combustion conditions. As a first step we will recompound Geon 84295 with another variety of calcium carbonate (Diamond Shamrock's Multiflex 1 DX) where an increase In surface area from 2.59 m? / g*(Atomite) to 7.35 m*/g* is realized. Previous studies indicate that this may be the key to the problem. In addition we are developing o model for arsessing the release, transport and conden sation of HCI during the burning of PVC coated wire and cable. Comparisons of existing systems with Geon 84295 will be mode. 1. M. M. O'Mora, J. Polym. Sei., 9, Art, 1387 - 1400, (1971). 2. M. M. O'Mora, I. F. G. C. C. Technical Report, IZ/KV74. 3. M. M. O'Mora, B. F. G. C. C. Technical Report, 9/16/74. 4. M. M. O'Mora, Combustion Toxicology, (August, 1974). MMO.RJ/dne 21383004 * Determined at ALGC by Argon absorption technique. T AO 6 98 APPENDIX 1383009 7  . Vi- - H1 '' :: yy ux V ' " \T;. : -u r.t.y vmt i r.- .: i.. t.r > * `it I** J* t.i. s >. t .- rpt; >f .-# f*-.- -ir r.i rf-r-- r, Lr.t>-nv 1 ir. fi*w> rr*l*t*r.t. 1- ' s*. ily v* . -jpound* v*. i *. . . j : '-yir. > "y ir .1 . i-.y^rl n ? \ Hyr*- yiri: pi- r . p__Qi . . * the condition of the ,"?lna';`his is the best copy that could be made. Py t- ,,--Nrrt'l . v i --ti it t .i t i-. vt#r 'it 8 <i\ - * 1:4. -tjx-itfs.1. t . ^ * s* i ' r-- f * . r p>*rt."a j..i* --Et.r-' t v- r- f. . t-. . v- *- r ti 't.- .w rr trj itn* t> t w iiw >il4 rut. .r| f yirii. for > w*rr*nt any f-irtr*rr - r. * turl tat* for tftr *op *>* - nioAt**-! .-in V** t` -n4 C J U'U4*~J r.*J. 1C 40o 3 3 ` >oodr >.r% C r\*m T