Document xjEyj1J37DQjd1ZxOYoYp7x3E

\ 4' IN TER -OFFICE CORRESPON OEN CE Et hyl Gaso l ine Cor por at ion Re s e a r c h La bo r a t o r ie s 723 Eas t Mi l w au k ek Av k n u e De t r o it ,Mic h ig a n June 8, 1938 Dr. Robert A. Kehoe College of Medicine University of Cincinnati Cincinnati, Ohio Dear Bob: I am sending you herewith a copy of our Report LTD 38-25 on the mechanism of decomposition of triethyllead bromide by steam distillation. As I told you during your last visit, vre are pursuing those studies, and I will send you reports from time to tine on the other alkyl lead salts. . Very truly yours, /? GC:AMC Enel. 0022263 LID #38-25 O.R.Kurt F.J.Dykstra 263-XP-15, pp 41' May 23, 1938. STEAM DISTILLATION 0? LEAD TRIE1BYL BBQMIDE PURPOSE Tills study ic part of the problem of dotoraining the volatility of lead ethyl salts and their decomposition products. The run reported here was aade by Mr. Dykstra in order to verify the results of previous runs (230-XP-19 and 230-XP-22 reported in LID #37-28 and 230-AP-23 reported in LID #38*1) and in order to acquire technique for subsequent runs on other lead alkyl salts. COHCLUSIOMS - The decomposition of PbEt^Br proceeds as follows. these results confirming those of previous runs* 2 PbBtgBr <* PbBtA + FfcBtoBrg 2 PbSt2Br2 * PbEtgBr + PbBrg + EtBr PbEtgBTg a PbBrg + 2 St ' orcent decomposed at 100 C Time of Run 230XP22 65fi 37 89 210 230XP23 84$ 16 81 160 This Run 83 17 72 160 SIMiARY Very good yields of pure EtBr and PbSfcA were obtained* The oexposition of the hydrocarbon products was not established because gas analysis results were unsatisfactory* The yield of gas, though measured more accurately than previously, is still much lower than theoretical. The relative extent of the two decomposition reactions of FbStgBrg was calculated from the data for PbEtgBr, FbEt4 , PoBrg and EtBr, which are in good agreement with each other. All f the lead was accounted for except 2.5 percent, which loss can bo ascribed to handling. LETAILS The apparatus and procedure were similar to those of runs 23Q-XP-22 and 230-XP-23. Several modifications were made wiich improved the products and reduced the work required. A 500 co, 3-neok round-bottom flask was used as a boiler. Ono side neok was fitted with a etopoooked inlet for nitrogen. The o th e r s i d e n e c k was c o n n e c te d to t h e r e f l u x c o n d e n s e r and P b Et. t r a p . The PbEt^ trap was fitted with a water return tube which worked satisfactorily. :,'o beads were used in the distilling column or in the condenser* The condenser was oooled with cold tap water at approximately 15C. A small tube connected the upper end of the condenser to a trap kept in a dry-foe acetone bath and then led to the gas burette. Ilia apparatus is sketched in the attached blueprint. The pbEtgEr was freshly reorystallised from dry mlxod hexane. 300cos distilled water was placed in the boiling flask, the air in the apparatus was replaced by nitrogen, the water was boiled until equilibrium was reached, then a filter paper thimble filled with FbStgBr was placed in the flask through the central neck. Consequently the increase in the gas volume in the apparatus, which is measured by the gas which passes into the burette, is all due to gas generated and not partly due to thermal ex pansion of the nitrogen ns in previous runs The gas was drawn into evacuated 100 cc. pipettes after being measured in the burette. The run v,-as stopped after 160 minutes* The system was flushed with, nitrogen to collect the gas in the system* The initial material end the products were analysed to obtain the results presented in the attached table. Analysis of the initial material gave two values for % ?b and one value for % Br equal within experimental error to those of pure FbBHBr* The other % Br seems unreliable in view of the method of purification so the most probable composition to aocept is that of pure FbEtgBr. The PbSt4 layer of the distillate had a specific gravity so close to that of pure FbEfc^ that it was not considered necessary to give it any further treatment, it was assumed to contain 1% EtBr. Analysis for Fb showed it to be quite pure, but the ft pb was used to calculate the iaols of Pblit^. The material which condensed in the dry ice trap was quite pure 3tBr according to the specific gravity and refractive index measurements* I'haro was/considerably larger amount than was obtained in previous runs, even after including the EtBr found condensed in the Fbt4 layer in previous runs. The water solution left in the boiler deposited beautiful crystals on cooling* presumably of PbBrg* It was rehaatod to put the pbBrjj and undceoaroaod FbEtgBr into solution, the flask was washed and an aliquot portion of the solution analyzed. A residue remained in the flask which was molten in hot water and solid at room temperature. All except a small portion of it and the residue on the filter paper were put into solution vith ether. The residue was dissolved in nitric acid and analyzed for lead. The gas was collected in three 100 oc. pipettes. The analyses were mostly unsatisfactory* The results on the first pipette did not chock each other. The results on the second pipette gave an lyQ ratio of 2.38. The contents of the third pipette were not analyzed. The molecular weight calculated from the weights of the gas in the pipettes,which is mostly nitrogen, indicate that the hydrocarbon gas may be CgKd end CgHg* 63.S oos. of evolved gas at 25C and 760 can. would be .00259 xaola. If the gaa were and CgHg this would account for .00259 mol Et* Borne Et might have polymerized to form higher hydrocarbons but then some of those should b found in the dry ice trap 7/ith the EtBr* The missing Efc calculated from the other pro duots, 0.00345 mol equals 0,100 g m ,, is small enough so that it might easily bo lost in the EtBr, the PbEt^, or absorbed in the rubber connections* The missing Pb is assumed to bo Pb2t4, -which, has either boon volatilised or held up ia the condenser, or docomposed to Pb in. the pbEt4 trap. It is also partly due to experimental errors in analysis. The relative extent of the two Pbt23r2 decomposition reactions are calculated from the data for PbEtgBr, PbBrg and Pbgt4 in the attached table. It will be noted that the sum of equations (4) and (5) necessarily equals (3) because all the Pb has been accounted tor by assigning the miss ing ?b to equation(4), Consequently there are really only two equations for solving for the two unknowns, a and b. The only check on the results lies in the fact that experimentally the Pb balance is quite good, Even so the 2,5^ of missing Pb is a large enough amount to cause a high uncer tainty in the values of a and b except for the fact that we can ascribe the missing Pb quite definitely to PbEt^, A high uncertainty could occur in the analytical results, so it is recommended that the technique be improved to recover all of the Pb, d'fEM DISTIilATIOB OP PbBfesBr - 263-XP-1S Initial PbBfcv3^Br ANALYSIS 14.055 gSi. S ^ T p b * ^ o ^ T s ? 55.44 21.43 Theoretical; 55.37 21.36 COMPOSITION- Cto,Mo ' 0,03990 PbEtjBr Distillate - PbEt4 layer 3.40 os 3 25 Sp.Gr. 1.64S, 1,843 (of, 1,8528) S3,40yi Pb 63.55 Theoreticals 64,06 0.01712 Fbt4 .00051 BtBr Dry ioe trap 0.35 oc Sfp.Gr.O 23.5, 1,445, 1,467 (of. 1.450) 0.00526 BtBr 5.1, 1.4233 (of. 1.4239) Still residue -water soIn. w n ether extract n it residua Volatile gas Missing Pb 0.01846 mol Pb, 0.028817 mol 3r .00312 mol Pb, .00311 mol Br .00023 Pb cos . #1 ml .wt, 27.85 16.6; combustible #2 27.95 12.7 ' #3 27.95 not analyzed 0.01036 PbBr2 .00810 PbEtgBr .00311 PbEtgBr .00023 (PbBT2 ? .00259 Bfc h/c z.m .00104 PbEt4 3a FbEtgBr 2b PbEtgBr FbEtgBr PbEt4 PbBr2 BtBr =*x 2 PbSt4 + PbBrg + BtBr 3a 0.02271 (1) 2b 0.00604 (2) 3a .02875 (3) 2a b - .01712 + .00104 .01816 (4) * a + b * .01036 + .00023 .01059 (5) a 0.00787 b 0.00302 * a .00757, Recoveredt .00051 + .00528 * .00529j Missing - 2b .00604,Recovered .00259^ Missing .00345 (7) ,00228 (6) X m o o to to ro 03