Document B8rpDKGZaOdGb41VYyVk8DVjL

, NO. 2 tank a icm.e presndrical 1 5 feet 'st and 00 C. issed a ure remator. ossible toruseating phosdried to advitha mber, phos- soline of 90 and a illate, Ifurio soda ough ease, d for 1 dis- ogen ith a the 1 i >lesi& the not ium BT )KB ed Sc. 1 3<*> ! S i % I i FEBRUARY, 1936 INDUSTRIAL AND ENGINEERING CHEMISTRY 193 phate industry. These developments relate to the concentra tion of low-grade phosphate ores by flotation and to the manu facture of phosphorus in the blast furnace. The production capacity of a blast furnace by the so-called one-step method of phosphoric acid manufacture by the Victor Chemical Works at Nashville, Ala. (#), is approximately 125 tons phosphorus pentoxide per day, but the furnace is designed to produce orthophosphoric acid. Some phosphorus pentoxide is ob tained in this furnace in the dust-collecting equipment. , One experimental blast furnace in Florida produced elemen tary phosphorus. As far as is known, no further development work has been done on this method of manufacture of phos phorus, and the plant is closed. Since there is no special call for phosphorus pentoxide for technical applications, the price is still on a high level of about $400 per ton. As with all pioneering and epoch-making in dustrial developments, it will take some time for the market prices to decrease. Even with this high price of phosphorus pentoxide the cost of refining some gasolines is very low because of the small amount required. This amount is 0.1 per cent of phosphorus and would be equivalent to 0.25 pound per barrel and a cost of 5 or 6 cents per barrel. The cost of coke (which can be recovered) is not large. Additional operating costs would then be the normal cost of redistilling gasoline which depends on the cost of fuel, running from 4 to 5 cents per barrel. The direct cost of the total treatment would be somewhere around 10 cents per barrel, excluding general overhead. j Literature Cited (1) Brown and Bqrry, J. Phya. Chem., 29,1312 (1925). (2) Easterwood, H. WTrana. Am. Inat. Chem. Bngra., 29, 1-20 (1933). (3) Malishey, B. W., /. Am. Chem. Soc., 57, 883 (1935). (4) Malishev, B. W., Petroleum Z., 28, No. 17, 7-10 (1932). Re c e iv e d August 20, 1935. Studies in the Drying Oils ROBERT S. TAYLOR AND JUDSON G. SMULL Lehigh University, Bethlehem, Pa. XIX. Oxidation of Linseed Oil1 HE nature of the chemical process by which 3. A conjugated system reacts with halogenating agents rela Tlinseed oil molecules, under the influence tively slowly (I). of oxygen, are built up to colloidal dimen 4. A conjugated system reacts readily with maleic anhydride, sions has been the subject of several theorieasn. d, sSochfaerifaesleis known, the nonconjugated system does not react, advanced the theory that film formation was entirely the result of the condensation of the oil molecules containing conjugated double-bond systems (10). That this is not a complete explanation of the film-forming process was pointed out by Scheiber (7) who subsequently developed this theory more fully (9). His theory states that the preliminary oxida tion of all the drying oils--with the exception of tung and related oils--results in the formation of conjugated double bond systems throughout the oil, and it is about these centers of activity that condensation and polymerization occur. The mechanism proposed for this reaction is: Data under the first three headings are submitted in so far as they help to confirm the more important results under the fourth heading. Complete data under the first three headings are not ob tainable because it is practically impossible, to prepare the nonconjugated oxidized isomers for comparison. Under the fourth heading, however, chemical reaction is shown when examination of the constants is carefully made. The curves show perhaps more clearly the decided changes that have taken place. These changes are explained later in more detail. --CH=CHCHr-CH=CH-- --> --CH=CH--C--CH==CH- h This study followed the changes in refractive index, heat of combustion, and iodine number resulting from the oxida tion of linseed oil. In addition, the reaction of Diels and Alder This theory has been extended by Scheiber and others to account for the changes which occur in the formation of "stand oils" (8, 6, 8). Kappelmeier believes that the condensation which results from the conjugation may be ascribed to the "diene synthesis" (5) of Diels and Alder. Conjugate systems, in general, show the following charac teristic properties: 1. A conjugated system has a higher refractive index than that of a nonconjugated isomer. 2. A conjugated system has a lower heat of combustion than that of a nonconjugated isomer. 1 The first seven articles of this series appeared as unnumbered papers in Inn. Ek g . Ch e m. as follows: 17, 138, 905 (1925); 18, 1245, 1262 (1926); 19, 62, 901, 903 (1927). Parts VIII to XVIII appeared in Volumes 20 to 26 (1928 to 1934), inclusive. 194 INDUSTRIAL AND ENGINEERING CHEMISTRY VOL. 28, NO. 2 Fig u r e 2. Ch a n g e in Re f r a c t iv e In d e x a n d Io d in e Va l u e w it h Ox y g e n Co n t e n t Fig u r e 3. Ch a n g e in Re f r a c t iv e In d e x a n d Io d in e Va l u e w it h Time oxygen addition to the molecule, the results given for calculated oxygen are empirical and yet, when compared with the determined oxygen values, the correlation is quite good. The method of calculation was to use the de termined oxygen value of 10.6 (1-day period) as the basis for the subsequent values. Then the iodine numbers of each sample were subtracted from the original 1-day values, and this difference was converted to oxygen, which was then added to the 10.6 to give the calculated figure. Be cause the 5-minute and 24-hour iodine values for samples (1-day) differ by 10 points, the cal culated oxygen values for the 5-day sample are different, even though the determined iodine values on the 5-day sample are identical. The Diels and Alder reaction was carried out involving maleic anhydride and conjugated compounds was applied to the oil in various degrees of oxidation (4). Methods The refractive index was determined by means of an Abb6Zeis refractometer at 25 C. Heats of combustion were measured in an Emerson bomb calorimeter. Ultimate analyses were made by the standard combustion method. Iodine numbers were determined by the Bolton-Williams method (pyridine dibromide sulfate solution in glacial acetic acid) (#). The oils used were alkali-refined linseed oils. Oxidation was carried out by bubbling air through the oil in glass percolators. The temperature was regulated by an electric resistance which was enclosed in glass and immersed in the oil. Results The results obtained are given ia-Table I and are plotted in Figures 1 and 2. In Tables I and II the 5-minute and 24-hour iodine-num ber columns refer to the time of reaction between the halogenating agent and the oil. The results given in Table II are plotted in Figure 3, those in Table HI are plotted in Figures 4 and 5, and those in Table IV in Figures 6 and 7. The columns headed "Calculated Oxygen" were computed on the assumption that one iodine atom was equivalent to one oxygen atom in the oxidation. Owing to the uncertainty as to the exact mechanism of Fig u r e 4. Ch a n g e in He a t o f Co mb u s t io n , Re f r a c t iv e In d e x , a n d Ox y g e n Ab s o r b e d w it h Time c.Ta b l e I. Da t a o n Mo n t h -Ol d Oil Sa mp l e s Ox id iz e d a t 50 Time Days 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Iodine No. 5 min. 24 hr. 166 166 i.67 iei i5S 1.58 i20 1.35 i09 iii ... Oxygen % 10.6 10.4 10.3 11.4 11.1 11.5 11,5 14.7 16.6 17.1 17.1 17.2 17.7 18.3 18.6 18.7 19.0 Calod. Oxygen 5 min. 24 hr. % 10.0 % 10.6 10. i io!o ii!e i 6!3 16'4 13^2 , ,IV8 li',5 Refractive Index 1.478 1.477 1.477 1.477 1.4778 1.482 1.4828 1.4835 1.4842 1.4843 1.4851 1...4..8...6.2 Ta b l e II. Da t a o n Tw o -Mo n t h -Ol d Oil Sa mp l es Ox id iz e d a t 45 C. Time Hr. 48 72 96 120 142 170 192 Iodine No. 5 min. 24 hr. 181 175 181 172 178 169 163 168 136 167 120 160 110 161 Refraotive Index 1.4766 1.4779 1.4778 1.4815 1.4832 1.4857 1.4868 Ta b l e III. Da t a 0 o n Imme d ia t e l y An a l y z e d Sa mp l e s o f Oil Ox id iz e d a t 25 C. Time Hr, 0 72 118 166 238 358 454 694 766 934 1103 1198 1318 1346 1366 1394 1414 1438 1462 1480 Oxygen % 11.2 ,, -- 12i6 12.9 13.2 li! 6 17.6 17.8 18!2 19i6 ,, Calcd. Oxygen % 11.2 11,2 11.1 11.2 11.2 11.4 11.4 liie 11.8 11.9 17.9 20.1 20.7 21.1 21.6 21.7 21.8 22.1 Iodine No. 170.6 170.6 170.8 170.1 170.4 169.2 168! 7 167 !4 105.5 165.1 117.0 99.7 95.4 91.7 87.5 87.0 86.1 84.0 Heat of Combus tion Cal./y. 8800 ,. 88....25 8880 87 io 7980 78..00 7740 Refractive Index 1.4777 1.4782 1.4782 1.4779 1.4781 1.4782 1.4782 1.4783 1.4783 1.4786 1.4793 1.4799 1.4879 1.4888 1.4890 1.4896 1.4898 1..4.8..99 From unpublished theses of R. Bortou, I. Mills, and C. Bigelow, Lehigh University, 1935. TO. 2 FEBRUARY, 1936 INDUSTRIAL AND ENGINEERING CHEMISTRY 195 Fig u b e . 6. Ch a n g e in Re f r a c t iv e In d e x , Ox y g e n Ab s o b b b d , a n d Io d in e ! Nt j mb e b w it h Time i f*. i =s= ) ! ;b d A 4 IL 1 si 1 gh on the samples tabulated in Table III. Varying amounts of maleic anhydride were'added to 10-gram portions [of oil, and the mixtures were heated tp 110 C. for 1.5 hours. The iodine numbers of these products were [then determined. The re sults are tabulated in Table V. It was found that maleic anhydride exhibited no iodine number when treated by this method so that the reduction in iodine number was caused either by reaction or by the mere presence of the anhydride as an inert material. The column headed "Calculated Iodine Number" was computed on the basis of the anhydride acting as an inert material.. The column headed "Discrepancy" lists the difference between this value and the actual value obtained. Any actual value greater than or equal to the calculated value indicates that no reaction has taken place, while values less than the calcu lated values indicate that reaction has taken place. Also, when the discrepancy reaches a constant value we can con- Ta b l e IV. Da t a 0 o n Imme d ia t e l y An a l y z e d Sa mp l e s o f Oil Ox id iz e d a t 60 C. Time Oxygen Calcd. Oxygen Iodine No. Refractive Index Hr. % % 0 11 24 49 62 70.6 80.6 94.5 11.2 li'.e 16.5 16.6 16.8 18.6 18.8 11.2 11.4 11.6 16.8 17.6 18.0 19.7 22.0 170.6 169.1 166.9 134.0 120.0 116.1 106.3 84.7 1.4777 1.4778 1.4780 1.4826 1.4839 1.4849 1.4861 1.4875 Data from unpublished theses of R, Borton, I. Mills, and C. Bigelow. Ta b l e V. Re s u l t s o f Die l s a n d Al d e r Re a c t io n Time Maleic Anhydride Iodine Calcd. Iodine No. No. DiBCrepancy Hr. 934 934 934 934 1198 1198 1198 1198 1346 1346 1346 1346 1394 1394 1394 1394 % O 2.6 5.0 7.5 0 6.0 7.5 10.0 0 7.6 10,0 12.5 0 7.6 10.0 12.6 167.7 164.5 159 157.5 166 167.6 166 166 99.6 86.3 83.5 81.6 91.7 78.3 74.8 72.6 163' 159 16. 6. . 157 164 150 92.6 90.6 88.6 85.3 83.3 81.6 + 1.5 0 + 1.5 +5 -6.3 -7.0 -7.0 -7.6 -8.5 -9.0 Fig u r e 7. Ch a n g e in Re f r a c t iv e In d e x a n d Io d in e Nu mb e r w it h Ox y g e n Co n t e n t elude that the reaction has been completed, and that the extra anhydride is merely acting as an inert material: Ho u t b of Oxidation 934 1198 1346 1394 Maleio Anhydride Reaoting 0 0 10 12.6 The point at which the oil begins to exhibit reactivity with maleic anhydride is coincident with the point at which index of refraotion, heat of combustion, etc., are changing rapidly. A series of determinations of the acetyl values shows that the maleic anhydride did not react with the hydroxyl groups. Acknowledgment This investigation was carried out under the Lehigh Insti tute of Research, the earlier part under the direction of J. S. Long and the later part under the direction of C. W. Simmons. Literature Cited (1) Boescken, J., and Gelber, E. T., Rec. trem. chim., 46, 168-71 (1927). (2) Bolton and Williams, Analyst, 55, 6 (1930). (3) Cutter, J. O., and Jordan, L. A., J. Oil Odour Chem. Assoc., 18, 5-11 (1935). (4) Diels, O., and Alder, K., Ann., 348, 31 (1906). (5) Ibid., 460, 98-122 (1928). (6) Kappelmeier, C. P. A., Farben-Ztg., 38,1018-20, 1077--9 (1933). (7) Soheiber, J., Farbe u. Lack, 1929, 477-8. (8) Ibid., 1929, 585-7. (9) Ibid., 1930, 51-3, 65. (10) Schiefele, B., Z. angew. Chem., 42, 787-90 (1929). Re c e iv e s August 1, 1936.