Document vVDJR9KGybxDjE315ZOQo48VY

960 INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 16, No. 9 Place a 0.5-gram sample of well-mixed borings in a 50-cc. iron crucible and add 2 grams of solid sodium hydroxide. Cover with a watch glass and moisten with 1 cc. of water. When the violent action ceases, add 9 cc. of water, place on a hot plate, and heat for 30 minutes. At the beginning the liquid boils rapidly, but later, when most of the water has been expelled, no ebullition is observed. Without cooling, transfer the contents of the crucible into a porcelain evaporating dish. Rinse the crucible carefully with water, 35 cc. of acid mixture, and then with water again. Collect the washings in the evaporating dish into which the melt was transferred. Evaporate to dryness and fume strongly for 5 minutes. Cool, take up with 50 cc. of sulfuric acid (1:10) and boil until the solution is complete. Filter on ashless filter paper. Wash at least six times with water. Ignite in a platinum cruci ble and weigh. Add a few drops of sulfuric acid', about 20 drops of hydrofluoric acid, evaporate carefully on edge of a hot plate, and ignite. The difference in weight will give the amount of silicon present expressed as SiOa. To prepare acid mixture, mix in the order given, 300 cc. sul furic acid (specific gravity 1.84), 300 cc. water, 300 cc. hydro chloric acid (specific gravity 1.19), and 100 cc. nitric acid (specific gravity 1.42). Blanks for deduction should be run on reagents used. Effect of Iron Oxide Pigments on Rate of Oxidation of Linseed Oils1 By F. H. Rhodes, C. R. Burr, and P. A. Webster Co r n e l l Un iv e r s it y , It h a c a , N. Y. HE effect of certain T paint pigments on the rate of drying of The red iron oxide pigments tend first to retard and then to ac celerate the oxidation of linseed oil in paint films. Partially hy drated iron oxide is more aciioe than is the anhydrous oxide, Pr o c e d u r e The apparatus used and linseed oil has been studiedwhile the presence of calcium carbonate in the pigment renders it the procedure followed in by Rhodes and Van Wirt.2 less active in accelerating the oxidation. Black, iron oxide is a determining the rate of oxi This investigation, how relatively inert pigment, although it retards slightly the oxidation of dation of the oil in the pres ever, was confined to the the oil. ence of the pigments were study of the effects of vari essentially similar to those ous white paint pigments, described by Rhodes and and no attention was paid to the iron oxides or to any of the Van Wirt.2 Paints were prepared by grinding together other colored pigments. In view of the great importance two parts by weight of the pigment to be studied and three and the extensive use of the iron oxide pigments, it was thought parts by weight of the vehicle. In each case the vehicle was advisable to make the study of the effect of these pigments the prepared by dissolving in the linseed oil a sufficient quantity of subject of a separate investigation. The present article' lead linoleate paste (17 per cent lead) to contain an amount of describes the results obtained with various typical iron oxide lead equivalent to 0.2 per cent by weight of the oil. The paints reds and with black oxide. The work on the yellow and brown were allowed to stand in sealed containers for at least 2 weeks iron pigments is as yet incomplete. before use. Weighed samples were then spread on cloth and Ma t e r ia l s exposed to an atmosphere of pure oxygen at 30 C., and the rate of absorption of oxygen and the rate of evolution of vola The materials used in this work were pure, refined linseed tile matter were measured. The rate of oxidation of the ve oil from North American seed, lead linoleate paste drier, and hicle alone (linseed oil with 0.2 per cent lead) was determined representative iron oxide pigments obtained from various manufacturers. The linseed oil showed the following analy sis: Specific gravity at 15.5 C....................., Refractive index at 25 C........................ Acid number............................................. Saponification number............................. Iodine number......... .......................... 0.934 1.4803 0.473 193.3 173. The analyses of the various pigments employed are shown in Table I. i. Received June 17, 1924. * Tins Jo u r n al , 15, 1135 (1923). in a similar manner. In each case at least two determina tions were made with each paint. The individual deter minations gave results which agree with each other within the limits of experimental error. Re s u l t s The results are shown graphically by the accompanying curves, in which the amounts of oxygen absorbed and theamounts of volatile matter evolved (each expressed in terms of percentage by weight of the oil in the paint) are plotted September, 1924 INDUSTRIAL AND ENGINEERING CHEMISTRY 961 against the lengths of time of exposure. For each pigment there is plotted only one curve, depicting the average re sults of the two' or more check determinations. On each diagram the graphs for the rate of absorption of oxygen by the vehicle alone and for the rate of evolution of volatile matter from the vehicle alone are shown for purposes of comparison. These graphs are drawn as dotted lines. In d ia n Red (Fig. 1)--This pigment first retards and then increases the rate of absorption of oxygen by the vehicle. The initial retardation does not seem to be due to the mere physical effect of the pigment in increasing the thickness of the film and preventing rapid diffusion of oxygen into the mass, but to a specific effect of the iron oxide in accentuating the initial period of induction of the oxidation reaction. One hypothesis which may explain this effect is that the finely divided ferric oxide absorbs some of the lead drier and thereby lowers the concentration of this catalyst in the oil. The subsequent acceleration of the oxidation is probably due to the reaction of the iron oxide with some component of the vehicle or with some oxidation product of the oil, with the resulting formation of a salt or soap of iron which acts as a drier. In this respect Indian red appears to behave somewhat like Carter white lead.5 duced during the 2 weeks' period of aging to which the paints were subjected before being exposed. That such was not the case is apparent from the form of the oxidation curve; the driers are formed after the oxidation starts. Apparently, therefore, the iron oxide reacts with certain acidic oxidation products formed during the drying of the oil. Of course it is possible that in a highly acid, unrefined oil iron driers may be formed in the paint before it is exposed to air. Bb ig h t Ox id e (Fig. 2)--In general form this oxidation curve resembles that obtained with Indian red, but the effect of the pigment in promoting the oxidation of the oil is much more pronounced than in the case of the Indian red. The "bright oxide" also more markedly increases the rate of evo lution of volatile matter. The differences in the amounts of iron oxide or in the amounts of impurities in the two reds are, at first glance, hardly sufficient tp explain the difference in.their effects on the oxidation. It will be observed, however, that the "bright oxide" contains a relatively large amount of combined water, while the Indian red contains but very little moisture. Apparently, the partially hydrated ferric oxide of the "bright oxide" is much more reactive and forms iron driers much more readily than does the more nearly anhydrous ferric oxide of the Indian red. Fe*Oa PXGMSNY Per cent Indian red Bright oxide Spanish red oxide Red oxide Tuscan rede Venetian red Black oxide* 98.77 93.30 87.92 86.52 14.75 9.24 96.96 a toss on drying 3 hours at 105 0 c. b Includes organic matter. SiOs . Per cent 0.83 1.10 6.93 6.57 0,13 1.22 1.10 Ta b l e I Free BaSOi AliOj MnO CaCOj MgCOj CaSO* HtOo Per cent Per cent Per cent Per cent Per cent Per cent Per cent 6.55 .. .. 1.55 0.70 3.27 0.22 .. 0.15 0.75 0.21 0.31 O.ll 0.43 0,75 ,. l'.Ol 5.88 85 .*93 ..1*01 0*52 6*. 46 0.52 0.46 0.77 Q.39 0.50 2.85 0.79 0.77 c Contains a considerable amount of a red organic coloring matter. * Calculated as FesCh. . Combined HtOb Per cent 0.10 3.84 1.38 0.31 3.65 1.27 3.84 ' If compounds of iron which act as driers in the oxidation of linseed oil were formed by the direct action of the ferric oxide on the oil itself, we should expect these compounds to be pro- Sp a n is h Red Ox id e (Fig. 3)--This gives results quite similar to those obtained with Indian red, although the initial inhibitory effect is somewhat more pronounced. The Span 962 INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 16, No. 9 ish oxide contains a considerable amount of combined water and might be expected, from analogy with the "bright oxide," to show a much more marked drying action than does the Indian red. The Spanish oxide, however, contains a small amount of calcium carbonate, while the "bright oxide" is free from such basic impurity. It is reasonable to suppose.that the calcium carbonate would tend to neutralize some, at least, of the acidic products of the oxidation of the oil and would thus decrease or prevent the formation of the iron drier. pigments. Because of the low percentage of ferric oxide in the Venetian rad the initial inhibiting effect on the oxidation is very slight, while the large amount of calcium carbonate in the pigment prevents the formation of any iron drier during the drying of the oil. Bl a c k Ir o n Ox id e (Fig. 7)--This decreases the rate of oxidation of the oil throughout the whole period of drying. The initial retardation, however, is not nearly so pronounced as in the case of Indian red. The black oxide appears neither to adsorb the lead drier to a marked extent nor to react with the oil to form iron driers. To a certain extent the black oxide resembles lithopone in its effect on the oxidation of linseed oil. Co n c l u s io n s Red Ox id e (Pig. 4)--This effect of calcium carbonate in inhibiting the drying action of the iron oxide is exhibited even more markedly in the case of the red oxide. This pigment, which contains 4.88 per cent of calcium carbonate, gives a paint the oxidation curve of whieh is very nearly identical with the curve for the oxidation of the vehicle alone. Tu s c an Red (Fig. 5)--With the Tuscan red the effect of the pigment in decreasing the initial rate of oxidation of the oil is very slight. This might be expected, for this pigment contains only a relatively small amount of ferric oxide. The ferric oxide present, however, is in a rather highly hydrated and very active form, so that the iron driers are formed readily and the pigment shows a very pronounced accelerating effect shortly after the beginning of the period of exposure. Ve n e t ian Red (Fig. 6)--This pigment gives results that are quite consistent with those obtained with the preceding The results obtained in this investigation justify the fol lowing conclusions'. 1-- The iron oxide reds (including Tuscan red and Venetian red! tend first to inhibit and then to accelerate the oxidation of linseed oil containing lead drier. The initial inhibiting action may be due to the adsorption of the lead drier, by the pigment, although we have no direct experimental evidence to support this hypothesis. The subsequent accelerating ac tion is due, presumably, to the formation of small amounts of ferric compounds which are soluble in the oil and which act as driers. 2-- The formation of the iron driers takes place during the drying of the oil and is due to the interaction of the pigment with acidic products formed during this oxidation. 3-- Iron oxide reds that contain partially hydrated ferric oxide are much more active in accelerating the oxidation of the oil than are the more nearly anhydrous oxides. 4-- The presence of basic substances--as, for example, calcium carbonate--tends to prevent the formation of iron driers during the drying of paints containing iron reds, and thus prevents the pigments from exhibiting their normal effect in accelerating the reaction. 5-- Black oxide shows only a slight effect in inhibiting the initial oxidation of the oil, and displays no tendency to react with the oil to form iron driers. Things Not Known about Rubber The souvenir number of the India Rubier Journal, which commemorates its fortieth anniversary and which was issued August 2, 1924, contains the following among "More Things Not Known about Rubber." What happens when raw rubber "freezes?" Why is frozen rubber opaque and thawed rubber transparent? Why do basic substances have an accelerating and acidic substances a retarding effect in vulcanization? (A simple rubber-sulfur mixture is here in mind.) Can rubber particles in latex be subdivided--i. e, further dispersed? Can vulcanized rubber, or crude rubber, or both,f be stabilized to such an extent that deterioration or "perishing" shall be no greater than in the case with, metals, such as gold or aluminium, or with organic substances, such as horn, leather, wood, and the like? Why is rubber that is kept in use--i. e., mechanically worked--less liable to deterioration than rubber that remains quiescent? Does rubber possess coarse--i. e,, nonmolecular--structure, and, if so, how dobs this structure bear on the mechanical properties? Or are the properties of rubber explainable solely by virtue of its molecular constitution and the grouping of the molecules to form definite aggregates? Why do finely divided, insoluble compounding ingredients reinforce rubber? . Why have the two forms of finely divided carbon--viz., lamp black and gas black--such widely different effects on rubber in which they are com pounded? To what extent does flocculation really occur among the particles of the compounding ingredients in rubber after vulcanization? Why does synthetic rubber age more rapidly than natural rubber? What are the effects of (a) air, (6) Light, and (c) temperature in the natural aging of vulcanized rubber? Why does vulcanized rubber age less rapidly'in moist air than in very dry air? How many other factors affect the aging qualities of manufactured rubber? Why do the soluble serum substances of latex offer such great resistance to removal in washing latex rubber? What alteration in the molecular or colloidal structure of rubber occurs in the softening of rubber by milling or by heating? To what extent is the effect of softening rubber by heat comparable with that of softening by milling? To what extent is the action of atmospheric oxygen an important factor in the softening of rubber by milling? Are the great differences observable between the resistance of milling of various natural rubbers caused by differences In the rubber hydrocarbon or by differences in the noncaoutchouc--e. g., pro tern--substances present? What hi the exact nature of the physical alteration in rubber on vulcani zation? - When will it be possible to produce "raw" rubber from vulcanized scrap by total removal of the free and uncombined sulfur and the fillers? Exactly what happens when rubber swells or "dissolves" in a liquid?