T H E J O U R N A L OF I N D U S T R I A L A N D ENGINEERING C H E M I S T R Y .
788
responsible for the corrosion, on account of their oxidizing action at the high temperature reached b y the metal in the presence of heavy scale. A particular instance of a water from a driven well in the city of New Haven may be mentioned. After this water had been used for a short time the boiler tubes were filled with small holes, and were coated with a very thick and hard irregular scale. The water gave the following analysis: Parts per million.
................ 85.7 ................. 26.8 ............... 28.1 .............. 99.6 .................. 83 .O ............... 117 .O ............... 25 .O ................. 9.5
Sodium chloride.. Sochum m e a t e . . Potassium nitrate.. Maanesium nitrate., Calcium mtrate. Calcium sulphate.. Ca€aum carbonate. Iros carbonate.. Silica..
.
..........................
20 .O
The quantity of nitrates (182. g parts per million of NO,) in this water is very great and although nitrates are very soluble in water, they would not fail to be deposited in the hot scale wherever the water penetrated into it. I n fact, small quantities of nitrates were found in the scale itself of this boiler, although they must have been destroyed by heat in the parts of the scale next to the metal. Another well water from a different locality in New J3aven was found to contain 39.85 parts per million of NO,, besides large amounts of carbonates and sulphates. This was not recommended for boiler purposes. Both of these well waters were characterized by being very free from organic matter and were neutral in reaction after the carbon dioxide had been boiled off,
-
Parts per million. NO.
Total solids.
......... ......... ........... ......... ........... 6........... 7........... 8 ........... 9........... 1.. 2.. 3 4.. 5
......... ......... ........... ........... ......... ........... ......... ...........
10.. 11.. 12 13 14.. 15 16.. 17 18 19.. 20.. .........
...........
........,
331 .oo 346.00 85.00 266.QO 185.00 243 .OQ. 301 .oo 434.00 106.00 345 .oo 172 00 261.00 81.00 119 .OO 220.00 95.00 257.00 164.00 170.00 275.69
Chlorine. 25 .OO 22.00 9 00 23.00 9.00 26 00 23.50 58 .OO 1 1 .oo 25.20 7.25 22.00 9.00 12.75 20.00 6.00 12.00 14.00 1 1 00 1 1 42
Nitrate NOS. 55.35 44.30 20.68 54.47 20 37 31.62 68.20 59 78 8.85 56.07 39.85 44.30 17.71 17.71 66.44 15.58 17.71 44.30 22.14 13.00
Further investigations are needed to reach a decision in regard to the amount of nitrates that may be permissible in a boiler water, and also in
Dec., 1909
respect to the protective effect of other constituents. The table of analyses given above comprises twenty waters from Connecticut, and serves to show the abundance of nitrates in them. As a comparison with the large quantities of nitrates shown in this table, i t may be observed that the average amount of NO, in the water supplies of Connecticut cities is only about I to 3 parts of NO, per million. SHEPPIELD LABORATORY. NEW HAVEN. CONN,
-
CRUDE PETROLEUM AS A REDUCING AGENT FOR ZINC ORES. BY HARRYH. HUGHES AND HARRISON HZLE Received September 17, 1909.
The process generally in use for zinc reduction is exceedingly cumbersome, slow and far from satisfactory. Comparatively small quantities of the roasted ore are mixed with coke and coal and heated in clay retorts which must necessarily be of considerable thickness, the zinc distilling over. The time required is about twenty hours and the expense for heat large. I n the best of coals there is quite an amount of matter which does not act as a reducing agent. Some of this is not only not helpful to the process but really injurious, as oxygen is furnished which hinders the reduction. These impurities also occupy space in the retort and consume heat the same a s ore. The high percentage of carbon and of hydrogen in crude oil and its resulting reducing power suggest it as a possible reducing agent. I t s extreme cheapness makes it all the more desirable. An objection to its use arises a t once in the fact t h a t a temperature of a t least IZOOO C. is required to practically reduce zinc oxide to metailic zinc and before such a temperature could be reached i n the usual furnace all the oil would be volatilized, leaving an insufficient amount of carbon to carry on the reduction. Evidently then, if crude oil can be used economically for zinc ore reduction it must be in a continuous process by which the furnace can be heated to a sufficiently high temperature and the mixture of oxide and oil fed into it. It would seem that under such conditions reduction should take place before any quantity of the oil can escape. To test this assumption a series of experiments was carried out in this laboratory. An ordinary gas pipe, 3/i, inch in diameter, w;ts connected at right angles with a cup from which
the mixture of ore and oil could be fed by means of a cock. A t first this pipe was placed in a Bunsen combustion furnace b u t no reduction occurred as the necessary temperature could not be reached. It was then passed through a Brown assay furnace, entering through a hole cut in the back and passing out the door with the end nearer the cup being slightly elevated. The coke fire around this gave ample heat without the use of a blast. The pipe extended some inches from the furnace acting as a condenser. Fifty grams of zinc oxide mixed with enough oil to make it pass through the cock was the usual charge. The oil required for this was in considerable excess of the amount needed for’ reduction. So-called “black oil,” the crude with kerosene and the lighter oils removed, was used at first and later the regular crude; either will answer. After the furnace was heated the charge was run in. An effort was made to pass i t in slowly, b u t with this rough apparatus usually without success as almost the entire charge passed in a t once. An evolution of gas from the oil on the heated surface followed. The charge remained in the heated pipe for twenty to thirty minutes when the pipe was removed and cooled and the contents examined. Beautiful specimens of zinc were found in the condensing portion of the pipe and there was a n almost complete absence of “blue powder.” Some of the zinc was feathery as if both the oxide and the oil were in a volatile state when reduction occurred. This reduction in a gaseous condition greatly shortens the time required as the reducing gas is in immediate contact with the oxide to be reduced. The gas generated by the excess of oil together with the carbon monoxide formed in the reduction was frequently lighted as i t passed out of a burner connected with the end of the pipe. I t should be possible to use the gases to heat the pipe or other retort containing the charge, largely reducing the fuel expense. A number of trials were made with uniformly good results. From our experiments we conclude: First, t h a t zinc ores can be successfully treated, after roasting, b y using crude petroleum a s a reducing agent. Second, that the theoretical advantages from the high reducing power of the oil and from the gaseous state of oxide and reducing agent, hold in a practical test.
We believe further that a continuous process based on these principles would be much cheaper, more rapid and more easily controlled than any process now in use, Arrangements have already been made for trying out the process on a large scale using the continuous process furnace devised by one of us.‘ WHITCOMB CHEMICALLABORATORY, DRURYCOLLEGE SPRINGFIELD. MISSOURI.
TINCTURE OF IODINE. By AZOR THURSTON. Received July 31, 1909.
Prior to the last edition of the United States Pharmacopoeia, tincture of iodine consisted simply of iodine dissolved in alcohol. Owing to the rapid loss of free iodine by the formation of either ethyl or hydrogen iodide the pharmacopoeial standard has been changed, b y the addition of potassium iodide, to prevent the conversion of the free iodine into iodides. I n the assay of the tincture determinations should be made for iodine, potassium iodide, ethyl and hydrogen iodides, and alcohol. Iodine is determined by the well-known U. $3. P. method by titrating five cubic centimeters with decinormal thiosulphate solution, whereby sodium iodide and sodium tetrathionate are formed : 2Na,S,O, I, = 2NaI Na,S,O,. Thenumberof cubic centimeters decinormal sodium thiosulphate used multiplied by 0.01259 will give the amount in grams of free iodine in 5 cc. of the tincture, and by multiplying the product by 2 0 gives the number of grams of iodine per IOO cc. Potassium iodide is estimated according to LaWall2 by evaporating the tincture on a water bath, adding several small successive portions of water, drop by drop, to aid in volatilizing the last portions of the iodine, and weighing the white crystalline residue in a tared watch-glass, which should be used for the experiment. The writer prefers to determine the potassium iodide a s follows: Place 5 cc. of the tincture in a platinum crucible and add 2 cc. dilute sulphuric acid; evaporate on a water bath until the alcohol and most of the free iodine are volatilized, then heat to dryness over direct flame, ignite to whiteness, cool and weigh. The residue will consist of potassium sulphate, and the weight obtained multiplied by I . 9 will equal the amount of potassium iodide
+
1
+
Hughes, Minrnr World. July 10, 1909. Proc. A . P A . . 1907, 159.
