sulfatc tends t o low results, while the nitrate has the opposite effect. Only by chance will t h caiicei one another.' A few determinalions made x i t h doniile the usual amounts of reagents seein t o show t h a t t!ie net error due t o occlusion is small. The present paper is a preliminary one, and i t is intended t o puhlish a more iietailed discussion of the n-liok subject later on. sI:sIIl.%RY
tile flasii is thoroughly shaken e v e r y haif hour for several hours, and aliowed t o settle. The petrolcum ethcr is decanted through a filter i n t o a IVilcy tube, a n d a sccoiid smaller quantity of the solvent is added t o t h e flask, sh:iken, settled, and filtered, as before. ,~ llie washing is continued in this way until rhe sulfur a n d the filter paper are free of oil, and the IViIey tube contxins about 1 7 5 cc. of solvent, or enough to fill tlic tube above the top of the coppcr roll or coil.
The desirability of investigating the possible eflect of sulfur conipounds on the properties of liihric,ating oils is pointed out. l h e disadvantages of the methods now in general use for the determination of sulfur are indicated. A new procedure, which gives results at least as acciirate as those by t h e bomh calorimeter, is described.
,.
DETERMINATION OF MINUTE QUANTITIES OF OIL IN SULFUR By L. S. Busbnelt and H. Smith Clark FX~POR SV TL P ~ WCOMPANY, R Fxrs~onr.TEXAS Received November 2s. 1919
The presence of very small amounts of oil in sulfur is objected t o by manidacturers of certain sulfur products. Tlic sulfur may contain carbonaceous residues from burned oil, and the method here described is not intended t o include these. The estimation of oil by simple extraction as ordinarily made with a volatile solvent-such as sulfuric ether or petroleum ether-will not suffice in thc case of sulfur, becausc, while t h e sulfur is only slightly soliil~le,there is usually such a small amount of oil prescnt (from 0.001 per cent t o 0 . 3 2 per cent, or more) t h a t Lhc catrnct contains considerably more sulfur than oil. in t.liis method extraction is made as usual, and the solvcnt purified from sulfur by boiling i t in a Wilcy coniiuuous extractor in which has been placed a roll of wpper foil. rll'l'ARATGS
A sheet of coppcr foil 4 in. X ij in. is rolled up in siicii fashion t h a t no two points of its surface c.ome i n
contact. I t is desirable t h a t the copper sheet be rolled uniformly so t h a t there will be a space of about threesistcenths inch between any one turn of the roll and the next turn. The sheet muy be cut so t h a t at one corner a t a b three-eighths inch square projects. A hole is punched in this t a b so t h a t the coil may be lifted from the Wiley tube q i t h a wire hook. The end x i t h this tab should be rolled first, and the t a b project from one end of the coil. A 100 cc. volumetric flask; a thin glass beaker, tall form, withdut lip, 300 cc. capacity; and a Wiley continuous extractor, without crucible or thimble, complete the required apparatus. PR0CEI)URE
Fifty grams of t h e pulverized sulfur (or a smaller quantity, if t h e oil content is known t o be high) are ~ e i g l r c dand transferred to the I O O cc. flask. On this are poured about 50 cc. of redistilled petrolcum ether; 1 Allen and Johnston. J. A m . Chcm. S o c , 82 (1910), 500; Johiision m d Adnma, Ibid., 83 (1911), 029.
l'hc coppcr coil is cleaned with dilute nitric acid, rpashed with water, then with nlcoliol, dried with ether, and placed in the Wiley tuhe. Tlic petroleum ether is then boilcd until all of the suliur has been deposited on the coppcr as coppcr sulfide. The solvent is transferred t o the tall beahcr, filtering if necessary t o eliminate the copper sulfide, which xi11 doubtless partially flake off. The coil, the Wiley
tube, ani1 the filter are waslied with smali quantities of pctrolcum ethcr, the combined filtrate and washings gently evaporated t o dryness at very low heat, and the residue rveighcd. NOTE ON THE SOLUBILITY OF BENZIDINE SULFATE IN WATER By C. S. Bisson and A. W. Christie U > I Y S K S I T Y "11 C I L I Y U R N I A , UCKltEI.LY,
CILiFOnnlA
Received September 9, 1919
One of the methods for t h e determination of sulfates which has lound considerable application consists in precipitation as benzidine sulfate, which may be quantitatively titrated with standard alkali or potassium permanganate. I n most of the articles describing such methods, the benzidine sulfate is said to be insoluble in water. %. review of the literature
T H E J O U R N A L OF I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y
4.86
failed t o reveal any definite d a t a on its solubility except a statement by Muller' which gives i t as from 0.01t o 0.03 per cent a t 2 j o C. We have determined its solubility a t different temperatures b y agitating an excess of benzidine sulfate with pure water in a constant temperature thermostat. The benzidine sulfate was prepared by adding a n excess of dilute sulfuric acid t o an aqueous solution of Kahlbaum's benzidine. The precipitated sulfate was filtered and washed with distilled water till free of acid. From 3 t o 5 g. were placed in a 500 cc. Pyrex flask containing pure water and the flask immersed in a thermostat which was regulated t o 0.5' C. The contents of t h e flask were continually agitated by a revolving glass stirrer. After a specified number of days t h e solution was removed and immediately filtered, t h e first 2 0 cc. of filtrate being discarded. Solubility determinations were made a t oo, z j O , s o o , and 80" C. Two methods for determining the solubility of t h e benzidine sulfate were used. One method consisted in evaporating a measured volume of I O O or 2 0 0 cc. t o dryness in a platinum dish on t h e steam bath. The residue was dried a t r o o o C., cooled in a desiccator, and weighed. The other method consisted in titrating a measured volume of the solution with 0.0; N potassium permanganate after the addition of 5 per cent of sulfuric acid. The latter method is described in detail elsewhere.2 1
Ber., 55 (1902), 1587. W. Christie and C. S. Bisson, THISJOURNAL, 12 (1920), 1 7 1 .
* A.
Vol.
12,
NO. 5
The solubility of benzidine sulfate in water as determined by the above methods is given in t h e accompanying table. SOLUBILITY OF BENZIDINE SULFATE IN WATERAT VARIOUSTEMPERATURZS Determination--B y Weight B y Titration G. per Liter G. per Liter 0.049 0.098 0.141 0.290
Temwure 0 25 50
80
The two methods gave concordant results a t all the temperatures except 80' C. T h e solutions for t h e determination of solubility a t o o and 2 5 O C. were stirred in the thermostat for four days. Those a t 50' and 80' C. remained in t h e thermostat only 2 4 hrs., since i t was found t h a t after several days the solutions became badly discolored, indicating t h a t t h e benzidine sulfate had undergone decomposition. The solution obtained a t 80' C. was slightly discolored even after 24 hrs., as was also the residue obtained on evaporation. This is probably due t o t h e partial oxidation of t h e benzidine a t t h e higher temperature, as evidenced by t h e low figure for solubility obtained b y titration. The residue obtained from t h e soo solution showed only a very slight discoloration a n d t h e results may be considered t o give t h e approximate solubility a t this temperature. From a consideration of these results it is evident t h a t in washing the benzidine sulfate in a quantitative determination of sulfate, t h e minimum amount of cold water should be used.
LABORATORY AND PLANT A CONVENIENT METHOD FOR THE DETERMINATION OF WATER 1N PETROLEUM AND OTHER ORGANIC EMULSIONS' By E. W. Dean and D. D. Stark CHEMICAL
SECTION,
PETROLBUM DIVISION, PITTSBURGH, P A .
u. s.
BUREAUO F MINES,
Received January 15, 1920 INTRODUCTION
One of the problems frequently confronting t h e petroleum chemist is the determination of emulsified water. Numerous methods have been devised, several of which have been described in a previous publication2 issued by t h e Bureau of Mines. Allen and Jacobs discuss in t h a t paper the advantages and limitations of various types of procedure and recommend as most desirable t h e scheme of distilling in a special electrically heated still. This type of still involves heating the entire outer surface of a 250 cc. distilling flask and its use permits t h e "breaking" of troublesome froth t h a t is usually formed when a viscous petroleum emulsion is heated. Certain difficulties in the construction and operation of this still have led t o discontinuing its use in favor of the more common method of diluting t h e emulsion before distillation with a solvent immiscible with water. The present Published by the permission of the Director of the U. S. Bureau of Mines. 2 See Bibliography, p. 490. 1
e
paper reports the work of t h e Bureau in modifying this latter method t o obviate certain of its admitted disadvantages. DISTILLATION
IN
THE
PRESENCE
OF
AN
IMMISCIBLE
SOLVENT
There are various conditions under which i t is desirable t o determine water i n petroleum emulsions and no one method is best in all cases. The method of gravity separation' is a convenient one a n d is satisfactorily reliable for emulsions t h a t are not too viscous or t h a t contain water i n not too fine a s t a t e of subdivision. For the estimation of extremely small percentages of water, in such products as transformer oils, special methods are necessary which are too delicate for ordinary needs. The distillation method is without doubt most generally applicable, highly reliable, a n d sufficiently accurate for all usual requirements. T h e use of a solvent not miscible with water offers several advantages, t h e most important of which is prevention of frothing. I t s principle disadvantages concern details of operation, and t h e modified procedure described in the present paper has been found t o obviate most of these difficulties. This method of determining water has been used in a 1 A common procedure is t o place a measured volume of oil in a suitable graduated tube, dilute with naphtha, and spin in a centrifuge. The water is driven to the bottom of the graduated tube and its volume can be read directly.
