July, 1928
INDUSTRIAL A N D ENGINEERING CHEMISTRY
769
Persulfate Method for Chromium Plus Vanadium in Chrome-Vanadium-Tungsten Steels' Modified Procedures H. H. Willard and Philena Young DEPARTMENT OF
CHEMISTRY, UNIVERSITY OB
T HAS been shown in a previous articleZ that a large quantity of tungstic acid, if dissolved in sodium hydroxide, may be added to an acid solution without precipitation, provided that a sufficient amount of ferric salt is present, and in this soluble form causes no interference in the bromate oxidation method for vanadium if the end point in the vanadic acid titration with ferrous sulfate is determined electrometrically. It is obvious that tungsten, thus kept in solution, would probably not interfere either in the persulfate or permanganate oxidation methods for chromium in chromevanadium-tungsten or in chrome-tungsten steels. The persulfate method3 was chosen as the simplest and most commonly used, and in the procedure which follows the directions for chromium in the presence of tungsten are giren.
I
MICHIGAN, A N N ARBOR, MICH.
5' C., add 25 cc. of ice-cold sulfuric acid, sp. gr. 1.5, and titrate the vanadic plus chromic acids electrometrically with 0.1 N ferrous sulfate. In Table I, 1.0- to 1.5-gram samples were used. Table I-Chromium in Chrome-Vanadium-Tungsten SteelsPersulfate Oxidation, Electrometric End Point CHROMIUM STEEL Per cenf B. of S. No. 50 (3.61% Cr) 3.60, 3.55, 3.60 3 4 10, 4 11 4 3 16, 3 16 5 2.83, 2 83
Use of Indicator
Diphenylbenzidine may be used to determine the end point in the titration of yanadic plus chromic acids with ferrous sulfate in such steels, though the procedure is somewhat Electrometric End Point longer, provided that the tungstic acid is completely removed, as a very small amount of it interferes with the deRecommended Procedure velopment of the indicator color. As tungstic acid is not (1) CHROMIUM IN CHROME-VANADIUM-TUNGSTEX STEELS precipitated quantitatively in a sulfuric acid solution, nitro-A 1- to 1.5-gram sample is convenient. Place it in a 400-cc. hydrochloric acid must be used, and the filtrate from the beaker, add 40 cc. of water and 10 cc. of sulfuric acid, sp. gr. tungstic acid evaporated with sulfuric to fumes to remove all 1.83. Warm gently until there is no further evolution of hydrochloric acid which would interfere in the oxidation of hydrogen. Boil over a free flame until salts begin to separate chromium. The amount of chromium in the tungstic acid is in order to assist in breaking up carbides, dilute to about ordinarily small and may be very easily estimated by fusing 50 cc., and boil until the salts have dissolved. To the boiling the tungstic acid with sodium carbonate, extracting with hot solution add nitric acid, sp. gr. 1.42, drop by drop, until water, filtering, and matching the color of the solution thus the violent action is over, then 5 to 6 cc. excess. Boil down obtained with an alkaline solution of equal volume to which until tungstic acid separates out in considerable amount; standard dichromate is added. For very accurate work it is dilute to 60 to 70 cc. with hot water, let settle, and filter off necessary to estimate the amount of vanadium in the tungstic the tungstic acid, using hot 1 per cent sulfuric acid to transfer acid, especially if the steel is high in tungsten and vanadium, the precipitate to the filter. Place a 150-cc. beaker under for in such a case considerable vanadium may be retained by the funnel, puncture the paper, and wash through most of the tungstic acid. The procedures for complete removal of the tungstic acid with water. Dissolve the remainder of the the tungstic acid, for the titration of vanadium plus chromaterial on the filter with hot 4 per cent sodium hydroxide. mium with ferrous sulfate using diphenylbenzidine as indiAdd 5 to 10 cc. more to the original beaker to dissolve any cator (which is the same as that for vanadium with ferrous precipitate which adheres to the glass, using only about 15 cc. sulfate), and for the colorimetric estimation of vanadium in in all. To the main filtrate from the tungstic acid, contained tungstic acid are given in the previous article.2 Furman,' in a 600-cc. beaker, add 1 gram of iron in the form of ferric who used diphenylamine as an indicator in this same titration, alum, and, stirring constantly, pour in the sodium tungstate states that the silver chloride must be filtered off as no definite which will contain some ferric hydroxide in suspension. end point can be obtained in its presence. The authors (It is convenient to have on hand a stock solution of ferric have found that silver chloride does not interfere at all with ammonium alum of such strength, 345 grams of the alum the sharpness of the color change of diphenylbenzidine a t per liter, that 25 cc. of the solution will contain 1 gram of Fe the end point, and its removal is therefore unnecessary. and 1 cc. HzS04, sp. gr. 1.83.) A clear solution will result Though the electrometric rather than the indicator method and the liquid should be diluted at this point to 300 cc. Heat for the end point is preferable in analyzing steels containing to boiling, add 10 cc. silver nitrate containing 2.5 grams tungsten, such is not the case with chrome-vanadium steels. Agn-03 per liter, and 5 grams of ammomium persulfate. If The procedure given below for the indicator method of titratno permanganate tinge appears in the solution on boiling, a ing chromic plus vanadic acids is rapid and accurate, and little more persulfate must be added. This difficulty will not the color change a t the end point is very sharp. The electrobe encountered if the correct amounts of sulfuric and nitric metric titration after a persulfate oxidation has been deacids, as indicated above, are used. Boil the solution for scribed by Kelley and C ~ n a n tand , ~ is very satisfactory if the 10 minutes to decompose the excess persulfate. Add 5 cc. solution is cooled to 5-6" C. at the time of titration. Table 1: 3 hydrochloric acid to reduce permanganate and boil I1 shows the results for chromium in chrome-vanadium steels. vigorously for 10 minutes to remove chlorine. Cool in ice to Recommended Procedure Received March 24, 1928. Willard and Young, IND.END.CHEM.,20, 764 (1928). *Re!!ey and Conant, Ibid., 8, 719 (1916); Lundel!, Hoffman, and Bright, Ibid., 16, 1064 (1923). 1
9
(2) CHROMIUMIN CHROME-VANADIUM STEELS-A 2gram sample is convenient when the steel contains not more 4
IND.ENG. CHBM., 17, 314 (1925).
llian 2 per cent chromium; for snrnples of larger clironiium wntent a 1-gram sample is suitable. Place it in a BOO-cc. beaker, add 15 ce. of water, 15 ec. ortlviphosplioric acid, sp. gr. 1.37, and run in a measured volume of sulfuric acid, sp. gr. 1.83, from a buret. i\llow 1.5 cc. of acid for ea.ch gram of steel and 3 ec. excess. After the steel has been completely decomposed, boil until a considerable quantity US salt, separates out, in order to assist in decomposing carbides. Dilute with 20 cc. of water and heat until tlie salts !iai-e dissolved. Add nitria acid, sp. gr. 1.42, drop by drop, l o the hot liqnid until the viulcnt oxidation of ferrous sulfate is oyer ( 2 t,o 3 cc. of acid are sufiicient; aroid any appreciable excess). Roil the solution to destroy oxides of nitrogen, dilutc to 150 cc., heat tu boili~ig,add 10 ce. of silver nitrate containirig 2.5 grams AgXOa per liter, and 1.5 grmu OS amiiiuniurri persulfate. If no perrriangnnate tinge appears in tlic solution on boiling, add inure ammoninm persulfate. In no case will more then 2 to 2.5 grams be required if samples of tlie weights suggestcd above are used. Boil the solution for 10 minutes t o decompose the excess persull'ate, dilute to 200 cc., heat to boiling, add 5 ce. of 1:shydrocliloric acid to reduce permanganate, and boil vigorously for 10 minutes to r m o r e elili~ri~ie,Add to tlre solirtion at rooin temperature
that quantity of crystdized sodium acetate ~vhiehis required to react with the excess of sulfuric acid used in dissolving tho steel. (1 cc. of concent,rated sulfuric acid is equivalent t o 4.8 grams of sodium aceiate trihydrate.) As soon as this has dissolred, add 0.6 to 0.8 cc. of 0.1 per cent diphenylbenaidine solution (prepared by dissolving 0.1 gram of the reagent in 10 cc. of coneent.rated suliuric acid and diluting this with '30 cc. of glacial acetic acid), nlloiv 5 miniitcs for the color t o develop, and titrate with 0.05 .V ferruiis snllilte. The corraction to be applied for the im1ic:itor is added to tlie volume of ferrous sulfate and :mounts to 0.015 cc. of 0.05 .\ ferrous sulfate per 0.1 cc. of indicator. In Table 11, 1.5- to %gram sarnplcs were u w l .
li.
Of
s. so. a0
!,',(1.02,A
cri
Per ceni I . 03 l.OZ
1.04 1
II.
Of
s. s o . 72
2.31 3.34 L 34
(0.81% Cd
Prr Cent 1 02
I.U%
1.04 2.33
2.34 0.Sl 0.01
Carbon Resistor Furnaces for Laboratory Use' 1% C. Kremers and L. U N I V B K S L T I OP
F. Yntema
IiLreurs, IlarraNr,
T
HMZE is ol'ten need for il furnace t!iat will mainbin a higher temperature than can be attained with nickelchromium alloy resistance units. Tungsten or molyhdenum may be employed, but the obvious disadvantage is that the units must be kept in a hydrogen or inert atmosphere. A plat,inum-vound furnace is expensive. A granular carbon resist,ance furnace offers the advantages of low initial cost; high t e m p e r a t u r e s , limited only by the n a t u r e of the refract o r i e s employed; and ease of operation and r e p l a c e m e n t of the h e a t i n g u n i t . Thc principle may be applied to tlie construction of either a pot or a tube furnace. For a pot furnace two c o 11 ccn i r i c alunduni tubes 12 incheslong (1) lire mounted vertically on a graphite bar ( 2 ) and the space between Figure I--Diagram of Furnace the tubes is filled to within 4 inches of t.he top, with a '/a- t o '/&inch granular carbon (3). . A grapiiite collar (4), to which one of t.he power leads @),isattached, makes contact with the granular graphite. The other lead (5) is attached to the lower end of the graphite bar. The graphite bar rests on an iron plate suspended from the *Received March 20, 1928. Presented before the Division of Iildustrial and Engineering Chemistry at the 75th Meeting of the Amrticim Chemical Society, Sf. Louis, Mo., Apr3 10 t o 18,1928.
ILL
b m holding t,hc insulation, ami an iron c d I u with wings (ti) rests on the graphite collar making contact vith the granulnr carbon. By changing the pressure on the collar the coliductivity of the furnace may be varied. In order to reenforce the outer of the two alundum tulles, an iron pipe ( i ) 2 inches greater in diameter is supported externally around the tube and the soace filled with alundum cement. The upper and lower enis of the box may conv e n i e n t l y be made of c i r c u l a r disks of transite and the side wall of heavy galvanized sheet iron. The insulation is an asbestos-Sil-o-Cel inixture. The furnace is best ruii with a step-doan transformer, with a variable s e c o n d a rg winding. A furnace m a d e of a l u n d u n i tubes of 3 and 5 inches i n s i d e diameter, respectively, r e q u i r e d 50 volts on the secondary t' give an inFigure >Furnace and Step-Down put of 3 kilowatts, by 'rransformer. ~ u r n a c c Censtrurfed of 3a n d 5-Inch Tubes a hmperature of 1000" C . may be reached in less than 1hour. Wlien that temperature is reached 30 or 40 volts on the secondary will hold it constant. A furnace constructed of 1.5- and &inch tubes requires proportionally less power to operate. A tube furnace vas made with a 2-inch outer tube 12
