Routine Colorimetric Determination of Titanium in Chromium Steels Separation of Titanium and Chromium by Perchloric Acid LOUIS SILVERMAN United States Navy Chemistry Laboratory, Munhall, Penna.
T
of the crucible were treated with sulfuric and hydrofluoric acida, the former in great excess. The crucible was heated till heavy fumes of fulfuric acid appeared. After being cooled and diluted with water, the crucible contents were transferred to a glass cylinder, and titanium was determined by the peroxide method, without filtering off the insoluble columbium and tungsten. METHODB. The steel (2.00grams) was decomposed by mixed hydrochloric and nitric acids, then heated to fumes with perchloric acid. After cooling, 100 cc. of watei were added, chlorine was boiled out, and the insoluble matter was filtered off on paper. Titanium was determined in the residue, as in Method A; and in the filtrate, the solution was evaporated to perchloric acid fumes, the beaker cooled, and the titanium filtered from the chromium and washed with “cold perchloric acid” through a fritted-glass crucible, and then determined by the peroxide method. METHODC. In the third method, used merely for comparison purposes, the steel (2.00 grams) was decomposed by mixed hydrochloric and nitric acids, fumed with perchloric acid, and cooled, and the titanium was filtered and washed, in the manner proposed for steel not containing columbium, tungsten, or molybdenum.
HIS paper maps a simple routine for the determination of titanium in certain steels, in which the sample is dis-
solved in mixed hydrochloric-nitric acids and fumed in perchloric acid, and the titanium is filtered off cold. The inadequacies of the method for certain steels are pointed out. Chromium trioxide is quantitatively insoluble in cold 69 to 72 per cent perchloric acid (4), and it can be shown that small amounts of titanium are soluble in the reagent. It is common practice to determine titanium (1 to 6 mg.) colorimetrically, using the peroxide method. In steels, titanium can be separated from chromium, nickel, and ferrous iron by cupferron (8), then determined colorimetrically. Titanium may also be separated from chromium b y the lead perchlorate method (6). The method proposed in this paper can be applied to steels of the 18-8 chromium-nickel type (Bureau of Standards No. 121) and to the 5 per cent chromium-0.5 per cent molybdenum type. The procedure is relatively simple. T h e steel is dissolved in aqua regia and evaporated to fumes in excess perchloric acid. When cold, the insoluble chromium trioxide is filtered off, and titanium is determined in the filtrate. Vanadium can be separated in a similar manner ( 4 ) , and the problem is now being further investigated.
Results Table I shows results with steels of the titanium-chromiummolybdenum type and the (18-8) titanium-chromium-nickel type, using Bureau of Standards KO.121 as the ultimate standard; the titanium was separated b y the cupferron method. Comparison of the cupferron method with the proposed method shows that all the titanium may not be recovered with five perchloric acid washings, even though this was accomplished with steel sample 3. Table I1 concerns two steels of the titanium-columbiumtungsten-molybdenum-chromium type, and the ultimate standard is Bureau of Standard Steel KO.121. The figures given in these tables were obtained by a routine analysis of the group. KO “corrections” were made by using
Experimental TITANIUM-CHROMIUM STEELS. One-gram samples of the steels were dissolved in mixed nitric-hydrochloric acids, and 30 cc. of i 2 per cent (technical) perchloric acid were added t o each beaker, The beakers were heated until chromium was oxidized to chromium trioxide. then several minutes more, and were then removed to a cool place in the hood. At the same time a wash solution of perchloric acid was made by heating 72 per cent perchloric acid to fumes and cooling to room temperature. For filters, Gooch crucibles with asbestos pads or fritted-gIass crucibles (Pyrex M style) were used in conjunction with a bell-jar type of suction apparatus. The cold steel solutions were filtered through dry crucibles. When using fritted-glass crucibles small glass stirring rods were used to agitate the insoluble red residues. Each crucible was washed five or more times with 5-cc. portions of the cold wash solution. The apparatus was disassembled, and the residues in the crucibles were n-ashed into the original reaction beakers with water. An additional 15-cc. portion of 72 per cent perchloric acid was added to each beaker of chromium trioxide, and the contents were heated to fumes of perchloric acid. After cooling, the solutions were filtered through dry crucibles, and the precipitates were washed with wash solution. Titanium was determined in each filtrate in the customary manner, by diluting with water, adding 5 cc. of 3 per cent hydrogen peroxide, and comparing with a standard titanium solution. For these comparisons, the cupferron method was used with Bureau of Standards Sample KO.121 (0.391 per cent titanium). TITAXIUM,COLUMBIUM, TUNGSTEN, MOLYBDENUM, AND CHROMIUM STEELS. Three procedures were investigated. METHODA. The steel (2.00 grams) was decomposed by hydrochloric acid, and the titanium precipitated by cupferron. After filtering, the residue is expected to contain titanium, columbium, and tungsten. The paper and contents were then treated with nitric (5:)and perchloric acids, and the beaker was heated to fumes of perchloric acid. After cooling and diluting with water, the insoluble matter was filtered off. Titanium was determined in the filtrate by the peroxide method. The residue was ignited in a platinum crucible and cooled. Then the contents
TABLE
I. DETERMINATION O F TITANIUM
Type of Steel
Titanium, Cupferron Method
Titanium, Proposed Method 1st 2nd filtrate filtrate
%
%
%
Nu. 1, 5 7 , Cr, 0.6% 310,0.1% C 0.50 0.49 0.02 No. 2, same type 0.48 0.50 0.02 No.3, 18% Cr, 10% Ni 0.34a 0.34 0.0 No. 4, B. of 8. No. 121 (0.394% Ti) 0.39b 0.37 0.02 a Manufacturer, 0.337,. b Ysed as color standard for cupferron method and for proposed method.
T.4BLE
11. COMPARATIVE RESULTSWITH TITAXIUM-COLUM-
BIUM-TTXGSTEN-MOLYBDENUM-CHROMIUM STEELS Titanium Sample 1 Sample 2
Method A T i in cupferron precipitate Soluble in dilute perchloric acid Recovered from residue Method B T i soluble in dilute perchloric acid T i recovered from residue Method C Ti, first filtrate Ti, second filtrate Manufacturer’s results
79 1
%
%
0.38 0.08
0.34 0.05
0.40 0.03
0.30 0.03
0.38 0.02 0.45
0.35 0.02 0.35
192
INDUSTRIAL AND ENGINEERING CHEMISTRY
results of analyses with more exhaustive washings and attentions other than routine.
Discussion The proposed method for the determination of titanium in steel is of special advantage for analysis of steels of the 5 per cent chromium-0.6 per cent molybdenum type. This alloy dissolves slowly in strong hydrochloric acid, but rapidly in mixed hydrochloric-nitric acids. The test solution is free of color, while with the cupferron method chromium, as carbide, remains with the titanium cupferron precipitate. A disadvantage is the fact that all the titanium may not be contained in the filtrate from the chromium trioxide (0 to 0.02 per cent). I n the case of the (18-8) steels containing titanium, the advantage of speed in solution of the steel is no longer marked. Some nickel will be found in the titanium filtrate, but this error is easily overcome by adding nickel sulfate to the test solutions, or standard, previous to peroxidation. When the alloying elements columbium, tungsten, molybdenum, chromium, and nickel are also present, difficulties are encountered. When the cupferron separation was used (Method A, Table 11) considerable titanium was left in the residues. Method A does not use an oxidizing acid attack. Method B involves an oxidizing acid attack, a perchloric acid attack, and a water extraction. Method C is proposed for titanium-chromium-molybdenum steels, but with a double extraction. Of the three procedures used, the best seems t o be Method A, unless the sample will not respond to hydrochloric acid attack. Method B is unusually long, while Method C may require three or more filtrations. I n a columbium steel, the columbium carbide is resistant to attack of aqua regia and fuming perchloric acid (3); and if a titanium-columbium steel is similarly treated most of the titanium is to be found in the filtrate (Method B); but if titanium and columbium salts are fumed with perchloric acid (1) a separation cannot be made. The apparent deduction is that columbium carbide is an insoluble entity and has
Vol. 14, No. 10
no solvent action on titanium (whatever the latter’s chemical state may be). I n contrast, tungstic and molybdic oxides can be expected t o retain small amounts of titanium. For this reason, one must expect to determine titanium in these alloyed steels by recovering the titanium in a soluble and in an insoluble portion of the sample, as indicated in Table 11.
Proposed Procedure for Titanium in ChromiumMolybdenum-Nickel Steels Transfer 1 gram of steel to a 150-cc. beaker, add 15 cc. of mixed acids (300 cc. of hydrochloric acid, 100 cc. of nitric acid 400 cc. of water), and warm to complete solution. Add 30 cc. oi 70 to 72 per cent (technical) perchloric acid, heat till red chromium trioxide ap ears, and several minutes more, and cool. Prepare a wasg solution by heating perchloric acid in a small beaker until heavy fumes appear. Cool. Use a bell-jar type suction or a filterin desiccator apparatus. Filter through a Gooch crucible with an as%estospad, or better, a Pyrex M filtering crucible. Wash with five or six 5-cc. portions of wash solution, using a small glass rod to agitate the precipitate, Catch the filtrate and washings in a 150-cc. beaker. Evaporate the fltrate to 10 to 15 cc., cool, add 25 cc. of water, and boil out chlorine. Dilute with 5 per cent sulfuric acid solution. Add 5 cc. of 3 per cent peroxide, and compare with a standard titanium solution. STANDARD. Bureau of Standards No. 121 (0.394 per cent titanium) may be prepared as above, except that the insoluble chromium trioxide residue is redissolved in water, fumed in perchloric acid, and cooled, and the washings are added to the first filtrate. Dilute to 197 cc. with 5 per cent sulfuric acid. Vanadium, if present, accompanies titanium. Decolorize the yellow titanium solution with hydrofluoric acid. A residual brown color indicates vanadium.
Literature Cited (1) (2) (3) (4)
Cunningham, T. R., IND.ENG.CHEM.,ANAL.ED.,10, 235 (1938).
Silverman, Louis, Chemist-AnaZyst, 23, No. 3, 3 (1934). Silverman, Louis, IND.ENG.CHEM.,ANAL.ED., 6, 287 (1934). Smith, G. F., “Mixed Perchloric, Sulfuric, and Phosphoric Acids and their Applications in Analysis”, p. 8, Columbus, Ohio, G . Frederick Smith Chemical Co., 1942. (5) Ibid., p. 42. (6) U. 9. Steel Chemists Committee, “Sampling and Analysis of Carbon and Alloy Steels”, p. 6 2 , New York, Reinhold Publishing Corp., 1938.
Removal of Adsorbents from Chromatographic Tubes JOHN TURKEVICH, Frick Chemical Laboratory, Princeton University, Princeton, N. J.
T
HE contents of a developed chromatographic tube are usually removed b y laying the tube horizontally on a
piece of paper and extruding the contents from the top of the tube b y pressing the bottom of the tube with a wooden stick or the l i e (1, 2). The method, however, involves considerable skill, since the material in the tube must be neither too dry nor too wet to slip out easily. The following method has been found particularly effective and has reduced the breakage of chromatographic tubes to a negligible point in undergraduate courses. After development of the chromatogram the solvent is allowed just to disappear from above the surface of the tube and suction is discontinued. The tube is placed horizontally on a table with paper to receive the contents of the tube. A rubber stopper of a size to fit the bottom of the chromatographic tube is attached to a compressed as line and by means of this stopper gas pressure is applied gentfy and periodically to the bottom of the tube The flow of gas t h o u h the tube not only presses against t i e solid contents of the tu%e but also dries the adsorbent to such a state that the contents slip out easily.
The pressure must be regulated either by a valve in the gas line or by placing the rubber stopper sideways in the bottom of the tube thus permittin part of the gas to leak through the space between the stopper a n i the tube. There is first a period of drying and packing the adsorbent, which normally requires greater pressure than the second period of extrusion. Consequently the pressure is applied intermittently for short periods, noting the results of each successive application. The column will finally start to move, often in segments. The pressure and the time of a plication must be decreased and the process is continued until t f e contents are pressed out of the tube. If the ressure is not controlled either by a valve or by periodic appEcation of the rubber stopper for short times to the bottom of the tube, the adsorbent will shoot out of the tube across the laboratory. If the adsorbed material is sensitive to oxygen, the gas may be oxygen-free nitrogen or carbon dioxide.
Literature Cited (1) Strain, “Chromatographic Adsorption Analysis”, p. 46, New York, Interscience Publishers, 19?,2. (2) Zechrneister and Cholnoky. Die chromatographische Adsorptionsrnethoden”, 2nd ed., p. 76, Berlin, Julius Springer, 1938.
