V O L U M E 2 7 , N O . 6, J U N E 1 9 5 5 the separations obtained for a mixture of five acids. The mixture of acids was passed through the anion exchange column prior to separation of the silica gel column. Similar separations were obtained when a tobacco sample was chromatographed to determine whether the solvent schedule was satisfactory for that particular concentration of tobacco extract. -1 typical analysis of grade P O 4 bright tobacco is shown in Figure 2. Analysis of several other types of tobacco, not presented here, showed similar separations using this solvent schedule. When a mixture of five acids (fumaric, succinic, oxalic, malic, and citric) was placed on a silica gel column prepared from unpurified Davison Grade 70 gel, only four acids were recovered, oxalic acid remaining on the column. I n another experiment unpurified silica gel was suspended in a saturated oxalic acid solution, following which excess oxalic acid was iemoved by passing the appropriate developer through the column. When a known amount of oxalic acid was added to this column, almost 90% was recovered. On the other h a d , purification of silica gel with hydrochloric acid, as described above, permits 96% recovery of oxalic acid. Use of 8-Quinolinol. The addition of 8-quinolinol to solvent bystenis used in paper chromatography should have wide application because of its fluorescence when exposed to ultraviolet light. I n the cases where 8-quinolinol moves with the solvent front, it facilitates the determination of the position of the front in volatile systems or in other systenls where the front is not readily distinguishable. Another advantage is that less time is consumed, as papers do not have to be dried and sprayed; instead they can be observed immediatelj after removal from the chamber. Complex mixtures could be identified in this manner on a single chromatogram-for example, both sugars and acids would show up under ultraviolet light; subsequently the paper could
931 be sprayed with an acid-base indicator to distinguish acid spots from those of sugars. ACKNOWLEDGMENT
The authors wish to thank the Research and Development Department of Philip Morris, Inc., for assistance during this study and for permitting the publication of this paper. LITERATURE CITED
Bryant, F., and Overell, B. T., Biochim. et Biophys. Acta, 10, 471 (1953).
Buch, hI. L.. Montgomery, R., and Porter, W. L., ANAL.CHEM., 24,489 (1952).
Bulen, W. -4..Varner, J. E , and Burrell, R. C., I b i d . , 24, 187 (1952).
Frohman, C. E., and Orten, J. AI., J . Biol. Chem., 205, 717 (1953).
Frohman, C. E., Orten, J. lI.,and Smith, A. H., Ibid., 193, 277 (1951).
Houston, F. G , and Hamilton, J. L., z k i a ~CHEM., . 24, 415 (1952).
Isherwood, F. A . , Biochem. J., 40, 688 (1946). Lugg, J. W. H., and Overell, B. T., Australian J . Sci. Research, 1 , 98 (1948).
llarshall, L. bl., Donaldson, K. O., and Friedberg, F., ANAL. CHEM.,2 4 , 7 7 3 (1952). Marshall, L. AI., Orten, J. ll.and , Smith, A. H., J . BioZ. Chem., 179, 1127 (1949).
llarvel, C. S.,and Rands, R. D., Jr., J . Am. Chem. Soc., 72, 2642 (1950).
Kumerof. P.. and Reinhardt. K.. ANAL.CHEM..2 5 . 3 6 4 (1953). Phillips, hI., and Bacot, A. lI.,J . dssoc. Ofic. Agr. Chemists, 36, 123 (1953).
Ihid., p. 504. Roberts, E. J., and Martin, L. F., ANAL.CHEM.,26, 815 (1954). Stark, J. B., Goodban, A. E., and Owens, H. S., Ibid., 23, 413 (1951). RECEIVED for review July 27, 1954. Accepted February 19. 1955. Presented before t h e Eighth Annual Tobacco Chemists’ Conference, Richmond, Va., November 11, 1934.
Determination of Uranium in Uranium Concentrates Use of Ethyl Acetate R. J. GUEST and J. B. ZIMMERMAN Radioactivity Division, Mines Branch, Department of Mines
A method is described for the determination of uranium in high grade uranium material. Uranium is separated from contaminants by means of an ethyl acetate extraction using aluminum nitrate as a salting agent. .4fter the uranium has been stripped from the ethyl acetate layer by means of water, colorimetric determination of the uranium is carried out by the sodium hydroxide-hydrogen peroxide method. The procedure is accurate, rapid, and easily adaptable to routine work.
C
LASSICAL methods of determining uranium usually employ a preliminary chemical separation before final determination of the uranium volumetrically or colorimetrically. Recently, fluorophotometric methods have been developed ( 2 1 , 23) which permit determination of uranium on many types of material without prior chemical separation. With some material, however, a preliminary chemical separation is necessary before a fluorophotometric method can be applied. Material high in uranium content (greater than 5% uranium oxide) ueually requires less preliminary treatment than material of low uranium content but, nevertheless, a separation is almost always necessary before a volumetric or colorimetric finish can be employed. Usually, one or more of such standard procedures as
& Technical Surveys, Ottawa, Canada
hydrogen sulfide separation, cupferron separation, ammonium or sodium carbonate separation, and mercury cathode separation are required (16). Such separations are, of course, time-consuming. A fluorophotometric method for uranium in high grade material (23) is in use in this laboratory. A preliminary separation ie seldom required, but there is difficulty in consistently attaining a precision of better than 1.5 to 2.0%. I n some cases a precision greater than this is required. A number of organic solvents for uranyl nitrate have been reported (19), and separations based on the use of some of these solvents have been described ( d , 18). I n this laboratory diethyl ether-nitric acid has been used in conjunction with the cellulose column (16). This method, while not as lengthy as other separation procedures, possesses all the hazards involved in using ether. The extraction of uranium by ethyl acetate from a nitric acid solution heavily salted with aluminum nitrate was first described by Grimaldi and Levine (6). Uranium was finally determined fluorimetrically or colorimetrically. Various fluorimetric applications using this extraction procedure have been described (6, 10, 22). I n an attempt to obtain greater precision and accuracy on uranium assays in uranium concentrates, the possibility of developing a method using this ethyl acetate extraction and a colorimetric or volumetric finish was investigated.
932
ANALYTICAL CHEMISTRY
Volumetric methods for uranium which have been used in this laboratory employ potassium dichromate as a titrant after reduction of uranium(V1) by means of a Jones reductor ( 7 , IT), lead reductor ( I ) , or stannous chloride solution ( 1 1 ) . A number of colorimetric methods for uranium have been described ( 3 , 4, 17). B method which has been used successfully in this laboratory is the sodium hydroxide-hydrogen peroxide method. As this method is rapid and will detect milligram quantities of uranium, its applicability after an ethyl acetate extraction vias investigated. REAGENTS AND APPARATUS
Reagents. Ethyl acetate (Merck, reagent grade). ALUMINUNNITRATE SALTING SOLUTION.Place approximately 450 grams of reagent grade (3Iallinckrodt) aluminum nitrate [.41(NOa)3.9H20] in a 600-ml. beaker and add 25 to 50 ml. of distilled water. Cover the beaker and heat the mixture on a hot plate. If a clear solution does not result after 5 t o 10 minutes of boiling, add 20 ml. of water, and continue the boiling for 5 more minutes. Repeat this step until a clear solution is obtained after boiling. Remove the cover glass and concentrate the solution by boiling until a boiling point of 130" C. is reached. Cover the beaker with a watch glass and either transfer the solution t o a constant temperature apparatus or keep the solution warm, finally heating t o just under boiling before use. If the solution is allowed to cool t o approximately 60' C., recrystallization of aluminum nitrate will take place. It is necessary. therefore, t o dilute the salting agent solution by about one third in order t o prevent recrystallization if the solution cools t o room temperature. Accordingly, if the solution is to stand overnight, add 35 ml. of distilled water per 100 ml. of salting agent solution, mix well, and cover. If the salting agent solution is to be stored, the following procedure has been found convenient. Adjust the solution t o the proper concentration (boiling point, 130' C.) and transfer t o a 100-ml. three-necked reaction flask set on a heating mantle. Adjust the heating so that t h e temperature of the Glution is kept a t about 110" C. I n one of the necks place a water condenser, in another neck a thermometer, and in the third neck a removable ground-glass stopper. This third neck is utilized for pipetting the salting agent solution. ALUMINUM NITRATERASH SOLUTION.Add 100 ml. of aluminum nitrate salting solution (boiling point, 130" C.) t o 73 ml. of distilled water and 4 ml. of concentrated nitric acid. Apparatus. Beckman D U spectrophotometer. Heating mantle. Three-necked reaction flask (1000 ml.). Water condenser. No. 0 rubber stoppers. Boil twice in ethyl acetate before use. Sixty-milliliter separatory funnels (Squibb, pear-shaped). PROCEDURE
Sample Dissolution. Place an appropriate quantity (1 to 5 grams) of the sample in a tared weighing bottle, stopper the bottle, and weigh the bottle and contents immediately. Carry out a moisture determination on a separate sample if uranium is t o be calculated on a dry weight basis. Bring the sample into solution in one of three ways: (1) nitric acid treatment, ( 2 ) .multiacid treatment, or (3) sugar carbonsodium peroxide fusion. For the nitric acid treatment, dissolve the sample in a suitable quantity of nitric acid and transfer the solution and insoluble residue into an appropriate volumetric flask and make up to volume. Regulate the dilution so that the aliquot chosen for extraction will contain between 10 and 30 mg. of uranium oxide if the final dilution for the colorimetric finish is to be 250 ml. Adjust the acidity of the sample solution t o about 5y0 in nitric acid. If nitric acid treatment is not sufficient, treat the sample with hydrochloric acid, nitric acid, perchloric acid, and finally sulfuric acid. If necessary, add a few milliliters of hydrofluoric acid. Fume the sample t o dryness and leach the residue with nitric acid, finally transferring the solution and residue t o an appropriate volumetric flask and adjusting to 5% in nitric acid as in the single acid treatment. If the sample is refractory, use t h e sugar carbon-sodium peroxide fusion method described b y Muehlberg ( 1 3 ) . After dissolution of the sample in this manner, transfer the acidified solution to an appropriate volumetric flask and dilute so that the final solution is 5y0 in nitric acid. Aliquot solution samples directly or dilute as required for an ethyl acetate extraction. If the sample is aliquoted directly for
an extraction, add 5 drops of concentrated nitric acid per 5-ml. aliquot of sample and standards before extraction. Where samples are diluted before aliquots are taken for extraction, adjust the acidity so t h a t t h e final volume is 5% in nitric acid. Ethyl Acetate Extraction. Place an appropriate aliquot (usually 5 ml.) in a 60-ml. separatory funnel, the stopcock of which has been lubricated with silicone grease. Add, by means of a graduated pipet, 6.5 ml. of aluminum nitrate s3lution per 5 ml. of sample solution. The aluminum nitrate salting solution should be added while hot (above 110' C.). Cool the solution to room temperature and add 20 ml. of ethyl acetate. Stopper the separatory funnels with pretreated rubber stoppers. Shake the mixture for 45 to 60 seconds. Occasionally crystallization will take place in the separatory funnel near the stopcock. I n such a case place the loIver part of the separatory funnel in a beaker of hot water until the solidified portion dissolves. After the layers have separated, drain off the aqueous (lower) layer. Occasionally a cloudiness will appear a t the boundary of the aqueous and organic layer. This cloudy portion should not be drained off. Add 10 ml. of aluminum nitrate wash solution t o the funnel and again shake the mivture for 45 to 60 seconds. Drain off the aqueous layer, once again being careful to retain the cloudy portion a t the boundary in the funnel. Rinse inside the stem of the separatory funnel with a stream of water from a wash bottle. Water Stripping of Uranium from Ethyl Acetate Layer Followed by Sodium Hydroxide-Hydrogen Peroxide Colorimetric Finish. Add 15 ml. of water t o the separatory funnel containing the ethyl acetate, stopper the flask, and shake the mixture for about 1 minute. After washing off the stopper with water, drain the aqueous layer into a volumetric flask of suitable size and wash the separatory funnel and ethyl acetate layer 4 or 5 times with 5-ml. portions of water by means of a wash bottle. Combine the aqueous fractions. Add enough 20% sodium hydroxide solution (w./v.) to neutralize the solution and dissolve any precipitated aluminum hydroxide, then add 10 ml. in excess per 100 ml. of final volume. Add 1 ml. of 30y0 hydrogen peroxide per 100 ml. of final volume and make up the volume to the mark with distilled water. Read the absorbance after 20 minutes on the Beckman DU spectrophotometer a t 370 mp against a reagent blank, using 1-em. Corex cells and a slit width of 0.2 mm. Compare the absorbances of the samples against the absorbances of standard uranium solutions which have been carried through the procedure a t the same time. Choose the standards so that they cover the range into which the samples are expected to fall, using a ratio of one standard to six samples. I n practice it is customary to work between the limits of 10 and 30 mg. of uranium oxide. This is arranged by estimating the required sample !wights and diluting and sampling accordingly. The final volume for colorimetric reading is usually 250 ml. Double Extraction of Uranium with Ethyl Acetate Followed by Application of Differential Colorimetry. Uranium determinations requiring the highest accuracy may be carried out by a double extraction of uranium with ethyl acetate followed by the application of differential colorimetry as described by Hiskey and othcrs ( 8 , 9 , 1 2 , l 4 , 2 0 ) . I n a such case it is recommended that between 100 and 150 mg. of uranium oxide be eutracted, and a wave length of 400 mp be used during the colorimetric finish. The procedure described below has been found satisfactory. Extract an appropriate aliquot of the sample solution with 20 ml. of ethyl acetate as described above. Drat?- off the aqueous layer into a second separatory funnel containing 10 ml. of ethyl acetate. Stopper the funnels and shake the mixture for 45 to 60 seconds. Drain off and discard the aqueous layer. Add 10 ml. of aluminum nitrate wash solution to the first ethyl acetate extract, stopper, and shake the mixture for 45 t o 60 seconds. Drain o f f the aqueous layer into the separatory funnel containing the second ethyl acetate extract, stopper, and shake the mixture for 45 t o 60 seconds. Drain off and discard the aqueous layer. Combine the ethyl acetate fractions. Rinse the second separatory funnel with 20 ml. of m-ater, draining the washings into the separatory funnel containing the combined ethyl acetate fractions. Shake the mixture for 1 minute. Continue the water stripping as described above, collecting the fractions in an appropriate volumetric flask. Finish colorimetrically as described previously, allowing the strongly colored solution t o stand 1 t o 2 hours to ensure stability before reading as a fading effect of about 0.005 absorbance (optical density) has sometimes been noted on freshly prepared samples. Read the absorbance of the sample solution on the Beckman DU spectrophotometer a t 400 mp against a reference solution
V O L U M E 27, NO. 6, J U N E 1 9 5 5 which contains a known amount of uranium and has been carried through the extraction and color development procedure in the same manner as the sample. .Slso carry along other standards containing slightly higher and lower amounts of uranium than the sample. Determine the concentration of uranium in the sample either by the calibration-curve method or the correction method, as described by S e a l (14). If the amount of uranium in the sample is not knoivn, make a test run by taking an aliquot of the sample solution and assajing for uranium by the more rapid single extraction method. The standard solutions to be used can then be chosen according to the result obtained. Removal of Interfering Thorium. After an ethyl acetate extraction, strip the uranium in v-ater from the ethyl acetate and collect the uranium fraction in a 250-ml. beaker. Add enough 20% (w,/v.) sodium hydroxide solution t o neutralize the solution and redissolve precipitated aluminum hydroxide. Then add 10ml. excess of 20yo sodium hydroxide solution and 1 ml. of 30% hydrogen peroxide per 100 ml. of final volume. Filter the solution through an 11-cm. 411% filter paper (Whatman), collecting the filtrate in a volumetric flask of suitable size. Wash the paper and precipitate once with 5 ml. of a solution of 2 7 0 sodium hydroxide containing 0 1 nil. of 3oy0 hydrogen peroxide. Redissolve the precipitate by nashing the paper with 10 ml. of 10% nitric acid solution, collecting the washings in the original beaker. Keutralize the solution with 20% sodium hydroxide solution, and add 2 ml. in excess. Add 0.5 ml. of 3070 hydrogen peroxide, and filter off the precipitate on the original filter paper, washing as before and collecting the filtrates in the original volumetric flask. If the precipitate on the paper is colored yellow, repeat this step. Make t h e solution in t h e volumetric flask u p t o volume and read the absorbance on the spectrophotometer. Carry standards through t h e same procedure as the samples. EXI'ERIMEhTAL
Z , series of standard uranium solutions was made up by dissolving high purity uranium oiide (UIOa) of known composition in nitric acid and making the solution up to volume in volumetric flasks. The acidity of the solutions was 5% in nitric acid. Appropriate aliquots of these solutions were transferred to volumetric flasks and color development b a s carried out by the sodium hydroxide-hydrogen peroxide method.
100
90
80
70
0"m 60 LL
0
50 Y
>
8
40
8 30
20
10
0 I L . AI(NO3l3
SOLUTION (13OOC. B P I PER 5 M L . SAMPLE
Figure 1. Effect of concentration of salting solution on extraction of uranium by ethyl acetate Quantity of aluminum nitrate salting solution (130' C.) required per 5 ml. of sample solution. Standard uranium solution added as uranyl nitrate. 25 mg. UaOs per 5 ml. sample. 45 t o 60 seconds' shaking time with 20 ml. of ethyl acetate
933 Similar aliquots of the standard solutions \?ere extracted with ethyl acetate using aluminum nitrate as a salting agent. The effect on the extraction, of pH, shaking time, and varying amounts of salting agent was investigated. Following removal of the uranium from the ethyl acetate layer, the uranium content of the svnthetic samdes was determined bv the sodium hvdroxidehydrogen peroxide colorimetric method. Several colorimetric and volumetric finishes were briefly compared.
n
130
120
10 10
0
z
0
10
20
30
40
50
60
70
80
90
100
% RECOVERY OF U30g
Figure 2. Effect of concentration of salting solution on extraction of uranium by ethyl acetate Concentration of aluniinuni nitrate required t o salt 6 ml. of sample. Standard uranium solution added as uranyl sulfate. 25 mg. UaOsin 5-ml. sample plus 6 . 5 ml. of salting agent. Other conditions as in Figure 1
Extraction tests were carried out on synthetic solutions containing common contaminating ions and the results were tabulated. T h e method Jvas then applied t o actual samples and results were compared with known values obtained by other methods. Procedures for the removal of interfering thorium Tvere investigated. A comparison of two methods of removing uranium from the ethyl acetate layer was made. A procedure was developed employing a double extraction technique for extracting larger amounts of uranium, finally determining the uranium by an application of differential colorimetry. Extraction of Uranium with Ethyl Acetate and Effect of Salting Agent. A series of aliquots of standard uranium solutions such that 5 ml. contained from 5 t o 100 mg. of uranium oxide was extracted with 20 ml. of ethyl acetate. The quantity of aluminum nitrate salting agent required to give complete extraction of uranium was first determined. It is shown in Figure 1t h a t a t least 5.5 ml. of aluminum nitrate solution, so concentrated that it boils a t 130" C., is required to give complete extraction of 25 mg. of uranium oxide ( U 3 0 8 ) contained in 5 ml. of solution. The most convenient way found to express the amount of aluminum nitrate added was in terms of the boiling point of its concentrated solution. Figure 2 indicates that to extract completely 25 mg. of uranium oxide from 5 ml. of solution n-ith 6.5 ml. of aluminum nitrate salting solution, the boiling point of the salting solution must be a t least 126" C. I t can be seen from these figures that the amount of salting agent added is critical. The use of a ratio of 6.5 ml. of salting solution, boiling point 130" C., to 5 ml. of sample solution was adopted as standard procedure. Tests indicated that, under the conditions of the procedure described here, a t least 50 mg. of uranium oxide can be extracted quantitatively with 20 ml. of ethyl acetate. The relationship is linear and results are reproducible. With a procedure using a double extraction of the uranium with ethyl acetate, a t least 150 mg. of uranium oxide appear to be quantitatively extracted. Effect of Acidity on Extraction. Quantitative extraction of uranium was obtained on samples varying from 1.5 to ll.Oyoin nitric acid using the described procedure. It has been found
ANALYTICAL CHEMISTRY
934
convenient to use an acidity of 5y0 in nitric acid for most of the work carried out here. Effect of Shaking Time on Extraction. Little difference in uranium extraction was found under the conditions of the procedure when the mixture was shaken for periods of time varying from 20 to 60 seconds. The deviation in recovery of 25 mg. of uranium oxide using 6% acidity and shaking time of 20, 40, and 60 seconds was approximately 1%. FINAL DETERMINATION OF URANIUM COLORIMETRICALLY AND VOLUMETRICALLY
Although the sodium hydroxide-hydrogen peroxide colorimetric method ( 1 7 ) was used in most of this work for final determination of uranium after the extraction, a number of other knishes were tried briefly. Two colorimetric methods, one using thiocyanate ( 8 ) and the other using thioglycolic acid ( 4 ) , are more sensitive than the sodium hydroxide-hydrogen peroxide method. The latter method, however, has shown greater reproducibility and is preferred here. The technique of differential colorimetry (8, 9, l a ! 1.4, 20) was applied in conjunction with the sodium hydroside-hydrogen peroxide method. This method, although more time-consuming than the standard colorimetric finish, gives more reproducible results. The gain in precision was approximately twofold on samples analyzed by both methods. .1 wave length of 400 mp was used for the differential colorimetry technique. Standard volumetric techniques, such as titration n4th potassium dichromate after reduction of uranium(V1) to uranium(IV), were tried briefly. Reduct,ion was carried b y both the lead reductor method ( 1 ) and the, stannous chloride method (11). It was first necessary to add sulfuric acid and take the samples to fumes to remove nitric acid before using either reduction method. These volumetric met.hods are more lengthy than the sodium hydroxide-hydrogen peroxide method and do not appear to offer any advantages in precision. I t was decided, therefore, that because of its speed and reproducibility the sodium hydroxide-hydrogen peroxide method was the most promising of the methods t,ried for the filial determination of uranium after an ethyl acetate extraction. EXTRACTION OF IONS OTHER THAN URANIUM
I n order to determine whether or not a number of common contaminants likely to be found in high grade uranium material are extracted with ethyl acetate, solutions containing these contaminants were made up and appropriate aliquots were put through the extraction procedure. The organic extract was washed with a xash solution containing a ratio of 6.5 ml. of saltr ing agent to 5 ml. of 5% nitric acid solution. This step is necessary to compensate for mechanical retention of the contaminating ion during separation of the organic and aqueous layers. The amount of contaminant found in the ethyl acetate laver was then determined colorimetrically, or gravimetrically in the case of calcium and msgnesium. Of the ions tested, only thorium was
Table I. Extraction of Ions Other than Uranium by Ethyl Acetate Contaminant Vanadium Iron Molybdenum Copper Arsenic Phosphorus (as Pa061 Cobalt Calcium Magnesium Manganese .4luminum Thorium
Contaminant Added, Mg. 10 10 10
10 20 20
10 10 10
10 800 (approx.) as salting agent 3.8
Contaminant Found, Mg.
