Analyst, July, 1973, Vol. 98, j!@. 525-528 525 An Ultramicro-scale Method for the Determination of the Uranyl Cation BY GLENN PETER WOOD (Department of Chemistry, University of San Andres, La Paz, Bolivia) A method that requires the use of very simple equipment has been developed whereby the uranyl cation can be determined at the micro-scale and ultramicro-scale levels with high precision. The method involves the use of the catalytic effect shown by the uranyl cation on the photo-decolorisa- tion of a naturally occurring carotenoid-type pigment. The interfering effects of several common anions and cations are also discussed. ANNATTO is the red colouring matter of the seed of the plant Bixa Orellana L.l and is a mixture of natural pigments that have similar chemical structures, the principal constituent being labile bixin- R = H ; R'=CH3 Solutions of bixin in acetic acid undergo photochemical decolorisation when irradiated with short-wave ultraviolet light.As nitrogen-purged solutions of bixin do not lose their colour when exposed to ultraviolet light, the reaction is thought to be based on its photo- chemical oxidation. As the catalytic oxidation of naturally occurring pigments has been used2 to determine ultramicro-scale amounts of cations, it was decided to investigate the possibility of discovering a cation that might catalyse the photochemical oxidation of labile bixin. About twenty cations were tested but only three had a noticeable effect on the velocity of the reaction, and of these three the uranyl cation had a much more pronounced effect than the other two cations.Investigation of this catalytic effect revealed that the reaction could be used to determine uranyl ions in the range 0.01 to 10 p.p.m. Several good techniques have been published for the determination of trace amounts of uranium by activation analysis,3 f l u ~ r i m e t r y , ~ ~ ~ fission-track counting6 and alpha-counting,' and although these methods offer good sensitivity a t low concentrations, all of them suffer from the disadvantage that they require the use of equipment or techniques, or both, not normally encountered in most chemical laboratories. Atomic-absorption spectroscopy is attractive but Perkin-Elmer only claim a lower detection limit of 30 p.p.m. for their Model 303 instrument. Florence and Farrar* described an excellent spectrophotometric method for the determination of uranium in ores that involves the use of the 2-(2-pyridylazo)-5-diethyl- aminophenol- zinc complex but their lower detection limit is 15 p.p.m.of uranium oxide (U,O,) in an ore. In our method, a standard graph of the bixin decolorisation gradients is prepared for the range of uranyl-ion concentrations of interest. Once this graph has been obtained a routine analysis can be completed in approximately 30 minutes (the time required to observe suitable decolorisation of the solution under ultraviolet irradiation) ; however, it should be realised that any number of solutions can be irradiated simultaneously according to the lamp facilities available. In the present work, one lamp was used to irradiate ten solutions simul- taneously and their final transmittance values were read consecutively. Each transmittance reading required approximately 30 s.EXPERIMENTAL cis-Bixin, in the form of the sodium salt, can conveniently be obtained by separation on a Sephadex columng and then precipitated as the free acid with dilute hydrochloric acid. @ SAC and the author.526 WOOD : AN ULTRAMICRO-SCALE METHOD FOR THE [Analyst, Vol. 98 In the present experiments the whole of the colouring matter from 40 g of Bolivian annatto was extracted into chloroform a t room temperature and the filtered extract was then mixed with anhydrous alumina (Merck, Extra Pure). The dyed alumina was then washed with chloroform until no further colouring matter could be removed. The only pigment that remains adsorbed on to the alumina is the cis-bixin, which was removed by placing the stained alumina on top of a chromatographic column containing clean alumina suspended in benzene.The column was then eluted with an acetic acid - ethanol mixture (1 + 9 V/V). The eluted product was recrystallised from glacial acetic acid (m.p. 185 to 186 "C and A,,, 468 nm). The yield was 400 mg, which was a sufficient amount to enable the entire set of experiments to be completed. 0.2 PREPARATION OF STANDARD DECOLORISATION CURVE- Previous experiments had shown that concentrations of water greater than 1 per cent. in the solutions to be decolorised reduced the rate of decolorisation and hence the sensitivity of the test, and so chromatographic acetic acid (99 to 100 per cent.) was used without further purification to prepare the bixin solutions containing uranyl ions in the following concen- trations: 0, 0.5, 0.1, 0.3, 0.7, 1, 2, 3, 4, 5, 7 and 10 p.p.m.In order to prepare these solutions, 0.00930 g of high-purity uranyl nitrate was weighed on a Mettler microbalance and then dissolved in 50 ml of the acetic acid, thus giving a standard solution containing 100 p.p.m. of uranyl ions. For the solutions containing from 1 to 10 p.p.m. of uranyl ions, aliquots of the standard solution (from 0.1 to 1 ml) were made up to 5 ml with the acetic acid and the solutions were then made up to 10 ml with a 2000 p.p.in. solution of bixin in acetic acid. In this way, an initial concentration of 1000 p.p.ni. of bixin was obtained for each solution, which ensured the presence of a large excess oi the pigment throughout the irradiations.The more dilute solutions containing from 0.05 to 0.7 p.p.m. of uranyl ions were prepared in the same way but by using aliquots of standard solutions containing 10 and 1 p.p.m. of uranyl ions. Kimax and Pyrex glassware (grade A) was used throughout; the temperature of the solutions was stabilised at 20 "C so as to ensure accuracy in volume measurement and all solutions were freshly prepared prior to exposure. 1 -4 - I I 1 I 1 *2 1.0 0.8 0.6 0.4 Irradiation time/minutes Fig. 1. Decolorisation graphs for bixin a t different uranyl concentrations: A, 0; B, 0.6; C, 1; D, 2; E, 3 ; F, 4; and G, 5 p.p.m.July, 19731 DETERMINATION OF THE URANYL CATION 527 Short-wave irradiations of the solutions were made with a Cromato-Vue 257.3-nm mercury lamp, which is operated at 0.36 A and 220 V; 2.7 ml of each solution were transferred by pipette into matched 1-cm quartz cells (Beckman), which were then irradiated at a distance of 25 cm from the lamp. After each 10-minute irradiation, the lamp was switched off and the cells were transferred into a Beckman DU-2 spectrophotometer, with which transmittance readings were taken at 468 nm.For the concentration range of uranyl ions above 7 p.p.m., the time of each irradiation was reduced to 1 minute or less. This step was found to be necessary in order to ensure that the corresponding transmittance - time graph was a straight line. For each concentration of uranyl ions the decolorisation graph of transmittance against irradiation time was plotted and found to be linear in each instance (Fig.1). In order to facilitate the use of these results for the practical determination of uranyl ions, it was decided simply to plot the gradient of each graph as a function of the uranyl-ion concentration. Four complete decolorisation series were run for each concentration of uranyl ions and the values shown in Table I are the mean values with the maximum divergence shown as the error. As expected, the greatest divergence was observed at low uranyl-ion concentra- tions. For the 0.05 p.p.m. solution the divergence was 3 per cent. and this error reduced to 2.7 per cent. in the 10 p.p.m. solution, although the error in the middle of the range was more frequently of the order of 1 per cent.TABLE I RELATIONSHIP OF URANYL-ION CONCENTRATION TO GRAPH GRADIENT [U022+], p.p.m. Gradient x lo3 [U022+], p.p.ni. Gradient x los 0.0 5.0 -J= 0.05 2-0 15.0 f. 0.15 0.05 6.0 rfr 0.10 3.0 17.5 f. 0.17 0.1 7.0 & 0.15 4.0 20.0 rfr 0.16 0.3 8.8 f. 0.15 5.0 22.5 & 0.2 0.7 11.0 f. 0.15 7.0 27.0 f 0.9 1.0 12.0 & 0.16 10 37.3 f. 1.0 By converting these errors into errors in uranyl-ion concentration the following con- clusions can be drawn: in the region of 10 p.p.m. an error of 2.7 per cent. in the decolorisation gradient indicates that the uranyl-ion concentration could be between 9-6 and 10-4 p.p.m., the deviation from the mean being of the order of 4 per cent. ; and in the region of 0.05 p p m . 35 30 25 20 15 10 5 I I I I I I I I I I 0 1 2 3 4 5 6 7 8 9 10 [uo22+1,p.p.rn.Fig. 2. Ursnyl concentration (0 to 10 p.p.m.) as a function of decolorisation gradient528 WOOD an error of 3 per cent. indicates that the concentration could be between 0.045 and 0.055 p.p.m., with a deviation of 10 per cent. These results are shown in Figs. 2 and 3, from which it can be seen that the relationship of gradient to concentration is linear between 1 and 10 p.p.m. of uranyl ions. P) z X 0 Fig. 3. Uranyl concentration (0 to 1 p.p.m.) as a function of decolorisation gradient In order to test for possible interfering effects of other ions, we carried out irradiations on a series of solutions, all of which contained 1 p.p.m. of uranyl ions, together with 500 p.p.rn. of any of the following cations : Fe2+, Cd2+, Pb2+, Ca2+, Naf, A13+, Mn2+, Co2+, Cr3+, Ni2+, Cu2+, Zn2+, Ag+ and Sn4+. As these cations were tested as the nitrate, sulphate, chloride, acetate or carbonate and no deviation was observed in any instance, we conclude that none of the above cations or anions interfere in the determination.Hg2+ and Mg2+, as the chlorides, caused an increase in the rate of decolorisation of about 5 per cent., but as they were present in a concentration 500 times greater than that of uranyl ions, we do not regard this increase as a serious interference. DISCUSSION The results show that uranyl ions have a catalytic effect on the photo-oxidation of labile bixin and that this effect can be used in the quantitative determination of the ions in micro- scale and ultramicro-scale concentrations. Although the relationship of decolorisation gradient to uranyl-ion concentration is non-linear at the parts per billion ( lo9) level, each point on the graph can be obtained simply and with a precision such that determinations of uranyl ions down to 0-05 p.p.m.can be made with a maximum error of 10 per cent. Four test decolorisations on a standard solution containing 0.01 p.p.m. of uranyl ions revealed that their concentration could be determined to an accuracy within 15 per cent. by using this method. The author is indebted to Justo Zapata and Jaime A d & for their help with the experi- ment a1 work. REFERENCES 1. 2. 3. 4. 5. Neumann, W. F., “National Nuclear Energy Series,” Div. VI, I, 11, McGraw-Hill, New York, 6. 7. 8. 9. Ingram, J. S., and Francis, B. J., Trop. Sci., 1969, 11, 97. Janjic, T. J., Milovanovic, G. A., and Celap, M. B., AnaZyt. Chem., 1970, 42, 27. Mackintosh, W. D., and Jervis, R. E., Rep. Can. Atom. Energy Cornm., AECL-481 and CRDC-704, Hoffman, J., Biochem. Z., 1943, 313, 377. April, 1957. 1949, p. 701. June, 1955. Carpenter, B. S., and Cheek, C. H., Anatyt. Chew., 1970, 42, 121. Campbell, E. E., Head, B. M., and Milligan, M. F., Rep. U.S. Atom. Energy Commn, LA-1920, Florence, T. M., and Farrar, Y. J., Analyt. Chem., 1970, 42, 271. Vera, H., and Wood, G. P., Trop. Sci., 1971, 13, 211. Received October 30th, 1972 Accepted February 12th, 1973