228 Analyst, April, 1968, Vol. 93, $9. 228-234 The Determination of Actinium-227 in Urine BY P. J. GOMM AND J. D. EAKINS (Health Physics and Medical Division, A .E.R.E., Harwell, Didcot, Bevks.) A method is described for the determination of actinium-227 in urine. After oxidation of the urine sample with nitric acid, actinium is co-precipitated on barium sulphate. The barium sulphate is converted into carbonate, dissolved in acid and the actinium co-precipitated on iron(II1) hydroxide to remove barium and radium. The iron(II1) hydroxide precipitate is dissolved in a mixture of nitric and hydrochloric acids and the solution passed through an anion-exchange column upon which iron, thorium and protactinium are absorbed, The column effluent, which contains the actinium, is essentially free from solids.Sources for #-counting may be prepared either by evapora- tion or by electro-deposition. Actinium recoveries of about 80 per cent. are obtained, with good decontamination from protactinium, thorium, radium, polonium and lead. ACTINIUM-227 occurs naturally as a member of the actinium (4n + 3) series of radioelements, shown in Fig. 1. It decays with a half-life of 22 years, predominantly by b-emission, 9808per cent., to thorium-227, and the remaining 1.2 per cent. of the decay is by a-emission to francium-223. Its immediate precursor in the series is protactinium-231, with a half-life of 3.4 x 104 years. As actinium-227 is a bone-seeking radionuclide giving rise to no less than five a-emitting daughters, its maximum permissible body burden (bone critical) set by the International Commission on Radiological Protection1 is only 0.03 microcuries, which is one of the most restrictive.In common with other bone-seeking radionuclides, actinium is only eliminated slowly from the body, and the I.C.R.P. allocate it a biological half-life that is identical with that of plutonium. If the excretion rates of these two elements are similar, then by comparison with Langham’s results2 for plutonium, an investigation level for actinium-227 in urine of 0.3 picocuries per 24 hours can be derived. There is little information available on the metabolism and excretion pattern of actinium. After intramuscular injection of actinium-227 in rats, Ba+ found that, after 256 days, 66 per cent. of the injected dose had been excreted in faeces and only 8 per cent.in urine. However, in a recent case4 of accidental intake of actinium-227 via a puncture wound, the urinary excretion rate was about twice that in faeces. The relative sensi- tivity of a counting technique is often considered as a function of E2/B where E is the counting efficiency and B is the background of the counter. The efficiencies of modern u- and 16- counters are roughly comparable, but the a-background is generally between one and two orders of magnitude lower than the /%background. Although the initial a-counting rate from separated actinium-227 represents only a small fraction of the disintegration rate, the in-growth of daughters is relatively rapid and after 20 days the a-counting rate will have increased by a factor of greater than 100.High sensitivity is a primary consideration in any bioassay procedure for actinium-227 and it can be seen from the above that this is best achieved by a-counting. If a-counting is to be used it is essential that good decontamination of the actinium-227 from its radioactive daughters is obtained. This is particularly important when analysing samples of excreta, which may contain daughters not only from the actinium-227 therein, but also from this isotope elsewhere in the body. It is possible to determine actinium-227 by either u- or p-counting. 0 SAC; and the authors.GOMM AND EAKINS : DETERMINATION OF ACTINIUM-227 IN URINE 229 Following a case of internal contamination with actinium-227, it became necessary to devise a specific method for its determination in samples of urine and faeces.The "gross a" methods previously used at Harwell for actinium-227 is not specific for this element. Various methods have been published for the determination of actinium. Petrow and Allen6 deter- mined actinium in uranium mill effluent by solvent extraction with di-(2-ethylhexy1)phos- phoric acid and Hagemann' and Hyde* used solvent extraction into 1-(2-thenoyl)-3,3,3- trifluoroacetone. However, low yields and poor decontamination factors were obtained when these procedures were applied to the analysis of urine samples. This paper describes the development of a specific method for actinium-227 in urine, in which recoveries of about 80 per cent. and good decontamination from both parent and daughter radioelements were obtained.A feature of the work is the use of actinium-228 as a y-tracer for assessing the recovery of actinium. This was prepared by a method similar to that described by Batki and Aldoff .9 23sU 7.1 x 1 0 8 ~ 418 to 456 I 231Th 254h 1 030,022 591 Others ='Th 18-4 2 2 7 A ~ 21.8~ 005 494 a 1.2% v 1 1 571 Others *)Fr 22m 6.81 Others I 2 1 s ~ ~ 1.8 10-3$ 7.36 - 21'Bi 2.16m 211Pb 36lm 207TI 478m I -44 Fig. 1. Decay scheme of the actinium (4n+3) series EXPERIMENTAL The determination of actinium-227 in urine may be conveniently considered in three stages: the initial separation of actinium from urine; the purification of the separated actinium; and the preparation of actinium-227 in a form suitable for a-counting. INITIAL SEPARATION OF ACTINIUM FROM URINE- As there did not appear to be any information on the chemical form of actinium in urine, samples were completely oxidised by evaporation with nitric acid and ashing the residue230 GOMM AND EAKINS: DETERMINATION OF ACTINIUM-227 IN URINE [Artalyst, VOl.93 in a muffle furnace. The ash was then dissolved in dilute hydrochloric acid and the actinium co-precipitated with barium sulphate. From previous work on the determination of radiostrontium in urinelO it was known that the concentration of inorganic sulphate in urine is about 0-02 N, and that maximum recovery of strontium, as sulphate, is obtained from solutions adjusted to 0-5 N with respect to sulphate. Before addition of barium carrier, therefore, the sulphate concentration was increased to this value.Initially ammonium sulphate was added, but it was found by the use of actinium-228 tracer that improved recoveries were obtained if the additional sulphate was added as sulphuric acid. A 3-ml volume of 3 M sulphuric acid was added to each sample and the recovery of actinium at this stage was increased from 70 to almost 100 per cent. It was also found with this procedure that the total sulphate concentration was not critical, and the volume of the initial urine sample could vary between 500ml and 2 litres without affecting the actinium recovery. Little calcium sulphate was precipitated under these conditions. Other members of the actinium seiies, particularly radium and thorium, also co-precipitated with barium sulphate, so that no significant decontamination was obtained at this stage.PURIFICATION OF THE ACTINIUM-2Z7- The barium sulphate precipitate was converted by metathesis to the carbonate by digestion with 50 per cent. w/v potassium carbonate solution containing 50mg of lead. This solution was used instead of the more usual sodium carbonate solution for the following reason. Many lead salts, including lead sulphate, are soluble in 50 per cent. potassium carbonate solution, so that lead isotopes, co-precipitated with the barium sulphate, dissolve when it is digested in this way. By using lead-212 as a y-tracer, it was found that 50 mg of lead hold- back carrier were required to obtain a quantitative decontamination of the barium carbonate. As many thorium salts also dissolve in potassium carbonate solution, some thorium is also removed by this method.The barium carbonate precipitate that carries the actinium, radium and some thorium and protactinium was dissolved in dilute hydrochloric acid and, after the addition of iron carrier, iron( 111) hydroxide was precipitated by the addition of ammonia solution. Actinium, thorium and protactinium are co-precipitated on the iron(II1) hydroxide but the barium carrier and radium-223 remain in solution. To ensure effective decontamination from radium, the iron(II1) hydroxide was dissolved in hydrochloric acid and re-precipitated after the addition of a further quantity of barium hold-back carrier. Danonll showed that thorium, in strong nitric acid solution, is absorbed on anion- exchange resins. In the presence of hydrochloric acid, iron is also taken up.Separation of actinium from the remaining thorium and iron carrier was achieved by dissolving the iron(II1) hydroxide precipitate in 7 N nitric - 3 N hydrochloric acid and passing the solution through a column of De-Acidite FF, previously conditioned with the mixed acids. Protactinium is also taken up on the column, as are some other actinides, but actinium is not retained. After washing the column with the mixed acids, the effluent and washings were combined. PREPARATION OF ACTINIUM FOR a-COUNTING- The most direct method of source preparation is by evaporation of the acid effluent from the ion-exchange column on a platinum tray. This effluent is almost free from solids and, after flaming, the source obtained is quite thin and suitable for counting in a zinc sulphide screen scintillation counter.This method was used to count actinium-227, isolated from spiked urine samples to determine the recovery. The background of the conventional scintil- lation counter (type 1093A) used at the U.K. Atomic Energy Research Establishment is about 5 counts per hour. If greater sensitivity is required, low background counters with silicon-junction diode detectorsl2 are available. These accept sources of 1 -cm diameter and to prepare thin sources of these dimensions, electro-deposition techniques are used. With actinium-228 as a convenient tracer, experiments were carried out with various electrolytes to find a suitable procedure for the electro-deposition of actinium. Sulphate,l3 nitrate,l4 acetate15 and fluoride16 solutions were tried.Recoveries varied between 50 and 70 per cent. and were not really satisfactory. Combinations of nitrate and ethanol, adjusted to pH 2 with ammonia solution, gave recoveries of as high as 95 per cent. on occasions, but consistent results could not be obtained, and no improvement resulted from the addition of microgram amounts of lanthanum carrier.April, 19681 231 The best results were obtained by using the acidic ammonium chloride electrolyte described by Mitche1l.l' At pH 2, with a current of 500 mA and a plating time of 1 hour, the mean recovery was 80 & 5 per cent. Increasing the plating time to 2 hours did not improve the recovery. GOMM AND EAKINS: DETERMINATION OF ACTINIUM-227 I N URINE RIEASUREMENT OF ACTINIUM-227 BY a-COUNTING- As mentioned previously, only 1.2 per cent.of the disintegrations of actinium-227 give rise to an a-particle. As there are five a-emitting daughters, all with relatively short half-lives, their in-growth occurs fairly rapidly, and their contribution to the total a-count may be estimated for any given time by the Bateman equation. The theoretical curve, A, showing the ratio of the a-activity of actinium-227 PZus daughters to that of the actinium-227 alone at various times after its separatioq is shown in Fig. 2. I L I I I I I 1 I I I I I I I I 0 20 40 60 80 100 I20 140 Days after separation Fig. 2. Growth of a-activity in separated actinium- 227 source : A, theoretical curve; B, experimental curve There are two advantages to be gained by allowing at least two or three weeks to elapse before a-counting the separated actinium-227.Firstly, the count-rate increases almost linearly over this period so that there is a considerable gain in sensitivity and secondly, if decontamination from its a-emitting daughters is incomplete, the actinium-227 activity will be over-estimated in counts made shortly after separation. This source of error will diminish rapidly in significance if counting is not carried out until later, so that a more accurate estimate of the actinium-227 is likely. In Fig. 3, the a-spectrum obtained from actinium-227 separated from a spiked urine sample is shown. This was measured over a period of 5 hours starting shortly after the anion- exchange separation. The presence of all the radioactive daughters is already evident and their contribution to the total a-count is already approaching that of the actinium-227 alone.Curve B in Fig. 2 shows a plot of the ratio of the a-counting rate of a separated actinium- 227 source, measured at various times, to the initial count made 5 hours after separation and corrected to zero time. Three factors influence the shape of this curve in relation to the theoretical one. Increased counting efficiencies for the actinium daughters caused by their greater a-energies will tend to raise the ratio above the theoretical value. The short half-life of the polonium-215 daughter (1.8 x 10-3 seconds) gives rise to many a-particles, which are virtually coincident with those arising from decay of its precursor, radium-219. These are232 GOMM AND EAKINS: DETERMINATION OF ACTINIUM-227 IN URINE [AutUbSt, VOl.93 not recorded as separate counts, and therefore the ratio will be depressed below the theoretical value. Also, any daughters not completely separated in the procedure will tend to reduce the value for this ratio below the theoretical. 6.6 MeV 7.3 MeV 0 10 20 30 40 50 60 70 80 90 I( I Channel number Fig. 3. cc-Spectrum of actinium-227 during the first 5 hours after separation from urine PROCEDURE- Pour the sample of urine into a 2-litre beaker, add 300rnl of nitric acid (sp. gr. 1*42), 5ml of octan-1-01 and a few glass beads and evaporate to about 100ml on a hot-plate. Transfer the solution to a 200-ml silica basin, evaporate to dryness under an infrared lamp, then put it in a muffle furnace at about 500" C for 10 minutes to complete the oxidation.Dissolve the ash in 70 ml of M hydrochloric acid and transfer with washings to a 100-ml centrifuge tube. Add 3 ml of 3 M sulphuric acid, heat the tube in a water-bath to about 80" C and add 20 mg of barium carrier (2 ml of a solution of 1.78 g of barium chloride dihydrate in 100 ml of water) dropwise, with stirring, then heat for a further 5 minutes. Centrifuge off the barium sulphate precipitate, wash once with 50ml of water, discard the washing and transfer the precipitate with 20 ml of water to a 150-ml beaker. Add 20 ml of 50 per cent. w/v potassium carbonate solution containing 50 mg of lead carrier (1 ml of a solution of 8-0 g of lead nitrate in 100 ml of water) and boil on a hot-plate for about 10 minutes to reduce the volume to 20 ml.Transfer the slurry to a 40-ml centrifuge tube with potassium carbonate solution and centrifuge off the precipitate, then wash with 10ml of potassium carbonate solution and 10 ml of water, discarding the washings. Dissolve the precipitate in 5 ml of 3 M hydrochloric acid and dilute to 20 ml with water. Add 2 mg of iron carrier (1 ml of a solution containing 1.0 g of iron(II1) chloride hexahydrate in 100 ml of 0.2 M hydrochloric acid) and ammonia solution (sp. gr. 0438) dropwise, with stirring, to precipitate iron(II1) hydroxide. Centrifuge off the precipitate and wash with 20 ml of 0.05 M ammonia solution, discarding the washing. Dissolve the precipitate in 5 ml of 3 M hydrochloric acid, add 10 mg of barium carrier and re-precipitate the iron(II1) hydroxide by the addition of ammonia solution (sp.gr.0+38). Centrifuge off the precipitate and wash with 20 ml of 0.05 M ammonia solution, discarding the washing. Fill a 1-cm diameter ion-exchange column to a depth of 5 cm with 200-mesh DeAcidite FF anion-exchange resin. Wash the column with 100 ml of 7 M nitric - 3 M hydrochloric acid solution. Dissolve the iron(II1) hydroxide precipitate in 5 ml of 7 M nitric - 3 M hydrochloric acid solution and transfer with a further 5ml of the acid to the anion exchange column. Allow the solution to pass through and wash the column with 20 ml of the 7 M nitric - 3 M hydrochloric acid solution. Combine the effluent and washing from the column, and evaporate to about 2 ml in a 100-ml beaker. Pour the solution from the beaker into a platinum tray, wash the beaker several times with water to ensure complete transfer, then evaporate to dryness, and flameApd, 19681 GOMM AND EAKINS: DETERMINATION OF ACTINIUM-227 IN URINE 233 the source.Allow the source to grow as long as possible before counting in a suitable or-counter. Calculate the actinium-227 content by reference to the theoretical growth curve of or-active daughters from actinium-227. DECONTAMINATION FACTORS- Decontamination factors were obtained for protactinium, thorium, radium, polonium and lead by using the actinium procedure. In each case six 1-5-litre urine samples were spiked with an appropriate tracer and, after processing the samples, the mean recovery of the tracer was calculated and expressed as a decontamination factor.The results of these measurements are shown in Table I. RESULTS TABLE I DECONTAMINATION FACTORS FOR THE ACTINIUM PROCEDURE Element Tracer used Decontamination factor Protactinium .. . . Protactinium-233 > loo* Thorium .. .. . . Thorium-230 260 Radium . . .. . . Radium-228 > loo* Polonium . . .. . . Polonium-210 160 Lead .. .. .. .. Lead-212 > loo* *These results were obtained by y-counting. The amount of tracer available and the background of the counter limited the minimum detectable activity to about 1 per cent. of that initially added. RECOVERIES- The recovery obtained up to, but not including, the electro-deposition stage was deter- mined by spiking six 1-5-litre urine samples with actinium-228 and analysing them by the previously detailed procedure.The actinium-228, separated in this way, was counted with a y-ray scintillation spectrometer and the recovery determined by comparing the counting rate in the O.9-MeV photopeak with that of an aliquot of the spike solution. The recoveries are given in Table 11. TABLE I1 RECOVERY OF ACTINIUM-228 FROM 1 *&LITRE URINE SAMPLES Sample No. Actinium-228 recovery, per cent. 1 82.2 2 84.9 3 86.1 4 86.6 6 80.7 6 78.6 Mean and standard deviation 83.0 f 2.7 As a final check, six l-5-litre urine samples were spiked with actinium-227 in equilibrium with its daughters. These samples were analysed in the usual way and the actinium-227 sources were counted 30 days after separation. From these counts the actinium recovery was determined by reference to the theoretical growth curve shown in Fig.3. The results are given in Table I11 and are in excellent agreement with the recoveries obtained with act inium-228. TABLE I11 RECOVERY OF ACTINIUM-227 FROM SPIKED URINE SAMPLES Sample No. Actinium-227 recovery, per cent. 1 82.2 2 86.4 3 84.7 4 87.6 6 84.6 6 86.6 Mean and standard deviation 86.2 f 2.0234 GOMM AND EAKINS With a radiochemical recovery of 85 per cent., a level of 0.3 picocuries of actinium in 10 per cent. by counting a l-5-litre (nominal 24 hour) urine sample can be measured to within the source for 8 hours, after it has been allowed to grow in for 30 days. CONCLUSIONS The work described in this paper shows that by using only three precipitations, and a single ion-exchange step, actinium-227 can be separated from urine samples in good yield and essentially free from its parent and daughter activities.In the development of the method, conventional procedures have been modified to achieve more than one objective in order to reduce the number of chemical steps required. The use of potassium carbonate solution containing lead carrier for metathesis of barium sulphate containing lead isotopes has not been reported before. The use of a mixture of hydrochloric and nitric acids for absorbing both iron and thorium simultaneously on an anion-exchange resin is also novel. Recoveries of actinium have been checked with actinium-228 y-tracer, as well as with actinium-227, and excellent agreement obtained. 1. 2. 3. 4. 6. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.REFERENCES “Recommendations of the International Commission on Radiological Protection,” I.C.R.P. Report of Committee I1 on Permissible Dose for Internal Radiation, Pergamon Langham, W. H., Hlth Phys., 1959, 2, 172. Barr, G., U.S. Atomic Energy Commission Report, UCRL-3268, 5, 1955. Newton, D., Rundo, J., and Sandalls, F. J., Paper presented at the “Symposium on Diagnosis and Treatment of Deposited Radionuclides,” Richland, U.S.A., May, 1967. Sponsored by the Batelle Memorial Institute, Pacific Northwest Laboratory, Hanford Occupational Health Foun- dation Inc., and U.S. Atomic Energy Commission. Brooks, R. 0. R., U.K. Atomic Energy Research Establishment Report, AM60, H.M. Stationery Office, London, 1960. Petrow, H. G., and Allen, R. J., Analyt. Chem., 1963, 35, 747. Hagemann, F., J. Amcr. Chem. Soc., 1950, 72, 768. Hyde, E. K., ‘‘Radiochemical separation methods for the actinide elements,’’ in Int. Conf. Peaceful Uses Atom. Energy, Geneva, 1955,7, 281. Bhatki, K. S., and Adloff, J. P., Radiochimica Acta, 1964, 3, 123. Eakins, J. D., and Gomm, P. J., U.K. Atomic Energy Research Establishment Report, R4853, H.M. Stationery Office, London, 1964. Danon, J., J. Amer. Chem. Soc., 1956, 78, 5953. Sandalls, F. J., and Morgan, A., U.K. Atomic Energy Research Establishment Report, R4391, H.M. Stationery Office London, 1964. Dupzyk, I. A., and Biggs, M. W., Paper presented at the 6th Annual Meeting on Bio-assay and Analytical Chemistry] Santa Fe, New Mexico, ;October i1960; 1U.S. Atomic Energy Commission Report TID-7616, 39, 1960. Iyer, R. H., Jain, H. C., Ramaniah, H. V., and Rao, C. L., Radiochimica Acta, 1964, 3, 225. Hansen, P. G., Hragh, J., Nielsen, H. L., Nucl. Instrum. Meth., 1964, 30, 161. Smith, G., and Barnett, G. A., U.K. Atomic Energy Authority Report AEEW-R386, 1964. Mitchell, R. F., Analyt. Chem., 1960, 32, 326. Publication 2. Press, Oxford, 1959. Received September 15th, 1967