A~zalyst, February, 1968, Vol. 93, #J. 101-106 101 Automated Neutron-activation Analysis of Biological Material with High Radiation Levels BY K. SAMSAHL* ( A B Atomenergi, Nykoping, Swedeii) A method has been deveIoped for the chemical separation and subsequent y-spectrometric analysis of the alkali metals, the alkaline earths, the rare earths, chromium, hafnium, lanthanum, manganese, phosphorus, scandium and silver in neutron-activated biological material. The separation steps, which are fully automatic, are based on a com- bination of ion exchange and partition chromatography and require 40 minutes for their completion. MANY procedures suitable for chemical separations of radionuclides by anion exchange from hydrochloric acid solutions have appeared in the literature. However, little information is available about simple and rapid sub-grouping of trace elements that are non-adsorbable by this method.The aim of the present work was to develop a method for routine separations of these elements in neutron-activated biological material. Because of the high level of sodium-24 radiation and sometimes also the long-lived alkaline earth activities, manual chemical processing may require the use of rigorous safety precautions or excessively long decay periods.lJ To overcome these difficulties and to hasten and simplify the chemistry involved, an almost fully automated, remote-control separation method is needed. This implies the exclusion of many conventional steps in analytical chemistry, e.g., extractions, precipitations, titrations and evaporations, as these are all more or less unsuited for auto- mation.A better solution to the present problem seems to be the application of ion exchange or related techniques. However, the high ionic strength of the sample solution (strong hydrochloric acid) severely limits the possibility of choosing many of the common separation procedures. Methods that do not often suffer from this disadvantage include separations from halogen acid solutions with anion-exchange resins, from nearly neutral or alkaline solutions with chelating resins and inorganic ion exchangers, and from a wide variety of acidic solutions with part it ion-chromat ographic technique. Since the pioneering work in partition chromatography with reversed-phases, carried out on different mixtures of inorganic constituents by Siekierski and Kotlinskaya,3 Siekierski and Fidelis4 and Gw6Zdi and Siekierski,6 the method has found an ever-increasing number of applications in radiochemical analysis.s The ability of chelating ion exchangers to separate traces of different cations from strong electrolyte solutions has been studied by Sides and Kenner.7 The ion-exchange properties of several inorganic materials have been studied by Kraus, Phillips, Carlson and Johnson, who developed extremely simple procedures for separations within the groups of the alkali metals and the alkaline earths.8 In the present method seven groups of chemical constituents are separated in the following sequence: hafnium and scandium are adsorbed from strongly acidic solution on a column of kieselguhr impregnated with di(2-ethylhexyl) orthophosphate (HDEHP) (Fig.1, column C) ; lanthanum and the rare earths are then adsorbed in an analogous way from weak acid (column D); traces of silver are separated as the bromide complex with anion-exchange resin (column E); a column of Chelex-100 chelating resin adsorbs chromium and manganese from weakly acidic solution (column F), whereas the alkaline earths are separated in a similar way from strong alkali (column G) ; and the series is ended with a column of zirconium ortho- phosphate, which selectively adsorbs the heavier alkalis, potassium, rubidium and caesium, from a weakly acid solution (column H). The remaining strong activities of sodium-24 and phosphorus-32 are passed to the effluent of the column series and fonn the seventh group of elements separated.* Present address : 8000 Munchen-45, Riemerschmidstrasse 6/111, Germany. 0 SAC and the author.102 SAMSAHL : AUTOMATED XEUTRON-ACTIVATION AXALYSIS [Ana&St, VOl. 93 The separations, which are based on selective sorption, are carried out simultaneously with a peristaltic pump, and are in direct relationship with the anion-exchange separation step from 8 N hydrochloric acid solution described e1sewhere.l The desired composition of the various influent solutions is obtained by introducing, between the columns, appropriate amounts of strong solutions of sodium acetate buffer, sodium hydroxide, sodium bromide, nitric and orthophosphoric acids. EXPERIMENTAL APPARATUS- A peristaltic pump (type Desaga, Heidelberg) was used for the separation and the working principles are shown in Fig.1. The pump is equipped with twelve Tygon plastic tubes with inner diameters of 1-6mm. Teflon tubes with 1-mm inner diameter are used to connect the columns in the series B, C, D, E, F, G and H. The various influent solutions are thoroughly mixed by pumping them through the mixing coils, J. The coils, which are made of Teflon tubing, each consist of eight turns with a diameter of 15mm. As shown in Fig. 1, the effluent solution from column B is led back to the suction side of the pump before entering the remaining columns in the series. This is done to decrease the flow-rate resistance in the system and to make the proportioning control easier. I 1 1 A = Peristaltic pump B = 5 x 200-mm Column of Dowex-2 (Cl-, 200 C = 5 x SO-mm Column of HDEHP-treated D = 7 x 50-mm Column of HDEHP-treated E = 7 x 50-mm Column of Dowex-2 (Cl-, 200 F and G = 16 x 50-mm Columns of Chelex-I00 H = 22 x 90-mm Column of Bio-Rad ZP-l (100 to 400 mesh) kieselguhr kieselguhr to 400 mesh) (Na+, 100 to 200 mesh) to 200 mesh) J = Mixing coils K = Point at which ION nitric acid - M sodium dihydrogen phosphate ( I + I ) is intro- duced L = Point at which 8N sodium hydroxide is intro- d uced M = Point at which 4M sodium acetate- ION sodium hydroxide - 5N sodium bromide (15 + 3 + 2) is introduced N = Point at which 8N hydrochloric acid -con- centrated sulphuric acid sample i s intro- duced 0 = Test-tube Fig.1. Scheme of the separation system The columns B, C, D, E , F, G and H, of borosilicate glass, are fitted with Teflon stoppers with a neoprene ring.The stoppers are kept securely in position with Teflon screw-caps. The distance between the lower end of a stopper and the top of the column is kept within 3 to 5 mm to minimise the amount of washing solution needed for a separation. The inner diameters of columns B and C are 5 mm, of D and E, 7 mm, of F and G, 16 mm, and that of H, 22 mm. The columns have glass filter discs of porosity G-1 or G-2 at the bottom. Details not shown in Fig. 1 are a lead-brick shielded calibrated flask used to receive the effluent from column H and bottles containing the stock solutions K, L, M and N.‘February, 19681 OF BIOLOGICAL MATERIAL WITH HIGH RADIATION LEVELS 103 PREPARATIONS AND CALIBRATION- Column B is filled by suction to a height of 200 mm with wet Dowex-2 x 10, 200 to 400- mesh, resin in the chloride form, and column E to a height of 50 mm with the same resin form.As indicated in Fig. 1, column B may also be sub-divided into three smaller, coupled columns, the first being 100mm high and the remaining two each 50mm high. This arrangement is made to allow for a further sub-grouping into five different groups of the many trace elements adsorbed on this column.1 Except for the thorough removal of colloidal particles, the commercially available resin in the chloride form may be used without pre-treatment. The resin is stored in distilled water. Columns C and D are both filled to a height of 50 mm with wet kieselguhr pre-treated with dk(2-ethylhexyl) orthophosphate (HDEHP) and are washed with a little water on a suction flask until the effluent ceases to foam and finally drained.The preparation of the column material was carried out as follows. Celite 545 (Johns-Manville) was mixed with water several times, decanting after each mixing, to remove colloidal particles. After drying, the kieselguhr was siliconised with dimethyldichlorosilane vapour and again dried; 15 g of the material were then stirred into a solution of 1.5 g of HDEHP in 50 ml of di-isopropyl ether. The bulk of the solvent was volatilised by allowing it to stand at room temperature, the last traces being removed under reduced pressure, The impregnated kieselguhr was stored in distilled water. Columns F and G are prepared with wet Chelex-100, 100 to 200 mesh, in the sodium form (Bio-Rad Laboratories).The resin is stored in distilled water after removing colloidal particles. To avoid shrinkage during operation the columns are washed with composite solutions, in each instance corresponding to the influent in practical runs, and are then drained by suction. The final height of both columns of resin mass should be 50 mm. The last column in the series, H, is filled to a height of 90 mm with dry ZP-1, 100 to 200 mesh (Bio-Rad Laboratories), which is washed with 4~ sodium acetate until the pH of the effluent is between 4 and 5. The column is finally drained. Columns B, C, D, E, F, G and H are then coupled in series and connected to the peristaltic pump, as shown in Fig. 1. They are then pre-equilibrated and the whole system calibrated by running the pump for some minutes.At the beginning of the running-in period the influent solutions to the last three columns, F, G and H, are by-passed to waste. This is done to decrease the time needed for the calibration and to obviate the risk of destroying the zirconium orthophosphate exchanger. At point N, 8 N hydrochloric acid is fed into the system through two Tygon tubes from a washing solution reservoir. The solution passes through column B and then back to the suction side of the pump via a small test-tube, 0, which continuously stores 0-5 to 1 ml. The solution is now pumped unchanged, and at the same speed, through column C. The effluent is automatically diluted to three times its volume before entering columns D and E. The dilution is effected by introducing through four different Tygon tubes at point M a solution, a stock of which is prepared by mixing 600 ml of 4 M sodium acetate, 120 ml of 10 N sodium hydroxide and 80 ml of 5 N sodium bromide.The influent to column F is obtained by injection of 8 N sodium hydroxide through one tube at point L. The resulting solution, which by-passes column F, is controlled for acidity with pH paper strips and, at the beginning of the running-in period, will show a strongly alkaline reaction. However, after a few minutes, when the air has been forced out of the system and the strong hydrochloric acid has reached full action at the end of the series, the pH will become stabilised at a value of 5 to 5-5. The pH of the influent to the foregoing columns, D and E, is about 36, at which pH the acetate system still has sufficient buffer capacity for the present separations.Column F is now attached to the series by turning the three-way stopcock ahead of the column. The effluent is mixed with a further portion of 8 N sodium hydroxide, introduced at point L. When it is strongly alkaline, the solution is passed through column G and finally brought back again to the buffering interval. This is effected by pumping in through two tubes at point K a solution consisting of 10 N nitric acid - M sodium dihydrogen orthophosphate (1 + 1). After the pH of the resulting solution has become stabilised at between 4 and 4.5 the last column, H, is also switched into the series. Its effluent should have an acidity falling within the same pH range, and the flow-rate is controlled at about 5 ml per minute.This corresponds to a flow-rate of about 1 ml per minute through B, the first column in the series.1 The pump is now stopped and the apparatus is ready for the separation.104 SAMSAHL : AUTOMATED NEUTRON-ACTIVATION ANALYSIS [ A nalyst, Vol. 93 The preparation of the columns, as well as the calibration step, is advantageously carried out before completing the irradiation of the samples, especially if the determination of short-lived nuclides is needed. PROCEDURE- Irradiation and #re-se$aration-The biological material, ranging in amount from about 15 mg for hard tissue, to 200 mg dry weight for soft tissue, is sealed in a quartz tube and irradiated for 2 days with a thermal neutron flux of about 2 x lofs neutrons per cm2 per second.The irradiation may be followed by a decay period of 1 to 3 days, according to the level of sodium-24 activity present. The sample may then be re-irradiated for 10 to 30 minutes to make possible also the determination of short-lived nuclides. Shortly after the second irradiation, it is transferred into a distillation apparatus, together with 60 pl of a 48 per cent. hydrobromic acid carrier solution containing 1 pg each of Ba2+, Ca2+, C9+, Cs+, Hf4+, K+, La3+, Mn2+, Na+, PO?--, rare earths, Rb+, Sc3+ and Sr2+; 1 pg of Ag+ is added from a weak sulphuric acid solution. For simultaneous determinations of elements, separated as chloride complexes on column B, an additional sulphuric acid carrier solution containing 1 pg each of Au3+, Cd2+, Co2+, Cu2+, FeS+, Ga3+, Ins+, Mos+,W6+ and Zn2+ is also added to the flask.l The organic matter is then destroyed with a mixture of concentrated sulphuric acid and 50 per cent.hydrogen peroxide, and radionuclides of the elements arsenic, bromine, chlorine, ger- manium, mercury, iodine, osmium, rhenium, rubidium, antimony, selenium and tin are distilled from the solution. Further details concerning the irradiation and the distillation procedures are given el~ewhere.~ tg After the distillation, concentrated sulphuric acid solution remaining in the flask is boiled down to 0.5 ml and diluted with 10 ml of 8 N hydrochloric acid. Any calcium sulphate precipitate is dissolved by gentle heating. After cooling, the solution is introduced at point N (Fig.1) and allowed to flow through the column series at the same rate, and with the same dilutions between the columns, as described above. On its way to the first column, B, the sample passes a small glass filter disc of porosity G 4 . Immediately following the sample, and separated from it by an air bubble, a wash solution, consisting of 15 ml of 8 N hydro- chloric acid, is also passed through the system in the same way. Column B is now removed from the system. As a result of preparing the column in three different parts, adsorbed trace elements may be easily sub-divided into five gr0ups.l A second wash solution, consisting of 15 ml of 8 N hydrochloric acid, is now passed through the remaining series of columns C, D, E, F, G and H in the same way The pump is then stopped, and each column attached to a suction flask for a further rapid rinsing.Columns C, D and E are washed with 10 to 20 ml of water, and columns F, G and H with the same amount of solutions, the composition of which corresponds in each instance to the influent. After a final draining, the various column materials are transferred into polythene tubes, homogenised and counted. A multi-channel y-spectrometer attached to a 3 x 3-inch well-type sodium iodide crystal is suitable for the measurements. RESULTS AND DISCUSSION RECOVERY AND REPRODUCIBILITY STUDIES- The results obtained in recovery and reproducibility experiments with seventeen different radioactive trace elements are given in Table I. The strongly irradiated samples shown in the first column of Table I were separately dissolved in concentrated sulphuric acid, or in hydrochloric acid plus hydrogen peroxide , and diluted. Various mixtures containing suitable activities of 6 to 8 of the trace elements and carriers, as described above, were then treated with mixtures of hot sulphuric acid and 30 per cent.hydrogen per~xide.~ Finally, each solution was heated to strong fumes of sulphur trioxide and boiled down to 0.5 ml, which was used for the separations. These were carried out with a peristaltic pump, as described above. The activities of the separated groups were compared with reference samples. The values given are the means of three experiments, and the standard error in no instance exceeded 3 per cent. DESTRUCTION OF ORGANIC MATERIAL AND TOTAL PROCESSING TIME- The mixture of concentrated sulphuric acid and 50 per cent.hydrogen peroxide acts not only as a simple and rapid wet-ashiiig agent for biological material, but will also quantitativelyFebruary, 19681 Irradiated material OF BIOLOGICAL MATERIAL WITH HIGH RADIATION LEVELS TABLE I RECOVERY VALUES FOR THE RADIOISOTOPES SEPARATED Mean value of yield, per cent. .A r Isotope Column measured B &NO, . . Silver-110 - RaC1, . . . . Barium-131 - CaC1, . . . . Calcium47 - Ce,(SO,), . . Cerium-141 - CrC1, . . . . Chromium-61 - CSCl . . . . Caesium-134 - Hf02 ., . . Hafnium-181 (-9) KC1 . . . . Potassium-42 - La(NO,), . . Lanthanum-140 - LU,O,. . . . Lutecium-177 - XlnSO, . . Manganese-56 (-2) NaCl . . . . Sodium-24 - NH,H,PO, . . Phosphorus-32 - RbCl .. . . Rubidium-36 - sc,o, . . . . Scandium-46 - Sm2% . . Samarium-153 - SrC1, . . . . Strontium-87 - Column Column C D - 91 - 100 (-2) 98 100 - - 100 105 convert the trace elements into appropriate oxidation states, which are ideally suited for subsequent chromatographic group separations. As it is a powerful oxidant in an acidic medium, hydrogen peroxide will oxidise many of the elements to their normally highest valencies, although a few of them, e.g., molybdenum and tungsten, are dissolved as peroxy- compounds. On the other hand, the biologically important elements, manganese and chrom- ium, are reduced from higher valency states to the bivalent and tervalent state, respectively, thus securing quantitative separation in one group of the scheme. The rare earths, including cerium, will be present in the tervalent state in the final hydrochloric acid sample solution, thereby making possible their separation, as a single group, from all other elements.The risk of precipitating typical trace elements, such as silver, tungsten and uranium, during the chemical processing may be regarded as non-existent, because of their low concentration in biological material, and the minute amount of carriers used. The destruction of organic matter can be carried out at a distance behind a lead shield in less than 10 minutes. If coupled to a distillation step for the simultaneous determination of volatile elements, as mentioned above, the time required to obtain the sample solution will increase to 30 to 40 minutes. Accordingly, the total processing time will vary from about 50 to 80 minutes, depending on the number of elements sought.Further, one non- scientific operator can process three samples simultaneously with three machines in 14 hours, provided the distillation step is omitted. In routine analysis of such samples no cleaning of the apparatus is needed, the preparation for a fresh separation consisting in mounting a new set of columns from stock, followed by a running-in period lasting for a few minutes. Chemical group separations similar to those described have also been carried out manually on human tissues. Under these conditions one skilled chemist could process only one sample at a time in 1 Q to 3 days.2JO APPLICABILITY OF THE METHOD- The use of the present method is restricted by the amounts of alkaline earths and silicon present in the samp1es.l The chemical separations were not extended to more than seven groups.This is probably sufficient for most routine applications, especially if advantage is taken of the different decay rates of the nuclides, and the measurementsare combined with a spectrum stripping technique. However, in the event of more detailed studies being required complementary separations will be needed. The determination of hafnium-181 in the presence of the high energy scandium-46 in the first group will certainly, in many instances, be difficult or impossible. However, a chemical separation has so far not been tried, because of, inter alia, the exceedingly long counting times needed for reliable hafnium determinations (at least with medical sampleslO).For lanthanum and the rare earths separated in the second group of the scheme, more detailed studies will probably always be impossible without a further chemical sub-grouping.106 SAMSAHL For example, in typical medical samples only the determination of cerium, lanthanum and samarium is possible.1° However, by splitting this large group of elements into two or more sub-groups, with selective elution steps,ll a much better resolution may be possible. As regards chromium and manganese in group IVY further chemical processing is rarely needed. An important exception occurs with blood samples, because of the formation of large amounts of manganese44 with fast neutrons. A simple chemical separation seems possible by selectively eluting manganese with weak hydrochloric acid.and the determination will probably not be affected by radionuclides of barium and strontium. Also, determinations of strontium based on strontium-87un will hardly cause difficulties. However, barium determinations, with either barium-131 or barium-139, may sometimes be impossible. Purification of barium activities after the elution of the alkaline earth group with weak acid is possible with ion exchange on zirconium orthophosphate.12 Individual chemical separations of the heavier alkali metals in group VI will hardly be necessary, provided a well-type detector is used in combination with spectrum stripping after the decay of potassium-42. The problem can be solved in a similar way in the analysis of phosphorus and sodium in group VII.The phosphorus-32 bremsstrahlung remaining after the decay of sodium-24 may, at least theoretically, be contaminated by the long-lived sodium-22. The spectrilm of this nuclide is then stripped away, the 1.02 MeV sum-peak being utilised in this connection. In contradistinction to an extensive chemical sub-division, the adding together of different groups may also be shown to be advantageous. For example, the relationship between silver-110 and chromium-51 is, in many kinds of neutron-bombarded biological material, sufficiently favourable to permit simultaneous measurement. Sometimes also the scandium-46 activity in group I may be added, thus saving long, expensive counting times. THE PROPORTIONIKG PUMP SYSTEM- Practical experience shows that the peristaltic pump used for the experiments will work in a reliable way, provided that careful calibrations are carried out.However, this type of apparatus seems to be rather sensitive to differences in flow resistance of the columns used for the separation, which may lead to slightly inexact volumes of liquid being supplied to the series. The same effect may also be caused by differences in the swelling of the plastic tubes transferring the strongly acid or alkaline solutions. In the present work, these difficulties were largely overcome in various ways, e.g., by frequently changing the plastic tubes of the pump, by using a large amount of acetate buffer for the separation, and by applying com- paratively coarse-sized and broad resin beds towards the end of the column series.The present separations may probably be improved by using a machine with a piston drive.1 With this kind of drive the apparatus will work accurately, being almost completely independent of flow-resistance problems and the concentrations of acid and alkali used. However, adjustments of a more minor nature in the composition at present used of the solutions injected into the system may be necessary. Because of the high degree of reliability obtained, it may also be possible to extend the number of columns in the series, e.g., to split the rare-earth group into sub-groups by selective sorption steps on small columns of HDEHP- treated kieselguhr.ll I am grateful to Dr. Erik Haeffner, Head of the Chemistry Department, for his interest in this work. Calcium-47, or sometimes calcium-49, can be used for calcium analysis in group V, I also thank Sigrid Hackbarth for skilful technical assistance. 1 . 2. 3. 4. 5. 6. 5. r 9. 10. 11. 12. REFERENCES Samsahl, K., Nukleonik, 1966, 8, 252. Samsahl, K., Brune, D., and “ester, P. O., Int. J . AppZ. Radial. Isotopes, 1965, 16, 273. Siekierski, S., and Kotlinskaya, B., Atomw i!??zerg., 1959, 7, 160. Siekierski, S., and Fidelis, J., J . Chromat., 1960, 4, 60. Gw6tdi, R., and Siekierski, S., Nztkleonzka, 1960, 5, 671. Hedrick, C. E., and Fritz, J. S., U.S. Atomic Energy Commissioii Report, IS-950, 1964. Sides, J. L., and Kenner, C. T., Analyt. Chem., 1966, 38, 707. Kraus, K. A., Phillips, H. O., Carlson, T. A., and Johnson, J . S., Int. Co~f. Peaceful Uses Atomic Samsahl, K., AnaZyt. Chenz., 1967, 39, 1480. Wester, P. O., Scafad. J . Clin. Lab. Invest., 1965, 17, 357. Sochacka, R. J., and Siekierski, S., J . Chromat., 1964, 16, 376. Maeck, W. J., Kussy, 31. E., and Rein, J. E., Analyt. Chem., 1963, 35, 2086. Enevgy, Geneva, 1958, 28, 3. Received March 28th, 1967