1190 Analyst October 1983 Vol. 108 $9. 1190-1194 Limits of Detection of Trace Elements in Biological Materials Analysed by Instrumental Neutron Activation Analysis Using X-ray Spectrometry and Magnetic Deflection of P-Rays Mariana Mantel Soreq Nuclear Research Centre Yavne Israel The limits of detection of 18 trace elements in blood urine and biological standards (from the International Atomic Energy Agency and the National Bureau of Standards) analysed by neutron activation followed by X-ray spectrometry and magnetic deflection were calculated. In addition to the usual parameters that affect the limits of detection obtainable by X-ray spectrometry the influence of the spectral interference (background) resulting from the complex matrices was also taken into account.The method is especially suitable for elements which after neutron activation emit X-rays with energies below 16 keV. Keywords Limits of detection of trace elements ; biological materials ; instru-mental neutron activation ; X-ray spectrometry ; magnetic deflection of 8-rays The application of X-ray spectrometry to instrumental neutron activation (INA) has been intensively studied in our 1aboratorylJ and its advantages and shortcomings pointed out. However in order to apply this technique to complex matrices it is necessary to suppress the background from the irradiated matrix. Because Si(Li) detectors used for the measurement of X-rays have a low sensitivity to photons with energies higher than 80 keV the interference is entirely due to P-rays. The latter produce a high background that in most instances, completely obscures the X-ray peaks.To overcome this interference we developed3 a method of background reduction by magnetic deflection of P-rays based on the fact that magnetic fields deflect the electrons of /3-rays but have no influence on X-rays. The sample is placed between the poles of a magnet located above the detector. The X-rays reach the detector unaltered whereas the /&rays are deflected according to their energy and to the intensity of the niagnetic field. A systematic study4J was carried out and the optimum conditions for the reduction in background by the removal of /3-particles established. Practical applications, such as the determination of bromine in blood serum6 and the determination of niobium in steel.^,^ were developed.In this work we calculated the limits of detection of 18 trace elements in various biological matrices blood urine and different biological standards provided by the International Atomic Energy Agency (IAEA) and the National Bureau of Standards (NBS). In particular, the influence of the background resulting from these complex matrices was taken into account as the effect on the limit of detection is considerable. The elements were determined by the method described. Experimental Materials Standards were prepared from high purity (Fluka or Johnson Matthey) compounds and the following biological standards (except blood and urine) provided by the IAEA and the NBS, were used. Milk powder. Animal mzlscle. IAEA reference material for multi-element analysis H-4.Bovine liver. NBS standard reference material 1577. Orchard Eeaves. NBS standard reference material 1571. Bowen’s kale. Blood and wine. Obtained from healthy volunteers. IAEA intercomparison sample for trace element analysis A-1 1. Provided by the IAEA MANTEL 1191 Apparatus The experimental set-up,6 shown schematically in Fig. 1 consists of (A) a Si(Li) diode (Seforad Israel) of 100 mm2 area 4 mm depletion depth and having a 1.0 mm thick beryllium window coupled through a pre-amplifier and an Ortec Model 485 linear amplifier to a 4096 channel analyser (Promeda Elscint Israel). The resolution of this system for 6.4 keV Fe K X-rays is 400 eV. B is an electromagnet that permits magnetic fields of up to 1.3 T (18-mm gap between poles) to be obtained.The magnet was designed in such a way as to permit positioning of the detector as close as 22 nim from the poles of the magnet. The smallest possible distance from the sample to the surface of the detector is 27 mm. Fig. 1. Schematic representa-tion of the experimental apparatus. A Be window; B sample holder; D detector; E electromagnet; and S sample. Sample Preparation Samples were prepared in special small polyethylene irradiation vials (1 cm id.) and sealed with polyethylene stoppers. The following procedures were used. Standards Soluble compounds. Aliquots of dilute solutions (a few micrograms per millilitre) in water, nitric acid or any other solvent that would not interfere with the irradiation were introduced into the vial and evaporated to dryness under an infrared lamp.Insoluble compounds. A few milligrams of metal powder or oxide were weighed into the vial. Biological standards. About 50 mg were weighed into the vial. Blood. Blood serum (0.5 ml) was introduced into the vial and evaporated to dryness under an infrared lamp. Urine. Urine (0.5 ml) was introduced into the vial and evaporated to dryness. A second aliquot of 0.5 ml was added and the procedure repeated so as to arrive at a total volume of 1 ml. Irradiation The irradiations were carried out in the pneumatic tube of the IRR-1 reactor at a thermal neutron flux of 1 x 1013 n cm2 s-l. For isotopes with half-lives of up to 12 h the samples were irradiated for a length of time equal to one half-life of the isotope obtained from the trace element to be analysed and counted for one half-life after one half-life cooling time.For half-lives longer than 12 h the maximum irradiation and cooling time was 12 h. Procedwe The standard samples were irradiated and counted using the experimental set-up shown in Fig. 1 and the activity obtained from 1 pg of the element (counts per microgram) was deter-mined. Blood urine and biological standards were irradiated and the background was measured in the same way. From these two values the limit of detection of each of the elements studied in every biological matrix was calculated 1192 MANTEL TRACE ELEMENTS IN BIOLOGICAL MATERIALS BY INAA Analyst VoZ. 108 Results The background obtained from the seven biological matrices studied was measured over an energy range of 2-20 keV.The values obtained were compared with those calculated based on the composition of the matrix the irradiation time and the reduction in background due to the magnetic field. Very good agreement was found between the measured and calculated values. Using pure compounds (standards) of each of the trace elements the activity per microgram of element obtainable by the present method without the interference of the back-ground was measured. These values (corrected for absorption in the matrix for up to 9 keV), together with the background values were used to calculate the limits of detection of 18 trace elements in seven biological matrices. The limit of detection was taken to be the smallest amount of the element that permits the attainment of a signal-to-noise ratio of 3dg(B is the background).The results obtained are shown in Table I. Where not otherwise stated the limit of detection refers to all seven matrices. The following conclusions may be drawn. In spite of the variations in composition the limits of detection for short-lived nuclides are practically the same in all the matrices studied. For nuclides with half-lives longer than 3 h the sensitivity will be higher in such matrices as orchard leaves which have a relatively low sodium content and lower in animal muscle with a high sodium content (values indicated by their respective footnotes in Table I). For longer half-lives the sensitivity will depend only on the concentration of phosphorus in the matrix. Only those trace elements that following neutron activation produce radioisotopes that decay by the emission of K or L X-rays with energies up to 16 keV were studied.For higher X-ray energies the background can be reduced with the same efficiency by plastic absorbers and the additional cost of the magnet is not necessary. Discussion One of the major problems associated with the application of instrumental neutron activa-tion and X-ray spectrometry to biological materials is the interference of #%particles emitted TABLE I LIMITS OF DETECTION OF TRACE ELEMENTS IN BIOLOGICAL MATRICES MAGNETIC DEFLECTION DETERMINED BY INA USING X-RAY SPECTROMETRY AND Samples of 50 mg (50 pl for blood) were irradiated for one half-life and counted for one half-life after one half-life cooling time. For half-lives longer than 12 h the samples were irradiated and counted for 12 h after 12 h cooling time .X-ray X-ray energy/ Limit of detection,* Element Isotope 4 measured keV Scandium Chromium Cobalt . . Copper . . Zinc . . Germanium Selenium Bromine Rubidium Strontium Yttrium. . Niobium Uranium Thorium Mercury Platinum Iridium . . Osmium ,. ,. . 20 s 27.8 d 10.5 min 12.8 h 13.8 h 46 s 3.9 min 6.2 min 1.06 min 2.8 h 3.1 h 6.3 min 23.5 rnin 22.4 rnin 65 h 80 rnin 14 h 1.45 rnin Sc Ka V Ka Co Ka Ni Ka Zn Ka Ge Ka Se Ka Br Ka Rb Ka Sr Ka Y Ka Nb Ka NP La1 Pa Lal Au La, Pt La, In La, 0 s LUl 4.5 5.4 6.9 7.5 8.6 9.9 11.2 11.9 13.4 14.2 14.9 16.6 13.9 13.3 9.7 9.7 9.1 8.9 p.p.m.5 x 10-2 3.4 x 10-1,t 1.1 x 10-1: 7 x 10-3 1.8 x lO-l,t 7 x 12.6,t 5.0: 2.8 x 1.5 x 1.4 x 1.0 1.3 x lo2 8.1 2 x 10-2 1 x 10-1 3 x 10-1 4.2,t 2.8$ 2.0 6 x 7 x 10-1,t 3 x 10-1: * Average of the results (range f40%) obtained for all seven matrices. t Limit of detection in animal muscle. Limit of detection in orchard leaves October 1983 USING X-RAY SPECTROMETRY AND MAGNETIC DEFLECTION OF P-RAYS 1193 from the irradiated sample.lP2 These /I-particles increase the dead-time of the Si(Li) detector, upset its resolution and produce a high background which may completely obscure the X-ray peaks. The interference is especially high in biological matrices where the major inorganic components are sodium potassium chlorine and phosphorus all strong /3-emitters following neutron activation.It follows that the possibility of non-destructively determining trace elements in these matrices will depend on the extent to which the P-ray interference can be reduced. As shown previously,6 the application of magnetic fields makes possible the non-destructive measurement of X-rays in neutron activated complex matrices. Theoretical calculations showed' that the intensity of the magnetic field necessary to deflect /I-particles depends on their energy and on the source - detector distance. For example at a distance of 2.7 em (the sample - detector distance in our set-up) see Fig. 1 a 0.2-T magnet is necessary to remove 91% of the /3-particles emitted by 32P (1.7 MeV) and a 0.5-T magnet is needed for a similar reduction of those emitted by 35Cl (4.92 MeV).TABLE I1 BACKGROUND OBTAINED WITH A 1.3-T ELECTROMAGNET Radionuclide . . . . 38P 48K 38Cl 24Na Background,* yo . . 0.8 3.6 5.6 15.8 * Average of values obtained for different energies (up to 16 keV) expressed as a percentage of the background obtained without the magnet. The background obtained from each of the four principal radioactive isotopes (24Na 42K, 38Cl and 32P) produced in an irradiated biological matrix was measured both with and without a magnet. The results showed that the decrease in background is not the same for all the nuclides and varies with energy for the same n~clide.~ Table I1 shows the background (expressed as a percentage of the background obtained without a magnet) obtained for each of TABLE I11 PERCENTAGE CONTRIBUTION OF THE MAJOR INORGANIC COMPONENTS TO THE TOTAL ACTIVITY (MEASURED WITHOUT THE MAGNET) OF THE IRRADIATED MATRIX [ISOTOPES OBTAINED BY THE (n,~) REACTION 24Na 42K Wl, 27Mg 32P and 45Ca] Matrix Animal muscleD .. . . Milk powderlo . . . . Bovine liver11 . . . . Orchard leaveslS . . Bowen's kale" . . . . Urine" 5 . . . . ~iood14 . . Matrix Animal muscleo . . . . Milk powderlo Bovine liver" . . . . Orchard leaves" . . Bowen's kalela . . . . UrineI4 f . . . . ~iood14 Element A Na K c1 I \ I A A I r 1 r -t % % % & & A-p.p.m. A* Bf C$ p.p.m. A B C p.p.m. A B C 2.1 x 10' 46 59 6 1.6 x lo4 28 37 1 1.9 x 10' 13 0.2 -4.4 x 10' 15 71 6 1.7 x lo4 5.2 22 1 9.1 x 10; 78 5 -8.2 x 10 0.8 5.5 0.1 1.5 x lo4 12 88 0.6 2.0 x 10' 58 4.7 -2.5 x loa 18 53 1.5 2.5 x lo4 13 43 0.5 3.4 x los 64 2.4 -4.0 17 88 71 3.0 1 5.3 1.5 7.5 2.4 x loa 23 70 6.5 9.7 x los 7 23 0.5 2.7 X lo* 67 2.6 -3.3 x 10' 24 95 81 1.7 x 10' 0.1 0.4 0.1 3.8 X 10' 76 4.0 -82 6.5 -Element MI? P Ca % & p.p.m.A B C 1.0 X 10' 10.3 - -1.1 X 10' 1.4 - -6.0 X 10' 2 - -6.2 X 10' 29 - -1.6 X 10' 4.6 - -2.2 x 10 - - -0.1 - - -- -7 p.p.m. A B C 9.5 x 10' 2.7 3.8 77 9.1 X 10' 0.4 2 38 1.1 x 104 1 4.4 92 2.1 x los 0.2 1.8 8.3 4.5 X los 0.4 1.6 9 1.4 x 10' - 0.6 10.5 0.5 - 0.2 27.2 , p.p.m. 1.9 x 10' 1.3 x 101 1.2 x 10' 1 x 10% 6 x lo-* 2.1 x 1 0 4 4.1 x 1 0 4 * A 5 min irradiation and 6 min cooling time.f B 3 h irradiation and 3 h cooling time. 3 C 4 d irradiation and 4 d cooling time. $Grams per 24 h average value 1194 MANTEL the four nuclides with a 1 -3-T electromagnet. The greatest reduction in background is seen to be due to 32P. This may be because for this nuclide which is a pure /3-emitter the background is entirely due to the effect of /3-particles that are efficiently deflected by the magnet. For the other nuclides which are /3- and y-emitters the Compton peaks of the y-rays will also contri-bute to the background. The smallest decrease in background is obtained for Z4Na which at the same time has the highest abundance (100%) of gamma rays.* From the results shown in Table 11 it follows that the reduction in background obtained with a magnet for a given matrix depends to a great extent on its composition.Thus the limiting factors in the limit of detection obtainable by the present method will be the composition of the matrix and the intensity of the magnetic field. In considering the influence of the background on the limits of detection of trace elements in biological materials one may group the trace elements according to the half-lives of the radio-isotopes used for their determination i.e. those with half-lives of a few minutes to 3 h 3 h 4 d, or more than 4 d. The amounts of the major inorganic components of the matrices studied in this work and the contribution of each to the total activity (measured without a magnet) of the irradiated sample, after (A) 5 min (B) 3 h and (C) 4 d of irradiation and equal cooling time are given in Table 111.In the first instance all the major components of the matrix contribute to the background the highest contribution being due to chlorine (except in animal muscle). In the second instance the background is essentially due to sodium and potassium and in the third instance to phosphorus and calcium. Conclusion X-ray spectrometry and magnetic deflection of p-rays were applied to instrumental neutron activation of seven biological matrices. The limits of detection of 18 trace elements were calculated taking into account the influence of the background which is specific for every matrix. These limits of detection are based on the experimental set-up available in our laboratory and are thus only a guide line for other workers interested in applying this method under their specific conditions.This work was supported by the IAEA Vienna. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. References Mantel M. and Amiel S. Anal. Chem. 1972 44 548. Mantel M. and Amiel S. in Amiel S . Editor “Nondestructive Activation Analysis,” Elsevier, Amiel S. Mantel M. and Alfassi 2. B. J . Radioanal. Chem. 1977 37 189. Mantel M. Alfassi 2. B. and Amiel S. Anal. Chem. 1978 50 441. Alfassi 2. B. Biran-Izak T. and Mantel M. Nucl. Instrum. Methods 1979 17 227. Rapaport M. S Mantel M. and Nothmann R. Anal. Chem. 1979 51 1356. Rapaport S. Mantel M. and Shenberg C. J . Radioanal. Chem. 1982 75 145. Lederer M. Hollander J. M. and Perlmann I. “Table of Isotopes,” Sixth Edition John Wiley New Parr R. M. “Intercomparison of Minor and Trace Elements in IAEA Animal Muscle (H-4),” IAEA/ Dybczynsky R. Veglia A. and Sushny O. “Report on the Intercomparison Run A-1 for the Deter-“Certificate of Analysis Standard Reference Material 1577 Bovine Liver,” National Bureau of Stan-“Certificate of Analysis Standard Reference Material 1571 Orchard Leaves,” National Bureau of Bowen H. J. M. J . Radioanal. Chem. 1974 19(2) 215. “Report of the Task Group on Reference Man,” International Commission on Radiological Protection, Received February 2nd 1983 Accepted April 21st 1983 Amsterdam 1981 pp. 25-41. York 1967. RL/69 Report No. 2 October 1980. mination of Inorganic Constituents of Milk Powder,” IAEA/RL/68 July 1980. dards Washington DC June 1977. Standards Washington DC January 1971. No,. 23 1974 p. 290