Analyst, August, 1967, Vol. 92, p p . 529-531 529 Fast-neutron Activation Analysis of Silicon in Sputum BY A. SARDI (Central Research Institute of Physics, Budapest, X I I I , Vaci ut 34, Huizgary) AND A. TOMCSANYI (“Kordnyi” National Institute f o r Tuberculosis, Buda$est, Hungary) A method for a relatively rapid and accurate determination of silicon in sputum by activation analysis is described. The method has been de- veloped to identify and separate the cases of silicosis from those of pulmonary tuberculosis. The sputum is digested by trypsin, then spun in a centrifuge three times a t room temperature, and once after heating with 5 per cent. trichloroacetic acid to remove phosphorus which interferes in the activity measurement. For irradiation, a neutron generator with a fast-neutron yield of 2 x lo9 n per cm2 per second is used.After exposing for 2.3 minutes and then waiting for 1 minute the activity measurements are made by using a multi-channel analyser for 4.6 minutes. The silicon content is evaluated from the aluminium- 28 activity produced by the reaction 28Si (n,p) 28A1 with a photopeak a t 1.78 MeV. The results show that the silicon content in the sputum of patients suffering from silicosis is, on the average, four times that of the samples used for reference. IT is a current problem in tuberculosis dispensaries to distinguish cases of silicosis from those of pulmonary tuberculosis, as the X-ray photographs in the two cases are often quite similar. Considering its nature, it seems reasonable to assume that the sputum of patients suffering from silicosis contains much more silicon than that of patients with tuberculosis.The deter- mination of silicon with conventional techniques being lengthy, laborious and inaccurate, investigations have been made to develop a method for the neutron-activation analysis of silicon in sputum. Neutron-activation analysis has been already carried out for several elements present in biological substances.l Silicon has not been included in these experiments mainly because it is not an essential biological constituent. Another reason for this is that silicon-31, produced by the reaction 30Si (n,y) 31Si in silicon exposed to thermal neutrons in a reactor, being a p-emitter of 2-6-hours half-life, cannot be identified by gamma spectroscopy, so that compli- cated and lengthy chemical procedures have to be used for the separation of silicon. The recent availability2 of neutron generators producing fast neutrons with energies oi 14 MeV offers the possibility of making use of the reaction 28Si (n,p) 28A1, the cross-section of which is 220mb.Considering the 92.17 per cent. abundancy of silicon-28, the 2.3-minute half-life and the 1.78-MeV photopeak of aluminium-28 activity for the transition 28A1 + 28Si* and 28Si* the product of this reaction lends itself well to gamma spectroscopic measurement after zimple separation from interfering activities. Fast-neutron activation analysis of silicon in inorganic substances has already been extensively dealt with in the literat~re,l0,11,12,~~,~~ The method is rapid, selective and of high sensitivity, e.g., 0.4pg per 100 d.p.m.for a fast-neutron flux of lo9 n per cm2 per second. An accuracy of 1 per cent. has been achieved by Martin, Mathur and Morgan12 for a 2-g sample with a 1 per cent. silicon concentration. Although fast-neutron activation analysis by the use of neutron generators has not yet been applied to biological substances, considering the above, the method also seems promising for their analysis. = 1.78M2 28Si that one 1-78-MeV gamma photon is emitted per P-parti~le,~ to EXPERIMENTAL PREPARATION OF SAMPLES- individual rate of secretion. The sputum of various patients was collected during 3 to 6 days, depending on the530 SARDI AND TOMCS~~NYI : FAST-NEUTRON [Analyst, Vol. 82 REAGENT- Tris(kzydroxymetlzy1)aminomethane - formic acid bufer solution-This consists of a mixture of 50 ml of 0-2 M tris(hydroxymethy1)aminomethane (24.2 g dissolved in 1 litre of distilled water) and 26.8 ml of 0.2 M formic acid made up to 100 ml with distilled water.DIGESTION- Trypsin was dissolved in 0.1 M tris(hydroxymethy1) aminomethane - formic acid buffer to give a 1 mg per ml concentration and the solution added to the sputum collected over 24 hours in amounts equal to that of the measured sputum. The well mixed solutions were digested for 24 hours at 37" C, during which period the viscous sputum became fully digested. REMOVAL OF PHOSPHORUS- To eliminate any interference from aluminium-28 produced in phosphorus by the reaction 31P (n,ol) 28 Al, with a cross section of 130 mb, this element had to be removed from the samples before activation.This is achieved by spinning the solutions three times in a centrifuge after acidifying them to 5 per cent. with 100 per cent. trichloroacetic acid. The opalescent supernatant liquid is discarded each time after spinning for 30 minutes and the precipitate suspended in 10 to 15 ml of 5 per cent. trichloroacetic acid. To remove the inorganic phos- phorus bonded to the macromolecule, the precipitate obtained in the third spin is hydrolysed with 5 per cent. trichloroacetic acid for 30 minutes at 80" C before spinning it a fourth time, the precipitate being suspended each time in 20 ml of ethanol, which removes not only the phosphoric acid released by the hydrolysis from the macromolecule, but also the phosphorus bonded to the lipoids of the cellular fragments, as well as the trichloroacetic acid.The final precipitate is evaporated to dryness. In the samples prepared by this method no phosphorus could be determined by the Fiske - Subba Row test.15 The detection limit for this test is 0-5 pg of phosphorus. In the samples prepared without the addition of trichloroacetic acid, the phosphorus concentration, checked by the Fiske - Subba Row method, was found to vary from 10 to 80 pg per 100 mg of dry substance. By "spiking" the sample with 1 mg of silicon dioxide we determined that after the phosphate separation the recovery of silicon was 95 per cent. IRRADIATION- The dried samples are put into 2 x 2-cm polythene bags and transferred in a cylindrical polythene holder to the neutron generator by means of a vacuum facility. Their arrival at the place of irradiation is checked by photo diodes, and causes the automatic start of the assembly that measures the neutron yield, and of the timer that times the operation.of the relay and valve by which the sample is returned to the place of measurement after the pre-set time for irradiation. The irradiations were carried out in the KFKI NA-1-type neutron generator producing 14-MeV neutrons at a maximum rate of 1O1O n per second.16 The neutron flux was measured by Gamma Production, Emmerich-type plastic phosphor fast-neutron detector and 2 to 3 minutes' irradiation times were applied with a neutron flux of about 2 x lo9 n per cm2 per second. ACTIVITY MEASUREMENT- Two different assemblies were used for the activity measurements.Assembly A con- sisted of a 3.5 by 2.75-inch sodium iodide - thallium crystal detector in a lead chamber con- structed from 10-cm thick lead bricks and a KFKI-type 128 channel analyser. The irradiated samples could be measured immediately after irradiation. The samples were left to cool for 1 minute to remove the 7-3-second half-life, 6 to 7-MeV nitrogen-16 activity produced by the l60 (n,p) 16N reaction. Assembly B consisted of a 6 by 4-inch sodium iodide - thallium crystal detector in an iron chamber of 15-cm wall thickness and Raduga-type 100 channel analyser. This assembly was located farther from the neutron generator, so that the activity measurement could not be started until 5 to 6 minutes after irradiation. It proved to have a lower background and a higher counting efficiency than assembly A.The measuring times were 4-6 minutes. RESULTS AND DISCUSSIONS The silicon concentration was evaluated from the 1.78-MeV peak activity measured in the repeatedly calibrated energy-channel width of 1.78 t 0.11 MeV. As the measurementsAugust, 19671 ACTIVATION ANALYSIS OF SILICON IN SPUTUM 53 1 TABLE I MEASURED SILICON ACTIVITIES AND CONCENTRATIONS IN FAST-NEUTRON ACTIVATION The values are normalised for 1-78 & 0.11 MeV to 100 mg of dry material, 2-3 minutes’ activation, 1 minute’s decay, 4-6 minutes’ measuring times and given neutron flux ANALYSIS OF SPUTUM FOR SILICOTIC AND NON-SILICOTIC PATIENTS Concentration of silicon in sputum, p.p.m. Ast, counts per minute A 7 7 - 7 Assembly A Asscmbly B Assembly A Assembly B - Background 91 & 10.5 89 & 9.3 - Silicosis No.1 . . . . 68 1. 13.4 253 & 34.5 17.6 1. 2.1 18.7 5 4.1 Samples 2 . . . . 82 & 20.2 222 5 29.2 18.2 & 2.8 15.6 & 1.8 3 . . . . 99 & 18.3 190 f 25-3 14.7 & 3.2 17.2 & 3.2 4 . . . . 80 & 12.5 210 f 37.2 20.1 -I: 2.3 21.3 f 1.9 Control 5 . . . . 27 8.6 49 f 14.0 3.9 f 1.5 4.2 & 1-7 6 . . . . 15 & 7-0 36 5 9.8 2.8 f 0.9 1.8 f 0.8 7 . . . . 25 & 6.7 92 & 15.2 5.1 1.7 3.7 f 0.43 were carried out under somewhat different conditions, the measured activities, Am, were normalised to 100-mg amples, 2.3 minutes’ activation, 1 minute’s waiting time and 4.6 minutes’ measuring time with a half-life of 2.3 minutes for the given reading on the meter measuring the neutron yield.The normalised activity values were taken as standard and have been calculated from the formula- where A , is themeasured activity A , uni.;tzus background counts; ~7 is the total time coefficient and equals qa.qw.Tm, the subscripts a, w and m denoting the time coefficients of the activation, waiting and measuring times, respectively; f is the reading of the neutron yield, meter counts; and w is the weight of the sample in milligrams. The average values of the standard activity obtained by using either of the assemblies are seen to be much higher for silicosis than for the samples used for the check (Table I): for assembly A, 83.3 counts as compared with 22.2 and for assembly B, 226 as compared with 59. Even the minimum values measured for silicosis are found to be more than twice the maximum value found for the control samples.The results thus seem definitely to support the diagnostic value of the silicon analysis in sputum. Samples from which the phosphorus had not been removed were also measured. The results obtained spectacularly confirmed the need for the separation from phosphorus. In view of the limited number of cases investigated to date, no definite conclusions can be drawn, but the results are promising enough for the experiments to be continued on a more extensive scale and under increasingly uniform conditions. We thank Professor M. Boszormhyi, Director of the “Korgnyi” National Institute for Tuberculosis, for suggesting the problem and his continued interest, and Miss Lucy Gonda for assistance with the experiments. ASt = 100 Aa/Tv. f.w 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. REFERENCES Berton, M., CEA, 1964, 50, 82. Meinke, W. W., and Shideler, R. W., Nucleonzcs, 1962, 20, (3), 60. Lbov, A. A, and Naumova, I. I., Atomn. Energ., 1959, 6, 468. Coleman, R. F., Analyst, 1961, 86, 39. Gillespie, A. S., and Hill, W. W., Nucleonics, 1961, 19, (ll), 140. Guinn, P., Ibid., 1961, 19, ( S ) , 81. Maddock, R. S., Meinke, W. W., T.I.D. Prog. Rep. 10, TID-14310, November, 1961, p. 61. Steele, E. L., “Proceedings of the International Conference on Modern Trends in Activation Kusaka, Y., Tsuji, H., Fujii, J., Muto, H., andMiyoshi, K., Bull. Cheun. SOC. Japa,, 1965,38, (4), 2. Turner, S. E., Analyt. Chem., 1956, 28, 1457. Caldwell, R. L., and Mills, 1:. R., jun., Nucl. lnstvuun. Meth., 1950, 5, 312. Martin, T. C., Mathur, S. C., and Morgan, J. I,., Int. J . Appl. Radiat. Isotopes, 1964, 15, 331. Wing, J,, Analyt. Cheun., 1964, 36, 559. Vogt, J. R., and Ehmann, W. D., Geochiwa. Cosmochiun. A d a , 1965, 29, 373. Colowick, S. P., and Kaplan, N. O., “Methods in Enzymology,” Academic Press Inc., New York, Pksztor, E., and Klopfer, E., KO@. Fiz. Kut. Int. Kozle., 1964, 12, 143. Analysis,” Texas, 1961, 161. 1957, p. 843. Received July 18th, 1966