ISSN:0003-2654    

DOI:10.1039/AN98914FX009

出版商:RSC    

年代:1989

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN98914BX011

出版商:RSC    

年代:1989

数据来源: RSC

摘要:

ANALYST, MARCH 1989, VOL. 114 SECOND INTERNATIONAL CONFERENCE ON NUCLEAR AND RADIOCHEMISTRY Brighton, UK, 11-15 July, 1988 This conference was the second in a series to be arranged by members of several of the Chemical Societies of the European Community. The Technical Organising Committee consisted of Professor Dr. H. J. Ache (Federal Republic of Germany), Professor Dr. J . P. Adloff (France), Professor Dr. K. Buchtela (Austria), Professor Dr. H . R. Von Gunten (Switzerland), Professor Dr. K. H. Leiser (Federal Republic of Germany), Dr. A. R. Ware (UK) and Mr. M. A. Crook (UK), supported by representatives from Japan, USSR and the USA. The first conference, held in Lindau, Federal Republic of Germany, in October 1984, was a considerable success and it was decided to hold a second conference to which scientists interested in nuclear and radiochemistry were invited to present and discuss the latest developments in these subject areas.The main aim of the conference was to allow scientists to discuss technical developments in nuclear technology and radioanalytical chemistry in a non-political way by concentra- ting on the technology and,the interpretation of results. This was of particular significance as the conference took place after the major nuclear incident at Chernobyl. This incident highlighted the fact that pollution can transgress national boundaries and it is imperative that the nations concerned can understand the measurements taken, can rely on each others’ results and are able to deduce their significance. In this context, exchange of results, standard methods of analysis and intercomparison of techniques are all of importance and were discussed.The conference attracted 100 papers grouped into Subject Sessions. Plenary Lectures were given on the following topics: “Chemistry of Heavy Ion Reactions,” “Short Lived Radio- nuclides ,” “Radioisotope Production and Labelling,” “Advances in Nuclear Analysis Methods,” “Chemistry of the Nuclear Fuel Cycle and Fusion Technology,” “Measurement of Radioactive Nuclides in the Environment,” “Migration of Nuclides in the Environment ,” “Radiochemical Methods in Space Research” and “Aspects of Radioanalytical Chemistry. ” The nuclear power industry has been placed in a high profile throughout the world, but the conference also recognised the advances in nuclear and radiochemistry in other fields including medical diagnosis and treatment, space research, engineering diagnosis, analytical chemistry and tracer studies.The delegates were welcomed to the conference by Mr. B. V. George, Director of Pressurised Water Reactors (PWR) 253 for the Central Electricity Generating Board. Whilst naturally referring to the role of nuclear and radiochemistry in the production of nuclear power in his opening remarks, he also predicted that greater efforts will be required in future to develop measurement and control techniques. This, he said, was because the current trend is for more stringent control of the use of radioactive material and this can only be assured by the availability of appropriately developed methodologies.The development of these methods would of necessity require good scientific quality assurance of the work and results. The conference was attended by delegates from many countries in Europe, the Americas, Japan and China. All benefited by the open discussion of the papers, some of which were presented in poster sessions. These sessions proved to be of considerable help and interest as it provided a better opportunity to review the work undertaken and discuss at length the results with the authors themselves. In support of the conference, there was an exhibition of the latest equipment and services offered by manufacturers. The development of proposed techniques and methods must be backed up by reliable working instruments and hardware, and the availability of such instruments was certainly demon- strated by those manufacturers present.Their support at the conference was much appreciated. Apart from the rigours of the conference and exhibition halls, the delegates found time for social events, which gave a further opportunity for scientists to get to know one another better. I, for one, will not forget trying to explain the origins of Morris Dancers to a group of delegates from Japan. At the end of the conference the Technical Committee met to consider the publication of its proceedings and the venue for the next conference. The Committee decided to offer papers submitted to them for publication, which has resulted in this special issue of The Analyst. As many papers were not strictly analytical chemistry, some authors have chosen to publish their work elsewhere. This edition does not therefore reflect the full proceedings of the conference. Finally, it was decided to hold a third conference in this series on Nuclear and Radiochemistry in Austria in September 1992. Please book the date in your diary, every attempt will be made to continue to build on the success of the previous conferences . Dr. A. R . Ware Chairman Technical Organising Committee (CEGB, Canal Road, Gravesend, Kent)

ISSN:0003-2654    

DOI:10.1039/AN9891400253

出版商:RSC    

年代:1989

数据来源: RSC

摘要:

ANALYST, MARCH 1989, VOL. 114 255 Measurement of Radioactive Nuclides in the B. F. Myasoedov and F. 1. Pavlotskaya V. 1. Vernadsky Institute of Geochemistry and Analytical Chemistry, StZ. 19, SU- I17975 MOSCOW, USSR Environment* USSR Academy of Sciences, Kosygin The sources of environmental pollution by natural and technogenic radionuclides (3H, 14C, *5Kr, 1291, 90Sr, 134Cs, 137Cs, 144Ce and the transuranium elements) have been considered. Data on the content of transuranium elements in environmental samples and the distribution of plutonium in surface ecosystems are presented. In addition, data concerning the radiation dose received by the human body due to natural and technogenic sources are given. Various radiochemical and radiometric methods for the determination of radionuclides in soils, sediments, natural waters and biological materials are discussed.Keywords: Natural and technogenic radionuclides; methods and measurement; environment The study of environmental contamination caused by anthro- pogenic impact and, primarily, by radioactive nuclides is one of the main scientific problems facing contemporary science. Human and living organisms can be affected by several sources of radiation. These include: (1) radiation from natural radionuclides; (2) environmental radiation from artificial radionuclides; (3) irradiation from the sources used in medicine; and (4) irradiation from working with sources of ionising radiation (professional irradiation). According to data from the National Council of Radiation Protection and Measurements (USA), the emission of radon, formed through the decay of natural uranium, is responsible for about 55% of the annual mean dose of radiation received by the general population.Medical sources contribute 15% , internal irradia- tion from natural radionuclides (4°K and uranium and thorium radionuclides) 11%, cosmic rays 8% and the inherent radioac- tivity of rocks and minerals 8%. The remaining 3% of this annual irradiation dose comes from professional irradiation and from global radioactive fallout as a result of the activities of the nuclear power industry. The dose produced by the testing of nuclear weapons in the atmosphere, which reached about 7% of the annual mean irradiation from natural sources in 1963 (Fig. l), now contributes less than 1%. By the early 1980s the contribution from the nuclear power industry had still not reached 0.I % and if the forecast of its development is correct it will contribute less than 1% to the natural background radiation by the year 2000. Based on reports by Soviet workers,1-3 the accident at the Chernobyl atomic power st3t.m will only increase the total dose of radiation received by the population in the European part of the USSR by a negligible amount. In order to be able to measure accurately the level of radioactive contamination in various natural materials, includ- ing the content and form of occurrence of various radio- nuclides, simple and efficient methods of determination are required. Radiometric methods are mainly used for this purpose, although in some instances, e.g., for the determina- tion of uranium, plutonium or neptunium , mass spectrometry, luminescence and neutron activation analysis are used.Contemporary radiometric methods, particularly those which include the automatic processing of the experimental data by computers, allow the direct determination of indivi- dual radionuclides in various materials with high sensitivity. Prior to their determination it is first necessary to extract and concentrate the radionuclides to be determined and then to separate them from other radioactive impurities. The methods most often applied for this purpose are extraction and chromatography and extraction - chromatography, in addition * Plenary lecture presented at the 2nd International Conference on Nuclear and Radiochemistry, Brighton, UK, 11-15 July, 1988.to precipitation and co-precipitation, chromatography on paper, sublimation and electrophoresis. Radiochemical analysis, based on the principles and methods of analytical chemistry, has its own unique features, which are connected with the separation of ultratrace amounts of a substance. As a rule, isotopic and non-isotopic carriers are used for this purpose and they behave in the same way as the radionuclides being determined.4 Recently, both extraction and chromatography have been used increasingly for the separation of radionuclides into a radiochemically pure state. These methods are characterised by their high selectivity, simplicity and high sample throughput; they virtually elimi- nate the adsorption processes that can occur during precipita- tion, and also the uncontrolled losses of radionuclides or contamination by impurities.When these methods are applied, the separation and radiochemical purification of radionuclides can be carried out in the absence of carriers. Methods involving the step by step separation of radio- nuclides are used for the simultaneous determination of several radionuclides in one sample (Fig. 2). When choosing a particular method of radiochemical analysis, the various forms of radionuclides that occur in environmental materials should be taken into account.6.7 For instance, in both the liquid phase of global fallout and in atmospheric precipitation, radionuclides are present in several forms (Table 1); the ratios of these forms are different for individual nuclides, they also vary with the seasons of the year.The content of neutral and anionic forms of radionuclides increases, with an increase in the amount of natural organic material present. The underrated data can also be obtained by 1945 1955 1960 1965 1970 1975 1980 1985 Year Fig. 1. Effective annual equivalent dose due to (2) medical utilisation of radiation, ( 3 ) nuclear explosions in the atmosphere and (4) nuclear power compared with (1) the background radiation dose256 ANALYST, MARCH 1989, VOL. 114 Bottom sediments 9OSr, 21°Pb, 229Th, 236U, 242Pu, 243Am HCI - HN03 - HF - HC104, HN03 A1 iquot I HN03, K2Cr04 solution SrC03 1 M HN03 Fe(OHI3 SrC03 Y, NH3 Y2b3 Aliquot I ~ M H C I '1 AG 1-X2 HCI I Nd(r9H: HCI AG 1-X8 i 4 1 - HCI Mean chemical yield, O h 210Pb 90Sr 243Am 229Th 242pu 60 65 -20 -20 -80 Fig.2. Scheme for the determination of fission products. natural and transuranium elements in bottom sedimentsi Table 1. Forms of occurrence of radionuclides in the liquid phase of global f a l l o ~ t . ~ , ~ All values expressed in YO. Values in parentheses are the means Form T3r 137cs C e Cationic . . . . 79-99(90) 50-98(72) 38-84(55) Anionic . . . . 0.5-22(6) 2-35 (16) &30 (18) Neutral . . . . &18(4) 0.7-27(12) 9-54 (27) the analysis of fresh water, provided that the radionuclide concentration is carried only on ionites (Table 2). Owing to a considerable improvement in both the instrumentation and techniques for measuring the half-lives of radioisotopes, radiometric methods for the determination of transplutonium elements (TPE) are now very precise.This is reflected in those methods for the determination of TPE that measure a-activity in which the best characteristics (low detection limit, high efficiency and small errors) are attained using a silicon semiconductor detector and a liquid scintillator. An alpha spectrometer equipped with a semiconductor detector (area 1-3 cm2) has an efficiency of nearly 50"/0 and a detection limit of 10-3 Bq (in the sample). As the energy resolution of such detectors reaches 20-30 keV, it is possible to determine radionuclides without preliminary separation. For complex a-spectra containing superimposing lines, differ- ent mathematical methods of processing the spectra are used. The electrodeposition of radionuclides on a backing material Table 2.Forms o f occurrence of transuranium element5 in groundwater.% All value5 expressed in Y' Form NP Pu Am Cationic . . . . . . . . 0.1 0.3 3.3 Anionic . . . . . . . . 76.4 34.7 0.1 Neutral . . . . . . . . 23.5 65.0 96.6 derived from aqueous or organic solutions (ethanolic, isopro- panolic, isobutanolic or mixed solutions), which gives a thin uniform coating on the target and a high resolution when measuring a-particles, is used more often on targets employed in a-spectrometry. However, good samples can still be produced without electrodeposition if a fine membrane filter is used as the backing material. Targets with sorption properties are also used, e.g., cation discs or discs of stainless steel with an ultra-thin layer of zirconium phosphate coated on the surface.If a sample contains only one a-active nuclide, or one a-active element with a particular isotopic composition (or the total a-activity is to be determined), then it is possible to use methods based on integral a-counting. Those methods in which the active sample is placed in the detector are the most efficient: a flow proportional 2n- or 4x-counter is used for dry samples and a liquid scintillation (LS) counter for solutions. The proportional counters have a counting efficiency of 50% and a very low background (about 10-3 counts s-1) making itANALYST, MARCH 1989, VOL. 114 257 possible to measure the absolute a-activity with an error of only 0.3%. The LS counters, which have a much higher background than the proportional counters, are also widely used for measuring the a-activity of radionuclides.The advantages of the LS method are the high efficiency of the counting (almost 1 OO./,) and the elimination of laborious steps such as the preparation of a dry sample for measuring. There are two methods of preparing samples for LS counting: (1) direct addition of an aqueous or organic solution of the sample to the liquid scintillator: and (2) extraction of the radionuclide from the sample with a liquid scintillator. In the latter method an extractant is added to the liquid scintillator solution. When biological materials were analysed using the LS method, the chemical yields of americium, curium and californium were 94-98, 93-99 and 93-94"/,, respectively,g and that of pluto- nium in soils was 75%. 10 The problem of counting (3-particles, which are of low energy.arises with the analysis of samples containing 3HH, 14C, etc. Normally the proportional flow counter (counting effi- ciency 5C-100%, background about 1 count s-1) or the LS counter is used for this purpose. Many radioactive isotopes can be determined using y-spectrometry in conjunction with semiconductor Ge detectors with multi-channel pulse analys- ers. Hence a method has been developed for measuring the abundance of 739Pu in various samples by detecting y-rays with energies of 414 and 129 keV, with an error of about 5%. Similarly. the abundance of 37Np can be measured from :,-rays with energies of 29 and 86 keV. A convenient method for the determination of nuclides that undergo spontaneous fission is to measure the neutron activity.Even the isotopes of plutonium, curium and califor- nium can be regarded as nuclides with a high neutron activity. The neutron counting method makes it possible to determine the concentration of these nuclides in various samples without destroying the samples. The influence of other a-active nuclides should be taken into account as these will give an additional flux of neutrons at the expense of nuclear reactions in a light nucleus of the type ( a : n). Mass spectrometry is widely used both for the determina- tion of the isotopic composition of uranium, plutonium, americium and curium and for the precise determination of these elements in the materials being analysed. In addition to its high sensitivity and precision, the main advantage of the mass Spectrometric method is that it does not require quantitative separation of the elements to be determined.Generally, 0.1-0.5 ng of plutonium or neptunium is sufficient for mass spectrometric measurements. The relative standard deviation is 0.2-0.3%, although it can be reduced to 0.1%. Natural Radionuclides There are more than 60 natural radionuclides in the Earth's biosphere and these can be divided into two categories: original and cosmogenic. The original natural radionuclides can be subdivided into two groups. The first group includes 11 radionuclides of elements from different positions in the Periodic Table, all of uhich have long half-lives (from 107 to 1015 years) (Table 3). The generation of some of these radionuclides ( 13xLa, lJ7Sm and 17hLu) is connected with the spontaneous fission of uranium in situ.The second group is composed of 32 radionuclides existing in the biosphere, i.e., the decay products of long-lived uranium and thorium isotopes. The cosmogenic radionuclides (Table 4) are formed mainly through the interaction of cosmic rays with the atmosphere and, to a lesser extent, with the Earth's crust. Once formed in the atmosphere, the cosmogenic radionuclides may reach the Earth's surface by means of atmospheric precipitation, they may interact in the gaseous form with components of the stratosphere or they may fall to Earth by becoming adsorbed on solid particles in the atmo5phere. Table 3. Characteristics of long-lived natural radionuclides in soils Energy/ Concentration1 Radionuclide years MeV gg-' 14 . .. . 1.3 x lo" . . . . 4.8 x 1010 . . . . 6.0 x 1014 . . . . 1.1 x 10" . . . . 1.2 x 10" . . . . 2.2 x 101" . . . . 6 x 101(1 . . . . 2.5 x 105 . . . . 7.1 x 108 . . . . 4.51 x 109 . . . . 8x104 . . . . 1.41 x 10"' . . . . 1620 . . . . 20 13 1.32 (88%) '1 1.46 (12%) p 0.27 ( 100%) p 0.63 (95%) -1 1.43 (70%) p 0.20(30%) y 0.81 (30%) a 2.2 (loo%) p (1.43 (100%) '1 0.31 (100%) a 4.77(72%) (3 0.40 a 4.72(28%) 4.18-4.56 '1 0.185 (55%) CY 4.19 (77%) y 0.05 (23%) a 4.68(76%) a 4.61 (24%) y 0.43 (12%) y 0.254 a 4.01 (76%) y 0.59 (24%) CY 3.95(24%) CY 4.76(95%) '1 0.187 (4%) (3 0.016 (85%) y 0.046 ( l5%) p 0.063 ( 1 5 % ) K, 2~ 10-2 Rb. 3 x 10-4 In, 1 x 10-7 La, 2 x Sm, 6 x lo-" LU, 7 x 10-7 Re. 1 x 10 Y 8 x 10- 1 1 7 x 1 0 - y u. 5 x 10-6 2 x 10- '1' 1 x 1 0 1 x 10-12 4 x 10-14 Table 4.Characteristics of natural radionuclides induced by cosmic rays Rate of form at i on/ -@"lax 1 Radionuclide atoms cm-2 s 1 4 keV 3H . . . . . . (I. 20 12.3 years 18 loBe . . . . 4.5 x 10-2 2.5 x 10hyears 555 22Na . . . . 8.6 x 10-5 2.6 years 545((3+) 32Si . . . . . . 1.6 x 10-4 700 years 210 14c.. . . . . 2.5 5700 years 156 26A1 . . . . . . 1.4 x 7.4 x 10'years 1.17 3 . i ~ . . . . . . 6.8 x 10-4 25 d 248 35s . . . . . . 1.4 x 10- 3 8 1 d 167 '"1 . . . . . . 1 . I x 10-3 3.1 x los years 714 81Kr . . . . 1.5 x 10-7 2.1 x ZOSyears Electron-capture 'YAr . . . . 5 . 6 ~ 10-3 270 years 565 Some of the most important natural radionuclides that occur in the atmosphere are the isotopes of radon, viz., 22oRn and 222Rn, and their daughter decay products.The mean abundance of these isotopes in the atmosphere is 1.8 Bq m-3, whereas in buildings it is much higher (15 Bq m-3). Owing to the shorter half-life of toron (55 s) compared with that of radon (3.8 d), the concentration of the toron decay products in the atmopshere is about ten times lower than that of the radon decay products. The concentration of radon is determined from its daughter decay products, which accumulate on the filters during filtration of a sample of air for analysis. The methods of Kusnetz12 and Thomas13 and their modifica- t i o n ~ , ' ~ . ~ ~ in addition to the rapid methods of Rollelh and Markov et af. 17 have been widely applied to the determination of the concentration of the equivalent equilibrium radon.These methods have been compared by Terentjev and Krisyuklx with respect to their productivity, the error in measuring concentrations and the sensitivity of the determina- tion. As can be seen from Table 5 , the method of Thomas is the most precise, whereas Rolle's method has the lowest precision. The best sensitivity (about 6.6 g Bq min-1) is achieved using the Kusnetz - Nazaroff method; Thomas' method has the lowest sensitivity. To achieve a sensitivity equal to that of the Kusnetz - Nazaroff method, it is necessary258 ANALYST, MARCH 1989, VOL. 114 to filter 32 times as much air when using the method of Thomas. The method of Kusnetz is the most precise and sensitive when the correct sampling time is used, i.e., the time from the beginning to the end of the measurement of the filter activity, and it requires only a small low-production air intake device.Hence this method has been widely applied to field measurements. The methods of Rolle and Markov permit the rapid determination of the concentration of equilibrium radon. The former method is about twice as sensitive. but it is the least precise of all these methods. An important feature of Markov's method is that by performing two measurements of the filter activity it is possible to determine the concentration of both radon and its daughter decay products. However, although this method provides more information and is not time consuming (15 min), it is less senqitive. The Concentration of cosmogenic radionuclides such as 'H, 'Be. IT and "Na, formed by the action of cosmic rays in the upper layers of the atmosphere, is negligible. However, in recent years 'H and I T have accumulated as technogenic products of the nuclear fuel cycle.Technogenic Radionuclides Radioactive material of technogenic origin, mainly fission products of uranium and plutonium, were found in the atmosphere after 1945, i.e., after nuclear tests and the production and application of radionuclides had begun, and from the rapid development of atomic power stations (after 1983). At present, atomic power stations make an essential contribution to the global production of electricity. According to data from the International Atomic Energy Agency (IAEA) ~ in 1987 31 7 atomic power-generating units were operating in 26 countries, producing a total of 285 X 10') W of electricity, while ten more qtations were being built.Data from the IAEA regarding future developments in atomic energy are presented in Table 6. As has already been mentioned. the contribution of contemporary nuclear energy, including the production of uranium. the processing of nuclear fuel and the disposal of radioactive waste, to the total doqe of annual irradiation does not exceed 0 . 1 YO. Most of the artificial radionuclides have reached the biosphere through nuclear tests. It is k n m n that an explosion of 1 Mt forms 3.7 x 1015 Bq of Y3r and 6.2 X 1OIi Bq of 137Cs. Table 7 gives data for the total amount of these radionuclides generated by nuclear explosions. Data giving the sources of some fission radionuclides are presented in Table 8. The data show that the main source of the radionuclides studied is global radioactive fallout formed as a result of nuclear tests.From report\ by Soviet workers,l.2i Table 5. Comparison of methods for the determination of the equivalent cquilibrium radon in air18 Difference in precision C,,,,, / Time/ Method values g Bq rnin - 1 min Kusnetz" . . . . . . . . 0.053 25.5 55 Kusnetz - Groerls . . . . 0.048 14.2 6 0 Kusnetz-Nazaroff14 . . . . 0.038 6.6 80 Rollel6 . . . . . . . . 0.102 25.5 17 Markov ei al. . . . . . . 0.044 48.8 15 Thomas' . . . . . . . . None 214.3 35 Table 6. Forecast of the power output of electricity and atomic power stations. All values expressed in gigawatts 1978 1979 1980 1985 2000 Electric power Atomic power APS,'% . . . . . . 5.8 6.4 7.3 12 20-27 stations .. . . 1830 1900 2030 245C2850 5230-6200 stations(APS) . . 106 122 148 290-350 1030-1650 the accident at the Chernobyl atomic power station will only slightly increase the total dose of radiation received by the population of the European part of the USSR. and the discharge of radionuclides such a s S5Kr, '"'Sr. 1311 and 137Cs contributes about 1% to the level of the global radioactive fallout. Some of the most dangerous contaminating com- ponents are the long-lived transuranium elements (TUE) (Table 9). As with other radioactive isotopes, the main sources of these isotopes are nuclear tests and accidents involving satellites. An explosion of 20 k t discharges 101-1 Bq of z-3qPu into the atmosphere. Overall. nuclear tests have produced about 5 t of globally scattered plutonium (of which 60% is 2-3vPu).The proportion of a-emitters in the collective equivalent irradiation dose is insignificant (<5% ) compared with radionuclides such as 137Cs, ' 9 r . 'IiZr and 1°(lKu. which contribute 53, 13, 9 and 9%, respectively. However, TUE isotopes have long half-lives and hence they remain poten- tially dangerous to mankind for a long period of time. Various -!-spectrometric and radiochemical methods have been developed for the determination of artificial radionu- clides such as '%r. '"Zr, "-5Nb, IO-~RU, 1O6Ru, l - 3 T s and 1-37Cs, in environmental materials. These methods are widely applied and have been described in a number of reviews and monographs.j.27 ~7 Hence, methods for the determination of some radioactive gases, mainly transuranium elements, t ill now be discussed.Determination of Radioactive Gases An important contribution to the global contamination of the biosphere is made by the long-lived radionuclides IT, \iKr. ?H and 1291. but effective methods for their concentration have not yet been developed. Table 10 gives data predicting the level of these gases in the biosphere from operations of nuclear power stations. In the calculations it is a\sumed that 100% of the T and XiKr and 1% of the l2'jI formed in the active zone of the nuclear reactor enter the atmosphere. Kryptorz Krypton-85 enters the biosphere after nuclear tests are over and mainly during the processing of nuclear fuel. A very small amount of X5Kr is also produced by atomic power stations. It is a (3-emitting fission product with a half-life of 10.7 years.Various procedures are used to detect and concentrate XiKr from waste gases. Quantitative separation can be achieved by extraction with fluorocarbons, by low temperature distillation and adsorption on charcoal or silica gel, by the use of membranes with selective permeability and by precipitation in the form of clathrates. The measurement of the specific activity of krypton samples, prepared by one of these methods, is carried out with the usual low background instruments employed for counting P-particles. Proportional gas counters with inner inflation are used most often. The detection limit in this instance is 2 X 30-5 Bq cm--3 for the analysis of a 50-1 air sample at atmospheric pressure.'X Table 7. Amount of ''(% and 17'Cs generated through nuclear tests in the atmospherel" Explosive Calculated artiount/lOl" Bq power1 Year of test 1945-51 .. . . . . 1952-54 . . . . . . 1955-56 . . . . . . 1957-58 . . . . . . 1959-60 . . . . . . 1961-62 . . . . . . Total: ' The nuclear explosive equivalent. Mt* OOSr ' 3'Cs 0.02 0.0075 0.0 127 1 .o 0.37 0.645 5.6 2. I 3.55 31.0 11.7 19.7 01 38.0 63.5 38.62 52.18 ri8.4 1 power i\ c h a r a c t e r i d by the trotyl - - -ANALYST, MARCH 1989, VOL. 114 259 Table 8. Sources of radionuclides in the environment Global radioactive fallout2(!.21/ Radionuclide 4 10l5 Bq 3H 14c 85Kr Y3r 106R~ 1311 134Cs 137cs 144Ce 12.3 years 243 x 103 5730 years 220 10.7 years - 28.5 years 600 2.06 years - 368 d - 8.02 d 700 x 102 30.2 years 910 284 d 300 x lo2 Annual discharge into the atmosphere/ 10IsBq Atomic power Processing stations2(] -22 23 plant s2O.24 3 x 10-3 44-3.5 7 x 10-7-2 x 10-2 2 x 10-5-4 x 10-3 7 x 10-6-2 x 10-4 1 X 10-5-3 X 10-1 z x 10-7 10-5 10- 10-10-4 10-6 10-6 10-5 - - 1.7 x 10-6 2.3 x 10-4 - - Annual discharge of liquid wastes from atomic power stations20,*2/ 10l5 Bq 3 x 10-6-2 x 10-1 - - 10-1 1-10-7 - 10- "1 0- 5 10- 10!-10-4 10-10-3 x 10-s - Discharge during the accident at Chernobyl J 5 / 10'5 Bq - 33 74 59 270 185 37 89 Table 9.Sources of transuranium elements in the atmosphere (Bq) Nuclear explosions in the atmosphere Radionuclide (before 1979)26 237Np - 239Np - ~ " P U . . . . . . 3.3 x 1014 23YPU . . . . . . 7.4 x 10'5 240PU . . . . . . 5.2 x 10'5 2 4 1 ~ ~ . . . . . . 1.7 x 1017 242Pu .. . . . . 1.6 x 10'3 241Am - 243Am - z42Cm - 244Cm . . . . . . 2.6 x 1011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Atomic power stations (1 GW per year)*6 - 1 x 107 1 x 107 1 x 107 2.8 x 108 4 x 106 - Processing of the irradiated fuel (1 GW per year)*6 7.3 x 103 2.4 x 105 4.0 x 107 7.8 x 106 6.7 x 106 1.8 x 104 4.1 x 104 2.4 x 107 8.1 x 107 3.3 x 107 1.5 X 109 Discharge during the accident at C he rno by1 1 $25 4.4 x 10'6 3.0 x 1013 - 2.6 x 1013 3.7 x 1013 5.5 x 10'5 7.4 x 10'" - 7.8 x 1014 Table 10. Forecast of the accumulation of 85Kr, T, 14C and 1291 due to emissions from nuclear power stations37 Forecast of atomic power Amount accumulated/lO15 Bq station output/ Year 1OYW 85Kr 3H '4C 1291 1980 146 8.9 x 103 4.6 x 102 9.6 x 10-1 25.5 X 10-3 1990 700 3.9 x 103 20.5 x 102 4.1 11.1 X 10-2 2000 1650 133 X 10' 70 X 102 15.2 4.07 X 10-I 2025 6650 706 X lo3 386 x 102 1073 28.4 X 10-1 1975 76 5 x 103 2.6 x 102 5.1 x 10-1 9.7 x 10-3 Tritium Natural tritium is formed in the atmosphere mainly as a result of the interaction of cosmic rays with atoms of nitrogen, oxygen and argon.The equilibrium global abundance of natural tritium is approximately 1.3 X 1018 Bq. About 99% of the total amount of natural tritium is converted into natural tritium water and tritium hydrogen. Much more tritium has entered the atmosphere through nuclear weapons tests. The thermonuclear tests carried out before 1979 discharged 120-160 x 1018 Bq of tritium, which greatly exceeds the level of natural tritium. Tritium is also generated and dispersed into the environment by atomic power stations and by the reprocessing of nuclear fuel.The half-life of tritium is 12.26 years and its decay is accompanied by the emission of p-particles with a maximum energy of 18.6 keV. These nuclear - physical properties of tritium govern its determination in various materials. The methods for the determination of tritium have been reviewed39 and information given on how to select and prepare samples and carry out radiometric measurements. In order to determine tritium in solutions, scintillation rad- iometry on either solid organic or plastic scintillators, and also on liquid scintillators, which are highly sensitive, is the method most often used; the efficiency of 3H counting is 40-60°/0. Proportional counters with inner inflation are used mainly for the analysis of gaseous samples or when high sensitivity in the determination of tritium is needed. Although these counters can measure tritium water vapour, it is better to convert the analyte into the gaseous form.Before tritium is determined it is usually concentrated, for example, by electrolysis and distillation (water), and by dry or wet combustion for solid samples. When tritium was determined in the waters of the Baltic Sea using a proportional counter, the sensitivity of the method was about 2.6 Bq 1-1 with a coefficient of variation of 13 and 4% for amounts of tritium ranging from 12 to 25 and from 120 to 1200 Bq 1-1, respectively.40 The use of liquid scintillation counting requires a preliminary purification of the samples in order to remove other radionuclides and various impurities which could interfere in the determination.This method has been applied recently to the determination of tritium in liquid organic products (with a sensitivity of about 1 Bq cm-*),41 in urine42 and in metals.43 Carbon Carbon-14 is a (3-emitter with an energy of 0.156 MeV and a half-life of 5730 years. The sources of 14C in the environment are of both natural (cosmic rays) and anthropogenic (nuclear weapons tests, nuclear power production, mineral fuel com- bustion, use of labelled compounds) origin. The generation of natural 14C takes place in the upper layers of the atmosphere at a rate of 2.2-2.3 atoms s-1 cm-2 or 1.5 X 1015 Bq per year. The total amount of cosmogenic 14C in the biosphere is calculated to be 8.5 x 1018 Bq.In the period from 1945 to 1980 the atmosphere received 0.2 X 101* Bq of 14C from nuclear weapons tests. In the discharges from atomic power stations I4C exists mainly as C 0 2 and, to some extent, as CO and hydrocarbons. Therefore, the method of measuring gaseous preparations of 14C by means of counters with inner inflation proved to be the simplest and most convenient procedure and also sufficiently sensitive. Carbon-containing gases such as C02, methane, ethane and acetylene, and also benzene vapour, can be used in this method. Liquid scintillation counting is now widely used for 14C measurements. Various countries manufacture these types of instruments, which260 ANALYST, MARCH 1989, VOL. 114 Table 11. Level of transuranium elements in the environment Sample z3’Np 23”u Soil (Bq g-1) .. . . . . . . . . 1.2 X 10 -6 0.07 Herbaceous (MBq km-2) . . . . . . . . . . - 0.3-8 plants (Bq g-1) . . . . . . . . . . - 4.5 x 10-4 (MBq km-2) . . . . . . . . - 0.4-0.7 Lichen (Bq g - 1 j Grain, vegetables (Bq kg-1) . . . . . . - (0.2-14) x 10 5 Sea water (yBq 1-1 j . . . . . . . . (0.3-1 1) x 10 Lake water (pBq 1-1) - - - . . . . . . . . . . - (Bq m-’) . . . . . . . . . . 0. 1 - - . . . . . . . . 239pu, 240pu Z4lAm 0.1-7 0.02 22-537 0.3-8 0.3-2 - 0.5- 12 - 4-10 0.7-2 - - (4-89) x - 0.7-52 0.2-8 0.1-29 0.46 typically have a measurement efficiency of about 80-90%, a sample cuvette volume of 15-20 ml and a background of 3&50 counts min-1. In a number of methods 1% is separated from the samples to be analysed as Ba14CO3 with subsequent counting on a flow 4m-counter or a counter with a window thickness of <1-3 g cm-2 of a thick layer.The sensitivity of the method is about 102 Bq.43 Iodine Iodine-129 is a (3-emitter ( E = 150 keV, half-life 17 X 106 years) and is the most important radionuclide from the biological point of view. Although it can be formed under natural conditions through spontaneous fission of uranium nuclei, the main source of this radioisotope in the atmosphere is the nuclear power cycle.4’,46 Despite the very low level of 1291 in the environment, its future contribution to the expected collective irradiation dose of the global population has been estimated to be significant .45 Methods for the determination of iodine in the environment involve steps such as its separation from the sample, concen- tration and radiochemical purification by means of precipita- tion, extraction, distillation and ion-exchange chromato- graphy prior to neutron activation analysis; the detection limit is 10 pBq.45 A y-spectrometric method for the determination of iodine has also been suggested with a detection limit of 15 mBq, which can be decreased after radiochemical purifica- tion, for example, down to 5 mBq 1-1 in the analysis of milk with a chemical yield of 65%.Determination of Transuranium Elements The main difficulties in the determination of transuranium elements in environmental materials are connected with their low concentration (Table 11) and the necessity to separate them from a large amount of sample having a complex chemical and radioisotopic composition.There are a number of different methods available in this context which involve several steps: (i) decomposition of samples such as aerosols, soils and bottom sediments or primary concentration from large volumes (up to 200 1) of natural waters; (ii) radiochemical separation from natural radionuclides and chemical elements, the level of which is usually several orders of magnitude higher; and (iii) final separation and measurement of the a-emission. The large number of reports devoted to soils is explained by the fact that these materials contain the largest amount of transuranium elements (Table 1226747) and, in addition, soils are the primary link in food chains. To separate transuranium elements from soil samples, bottom sediments, plants and biological materials, two methods are commonly used: the first method involves direct leaching by mineral acids from samples calcinated at about 550 “C, whereas the second involves preliminary treatment with hydrofluoric acid, fusion with soda, persulphate, etc.When soils are contaminated by global radioactive fallout, Table 12. Plutonium distribution in bi~~geocenoses.2~,~7 All values expressed in % Mixed Flood- Component forest lands Soil . . . . 99.9 82.4 Bedding . . . .1.6 x 10 8.2 Herbaceous 0.2 Trees: Leaves . .1.1 x 10 - Roots . . . . 8 X 10-2 - Invertebrates and small animals . .1.3 X 8.9 plants - . . . . Water Various meadow biogeocenoses 99.7 98. (-99.8 1.9 x 10-2 3.2 X 10-2 - (2-3) x 10-5 2 x 10-4 - 1.4 x 10-6 - 1.9 x 10-5 7 X 10-5-6 X lopy Table 13.Distribution of plutonium between amorphous and silicate compounds47~48 Time and conditions of occurrence Silicate Content in soil, % Soil in soil Amorphous (mineral part) Turfpodzol . . . . Sd-3years; 97.6 i 0.7 0.8 k 0.3 Greysoil . . . . 5d-4years; 96.0 t 0.6 1.0 t 0.02 laboratory experiments laboratory experiments Leached black earth . . . . . . 17 h-3 years; 98.2 i 0.8 1 .O i 0.3 laboratory experiments Leached black earth . . . . . . Cu. 10years; 97.0k0.8 9.0k 8.0 experimental area plutonium Dustyloam . . . . Global 91.0 2 0.8 9.0 t 8.0 then for model experiments, leaching by hot 7-8 M nitric acid is sufficient. Data obtained for the transfer of plutonium into solution from the acidic degradation of soil samples and other environmental materials, contaminated by relatively soluble plutonium compounds, are given in Table 13.The data show that the level of plutonium in the mineral fraction does not exceed 1%. Further, this level does not exceed 15% even when high relative molecular mass organic materials (humines) are present in the mineral fraction.47.18 However, if the forms in which the transuranium elements occur in soils are unknown, and if highly insoluble compounds and oxides are expected to be present, then it is difficult to obtain reliable results unless the samples are decomposed using oxidants or hydrofluoric acid. The concentration of transuranium elements from large volumes of sea and fresh water is usually carried out by co-precipitation with carbonates of the alkaline earth metals or with iron hydroxides.49--52ANALYST, MARCH 1989, VOL.114 2*2Pu, 2MCm leaching or coprecipitation 261 Evaporation 1 M HN03 - 96% MeOH To separate transuranium elements from other radio- nuclides and chemical elements present in samples, methods such as distribution chromatography on ionites, extraction or extraction chromatography after their combined separation from solution on a common collector [BiP04, CaC204, Fe(OH)3] or on ionites, are most widely ~sed.50~52-63 Tri- isooctylamine (TIOA) , tributyl phosphate (TBP), trioctyl- amine (TOA), thenoyltrifluoroacetone (TTA) and di(2-ethyl- hexy1)phosphoric acid (DEHPA) are used as extractants. Re-extraction of the transuranium elements from the organic phase or their desorption from the ionites is carried out using nitric or hydrochloric acid of different strengths and by adding hydrofluoric or hydroiodic acid, hydrogen peroxide, ammo- nium iodide or other reagents that can reduce PuIV to PuIII or facilitate the formation of complex compounds.The elec- trodeposition method is generally used for the final separa- tion. The schemes developed for the separation of transuranium elements prior to their determination are sophisticated and involve many steps. It is often necessary to repeat different operations in order to obtain individual transuranium ele- ments in a radiochemically pure state. The chemical yield varies within a wide range, being 20-100% for plutonium and 15-100% for americium and curium; however, their mean values show a smaller difference.Table 14 gives some data on Table 14. Detection limits for transuranium elements Sample Detection limit Soil, bottomsediments . . . . 37 x 1 k 3 B q kg-' Bottomsediments , . . . . . 3.7 x lO-'Bqpersarnple Fallout . . . . , . . . . . 3.7 x 10-'Bqm-2 Atmospheric dust , . . . . . 2 x 10-7Bq m-3 Natural waters . . . . . . . . 18 x 10-6 Bq 1-1 Biological materials (raw mass) . . 3.7 x lO-3Bq kg-1 Biologicalmaterials(ash) . . . . 1 X 10-4Bqg-1 the detection limits for transuranium elements, indicating the high sensitivity of the methods used. Among the transuranium elements plutonium has received the most attention and a large number of different methods have been developed for its determination based mainly on the application of extraction and distribution chromatography (Fig.3). Methods by which plutonium can be determined without its separation and radiochemical purification are known.64.65 A dry sample (soil, fallout, bottom sediments, etc.) is thinly coated on the substrate (by cathode sputtering or some other means) and placed in the ionising chamber. In the analysis of atmospheric precipitation and weighed samples, the reported detection limits are 10-5-10-3 Bq g-1 and 6 X 10-12 Bq 8-1 for soils and 10-6 Bq m-3 for aerosols of atmospheric air.66 This method has a number of disadvantages, hence it has not been widely used. For instance, there must be no other nuclides in the analytes which emit a-particles with similar energies to those emitted by plutonium, or, if present, these other nuclides must be at a level that is 2-3 orders of magnitude lower than that of plutonium.The determination of plutonium in plants can be carried out both by measuring the 241Am y-emission and by taking into account the ratio of Pu to 241Am.67 Fewer methods have been developed for the determination of americium and curium in environmental materials. The transplutonium elements are concentrated by their co-precipi- tation with cerium and lanthanum hydroxides or fluorides, or by solvent extraction; radiochemical purification is carried out using distribution chromatography from aqueous and aqueous alcoholic media, and also by extraction, extraction chromato- graphy, etc., the final separation being achieved by electro- deposition (Fig. 4). The mean value of the chemical yield of americium and curium varies from 50 to 80% ; when they are determined in large volumes of sea water, this value decreases to 20-50%.The detection limit for americium and curium is I CaC204 I "9": M HCI I I Bio-Rad 1-X8 lpul - 80% MeOH - H20 - 86% MeOH Chemical yield, Yo Sample Pu Cm 6 6 k 13 7 9 f 11 Soil Vegetables 8 0 2 13 7 8 f 14 Sea water 7 0 + 9 86+ 12 Sea sediments 7 5 k 9 7 6 f 9 Algae 80 k 14 81 f 13 Fish 72 k 13 8 6 k 10 Fig. 3. Scheme for the determination of transuranium elements in aerosols, bottom sediments, sea water and biological materials59262 BiP04 HN03, 242Pu, (236Pu), *43Am, NH20H.HCI, Bi, H3 PO4 6~ HCI, 0 . 7 5 ~ HCI ANALYST, MARCH 1989, VOL. 114 Aqueous Soil phase Solution I l l Dowex 50-X8 HCI, evaporation HN03, NaN02 Dowex 1-X8 I l l - Chemical yield, YO Sample Pu Am Fresh water 8 2 + 9 7 0 + 7 Sea water 8 5 + 7 77 k 16 Fig. 4.natural waterss7 Scheme for the determination of transuranium elements in 8 x 10-4 Bq.68 To increase the americium concentration (e.g., when analysing soil from the Nevada testing ground), the sample matrix is first decomposed and the americium is extracted with a quaternary ammonium salt and then with TTA. The organic phase is evaporated on a stainless-steel target and the photon emission measured ( E = 60 keV) by a germanium detector with a beryllium window. 11 The detection limit is 7.4 X 10-3 Bq. Even fewer methods are available for the determination of neptunium; however, this element has recently received much attention. Although the levels of 237Np in environmental materials are not high, nevertheless it is the most long-lived nuclide of the transuranium elements and the most mobile in the ecosystems.Its contribution to the total irradiation dose received by humans (in foodstuffs) is comparable to that of plutonium and even higher in a number of instances.69 The methods used for the determination of neptunium involve the same sequence of operations for its concentration and radiochemical purification as those used for the determi- nation of americium and curium (Fig. 5 ) . Thermolens spectrometry has been proposed as a method for the determination of neptunium. The method is based on the local absorption of the laser induced fluorescence, with subsequent conversion of the absorbed energy into a temper- ature or refraction index gradient.The detection limit of the method is 2-100 pmol 1-1 of neptunium.7" As there is a need for sensitive and less time consuming methods, consideration should be given to a recently devel- oped method for the determination of neptunium based on the luminescence of crystallophosphors.7~ This method permits the detection of up to 10-11 g of neptunium in the presence of background levels of other radioactive elements, hence making it possible to reduce the number of chemical operations required. In addition to the determination of neptunium, this method allows all the plutonium isotopes to be determined. Conclusion Many methods for the direct determination of radiochemical nuclides together with methods involving their preliminary separation into a radiochemically pure state have recently been developed.These methods allow the determination of 1 239Np, 10 M HCI Diisopropyl ether I 'I Aqueous phase I 0.01-0.05 M HF OToA Organic phase 10 M HCI I M HCI - 0.1 M HF LaF3 8 M HN03 - AI(N0313 Dowex I-X8 Evaporation with HC104 1 M HCI then 0.5 M HCI I'I Dowex 50-X8 Fig. 5. Scheme for the determination of neptunium66 both natural and technogenic radionuclides and enable reliable measurements of the level of these elements in various environmental materials to be made. The large amount of accumulated experimental material makes it possible to understand the fate of radionuclides in various ecosystems, to follow the dynamics of their variation and to obtain reliable data on the forms of occurrence of radionuclides in the samples to be analysed.However, in a number of instances, particularly where the determination of transuranium ele- ments is concerned, these methods tend to be rather compli- cated, to involve many steps and to be time consuming. Therefore, one of the challenges facing radiochemists today is to develop much simpler and more reliable, sensitive and selective methods for the determination of radionuclides in environmental materials. 1. 2. 3 . 4. 5 . References At. Energ., 1986, 61, 301. Sivintsev, Yu. V., At. Energ., 1988, 64, 46. Izrael, Yu. A , , Sokolovsky, V. G., Sokolov, V. E.. Vetrov. V. A., Dibobes, I. K . , Trusov, A. G.. Riabov, I. N.. Aleksakhin, R . M., Povajaev, A. P., Buldakov, L. A., and Borsilov, V. A , , At. Energ., 1988, 64, 28. Lavrukhina, A.K., Malysheva, T. V., and Pavlotshaya. F. I.. "Radiochemical Analysis," Academcy of Sciences, Moscow. 1963. Inn. K. G., J . Radioanal. Nucl. Chem.. 1987. 115, 91.ANALYST, MARCH 1989, VOL. 114 263 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. Pavlotskaya, F. I . , “Migration of the Global Radioactive Fallout in Soil,” Atomizdat, Moscow, 1974. Pavlotskaya, F. I., DSc Thesis, Moscow. 1981. Avogadro, A., Murray, C. N., de Plano. A., and Bidoglio, G., in “Environmental Migration of Long-lived Radionuclides,” IAEA, Vienna, 1982, p. 527. Micglo, J . J., J . Appl. Radiat. Isot., 1978, 29, 581. Wilkins, B. T., Stewart, S. P.. and Major, R . 0.. J . Radioanal. Nucl. Chem., 1987, 115, 249.Larsen. J . L.. and Lee. S. V., J. Rudiounal. Chem., 1983, 79, 165. Kusnetz, H., Am. Znd., Hyg. Assoc. Q., 1956, 17, 85. Thomas, J., Health Phys., 1970, 19, 691. Nazaroff, W. W., Health Phys., 1980, 39, 683. Groer, P.. “Annual Report, Alamos National Laboratory,” ANL-7860. Part 11, 1972, p. 285. Rolle, H . , Health Phys.. 1972, 22, 233. Markov, K. V., Rybov. N. B., and Stas, K. N., in “Report of the Central Scientific Research Institute of Instrument Engin- eering, Part 2,” Atomizdat, Moscow, p. 183. Terentjev, M. B., and Krisyuk, E . M., At. Energ.. 1983, 5 5 , 310. “Report of the United Nations Scientific Committee on the Effects of Atomic Radiation,” United Nations, New York, 1964. “Ionizing Radiation: Sources and Biological Effects of Atomic Radiation,” Report to the General Assembly.United Nations, New York, 1982. “Irradiation from the Military Nuclear Tests and Nuclear Fuel Cycle” (prepared by the Secretariat), UN Scientific Committee on the Effects of Atomic Radiation, 1985, A/AC.82/R.433. Voroviev, E. I . , Iljin, L. A., Turovsky, V. D., Gusev, N. G . , Pavlovsky. 0. A , . and Parkhomento, G. M., At. Energ., 1983, 54. 277. Moiseev, A. A . , “1-37Cs-Environment-Man.” Energoatomiz- dat, Moscow, 1985, p. 120. “Irradiation from Nuclear Tests” (prepared by the Secretariat), Scientific Committee on the Effects of Atomic Radiation, Vienna, 1984, AIAC. 821R.422. ‘-Information about Chernobyl’s Accident and its Conse- quences,“ State Committee on Atomic Energy, USSR, 1986. Pavlotskaya. F. I.. and Polikarpov, G. G ., in Aleksachin, R . M., Editor. ‘.Itogi Nauki i Techniki, Radiatsionnaya Biologiya, Volume 4, Problemy Radioevologii,” VINITI, Moscow, 1983, p. 99. Kogan, R. M., Nazarov, I. M., and Fridman, Sh. D., “Osnovy Gamma-spectrometrii Prirodnykh Sred,” Atomizdat ~ Moscow, 1976, p. 366. “Proceedings of the Symposium on Low-level Measurements o f Actinides and Long-lived Radionuclides in Biological and Environmental Samples, Lund, Sweden, June 9-13,1986. Parts I and 11.’- J . Rudioanul. Nucl. Chem., Articles, 1987, 115, Nos. 1 and 2. “Sbornik Metodik PO Opredeleniyu Radioactivnosti Okruz- hayushei Sredy,“ Gidrometeoizdat, Moscow, 1966. “Sbornik Metodik PO Opredelcniyu Radioactivnosti Okruz- hayushei Sredy,” Gidrometeoizdat, Moscow, 1968, p. 26. Silantiev. A. N.. “Spectrometricheskii Analiz Radioactivnykh Prob Vneshnei Sredy.” Gidrometeoizdat, Moscow, 1969. “Metodicheshie Recommendatsii PO Sanitarnomu Controliu za Soderzhaniem Radioactivnykh Veshestv,” Department of Health Protection, Moscow, 1980.Food and Agriculture Organization, International Atomic Energy Agency and World Health Organization, “Methods of Radiochemical Analysis.” WHO, Geneva, 1967. Rovinski, F. Ya, Iokhelson, S . V., and Yushkan, E. I., **Metody Analiza Zagriazneniya Okruzhayushei Sredy,” Ato- mizdat, Moscow, 1978. “Reference Methods for Marine Radioactivity Studies.” Tech- nical Reports Series No. 118, IAEA. Vienna, 1970. Melnikov, V. A . , Moskvin, L. N., and Epimahov, V. N . , Radiokhimiyu, 1984, 26. 783. Babaev, N . S . , Demin, V. F.. Ilyin, L. A.. Kniznikov.V. A . , Kuzmin, I. I.. Legasov, V. A , , and Sivinzev. Yu. V., 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 39. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. “Yadernaya Energiya: Chelovek i Okruzhayushya Sreda,” Atomnergoizdat, Moscow, 1984. Lo, J.-G., Radiat. Eff., 1986, 96, 81. “Tritium Measurement Technique,” NCRP Report No. 47, Washington, DC, 1976. Blinov, V. A., Gedeonov, L. I., Titkov, V. P . , and Anisimov, V. V., Radiokhimiya, 1984, 26, 81. Geverch, V. B., Palladiev, A. A., and Pugova, T. Yu., Radiokhimiyu, 1986, 28, 607. Momoshima, N., Nagasato. Y., and Takashima, Y., J . Kadiounal. Nucl. Chem. Lett., 1986, 107, 353. Sugisaki, M., Idemitsu, K., and Furuya, H . , Radiochem. Radioanal. Lett., 1982, 51, 283.Gaiko. V. B., Korablev, N. A.. Soloviev. E. N.. Trosheva, T. I., Shamov, V. P., Umanetz, M. P., and Sherbina. V. T., At. Energ., 1985, 59, 144. Cholina, Yu. B., in Aleksachin, R. M., Editor, “Itogi Nauki i Techniki,” Volume 4, VINTTI, Moscow, 1983, p. 43. Filistovich, V. I., and Nedvetskaite, V. K., Radiokhimiya, 1986, 28, 390. Myasoedov, B., Pavlotskaya. F., and Goryachenkova, T., in Mizuike, A . , Editor, “Proceedings of the Third Japan - USSR Joint Symposium on Analytical Chemistry, November 5-9, 1986, Nagoya, Japan,’‘ Nagoya University, Nagoya, 1986. Pavlotskaya, F. I., and Goryachenkova, T. A., Radiokhimiya, 1987. 29, 99. Holm, E . , Ballestra, S . , and Fukai, R . , Talanta, 1979,26, 791. Yamato, A . , J . Radioanal. Chern., 1982, 75, 265. Pillai, K. C . , and Matkar, V. M., J . Radioanal. Nucl. Chem., 1987, 115, 217. Suzuki, S . , Shirabashi, J . , and Nagasawa, K., in “Quick Methods for Radiochemical Analysis,’’ Technical Reports Series No. 95, IAEA, Vienna, 1969, p. 35. Bernabee, R. P . , Health Phys., 1983, 44, 688. Bernabee, R . P., Percival, D. R . , and Hindman, F. D . , Anal. Chem., 1980, 52, 2351. Schulz, R., Wink, G. T., and Fujii, L. M., SoilSci.. 1981, 132. 71. Gasmini, S . , J . Radioanal. Chem., 1980, 55, 253. Mathew, E., Matkar, V. M., and Pillai, K. C., J . Radioanal. Chem., 1981, 62, 267. Yamamoto, M., Komura, K., and Sakanoue, M., Radiochim. Acta, 1981, 29, 205. Hayashi. N., Ishida, J . , Yamato, A . , Iwai, M., and Kinoshita, M., J. Rudiounul. Nucl. Chem., 1987, 115, 369. Butler, F. E., in “Quick Methods for Radiochemical Analysis,” Technical Reports Series No. 95, IAEA, Vienna, 1969. p. 37. Holm, E., and Fukai, R., Tafantu, 1976, 23, 835. Bunzl, K.. and Kracke, W . , J . Radiounal. Nucl. Chem., 1987, 115, 13. Dellesite. A , , and Marchionni, V., in “Techniques for Identify- ing Transuranic Speciation in Aquatic Environments,” IAEA, Vienna. 1981, p. 49. Astafurov. V. I., and Zemlyanukhina. N. A . , Radiokhimiyu, 1984, 26, 119. Yakunin. M. I.. and Dubasov, Yu. V., Radiokhimiyu, 1986,28, 210. Sakanoue, M.. Radiochim. Acta, 1987, 42, 103. Pshelkin, V. A , , and Sviderskii, M. F . , Radiokhimiya. 1984, 26, 694. Sakanoue, M., Yamayo, M., and Komura, K., J . Radioanal. Nucl. Chem., 1987, 115, 71. Eisfelk, K., Matthies, M., Paretzke, H . G . , and Wirth, E . , in “Environmental Migration of Long-lived Radionuclides,” IAEA, Vienna, 1982, p. 701. Beltz, J . V., and Hassler, P., Nucl. Technol., 1980, 51. 169. Nivikov. Yu, P., Karyakin, A. V., Myasoedov, B. F., Gliva, V, B., and Ivanova, S . A., Speclrochim. Acta, Part B , 1986,41, 777. Paper 8102973C Received July 21st, 1988 Accruted August 16th, 1988

ISSN:0003-2654    

DOI:10.1039/AN9891400255

出版商:RSC    

年代:1989

数据来源: RSC

摘要:

ANALYST, MARCH 1989, VOL. 114 265 Aspects of Radioanalytical Chemistry* Alan Dyer Department of Chemistry and Applied Chemistry, University of Salford, Salford M5 4 WT, UK A review is presented of recent progress in the more traditional techniques for radiochemical analysis, including solvent extraction, ion exchange and ion chromatography together with some isotope detection methods. Prominent in the detection studies are liquid scintillation spectrometry and its associated sample preparations. Less familiar approaches are also considered, for example, high-performance liquid chromatography of beta isotopes and inductively coupled plasma mass spectrometry. Keywords: Radioanalysis; solvent extraction; liquid scintillation; ion exchange The subject of radioanalytical chemistry covers a wide range of techniques, both instrumental and non-instrumental. The most prominent instrumental method is that of neutron activation analysis. 1 Another major area involves the determi- nation of isotopes in biological and in environmental sam- ples.2 However, this paper reviews the progress made in the more traditional radioanalytical techniques together with the developments that have taken place in less familiar instrumental methods of analysis. Traditional Met hods The recent literature catalogues the continued use of solvent extraction techniques as an aid to the separation of and subsequent analysis of radionuclides.Examples of such are the use of N-benzoyl-N-phenylhydroxylamine for actinide extraction3 and studies on the effects of complexing agents on the extraction of lanthanides by di(2-ethylhexy1)phosphoric acid.4,' More novel extractants, for example, for the uptake of I- 6 and the determination of T c , ~ are the crown ethers such as dibenzo-18-crown-6.In one instance the crystal structure of a UIV crown ether complex has been determined.8 As a further example, Grases and Fais have developed a method, based on the use of tetrahydroxy-1,4-quinone, which is specific for the determination of Tc in vegetation. Similar applications have been reported by Zakareia et al. ,lo Chakravortty et af. ,I1 Gehmecker et a1.12 and Ilic and Milonjic.13 The other traditional nuclide separation method is ion exchange and, as an example of this, Le Van S0'4J5 has described the use of silica distributed on amorphous zirconium phosphate (Si - ZrP).A more analytical approach was taken loo t , b , l 10-1 1 0.01 0.1 1.0 10 [HCI]/M Fig. 1. on to zirconium phosphate (ZrP-1)16 Absorption of Pu3+, Pu4+ and Pu022+ from solutions of HCI * Plenary lecture presented at the 2nd International Conference on Nuclear and Radiochemistry, Brighton, UK, 11-15 July, 1988. by Gehmecker et al.,16 who investigated the use of several substrates to separate the different oxidation states of Pu by ion-exchange chromatography. One of the substrates studied was zirconium phosphate (ZrP-1) and its effectiveness is illustrated in Fig. 1. As regards detection methods, the technique of liquid scintillation counting (LSC) has been developed in various ways. Mentasti et af. 17 found it convenient to determine Pb in the concentration range 1-10-6 mg 1-1 using a chelating cellulose filter, counted in Instagel.Abuzeida and co-work- ers18J9 have developed the use of LSC for application to the determination of Pu and U by using differential solvent extraction combined with a-spectrometry ; examples of their results are given in Table 1. Modern liquid scintillation (LS) instruments have been developed to provide sophisticated pulse-height analysis, which specifically quantifies the changes in @-ray spectra caused by quenching agents. This facility can be used to resolve other spectra as Schonhofer and Henricii20 have demonstrated. They showed clearly the value of using a state of the art LS spectrometer in the determination of low-level activity generally (e.g., environmental 3H and 14C) and, in addition, described the determination of 222Rn and 22hRa in mineral waters, without sample preparation (direct incorpora- tion into a gel scintillant). Fig.2 shows one of the observed spectra. Further simultaneous determination of isotopes (238Pu and 241Am) has been described by Miglio.21 This method involved the separation of 241AmlII into an aqueous phase (0.15 M HN03 - dioxane-based scintillator) and the retention of 23SPuIV in a hydrogen bis(2-ethylhexy1)phosphate - toluene phase. The use of commercial LSC enabled the channels to be set to accommodate the different isotope energies. Similarly, 90Sr has been determined by LSC spectral shape analysis.22 Table 1. Determination of uraniumlx,I9 Concentration of U/mg ml-I Sample No.LS* ESF-II- ESF-111- 1 0.072 0.079 0.076 2 0.24 0.29 0.28 3 0.73 0.80 0.75 4 1.26 1.58 1.40 5 1.90 2.20 1.90 6 0.23 0.24 0.23 7 2.8 3.3 2.9 8 6.8 7.5 6.7 9 6.4 7.0 7.0 10 8.3 9.0 8.3 11 7.2 8.0 7.0 12 7.0 7.7 7.6 * LS = liquid scintillation method. 1- ESF = energy-dispersive X-ray fluorescence method.266 ANALYST, MARCH 1989, VOL. 114 loo I 0 500 1000 1500 2000 Channel number Fig. 2. Pulse-height spectrum of mineral water2() Isotope Detection and Measurement Perhaps the most widely used method of radionuclide measurement is that involving the creation of a colloidal system whereby aqueous samples can be incorporated into an organic liquid scintillant via the use of a detergent. Clearly, in such a system the isotope can either be in the bulk organic phase or in the micellar aqueous phase (usually the counting system is the “water in oil” type of micro-emulsion).The presence of a quenching agent in these systems may or may not be effective, depending on its solubility in the aqueous or organic phase. Bush23 pointed out that a plot of the sample channels ratio (SCR) versus the automatic external standard channels ratio (AESCR) should indicate whether or not the quencher is in the same phase as the isotope to be determined. The questions then arise of whether, if the quencher is not in the same phase as the isotope, it still affects the radionuclide determination and, if so, it does so in a regular and therefore correctable manner. Recent studies24 have involved the use of a quencher that is soluble in water and not in toluene (phenolphthalein) and one that is soluble only in toluene (Sudan 111) to provide information on the determination of [3H] as THO and [Wltoluene and also [“Cltoluene.To summarise, the results showed that phenolphthalein solutions could be corrected back to a standard “double-ratio” plot regardless of isotope energy or isotope phase location. Conversely, no correction could be envisaged when Sudan 111 was present. A detailed interpretation of these results emphasises the contribution from colour quenching (Sudan III), which, clearly, cannot readily be compensated for in colloidal systems, and also the effect of quenching agents on the size of micelles and their stability. Those using the “colloid” counting method are encouraged to take particular care by preparing standards for comparison with the “real” samples. The effect of temperature on the fluorescence created by LSC has been described by other workers.25 The use of LS systems to detect and quantify Cerenkov light is well established as a convenient and cheap radioisotope detection method.Carmon and Dyer2h-27 have extended this use by coupling a multi-channel analyser to a standard commercial LS spectrometer. Hence, quantitative Cerenkov spectrometry can be carried out provided the peak resolution is sufficient; this is illustrated in Fig. 3, which shows the spectra of W r and 137Cs. The method has been applied to the determination of *9Sr, ”Sr - SOY, 10hRu and 137Cs in each other’s presence. Measurements carried out in different concentrations of nitrates and nitric acid indicate the possible applications of this method to the analysis of aqueous nuclear waste; it is possible to use the channel settings available in the LS counter to compensate for the quenching of Cerenkov spectra by the NO3- ion.The technique has been used successfully28 to determine the uptake of 137Cs, 9oSr and 90Y on to zeolites from mixed isotope solutions in the pH range 2.5-11 and in the cation concentration range 10-3-10-5 M. Simonnet et af.29 suggested that determinations of 32P are 1200 800 $ 400 .- S 3 L 5 0 160 320 480 640 800 960 1120 ui E 640 8 480 3 320 160 0 160 320 480 640 800 960 1120 1280 Channel number Fig. 3. (a) Cerenkov pulse-height s ectra of (1) W r - SOY in 5 M NaNO,; (2) W r - YOY in 1 M NaNO,; f3) ‘M Sr - SOY in 0.05 M NaNO,; and (4) ‘I0Y in 0.05 M NaN03.( b ) Cerenkov pulse-height spectra of (1) *%Sr and (2) 137Cs in 0.1 M NaNO, (unquenched)*h.” Table 2. Determination of 23XU in urine37 [ Z ’ W ] , p.p.m. a-Ray Sample ICP-MS* spectrometry+ Nominal A . . . . . . 1 . 1 k 0 . 1 1.2k0.2 1 .0 B . . . . . . 2 . 1 + 0 . 2 2 . 3 2 0 . 3 2.0 C . . . . . . 5.0 2 0 . 4 5 . 8 k 0 . 6 5.0 D . . . . . . 10.2 f 0 . 8 11.5k0.7 10.0 E . . . . . . 19.5 kO.8 22.1 k 1.9 20.0 F . . . . . . 0.15 20.02 0.5 2 0 . 2 0.1 Limit of detection -0.05 -0.15 - * Sample (2 ml) diluted with water (1 + 3); 10 counts each of 1 s t Sample (1 I), 1 count of 2000 min 2 2SD. Specific activity of *3XU duration 5 2SD. = 1.2 x 104 Bq g-1 of U. better when carried out in vials impregnated with LSC cocktail (PPO/POPOP) rather than by using a Cerenkov method.These same workers have also monitored 1251 via a type of “Garnmavial.”30 Miscellaneous Tham and Preiss31 have described a method of analysis for Fe whereby the iron content of dolomites was determined (ca. 0.lY0) using X-ray fluorescence induced by a S-mCi loTd source. This method was given the acronym RIXRF (radioiso- tope induced X-ray fluorescence) and used a Si(Li) detector. Aggarawal et uf.32 have developed the technique of non- isotopic diluent spectrometry (NIDAS) to measure Pu con- centrations. The diluent used was *W and a precision and accuracy of 1% were achieved using a large-area (300 or 4.50 mm2) Si surface barrier detector. Potentially, the measurement of 0 radiation in high-perfor- mance liquid chromatography effluents is an important advance as regards analysis; van Nieuwerk and co-work- ers33.34 have recently described this technique.They used UV and Isoflo (Edinburgh) detectors and determined [’%]para- thion at 0.4 and 0.02 Bq at flow-rates of 0.2 ml min-1 and 10 yl min-1, respectively (the lower limit corresponds to 0.01 mg). As regards future developments, Dale35 have reviewed the potential analytical application of positrons, referring to various analyses of zeolties, graphite, silica gel and alumina. Although the review contains no analytical results, Dale35 suggest that the method could be improved by the use of an accelerator beam rather than the 68Ge source that is used currently. Alternatively, Gaggeler et af.3h have discussedANALYST, MARCH 1989, VOL.114 267 vacuum thermochromatography. They described a theoretical model, which predicts that the method can be used for rapid chemical separations of volatile carrier-free species. The tehnique involves a molecular flow down a chromatographic column lined with Cu foil and, with temperatures ranging from ambient to 1230 K (temperature gradient, 18 K cm-I), provides analyses for 21"Pb species by subjecting the Cu foil removed from the column to a-ray spectrometry. Finally, radionuclides have been determined successfully by non-radiometric methods. For the most part, radiofrequency inductively coupled plasma sources have been used to quantify long-lived isotopes and, for example, Hislop et a1.37 have recently described the determination of 99Tc, 1291 and 238U using inductively coupled plasma mass spectrometry (ICP- MS); examples of their results are given in Table 2.The author thanks Analytical Division, asked to present this 1. 2. 3. 4. 5 . 6. 7. 8. 9. SO. 11. 12. 13. 14. 15. Huang, K. S.. J. the Radiochemical Methods Group, Royal Society of Chemistry, for being paper. References Radioanal. Nucl. Chem., 1987, 112, 193. Ballesyra, S., Barci. G . , Holm, E., Lopez, J., and Gastaud, J., J. Radioanal. Nucl. Chem., 1987, 115, 51. Mathur, J . N., and Khopkar, P. K., Rudiochim. Actu, 1986,39, 77. Bhattacharyya, S. N . , and Ganguly, K. M., Radiochim. Acta, 1986. 39, 63. Bhattacharyya, S. N., and Ganguly, K. M., Radiochim. Acta, 1986, 40, 17. Jalhomme, M. G . , Radiochim. Acta, 1986, 40, 203. Jalhomme, M. G., J . Radioanal.Nucl. Chem. Lett., 1986, 104, 131. Wang, W., Lin, J . , Shen, H . , Peiju, Z . , Wang, M., and Wang, B., Rudiochim. Acta, 1980, 40, 199. Grases, F., and Fai, G . , Radiochim. Acta, 1986, 39, 81. Zakareia, N., Khalifa, S. M., Daoud, J . A . , and Aly, H. F., Radiochim. Acta, 1986, 39, 89. Chakravortty, V., Dash, K. C., and Mohanty, S . R., Radio- chim. Acta, 1986, 40, 89. Gehmecker, H., Trautmann, N., and Herrman, G.. Radio- chim. Actu, 1986, 40, 11. Ilic. Z., and Milonjic, S . , J. Radioanal. Nucl. Chem., 1986, 99, 279. Le Van So, J . Radioanal. Nucl. Chem., 1986, 98, 225. Le Van So, J . Radioanal. Nucl. Chem., 1986, 99, 55. 16. 17. 18. 19. 20. 21. 22. 23, 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35, 36. 37. Gehmecker, H., Trautmann, N., and Herrmann, G . , Rudio- chim.Acta, 1986, 40, 81. Mentasti. E., Gennaro, M. C., and Volpe, P.. J. Radioanal. Nucl. Chem., 1986, 97, 301. Abuzeida, M., Abouzreba, S . , Almedhem, B., Zolotarev, Y. A., and Komarova, N. A . , J. Radioanal. Nucl. Chem., 1987, 111, 3. Abuzeida, M., Arebi, B. H . , Zolatarev, Y. A , , and Komarov, N. A . , J. Radioanal. Nucl. Chem., 1987, 116, 285. Schonhofer, F., and Henricii, E., J. Radioanal. Nucl. Chem., 1987, 115, 317. Miglio, J., Anal. Chem., 1986, 58, 3215. Mishra, A. C., and Parks, N . J . , Appl. Radiat. Isot., 1987.38, 455. Bush, E . l., Int. J. Appl. Radiat. Isot., 1968. 19, 447. Dean, J. C. J . , PhD Thesis, University of Salford, 1987. Homma, Y., Murase, Y., and Sonehara, K., Appl. Radiat. hot., 1987, 38, 91. Carmon, B., and Dyer, A., J. Radioanal. Nucl. Chem., 1986, 98, 265. Carmon, B., and Dyer, A., J. Radioanal. Nucl. Chem., 1987, 109, 229. Carmon, B., Dyer, A . , and Keir, D . , J. Radioanal. Nucl. Chem., 1988, 125, 135. Simonnet, F., Combe, J . , and Simonnet, G . , Appl. Radiat. Isot., 1987, 38, 311. Simonnet, F., Combe, J . , and Simonnet, G., Appl. Radiat. Isot., 1987, 38, 571. Tham, F. S., and Preiss, I. L., J. Radiounal. Nucl. Chem., 1986, 99, 133. Aggarawal, S. K., Duggal, R. K., Rao, R., Ramasubramanian, P. A., and Jain, H. C., Rudiochim. Acta, 1986, 41, 23. van Nieuwerk, H. J . , Das, H. A , , Brinkman, U. A. T., and Frei, R. W., J. Radioanal. Nucl. Chem., 1986, 99, 427. van Nieuwerk, H. J . , Veltkamp, A. C., Das, H. A., Brinkman, U. A. Th., and Frei, R . W., J. Radioanal. Nucf. Chem., 1986, 100, 165. Dale, J. M., J. Radioanal. Nucl. Chem., 1987, 110, 403. Gaggeler, H., Eichler, B., Greulich, N., Herrman, G . , and Trautmann, N., Radiochim. Acta, 1986, 40, 137. Hislop, J . S . , Long, S. E., Brown, R. M., Morrison, R . , and Pickford, C. J . , paper presented at the ICRM Low Level Techniques Group Meeting, Wurenlingen, 1987; submitted for publication in Environ. Int. Paper 8i02975J Received July 21st, 1988 Accepted September 8th, 1988

ISSN:0003-2654    

DOI:10.1039/AN9891400265

出版商:RSC    

年代:1989

数据来源: RSC

摘要:

ANALYST, MARCH 1989, VOL. 114 269 Advances in Nuclear Analysis Methods* Emile A. Schweikert Center for Chemical Characterisation and Analysis, Texas A and M University, College Station, TX 77843-3144, USA Recent developments in excitation, detection depth profiling and ultratrace analysis are discussed. On the topic of ultratrace analysis, the coupling of neutron activation followed by mass spectrometric detection of the activation products opens the possibility of determinations based on stable reaction products. The feasibility of neutron activation - mass spectrometry (NA- MS) has been demonstrated by its application to the determination of both 6Li(n,3H)4He and 10B(n,a)7Li. In the realm of depth profiling, resonant elastic scattering is a valuable tool for tracking 1 6 0 via 160(a,a)160 with a depth resolution of 6 pg cm-2.The technique of recoil-nucleus time-of-flight neutron depth profiling is discussed. The availability of cold neutron beams will result in a signal intensity increase of two or more orders of magnitude over thermal neutron beams. Hence cold neutron beams will enhance the capabilities of neutron capture, of prompt y-ray activation analysis and of neutron depth profiling. Finally, heavy ion induced desorption with time-of-flight mass spectrometric detection of desorbed species presents a unique capability for the analysis of surfaces, mixtures and, in particular, large biomolecules. Keywords: Nuclear analysis methods; excitation; detection depth profiling; ultratrace analysis; recent developments Nuclear analysis techniques have long played a prominent role at the forefront of analytical chemistry, e.g., neutron activa- tion in trace analysis or Rutherford back-scattering for thin-layer characterisation and depth profiling.Some recent advances and prospects for further developments in nuclear analytical chemistry are discussed in this paper, which is not a review (an excellent one has appeared recently’) but aims to highlight trends and to place them in the panorama of chemical characterisation. Also, the term “nuclear” is taken here in a broad sense , i. e., it includes techniques derived from nuclear science. The topics discussed in this paper concern applications of nuclear techniquec in two important analytical areas, ultratrace analysis and depth profiling.and analytical prospects with excitation modes using beam> of cold neutrons and heavy ions. Ultratrace Analysis The domain of ultratrace determination forms one of the frontiers in analytical chemistry. A summary of the “best” detection limits currently claimed for a range of excitation techniques using photons, ions or neutrons is given in Fig. 1, which highlights relative capabilities and trends. One import- ant point is that nuclear and X-ray techniques are quantitative even when the species to be measured is present at a level only one order of magnitude larger than the detection limit; this is an advantage when such precision is required. A discussion of the various ultratrace techniques is beyond the scope of this paper although that which combines neutron activation and the mass spectrometric detection of the activation products’ deserves mention.This approach allows the analyst to consider procedures based on very long-lived or stable isotopes produced by neutron irradiation. In particular, neutron activation - mass spectrometry (NA - MS) has been used to determine lithium via sLi(n,”H)4He and the mass spectrometric detection of 3He produced by the (3-decay of 3H. A detection limit of 10 ng g-1 has been reported. The method has been applied successfully to the analysis of a series of complex biological and inorganic samples (e.g. , oyster tissue, coal dust and marine sediments) and to the determination of * Plenary lecture presented at the 2nd International Conference on Nuclear and Radiochemistry, Brighton, UK, 11-15 July, 1988.c- RBS .I- PIXE f- NAA I t AMS c- SIMS Synchrotron X-ray microprobe + LIMS +RIMS c- RIS 1 , I -21 -19 -17 -15 -13 -11 ag fg PS Detection sensitivity (orders of magnitude) Fig. 1. “Best” detection limits for isotope and elemental trace analysis techniques. RIS, resonance ionisation spectroscopy; RIMS, resonance ionisation mass spectrometry; LIMS, laser ionisation mass spectrometry; AMS, accelerator mass spectrometry; SIMS, secondary ion mass spectrometry; NAA, neutron activation analysis; PIXE, particle-induced X-ray emission; and RBS, Rutherford back-scatter- ing spectrometry *oB via 10B(nla)7Li and detection of 4He by mass spec- trometry.2 Although mass spectrometric measurements alone have been used to investigate nuclear reaction products,3--” the NA - MS technique could certainly be extended to other nuclear activation modes and products.Depth Profiling Two nuclear methods that have contributed to recent develop- ments and have prospects for further developments are now discussed. Resonant Elastic Scattering The possibility of sensitive and spatially well resolved detec- tion of 1 6 0 with resonant elastic scattering using a beam of 4He ions has been recognised for some time.”’() The experimental prerequisite is an a beam with energy, E 3 3.045 MeV; measurements are made using the usual Rutherford back- scattering spectrometry (RBS) set-up.11 An example of an a back-scatter spectrum is shown in Fig. 2. The peak resonantly270 ANALYST, MARCH 1989, VOL. 114 600 1 cn c 3 c O0 300 s s 200 loo 1 0 ‘ ’ U 800 1000 1200 1400 1600 1800 Energy/ keV Fig.2. oxidised Si target. Incident energy, 3.0638 MeV Energy spectrum of alpha particles scattered at 165” from an 2.1 1 0 20 40 60 80 Depthlpg cmP2 Fig. 3. polished Ti sample left exposed to air Stoicheiometric ratio (OiTi) as a function of depth of 700 - 650 5 600 5 550 - a 0 -0 I] m - - - 0 500 450 - .- - c m 2 350 0 300 250 0 400 800 1200 1600 2000 Depth/A Fig. 4. Oxygen atomic abundance in the thin surface layer of the superconducting sample as a function of depth. The full line is a smoothed trace through the experimental points scattered a particles are imposed on a background of non-resonant scattering from oxygen; in turn this background sits on the continuum of elastic scatteriDg by the silicon nuclei in the unoxidised part of the target.An example of the use of 160(a,a)160 is shown in Fig. 3 as a profile of Ti02. The depth resolution for a normal incidence beam, with back-scattered a Table 1. Reactions suitable for neutron depth profiling on stable targets Abundance, Energy of emitted Fermi cross- section Reaction [Yo part iclesikeV ?He(n,p)’H . . “Li(n,cu)’H . . 1(JB(n,a)7Li . . I4N(n,p)l4C . . 170(n,a)14C . . 33S(n,cu)i0Si . . T l ( n , p ) T . . 40K(n.p)40Ar 0.000 1 4 7.5 19.9 99.6 0.038 0.75 0.012 75.8 572 2055 1472 5 84 1413 308 1 598 223 1 191 2727 840 42 404 41 1 17 56 5533 940 3837 1.83 0.24 0.19 0.49 4.4 particles observed at 165”, is ca. 6 pg cm-2 for Ti02. A further application of the use of lhO(a,a)lhO is shown in Fig. 4. This oxygen profile of a superconducting Y - Ba - Cu - 0 specimen indicates that the oxygen stoicheiometry required for super- conductivity ([OJ 3 6.5%) occurs only at a depth of several hundred Angstroms inside the specimen.12 Neutron Depth Profiling (NDP) The principles and physical basis of NDP are described in the literat~re.13.1~ The technique is useful for the profiling of boron implants in silicon and helium implants in metals‘”lh and several other applications have also been described.” A summary of the reactions suitable for NDP and their characteristics is given in Table 1. Surprisingly, none of the NDP work reported so far has relied on the detection of the recoil nuclei. These have shallow escape depths by virtue of their low recoil energies and high masses and hence constitute excellent probes for the very near surface region.Better depth resolution can be achieved with recoil nuclei than with charged particles. The energy of the detection system presents an instrumental limitation to the depth resolution that might be obtained. The low-energy recoil nuclei have long flight times. and high- resolution time-of-flight (TOF) measurements are the obvious method of choice for their detection. In TOF detection, species are identified by their charge to mass ratio (qim). For maximum sensitivity and selectivity, recoil nuclei should have a single-charge state on emission from a surface. In practice the situation is not as clear cut, but remains favourable. For example, the 2OhPb recoils from the cw-decay of *loPo have a distribution of charge states that is approximately 10: 1 in favour of the +1 state over the +2 state.18 An experimental set-up for recoil-nucleus time-of-flight neutron depth profiling (RN-TOF-NDP) is shown schematic- ally in Fig.5. The start signal for the flight-time measurement is obtained from the electrons emitted when the recoil nucleus emerges from the sample surface. These electrons are accelerated towards a microchannel (annular) plate detector to generate a start signal. The recoil nucleus, after flight, impinges on a second microchannel plate detector and generates a stop signal. The RN-TOF-NDP experiment has been proposed only recentIy.19 Cold Neutron Techniques Cold neutroons are defined as neutrons having wavelengths in excess of 4 A or energies below 5 meV. Such wavelengths and energies are comparable to interatomic spacings and to the vibrational frequencies of atoms, respectively.Hence cold neutrons significantly enhance the structural characterisation capabilities that are available with thermal neutrons and X-rays.*() Moreover, there are waveguides available for cold neutrons. Consequently, they can be transported away from the reactor without prohibitive loss in intensity and they can be “focused” by judicious use of curved waveguides, produc- ing intense neutron beams on small spots. Hence, in contrastANALYST, MARCH 1989, VOL. 114 I n \\\ Grid stop Start Fig. 5. 50 40 30 20 10 v) c 3 +4 8 0 p 20 15 10 5 0 ..... . . . - . .I . .. . F Recoil . . I . . Gate and delay generator Timing electronics Schematic diagram of experimental set-up for RN-TOF-NDP 200 400 600 800 y-Ray energylkeV Fig.6. Prompt gamma neutron activation analysis (PGAA) of an NBS Standard Reference Material No. 1572, Citrus Leaves, exposed to (a) a thermal and ( b ) a cold neutron beam. Spectrum (a) was taken at the PGAA facility at NBS, (b) using a cold neutron beam at KFA Jiilich (spectra courtesy of R. Zeisler, NBS) to thermal neutron beams, cold neutrons can be delivered as intense beams in a low background environment (far away from the reactor). This provides favourable conditions for neutron-capture prompt y-ray activation analysis (PGAA). Several possibilities for the application of PGAA using thermal and cold neutrons are shown in Fig. 6.21 The count rates for the two prompt y-ray spectra are not comparable as the experiments were performed with different reactors; however, the signal to noise ratios are enhanced significantly when cold neutrons are used. For PGAA, an improvement in signal strength of two or more orders of magnitude is anticipated when using cold, as opposed to thermal, neutron beams, and hence the scope of PGAA should be expanded.22 Several possibilities remain to be explored.Neutron-cap- ture cross-sections are much larger for cold than for thermal neutrons and hence the use of focused cold neutron beams as “neutron microprobes” with unique analytical potential can be foreseen.23 271 1 10 100 (dE1dx)lMeV mg-’ cm-2 Fig. 7. Desorption yield for ergosterol ions (Mi-) for different primary ions as a function of the corresponding electronic stopping power plotted in a log - log profile.The straight line indicates a yield proportional to the square of the sto ping power.31 (m) 1271; (0) 63Cu; (A) 3 2 s ; (0) 1 6 0 ; (a) W; and (Ap7Li Microchannel plate detector k@\ Start detector \\zi lens ---I 252Cf - - - <?O source Y Constant fraction I discriminator Sample holder discrir stop I converter ND 66 Analyser + I P D P l l 24 I Fig. 8. Schematic diagram of experimental set-up for PDMS Heavy Ion Techniques The advantages of using heavy ions (atomic number, z b 3) have been demonstrated in a number of instances. The 2 2 dependence of their stopping power makes heavy ions the probe of choice for detecting thicknesddensity changes in opaque samples by absorptiometry.24 Similarly, the 2 2 depen- dence of their elastic scattering cross-section yields greater detection sensitivity and refined mass and depth resolu- tion.25J6 Heavy ion induced nuclear reactions are the preferred approach for the quantitative depth profiling of hydrogen or for trace determination in bulk materia1.27.28 It should be noted that radioactive beams are included under the heading of heavy ions.They provide alternatives to nuclear reactions that can be induced by stable ion bombardment. An examination of q values alone reveals potentially useful reactions for the detection of low z isotopes.29 This topic remains to be explored experimentally. A technique based on heavy ion induced desorption has analytical appeal. This phenomenon does not involve nuclear interactions directly but can be characterised as a “peripheral- nuclear” technique.The desorption process, which was first observed by Macfarlane and Torgerson,3” consists of the272 ANALYST, MARCH 1989, VOL. 114 3600 3000 2400 1800 1200 600 VI C 3 + s o -I+ 42+ C3H3+ 39 91 C7H7+ + monomer 0.4 0.8 1.2 1.6 2.0 2.4 2.8 600 n C, c4\ 2 0.5 0.9 1.3 1.7 2.1 2.5 2.9 3.3 3.7 Flight timeips Fig. 9. polystyrene. The spectrum was acquired for 7 h at -7 kV (a) PDMS positive-ion spectrum o f polystyrene. The spectrum was acquired for 3 h at +7 kV. ( h ) PDMS negative-ion spcctruni of emission of atomic and molecular species from a solid when the latter is bombarded by fast heavy ions ( E > 0.1 MeV u-I). The desorbed ions (or secondary ions) are identified in turn by The number of ions and neutral species desorbed from the target depends on several parameters, in particular the number, nature and energy of the bombarding projectiles.The desorption yield, Y , expressed as the percentage ratio of ions desorbed versus primary projectiles, can reach and even exceed 100%. The high yield makes this desorption process attractive for highly sensitive chemical analysis. The desorp- tion yield increases with the square of the stopping power (Fig. 7) and hence has a dependence on the nature of the projectile approximating to 25.31 Owing to this phenomenon, desorption TOF-MS. experiments are carried out with ions having the highest possible value of z and energies in the range 0.5-1 MeV u-1 (at approximately maximum dEidx). Obviously such ions can be generated by accelerators, but the above requirements are also remarkably well met by fission fragments.In practice, 252Cf is a convenient source of fission fragments to be used for inducing desorption. An experimental set-up using fission fragments from a 252Cf source and TOF-MS for identification of the desorbed ions is shown in Fig. 8. Typically, the analytical technique based on ion-induced desorption is referred to in the literature as “particle” or “plasma induced desorption mass spectrometry” (PDMS) .32 In the set-up shown in Fig. 8, the technique is operated in a single ion counting mode. It is also useful to recall thatANALYST, MARCH 1989, VOL. 114 273 390 ( a ) 260 325 I 0 500 lai 3 63 171 3 1500 2000 2500 360 300 240 180 120 60 0 N H+ + 27 *k IL-. 149 500 1000 1500 2000 2500 Flight timehs 240 200 160 120 130 4c c ( C) H+ 59 500 .I000 1500 2000 2500 3000 Fig. 10. ( u ) Time-of-flight spectrum of an untreated NBS Glass 617. Collection time was 1 h at 5 kV acceleration and 2 kV einzel lens potcntials. ( h ) Positive-ion spectrum for NBS Glass 617, cleaned with dilute nitric acid (10'/0). Collection time was 1 h at 5 kV acceleration and 2 kV einzel lens voltages, (c) Positive-ion spectrum for NRS Glass 617, etched with 10% hydrofluoric acid. Collection time was 1 h at 5 kV acceleration and 2 kV einzel lens voltages 150 125 100 v) C 4-s 2 75 0 50 25 0 150 125 v) 100 : 75 c K 0 50 25 I I C' Li 0 I t 375 425 475 525 575 625 675 725 775 Flight time/ns Fig. 11. NBS Glass 61 1 ; and ( h ) NBS Glass 617 Mass region miz 5-20 showing desorption of lithium: ( a ) TOF-MS is a simultaneous multi-mass analysis technique.With recent advances in ion optics and timing electronics, mass resolution (Arnirn) in excess of 10000 has been achicved.33 The features of PDMS can be summarised as follows: it is suitable for isotopic and molecular analysis by \.irtuc of the TOF-MS procedure; mass spectra are obtained with a total primary ion dose of S l O h ions, hence sample consumptionidamage is negligible; and PDMS has a shallow sampling depth (several to 100 monolayers depending on the nature of the target). i.e., the technique is suitable for outermost surface analysis without interference from the substrate. The amount of literature on PDMS is growing rapidly; an overview is provided by references 34-38.The features of PDMS can be illustrated by a few examples. An application to qualitative analysis is the use of PDMS to fingerprint polymers, see Fig. 9. Surface sensitivity can be investigated, e.g.. on a glass specimen that has been tested both before and after surface cleaning and etching (Fig. 10). In terms of sensitivity, the desorption yield for a given species is depen- 100 80 v) 60 c K 3 0 0 40 20 20 40 60 I 80 100 0 6Li, o/o Fig. 12. Error bars refer to counting errors Plot of "Li concentration versus the yield of hLi and 7Li. dent on the chemical environment from which it has been desorbed, indeed, as little as several tens of p.p.m. (picogram amounts in absolute terms) have been detected (Fig. 11). At present the database for PDMS is insufficient to allow a complete understanding of all the parameters affecting the desorption yield.39 Hence quantitative assays must be handled on a relative basis. The results of one example of a relative quantitative determination using the standard additions method are shown in Fig.12. PDMS has been applied as both macro- and micro-analysis techniques on a range of specimens from rocks to organic dye mixtures. The best known applica- tion is to the determination of large involatile biomolecules; relative molecular masses in excess of 20000 u have been detected using PDMS.37 Outlook Novel features in nuclear methods of analysis are emerging from different excitation modes using either low-velocity low-mass projectiles, viz., cold neutrons, or high-velocity heavy projectiles, viz., heavy ions of MeV u-1 energies.Both types of probe show promise for the enhancement of chemical and physical knowledge at atomic and molecular levels. One continuing challenge is to identify and demonstrate the role that these and other nuclear analysis techniques ought to have in physico-chemical characterisation. The author gratefully acknowledges comments and contribu- tions from H. L. Rook, R . F. Fleming, K. G. Downing (all of NBS), C. V. Barros, E. F. da Silveira, B. K. Patnaik (all of PUC-Rio) and from former and present students, M. U. D .274 Beug-Deeb, B. L. Grazman, L. Acevedo-Maldonado and L. Quifiones. Parts of this work were supported by the Robert A. Welch Foundation (Grant A-944) and the National Science Foundation (Grant INT-8602288).1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. References Ehmann, W. D . , and Yates, S . W., Anal. Chem., 1988, 60, 42R. Clarke, W. B., Koekebakker, M., Barr, R . D . , Downing, R. G . , and Fleming, R . F., Appl. Radial. Isot., 1987, 38, 735. Macnamara, J . , Collins, C . B., and Thode, H. G., Phys. Rev., 1950. 78, 129. Balestrini, S. J . , Phys. Rev., 1954, 95, 1502. Reynolds, J . H., Phys. Rev., 1950, 79, 789. Cameron, J. R., Phys. Rev., 1950, 90, 839. Skeldon, P., Shimizu, K., Thompson, G. E . , and Wood, G. C., Surface Interface Anal., 1983, 5, 252. Davies, J . A . , Domeij, B., Pringle, J. P. S., and Brown, F., J . Electrochem. SOC., 1965, 113, 675. Amsel, G . , Nadai, J . P., Artemare, E . D., David, D., Girard, E., and Moulin, J., Nucl.Instrum. Methods, 1971, 92, 481. Peterson, S . , Norde, H . , Possnert, G., and Orre, B., Nucl. Instrum. Methods, 1978, 149, 285. Chevarier, A . , Chevarier, N., Deydier, P . , Jaffrezic. H., Monocoffre, N., Stern. M., and Tousset, J . , J . Trace Micro- probe Technol., 1988, 6, 1. Patniak, B. K.. Barros Leite, C. V., Baptista, G . B., Schweikert, E. A., Cocke, D . L., Quinones, L., and Magnus- sen, N., Nucl. Instrum. Methods B , in the press. Ziegler, J. F., Cole, G. W., and Baglin, J . E . E.. J . Appl. Phys., 1972, 43, 3809. Biersack, J. P . , Fink. D . , Henkelmann, R., and Muller, K., Nucl. Instrum. Methods, 1978, 149, 93. Downing, R. G., Maki, J. T., and Fleming, R . F . , in Casper, L. A., Editor, “Microelectronics Processing: Inorganic Materials Characterisation,” ACS Symp.Ser., 1986, No. 295, p. 163. Lennard, W. N., Geissel, H . , Alexander, T. K., Hill, R . , Jackson, D . P.. Love. M. A., and Phillips, D., Nucl. Instrum. Methods Phys. Res., 1985, BlOAl, 592. Fink, D . , Biersack, J. P., and Tjan, K., Nucl. Instrum. Methods, 1982, 194, 105. Ito, S . . and Maeda, N., Nucl. Instrum. Methods Phys. Res., 1987, B29, 331. Grazman, B. L., Texas A and M University, personal communication, 1988. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. ANALYST, MARCH 1989, VOL. 114 Greene, G. L., Editor, “The Investigation of Fundamental Interactions With Cold Neutrons,” NBS Spec. Publ., 1986, No. 711. Lindstrom, R. M., Zeisler, R., and Rossbach, M., J. Nucl. Radioanal. Chem., 1988, 114, 277.Fleming, R. F., National Bureau of Standards, personal communication. Rook, H. L., National Bureau of Standards, personal commun- ication. Keith, D. J . , and Schweikert, E. A , , J . Trace Microprobe Technol., 1982, 1, 187. Sullins, R. T . , Barros Leite, C. V., and Schweikert, E . A , . J. Radioanal. Chem., 1983, 78, 171. Sullins, R. T . , Barros Leite, C. V., and Schweikert, E. A , , J . Radioanal. Chem., 1983, 78, 181. Cachard, A., Thomas, J. P., and Ligeon, E., in Thomas, J . P., and Cachard, A , , Editors, “Material Characterisation Using Ion Beams,” Plenum Press, New York, 1978, p. 367. McGinley, J. R . , Zikovsky, L., and Schweikert, E. A , , J . Radioanal. Chem., 1977, 37, 275. Friedli, C., Schweikert, E. A., and Lerch, P., J . Nucl. Radioanal. Chem., 1989, in the press. Macfarlane, R. D . , and Torgerson, D. F . , Science, 1976, 191, 920. Hakansson, P., PhD Thesis, University of Uppsala, 1981. Macfarlane, R . D., Acc. Chem. Res., 1982, 15, 268. Benninghoven, A . , paper presented at the Pittsburgh Confer- ence on Applied Spectroscopy, February 1988, paper No. 615. Koppenaal, D . W., Burlingame, A . L., Maltby, D., Russell, D. H . , and Holland, P. T., Anal. Chem., 1988, 60, 113R. McNeal, C. J . , Editor, “Mass Spectrometry in the Analysis of Large Molecules,” Wiley, New York, 1986. Hilf, E. R., Kammer, K., and Wien, K., Editors, “Lecture Notes in Physics: PDMS and Clusters,” Springer-Verlag, Berlin, 1987. McNeal, C. J . , Editor, “Texas Symposium on Mass Spec- trometry: IV. Analysis of Peptides and Proteins,” Wiley, New York, in the press. Le Beyec, Y., Editor, “Proceedings of the Second International Workshop on MeV and keV Ion and Cluster Interactions With Surfaces and Materials,” J . Phys. (Paris), 1989, in the press. Beug-Deeb, M. U. D . , da Silveira, E. F . , and Schweikert, E. A., J . Trace Microprobe Technol., 1989, in the press. Paper 8102979B Received July 21st, 1988 Accepted August 24th, 1988

ISSN:0003-2654    

DOI:10.1039/AN9891400269

出版商:RSC    

年代:1989

数据来源: RSC

摘要:

ANALYST, MARCH 1989, VOL. 114 275 Comparison of Two Radiometric Methods for the Determination of Americium-241 in Plutonium Containing Materials* Jozef L. Parus, Wolfgang Raab, Norbert Doubek, Peter Zahradnik and Stein Deron Safeguards Analytical Laboratory, International Atomic Energy Agency, P. 0. Box 100, A- 1400 Vienna, Austria Two methods for the determination of 241Am based on gamma-ray measurements are presented and compared: ( i ) isotope dilution gamma-ray spectrometry (IDGS) with 243Am and ( i i ) direct high-resolution gamma-ray spectrometry (HRGS). In the IDGS method, a spike was added and the Am was separated by partition chromatography. The intensity ratio of the gamma-ray energies was measured at 59.537 and 74.67 keV for 241Am and 243Am, respectively. The 241Am content was calculated from calibration graphs obtained for six standard solutions prepared by the addition of known amounts of 241Am reference solutions to weighed aliquots of 243Am spike solutions.The 241Am content in several Pu containing materials was calculated with respect to Pu, which was determined independently by titration. Eight samples were analysed, six measurements being carried out for each sample. The z4lAm content ranged from 0.9 to 3.5% with a relative standard deviation from k0.12 to 20.35%. In the HRGS method, the gamma-ray spectra of all the chemically untreated solid samples and of all solutions containing about 4 mg of Pu in 100 ml of solution were measured in the energy range 0-307 keV. The spectra were evaluated with a multi-group analysis program and the 241Am content was calculated with respect to Pu from the gamma-ray energy intensities in the 94-104 keV series of peaks.The relative standard deviation was within k l 0 / o and there was no systematic difference in comparison to the isotope dilution method. Keywords: Americium-24 1; americium-243; isotope dilution; gamma-ra y spectrometry; plutonium contain- ing materials The '-+'Am produced by the beta-decay of 241Pu represents a significant impurity in Pu containing materials, affecting the nuclear properties of the material and measurements such as those for nuclear materials accountability. Because of its relatively short half-life and high yield of low-energy gamma- radiation, 2"Am is a major contributor to the total activity of Pu containing material. It can be determined by both alpha- and gamma-radiation measurements. *,2 In order to obtain the most accurate results using alpha- radiation measurements, quantitative separation of 2"Am from Pu is required because 23SPu emits alpha-particles of very similar energy to those of 24lAm.Hence the preparation of a purified alpha-source for spectrometric measurement is a lengthy procedure. The main limitation in making direct, accurate alpha measurements is the difficulty in obtaining a reproducible geometry, owing to possible inhomogeneities in the distribu- tion of the analysed material in the alpha-source. This can be eliminated by using isotope dilution with 243Am, but the accuracy of the measurement is not better than 1-2%.7 The aim of this work was to develop a routine method for the determination of 241Am having an accuracy of better than 2 1%.Isotope dilution gamma-ray spectrometry (IDGS) was chosen for this purpose because a method based on activity ratio meacurements should give greater accuracy than a method involving a comparison with external standards. In our laboratory high-resolution gamma-ray spectrometry (HRGS) is used for the determination of Pu isotopes in all incoming samples containing this element. The procedure used is that developed by Gunnink4; this procedure also yields a measure of the 241Am concentration. Direct HRGS was used in parallel to obtain a measure of its accuracy. Experimental Materials Eight different Pu containing materials were analysed; viz., * Presented at the 2nd International Conference o n Nuclear and Radiochemistry.Brighton, UK, 11-15 July, 1988. three plutonium oxides (POX) and five mixed uranium - plutonium oxides (MOX). All materials were obtained in powdered form and two sub-samples of each material were taken for analysis. The ratio of U to Pu varied from about 1.8 to 20.0 and the amount of POX was 0.6 g (0.5 g of Pu) and the amount of MOX varied from 3 to 10 g (0.42-0.93 g of Pu) per su b-sample. The POX samples were packed in 2.5-ml Erlenmeyer flasks and the MOX samples in BC4 containers; the BC4 is a cylindrical brass cannister with an external diameter of 13.5 mm, a height of 40 mm and a wall thickness of 0.5 mm. It has a threaded brass screw-cap fitted with a plastic gasket. An open quartz vial 12 mm in diameter and 39.4 mm high is placed inside the brass cannister to hold the powdered sample.Isotope Dilution Gamma-ray Spectrometry (IDGS) Preparation of sample aliquots Plutonium- and mixed uranium - plutonium oxide powders were dissolved in a mixture of concentrated HN03 and 0.1 M HF (about 10 ml per gram of sample) and diluted with 2 M HN03 to obtain a solution containing about 4 mg ml-1 of Pu. A 1-ml volume of each stock solution was diluted further by mass to give a final concentration of about 50 pg ml-1 of Pu. ,Three weighed aliquots of these solutions, each containing about 80 pg of Pu, were added to three weighed aliquots of the 243Am spike solution. Preparation of spike solution About 8 mg of 243Am203 powder (ORIS, France), containing 1.27% of 2"Am (supplier's specification), were dissolved in 80 ml of 3 M HN03.This stock solution was diluted by a factor of 20 under mass control. The dilution factor selected was that which gave the same count rate at 74.67 keV (Z43Am) as that observed at 59.54 keV with a 0.5 pg ml-1 solution of a 24lAm standard. This count rate is obtained for a concentration of 4.5 pg ml-1 of 243Am in the spike solution, when the ratio of the concentrations is approximately equal to the ratio of the half-life of 241Am to that of 243Am.276 ANALYST, MARCH 1989, VOL. 114 Spiking and separation A 2-ml aliquot of the sample solution was mixed with 2 ml of the spike solution in a penicillin vial and the respective masses were recorded. The mixture was placed on a hot-plate at 120°C for 3 h to ensure complete homogenisation of the solution after which the solution was transferred on to a TOPO-FRACTOSIL column to separate the Am,5 which was eluted with 2 ml of 3 M HN03 into a 73 x 25 mm polyethylene vial.The volume of the eluate was then adjusted to 10 ml with 3 M HNO3 and the solution was used for gamma-ray spectrometric measurement. Calibration of gamma-ray spectrometer Eight 2-ml aliquots of the "Am spike solution, prepared under mass control, were placed in polyethylene vials of the same size as those used for the sample aliquots. Then, 0.4,O.g and 1.6 pg of the 241Am standard were added in duplicate to six vials containing the 24iAm spike and the volume of each solution was adjusted to 10 ml with 3 M HN03. The intensity ratio of the gamma-ray energies of 59.537 keV (2"Arn) and 74.67 keV ('J'Am) was measured for each solution.The following calibration relationships were obtained on the basis of a linear regression analysis: y = -0.05246 + 0.4539% for a 1500 mm2 detector; and y = -0.05153 + 0.39999~ for a 500 mm' detector, where y is the concentration of 2JIAm (pg g-1) in the spike solution and x is the ratio of the intensity of 2"Am to 243Am. The residuals had a maximum value of 0.25% relative to the known value. Gamma-ray Spectrometric Measurement Spectrometer with 500 mm2 high-purity germuniurn ( H P G € ) detector Each aliquot of the Pu solution was measured three times. The total number of counts was set at about 500000 for the less intense peak. The spectra were accumulated on 2048 chan- nels. Twenty channels were taken for peak-area calculation and two channels on each cide of the peak for linear background subtraction.The relative standard deviation for three counts ranged from 0.05 to 0.4%. The container with the measured specimen was inserted into a brass cylinder (72 X 25 mm i.d.; 72 x 75 mm o.d.) placed directly at the top of the detector so that the distance to the detector window was about 5 mm. The counting time was about 4600 s for a spike solution and between 700 and 1000 s for the other solutions. The dead-time of the multi-channel analyser was 15% maximum. No pile-up rejection system was used. Spectrometer with 1500 mm2 HPGE detector The measurements were performed with an automatic sample changer for a fixed time of 4500 s. The measured solutions were placed in similar brass cylinders at a distance of 40 mm from the detector window.The 2048 channels were used for data accumulation together with a pulse pile-up rejector connected to the spectrometer. All standards and analysed aliquots were measured once. Twenty channels were used for peak-area calculation and three channels on each side of the peak for background subtraction. High Resolution Gamma-ray Spectrometry (HRGS) Method description The method developed by GunninkJ and Gunnink and Rutherh was used for plutonium isotope composition analysis, utilising gamma-ray intensities in the range 94-208 keV. In addition to giving the plutonium isotope composition the method also allows the determination of the 24lAm content in relation to the total Pu, based on the intensity measurement o f two 24lAm gamma-ray energies at 98.95 and 102.96 keV.In the gamma-ray spectrum of Pu these two gamma-ray energies overlap strongly with the K X-rays of Pu and Np and the gamma-rays of 2-Tu and 23gPu. Calculation of their intensities requires deconvolution of the series of peaks in the range 94-104 keV. The method does not utilise any standards and only requires knowledge of physical constants to interpret the spectrum. The only calibration required is characterisation of the peak-shape constants for the measuring system, The Multi- group Analysis (MGA) computer program (developed by Dr. R. Gunnink) used in this work is also used on a routine basis in our laboratory for the evaluation of gamma-ray spectra of Pu containing materials.' Measurements All samples were measured initially as received at the laboratory. The POX samples were contained in 25-ml Erlenmeyer flasks and the MOX samples in Type BC4 containers and doubly sealed in plastic bags.For the measure- ment, the samples were placed on the top of the detector, which was covered with a cadmium absorber, 1 mm thick, to attenuate the 59.5 keV peak of 2"Am to approximately the height of the most intense peaks in the 94-104 keV series of peaks. The gamma-ray spectrometer was equipped with a 1500 mm2 HPGE detector, 13 mm thick, and had an energy resolution of 565 keV at 122 keV. A pile-up rejector and a spectrum stabiliser were included. The spectra covered an energy range of 307 keV on 4096 channels, which corresponds to 75 eV per channel as required by the MGA program.The counting time was 60 min and three spectra were recorded for each container. All samples were measured again after dissolution. A 1-ml volume of the solution containing about 4 mg of Pu was diluted to 100 ml in a polyethylene bottle with 3 M HN03, the bottle closed with a stopper and doubly sealed in a plastic bag. In this instance a 500 mm2 HPGE detector, 13 mm thick, with an energy resolution of 530 eV at 122 keV was used. The detector was covered with a cadmium absorber (1 mm thick). Three aliquots from each sub-sample of MOX and one aliquot from each sub-sample of POX were measured. Results and Discussion The results for the determination of 231Am are presented in Table 1 and are expresced in mass per cent. with respect to Pu.The concentration of Pu in solution was determined by A g o titration.8 The term IDGS 3500 in Table 1 refers to the result5 for isotope dilution measurements performed with the 1500 mm2 HPGE detector, IDGS 500 to the isotope dilution results obtained with the 500 inm2 HPGE detector, HRGS SOL to the results of high-resolution gamma-ray spectrometric Table 1. Results for the determination of '-"Am IDGS 1500 IDGS 500 HRGS SOL HRGS BC3 x, x, x. x. Sample Y' RSD, % RSD. 'XI KSD, Y o RSD. No. Material mlrn YU mlm 'YO mlrn (70 r r i i r n 'XU POX POX POX MOX MOX MOX MOX MOX 1.012 0.24 1.012 0.18 1.408 0.13 1.403 0.16 1.348 0.21 1.343 0.19 3.533 0.13 3.533 0.17 1.974 0.35 1.975 0.34 1.918 0.12 1.916 0.18 0.900 0.33 0.898 0.18 2.524 0.12 2.521 0.12 1.008 0.35* 1.398 0.61 * 1.345 0.71* 3.536 0.33 1.974 0.58 1.928 0.87 0.907 0.33 2.532 0.65 1.022 0 61 1.421 0 46 1.353 0 34 3.470 0.53 1.961 0.30 1.901 0.63 0.913 0.31 2.496 0.99 Average RSD: 0.20 0.19 0.55 0.53 * Result of two replicate measurements.ANALYST, MARCH 1989, VOL.114 277 Table 2. Comparison of 241Am results obtained with different procedures. Values expressed as measured 241Am contentP41Am content obtained by the IDGS 1500 technique Sample IDGS 1500 IDGS 500 HRGS SOL HRGS BC4 No. 1 .0000 1.0000 1 . 0000 1,0000 1.0000 1.0000 1 .0000 1.0000 I . 0000 0.9964 0.9970 1 . 0000 1.0005 0.9990 0.9978 0.9988 0.9965 0.9929 0.9978 1 .0008 1 . 0000 1.0052 1.0078 1.0032 1.0099 1 .(I090 1.0036 0.9823 0.9934 0.9912 1.0144 0.9888 measurements, evaluated with the MGA program, for solu- tion samples and HRGS BC4 to the HRGS results for powdered samples measured in BC4 containers. The arith- metic mean, x, is calculated from six replicate measurements and is expressed in mass per cent., unless indicated otherwise.The relative standard deviation (RSD) is expressed as a percentage of the mean. The IDGS 1500 and IDGS 500 results have virtually the same range of standard deviations, viz., from 0.12 to 0.35%; the average standard deviation, 0.20%, is also the same for both sets of results. Hence the standard deviation is independent of both the measuring system and the method of counting used. Increasing the number of counting operations to three for each aliquot in the IDGS 500 measurements did not increase the precision in comparison to IDGS 1500 where there was only one count per aliquot.Also, the use of a pile-up rejector and a dead-time correction system had no influence on the results. This was expected because counting losses are equally probable for both peaks, hence their ratio is not affected. The RSD calculated from counting statistics was in the range 0.2&0.26%; the measured RSD is above this range only in two instances for the IDGS 1500 results and in one instance for the IDGS 500 results. The results of HRGS for both solutions and powders have the same range of RSD values, viz., from 0.3 to 1%. This is about three times higher than for IDGS. The higher values of the RSD can be attributed mainly to errors in the dewnvolution process applied to the overlapping peaks. The results for the 241Am content in the samples are compared in Table 2, in which the values obtained by IDGS with the 1500 mm2 detector are taken as a reference.As can be seen, there is no systematic difference in the 241Am content found by the various procedures. The maximum difference for both IDGS procedures is 0.36%, which is within exactly one standard deviation. The results of HRGS measurements for solutions are within 0.78%, also very close to the value for one standard deviation (0.87%). The largest difference in the HRGS results for the chemically untreated samples in BC4 containers is 1.44%, which corresponds to 1.5 times the standard deviation for this procedure. The accuracy of the methods could also be checked independently, because the date on which the Am was separated from the POX samples was known, as was the 241Am concentration on that date (70 p.p.m.).The agreement between the calculated 241Am content and that found by IDGS was within 0.7%, i.e., within two standard deviations. Based on the results obtained, the 241Am content in the POX and MOX samples could be determined at the 95% probability level with an accuracy of 50.7% by IDGS, +1.5% by HRGS in solutions and +2O/0 by HRGS without any sample handling. 1. 2. 3. 4. 5. 6. 7. 8. References De Regge, P., Int. J . Appl. Radiat. Isot., 1984, 35, 251. Amoudry, F., and Silly, M., Int. J, Appl. Radiat. Isot., 1984, 35, 259. Rebagay, T. V., Huff, G. A., and Lee, R. S . , Nucl. Mater. Manage., 1983, 12, 168. Gunnink, R., “Determination of Plutonium Isotopic Abun- dances by Gamma-ray Spectrometry,” UCRL-52879, Law- rence Livermore Laboratory, Livermore, CA, 1980. Delle Site, A , , Doubek, N., Fiedler, R., Raab, W . , Swietly, H., Bagliano, G., and Deron, S . , in Laing, H. R., Editor, “Analytical Chemistry Instrumentation,” Lewis Publishers, Chelsea, MI, 1986, p. 325. Gunnink, R., and Ruther, W. D., “New TASTEX Gamma Spectrometer System for Plutonium Isotopic Analysis,” Part 1 Operations Guide, M-169 Part 1 (TSPO-247), Lawrence Liver- more National Laboratory, Livermore, CA, 1985. Parus, J., and Raab, W., Appl. Radiat. Isot., lnt. J . Radiat. Appl. Instrum., Part A , 1988, 39, 315. Drummond, J. L., and Grant, R. A., Talantu, 1966, 13, 477. Pug er 81 03384 F Received Ai.tgust i2nd, 1988 Accepted September 20th, I988

ISSN:0003-2654    

DOI:10.1039/AN9891400275

出版商:RSC    

年代:1989

数据来源: RSC

摘要:

ANALYST, MARCH 1989, VOL. 114 Measurement of Proton-induced Prompt Low Energy Photons by High Resolution Spectrometry* 279 Max Peisach National Accelerator Centre, P. 0. Box 72, Faure 7131, South Africa Lotter Geoffrey Lackay and Dherendra Gihwala Department of Physical Sciences, Peninsula Technikon, Bellville 7535, South Africa Prompt photons with energies of between 20 and 200 keV are not often favoured for analytical use because the resolution of Ge(Li) detectors is insufficient to cope with the high density of photopeaks that occur in that energy region of the spectrum, and the efficiency of Si(Li) detectors is very low for photons with energies above about 30 keV. Using thin intrinsic germanium detectors, a survey was carried out of the photons emitted by 77 elements under proton bombardment, either in the pure form or as pure compounds.From the survey, those photons that were potentially useful for analysis were identified and their yields measured for proton bombarding energies from 3.5 to 6.0 MeV. Some subjects for intensive study were identified. Because of the known high yields of the 110- and 197-keV gamma-rays from fluorine, the determination of this element in cements is reported. Several transition metals could be determined through the spectrometry of their low-energy prompt gamma-rays. The simultaneous use of prompt K X-rays and low energy gamma-rays made it possible to determine some rare-earth elements. The same approach was used for the determination of alloy metals in a gold matrix. Keywords: High resolution spectrometry; prompt photons; cements; alloys; geological ores The potential of proton-induced prompt gamma-ray spec- trometry as a rapid and simple analytical technique has been exploited for more than a decade.but most of the applications are confined to the use of relatively high energy gamma- rays.'-5 As most analysts rely on the use of large Ge(Li) or intrinsic Ge detectors, which have a resolution of more than 1 keV. the separation of individual gamma-rays is not possible where an appreciable concentration of gamma-rays exists over a short energy region, for example, for low energy gamrna- rays below 200 keV. An alternative to large germanium-based detectors is the Si(Li) detector. which is used extensively for X-ray spec- trometry. Although in principle this detector can be used for Spectrometry of low energy gamma-rays, the efficiency of Si(Li) detectors falls dramatically when the radiation energy exceeds about 30 keV. Hence an energy region exists between about 20 and 200 keV in which neither the Si(Li) nor the large Ge(Li) or intrinsic Ge detectors can be used effectively for analysis.At present this energy region, which is best served by thin intrinsic Ge detectors, has not been exploited extensively by analysts. A notable exception is the determination of fluorine through its 110- and 197-keV prompt gamma-rays,b.' because these radiations are particularly intense and the matrix in which fluorine is commonly determined does not generate close-lying interfering gamma-rays. To extend the full analytical significance of proton-induced prompt gamma-ray spectrometry, even into this low-energy region, it was necessary to study as many elements as possible.In this work all the stable. non-gaseous elements were studied under bombardment with protons from 3.5 to 6.0 MeV, even if they were known not to produce prompt gamma rays in the energy region 20-200 keV. After the study had been com- pleted, promising areas of application were identified. Experimental Materials and Standards Where possible. pure elements in their uncombined state were used. When pure elements were not available or if they were Presented at the 2nd International Conference on Nuclear and Radiochemistr!. Brighton. UK. 11-15 July, 1988. chemically reactive in air, pure, simple compounds such as oxides, carbonates, and nitrates were used instead.These compounds are listed in Table 1. Table 1. Compounds used as target materials Target Element Compound Li . . . . . . . . LizO N . . . . . . . . Kapton (CzZHIoNZ04)n 0 . . . . . . . . Y,03 Mg . . , . ' ' Mg(OH12 F . . . . . . . . GdF, N a . . . . . . . . NaCl P . . . . . . . . Hg3P0, CI . . . . . . . . BaCIz K . . . . . . . . KBr Ca . . . . . . . . Ca(OH), As . . . . . . . . As,03 Br . . . . . . . . KBr Rb . . . . . . . . Rb,SO, Sr . . . . . . . . Y . . . . . . . . Nb . . . . . . . . c s . . . . . . . . Ba . . . . . . . . La . . . . . . . . Ce . . . . . . . . Pr . . . . . . . . Nd . . . . . . . . Sm . . . . . . Eu . . . . . . . . Gd . . . . . . Tb . . . . . . . . Dy . . . . . . . . Ho . . . . . . Er . . . . . . . .Tm . . . . . . Yb . . . . . . . . Lu . . . . . . . . Hg . . . . . . . . TI . . . . . . . . Th . . . . . . . . u . . . . . . . .280 ANALYST, MARCH 1989, VOL. 114 Cement and steel standard reference materials were obtained from the US Bureau of Standards, Washington, USA. Standard gold alloys were obtained from the Depart- ment of Dental Technology of Peninsula Technikon, while geological ores were obtained from the Geology Department of the University of Cape Town. Preparation of Targets The thick metal targets consisted of discs 13 mm in diameter. Although the thickness of the discs varied, the thickness exceeded the range of 6 MeV protons in all instances. Elements available in powdered form, such as Be, C, S, Se and Os, and selected compounds, were pressed into tablets 13 mm in diameter and 1-2 mm thick.Standard cement reference materials and geological ores were treated similarly. Standard steels were machined into discs of the same diameter and about 2 mm thick, polished to a mirror-like finish and washed with ethanol, acetone and water under ultrasonic agitation. Gold standards were used in the form of 4 x 4 mm discs whic-h were 2 mm thick. Irradiation and Measurement All irradiations were carried out in a multi-purpose scattering chamber8 at the 6-MeV van de Graaff accelerator at Faure. A constant geometrical arrangement of targets and detectors was rigidly maintained. Proton beams with energies of between 3.5 and 6.0 MeV were used for the survey, while analyses were performed at 4.5 MeV. Beam currents were adjusted so that the dead-time of the entire counting assembly did not exceed 10%.Irradiations usually lasted about 20 min, but for analytical measurements, irradiations varied according to the elemental concentrations and precision required. Gamma-ray spectrometry was carried out with an intrinsic germanium detector with an active area of 25 mm2 and a thickness of 5 mm, mounted inside the scattering chamber at an angle of 45" to the bombarding beam. The resolution of the detector was 147 eV at 5.9 keV and 500 eV at 122 keV. Data were recorded on magnetic tape and processed off-line. Table 2. Identified gamma-rays from the proton bombardment of a tablet of pure Gd203 Peak Energy/ Peak Energy/ No. keV Assignment No. keV Assignment 1 9.1 2 21.3 3 24.5 4 54.6 5 60.0 6 63.9 7 64.4 8 73.7 9 75.3 *5'Gd p( 1,O) 156Gd n( 1,O) 156Gd n(2,l) 157Gd p( 1,O) l55Gd p( 1,O) I5'Gd p(2,O) 157Gd p(5,3) 1sgGd n( 1,O) 16"Gd p( 1,O) 10 76.9 11 79.5 12 86.5 13 89.0 14 95.6 15 110.0 16 116.3 17 123.0 18 131.4 157Gd p(4,2) 158Gd p( 1,O) l55Gd p(2,O) 156Gd p( 1,O) I5'Gd p(6,4) 157Gd n(2,l) 1j7Gd p(5,2) 154Gd p( 1,O) I5'Gd p(4,O) Results and Discussion Survey of the Elements Although there is a vast amount of literature on the decay of the excited states of atoms and nuclei by photon emission, most of this information was accumulated under conditions that make comparison of the X-ray and gamma-ray intensities from different elements in a single matrix difficult, if not impossible.For analytical purposes, therefore, it was neces- sary to compile data obtained under comparable conditions.In this work, intensities were derived from spectra measured at selected proton energies under irradiation with the maxi- mum acceptable current. Subsequently all data were normal- ised to unit integrated charge. A compilation of all the data' provides a means of assessing which X-rays and gamma-rays may be used for analysis and which are needed to evaluate possible interferences. For the latter, even low intensity radiations are significant in those instances where the target element forms a major component of the matrix. Throughout, the convention employed in previous work10 for naming prompt gamma-rays has been used, because the analyst needs to know the exact target nucleus. ' - 0 100 200 300 400 500 600 700 800 900 1000 1100 1200 Channel number Fig.1. Prompt photon spectrum from pure metallic titanium bombarded with protons (EP = 4.5 MeV). The photons are identified in Table 3 Ko; 10' 0 60 120 180 240 300 360 420 480 540 600 660 720 Channel number Fig. 2. Prompt photon spectrum from a tablet of pure Gd203 bombarded with protons ( E , = 4.5 MeV). The photons are identified in Table 2 Table 3. Yield and sensitivity data for identified gamma-rays from the proton bombardment of pure titanium Yield/counts yC-1 Energy/ Peak No. keV Assignment 3.5 MeV 4.0 MeV 4.5 MeV 5.0 MeV 5.5 MeV 6 MeV 1 32.9 5"Ti n(4,3) 39 790 2800 4400 7500 11800 2 35.3 "Ti n(3,2) 47 1300 5000 8500 13100 21300 3 58.2 47Ti n(2,l) 68 58 1700 3500 6400 20700 4 62.2 47Ti n(2,O) 96 690 1760 2500 3200 4700 5 87.5 47Ti n( 1,O) 22 200 3400 7600 14000 22300 6 90.7 4yTi n(1,O) 830 3500 8700 13100 19000 27000 7 94.1 s"Ti n(2,l) 40 2000 8300 17100 20300 48400 8 98.0 4RTi n(4,2) 8 52 74 210 410 7900 9 112.5 48Ti n(2,l) 6 18 38 66 4700 30000 10 152.9 4yTi n(2,O) 110 380 1000 1800 2100 2500 11 159.8 4xTi p( 1,O) 370 1500 3700 5700 8000 11500 12 210.8 48Tin(3,1) 10 19 36 43 180 2500 13 226.2 s"Tin(1,O) 10 290 1300 2700 4900 8900 Sensitivity/ pgg-' mC-1 at 6 MeV 640 350 710 2760 400 300 180 920 230 6280 420 7390 450ANALYST, MARCH 1989, VOL.114 281 Table 4. Sensitivity limits for gamma-ray analysis using protons with an energy of 6 MeV and an intrinsic germanium detector. X-ray sensitivity limits are based on the Kcc yields Sensitivity limit Sensitivity limit Energy! Gamma-ray yield/ (gamma-ray)/ (Kcc X-ray)/ Element' keV Assignment counts UC-1 pg g-1 mC-1 pg g-1 mC-1 - F .. . . . . . . 197.2 IyF ~ ( 2 8 ) 2.647 x 10s 11 Al . . . . . . 170.8 27~1p(2.1) 3.083 x 10-1 1420 K 266.7 41Kp(3,2) - 8.126 x 10 11 900 - Ca . . . . . . 121.5 %a n( 1 .O) 3.173 X 10' 290 Sc . . . . . . 292.8 'Sc n(3.1) 8.318 X lo2 1600 2370 Ti . . . . . . . . 94.1 "'Ti n(2.1) 4.844 x 10' 180 2050 v . . . . . . . . 315.6 'lVn(5.3) 5.949 x 103 820 1780 Cr . . . . . . 54.7 'JCr n( 1 .O) 6.529 x 10' 660 600 Mn . . . . . . 125.9 % I n p(1,O) 3.299 X lo4 230 570 Fe . . . . . . 127.6 S7Fe n(3,2) 7.057 x 10' 4100 510 . . . . . . 270 460 c o 339.1 T o n( 1,O) 1.444 x 104 Xi . . . . . . 159.3 6 4 ~ i n(1 ,0) 8.009 x lo2 2700 370 Zn . . . . . . 174.9 hXZn n( 1,O) 1.001 x 104 360 58 .. . . . . . . - c u . . . . . . 115.1 WU 42.0) 2.653 X lo4 260 150 Ga . . . . . . 174.9 71Ga n(l.O) 1.024 x 104 180 55 Ge . . . . . . 121.1 76Ge n(3.0) 6.626 x 10' 920 53 As . . . . . . 112.5 '?As n( 1,O) 8.108 X lo4 130 51 Se . . . . . . 219.0 76Se p(2,l) 2.611 x 10' 1710 49 Br . . . . . . 183.0 "Br n(3,0) 8.921 x 103 1040 43 Sr . . . . . . . . 231.9 XHSr n( 1 ,O) 5.498 X 10' 1910 35 Y . . . . . . . . 357.1 xyY n(3.2) 2.719 x 10' 6500 33 Rb . . . . . . 231.7 XsRb n( 1 .O) 5.573 x 103 400 36 Zr . . . . . . 92.0 Y2Zr n(2,l) 9.560 X 10' 640 30 Nb . . . . . . 123.4 "'Nb g(7,5) 8.554 x 102 2470 29 Mo . . . . . . 203,8 y5Mop(l.0) 2.344 x lo3 1370 28 Ru . . . . . . 127.2 I(jlRu p( 1,O) 3.009 x 103 1430 27 Rh . . . . . . 118.7 I"3Rh n(l.O) 1.380 X lo4 350 27 Pd .. . . . . 113.0 lL'yPd p( 1 ,0) 6.112 x 10' 790 23 Cd . . . . . . 246.3 II'Cd p( 1.0) 1.253 X lo2 9560 170 In . . . . . . . . 115.4 IIsIn n(2.1) 7.066 X lo3 920 144 Sn . . . . . . 61.1 IQSn n( 1,O) 2.788 x 10' 1010 30 Sb . . . . . . 212.2 I2lSb n(l,0) 2.631 X lo2 1210 27 Te . . . . . . 109.3 IZsTe p(2,l) 1.236 X lo3 1950 29 cs . . . . . . 160.9 133cs p(2,O) 5.766 x 10 22 100 27 Ba . . . . . . 155.4 I"Ban(1,O) 1.261 x lo2 7300 30 La . . . . . . 254.7 3YLa n(1,O) 7.754 x 10 4980 25 Ce . . . . . . 109.0 l W e n ( 3 , l ) 5.766 X 10 19 400 26 Pr . . . . . . 145.4 1"Pr p(1,O) 1.598 x lo? 1320 30 Nd . . . . . . 130.0 ls[)Nd p( 1,O) 4.536 x 102 1100 28 Eu . . . . . . 110.5 Is1Eu p(4,2) 1.353 X 10' 640 36 Gd . . , . , . 26.5 IssGd p(2,l) 4.468 x 103 670 42 Ag . . . .. . 205.0 I'J7Ag n( 1,O) 8.393 X 10' 2350 54 I . . . . . . . . 124.7 1271 n(l.O) 6.989 x lo3 480 24 Sm . . . . . . 121.8 I'*Sm p(1,O) 1.896 X lo3 670 34 Tb . . . . . . 137.4 i5yTb p( 1,O) 7.738 x 102 1330 54 Dy . . . . . . 73.4 164Dyp(1,0) 5.526 X 10' 3610 57 Ho . . . . . . 94.7 l6sHo p( 1,O) 7.359 x 102 1830 57 Er . . . . . . 80.6 166Er p( 1,O) 1.172 X lo2 2680 65 Tm . . . . . . 109.8 IhyTm p(2,l) 3.155 x 103 490 58 Yb . . . . . . 76.5 174Ybp(1,0) 7.778 X lo2 2260 58 Lu . . . . . . 113.8 175LU p( 1 ,0) 2.167 X lo3 650 75 Hf . . . . . . 180.9 I7YHf n(3,0) 1.178 X lo4 450 110 Ta . . . . . . 136.2 lglTa p(2,O) 9.043 x loz 1050 220 w . . . . . . 122.3 lX6W p( 1 .O) 3.504 X lo2 1390 210 Re . . . . . . 134.2 lX7Re p( 1,O) 1.140 X lo3 320 80 0 s . . . . . . 156.7 1x70s p(8,4) 7.921 x 102 1320 190 I r .. . . . . . . 139.0 1931r p(3,o) 8.713 X lo2 1120 130 Pt . . . . . . . . 211.3 lYsPt p(4,O) 1.102 x 102 1520 120 Au . . . . . . 279.0 IY7Au p(3,O) 7.452 x 10 3360 93 Hg . . . . . . 158.4 lYYHg p(1,O) 3.471 x 10 6400 83 T1 . . . . . . . . 185.2 2OsT1 n(5,3) 5.192 x 10 8300 66 58 Pb . . . . . . Th . . . . . . 112.8 232Thp(2.1) 1.095 X lo2 2060 52 - - - - Bi . . . . . . 195.0 2("Bi n(9,5) 3.131 X 10 6400 55 u . . . . . . . . 201.0 2 3 x u ~(11.1) 8.714 X 10 9060 50 No prompt gamma or X-rays were observed for the elements Li, Be, B , C, N, 0, Na, Mg, Si, P, S and C1 under bombardment with 3.5-6.0- MeV protons.282 ANALYST, MARCH 1989, VOL. 114 Table 5. Data for the determination of fluorine in cements using 4.5-MeV protons and the 110-keV IyF p(1.0) gamma-ra) Fluorine Mean SRM content.count! 639 0.02 1298 635 0.04 2512 637 0.04 2460 638 0.04 2473 636 0.06 3809 633 0.08 5074 634 0.08 4965 cement % m/m pC-' No. of analyses: 33 Mean count (for 1% F): 62 857 k 1005 pC-1 Relative standard error: 1.6% Root mean square error: 0.003% mlm Count for 1% mim Fi pC-' 64 900 62 800 61 500 61 825 63 483 63 425 62 063 Fluorine found. % 0.021 0.040 0.039 0.039 0.061 0.081 0.079 No of analyses 4 6 4 5 5 5 4 Error Absolute. YO mim + 0.0007 -0.0004 -0.0009 - 0.0007 +0.0005 +0.0007 -0.0010 Relative, o/o +3.5 -1.0 -2.3 -1.8 +0.8 +0.9 -1.3 Table 6. Summary of results for the analysis of cement with protons of 4.5 MeV using an intrinsic germanium detector Element F . . , . , , Al . . . . , . Ca . . . . , . Sr .. . . , . Concentration range, YO Photon used 0 02-0 08 110-keV IyF p( 1 ,0) 197-keV l"Fp(2.0) 1 5-3 3 171-keV"AI p(2,l) 42 5-47 2 122-keV4hCa n(2.0) 131-keV 4XCa n(3.0) 0 03-0 3 14 2-keV Sr Kol X-ra! Yield for 1"L of element countsuC I 6 286 x 10' 3 376 x loJ 4 328 x 10' 2 820 x 101 2 584 x 10; 3 299 x 104 Relative standard error, "/u 1.6 4.3 3.6 5.7 3.9 3.8 Root mean square error pgg-l 33 195 2100 770 240 52 - 104 5 1 0 3 a, t m L a, a g 102 2 0 10' Sr KO I Sr 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 Channel number Fig. 3. SRM 633 Cement (15, = 4.5 MeV) Prompt photon spectrum from the proton bombardment of The prompt photon spectra from the proton bombardment of pure Ti and pure Gd203 are shown in Figs. 1 and 2. respectively. Photon Yields and Sensitivity of Analyses The sensitivity of the analysis is determined not only by the yield of the X-ray or gamma-ray, but also by the background against which the photopeak has to be measured.The detection limit" of approximately three times the standard deviation of the background in the spectrum obtained from a target of. ideally, a pure element. was used as a measure of the sensitivity. In a thick target this value expresses the interfer- ence-free detection limit of the element. because in an ideal matrix containing no other elements. the spectrum obtained from the element under investigation will replicate the shape of the spectrum obtained from a pure element. In terms of the above definition, sensitivities and yields were calculated from gamma-rays and X-rays for each element, obtained during bombardment with protons from 3.5 to 6.0 MeV in steps of 0.5 MeV.Photon yields were calculated as direct counts uC-* of current. The yields of identified gadolinium and titanium gamma-rays shown in Tables 2 and 3, respectively. should serve as an example of similar data obtained for all the elements studied in the survey. Approxi- mate interference-free sensitivity limits for gamma-rays and X-rays were calculated similarly as pg g-1 for 1 mC of current. Table 4 shows the sensitivity limits for both gamma-ray and X-ray analyses using protons of 4.5 MeV. The X-ray sensitiv- ity limits are based on the combined Karl and Kor2 X-ray yields. Application of Prompt Photon Spectrometry Analysis of cements The manufacture of cements involves the processing of geological material from different sources.with the main components being calcium carbonate and aluminium silicate. The exact composition of the ores may have a profound effect on the performance of the final product. Therefore, it is obvious that rapid analysis of both new materials and the product is vital for process control. The technique described in this work adds to the arsenal of the analyst in that it is rapid. non-destructive and experimen- tally simple. Its application was checked with SRM cements. A typical spectrum obtained from SRM 633 Cement after a 10-min bombardment with protons of 4.5 MeV is shown in Fig. 3. Strontium is a minor component of cements and is commonly present in concentrations of 0.03-0.3%. The Kor X-ray of 14.16 keV was used for the determination of this element.Fluorine. normally a minor component in cements. can be determined at the ug g-I level using gamma-rays of 110 and 197 keV, both of which are generated by Coulomb excitation. Although aluminium is best determined through the use of several intense gamma-rays at high energies measurable with a Ge(Li) detector. the gamma-ray of 171 keV was found to be sufficiently intense when measured with an intrinsic germanium detector. It was therefore used for the determination of this element. Calcium. the major component of cements, was determined through its heavier isotope of relative atomic mass 48. Despite the fact that this isotope is present in nature in abundances of only 0.19 atom-%, the high concentration of calcium makes it possible to use its gamma- rays of 122 and 131 keV to determine this element.The results given in Table 5 for the determination of fluorine provide an example of similar data obtained for the other determinations. The precision, sensitivity and accuracy for the determination of fluorine were evaluated from the analysis of seven cement standards. Between four and sixANALYST, MARCH 1989, VOL. 114 replicate analyses were performed on each standard and a mean value (from 33 observations) of 62857 i 1005 counts & - I for lolo mlm of F was obtained. The relative standard error of 1.6% gives an indication of the precision of the analysis over the cited concentration range. The root mean square error of 0.003°/~ mlm indicates that the method can be used for the determination of fluorine over concentration ranges usually associated with the proton induced X-ray emission (PIXE) technique.This is particularly interesting because fluorine cannot be determined by the latter tech- 283 0 100 200 300 400 500 600 700 800 900 1000 Channel number Fig. 4. gold allo) ( E , = 4.5 MeV). The photons are identified in Table 7 Prompt photon spectrum from the proton bombardment of a Table 7. Identification of photons from gold alloys Photon Peak Energ) No keb Assignment 1 9 44 Pt Lor, X-raq 2 9 71 AuLa,X-raq 3 11 07 PtLf3,X-ra) 3 13 38 AuLy, X-ra!, 5 19 42 Au La, + Au La, X-ra) 6 22 16 Ag L a , X-raq 7 8 9 1 0 11 12 13 14 15 16 17 18 19 20 22.88 Au L a i + Au Lf3, X-ray 24.93 Ag Kp, X-ray 54 hiCun(1.0) 61 Unidentified 65.11 Pt Ka, X-ray 66.82 Pt Ka, X-ray 68.79 Au K a , X-ray 75.74 Pt KB, X-ray 77.97 Au K/3, X-ray 80.17 Au KBz X-ray 115 ~ ' C U n(2.0) 153 6 ~ C u n(3.1) 194 "Cun(1.0) 201 Wkn(8.6) Remarks Au LP,interference Sum peak Au L a , + Au Lp, interference Sum peak Used for analysis Low intensity Used for analysis Au Ka, interference Used for analysis Pt Kpz interference Used for analysis Used for analysis Used for analysis Low intensity Low intensity nique.A summary of the analytical parameters for the other photons is given in Table 6. The differences in the precision obtained in the use of the respective photons reflect the statistical errors in counting, the intensities of the correspond- ing backgrounds and the relative intensities of the photons. Analysis of gold alloys Small amounts of metals can be introduced into fabricated gold objects in order to debase the gold content of the object or to produce alloys with properties to suit a particular purpose.The samples chosen in this study were dental gold alloys. specifically used for casting tooth inlays, and attach- ments for denture bridges and crowns. Although PIXE, using Si(Li) detectors, seems to be the a priori method of choice, the high L X-ray flux of gold could lead to difficulties in detecting a minor component such as platinum because of interfering L X-rays. In addition, the presence of a base metal such as copper would produce K X-rays that are in the same energy region as the gold and platinum L X-rays, thereby enhancing the interference effect. On the other hand, if an intrinsic germanium detector is used, Table 9.Summary of results for the analysis of gold alloys with protons of 4.5 MeV using an intrinsic germanium detector Yield for Root 1% of mean Con- element Relative square centration counts standard error/ Element range, YO Photon used pC-1 error.% ugg-1 Cu . . 10-15 1l5-keVh5Cun(2,0) 81 626 1.7 270 153-keV~Xun(3.1) 26033 3.9 590 Ag . . 13-41 24.942-keVKpX-ray 102764 1.0 946 Pt , . 6-10 65.111-keVKa2X-ray 97080 4.7 110 75.736-keVKaI X-ray 23 412 1.4 479 Au . . 40-90 68.794-keVKaIX-ray 89314 2.1 103 80.165-keV KBz X-ray 35 785 3.9 528 Table 10. Identification of photons from geological ores Peak KO. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Energy/ keV 14.14 15.85 33.30 34.57 37.19 38.73 39.26 40.23 42.27 43.30 110.0 121.3 125.9 130.9 Assignment Sr K a X-ray Sr Kp X-ray La K a X-ray Ce K a X-ray Nd Ka X-ray La KP2 X-ray Ce KP, X-ray Ce Kp2 X-ray Nd Kp, X-ray Nd KB2 X-ray "Fp(l.0) 4XCa n(2,0) 54Mn p( 1,O) 4XCa n(3.0) Remarks Used for analysis Used for analysis Used for analysis La KP, X-ray interference Low intensity Used for analysis Low intensity Ce, Th interference Fe interference Used for analysis Nd interference Table 8.Data for the determination of silver in gold alloys using 4.5-MeV protons and the 24.942-keV Ag Kf3 X-ray Error Count for 1 ";b Ag content, Mean count/ m.'m Ag.' Ag found, No. of Absolute. Relative. Alloy "0 m,'m p c - I gC-I Yo analyses YO mim YO DA . . , . . . . . 11.5 1 166 479 101 433 11.35 4 -0.15 - 1.30 S G L . , . . . , , . 22.5 2311 448 102 731 22.49 4 -0.01 -0.03 DII ., . . . , . . 13.5 1 394 240 103 277 13.57 4 +0.07 +0.50 P3 . . . . . . . , 40.4 4 196 407 103615 40.84 4 +0.34 +0.83 So. of anal>ses: 16 Mean count (for 1% Ag): 102 764 k 959 uC-I RelatiLe standard error: 1.0% Root mean square error: 0.095"/0 m'in284 ANALYST, MARCH 1989, VOL. 114 2 0 Table 11. Data for the determination of lanthanum in geological samples using 4.5-MeV protons and the 33.30-keV La Ka X-ray Error 27 24 I Count for La contenti Mean count/ 1% La/ AFC056 . . . . . . 999 12 723 127 357 AFC036 . . . . . . 636 8 508 133 772 AFC010 . . . . . . 361 4 684 129751 FC071 . . . . . . 196 2 706 128 061 AFC231 . . . . . . 121 1592 131 570 Sample Pg g-' UC-' pC-' No. of analyses: 25 Mean count (for 1% La): 130 102 L 2620 pC-' Relative standard error: 1.5% Root mean square error: 6 pg g-' - La found/ No.of 978 5 654 5 360 5 193 5 122 5 Pg g-l analyses Absolute/ Relative, -21 -2.1 + 18 +2.8 $g g-' % -1 -0.28 -3 -1.53 -1 -0.82 Table 12. Summary of results for the analysis of geological samples with protons of 4.5 MeV using an intrinsic germanium detector Mn , Sr . La . Ce . Nd . Concentration range/ Element Pgg-' Photon used , , , , , , , 2 x 103-10 x 103 126-keV 54Mn p(1,O) . . . . . . . 9OC-5000 14.14-keV KaX-ray . . . . . . . 120-1300 33.30-keV KaX-ray . . . . . . . 300-2700 34.57-keV KaX-ray . . . . . . . 200-1000 42.27-keV KP, X-ray Yield for 1% of element/ counts pC-1 12 922 335 054 130 102 84 724 87 206 Relative standard error, 2.7 1.6 1.5 3.3 3.6 Root mean square error/ wg g-1 74 45 6 91 121 Table 13.Identification of photons from SRM steels Peak N O . 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Energy1 keV 15.774 16.614 17.478 18.621 19.786 24.9 25.2 28.0 54.7 58.6 67.4 90.1 91.3 92.0 109.1 111.5 115.1 122.1 125.8 127.2 153.0 156.3 159.3 197.1 249.0 349.0 377.9 Assignment Zr Ka X-ray Nb Ka X-ray Mo Ka X-ray Nb KP X-ray Mo KP X-ray 5*Fen(1,0) Unidentified 51V~(3,2) 54Cr n( 1 .O) 58Fe n(3,2) 61Ni p(1,O) 58Fe n(6,4) 55Mn n(4,3) 94Mo n(2,O) 58Fe n(3,O) 65Cu n(2,O) 57Fe p(2,l) 55Mn p( 1,O) 5'Fe n(3,2) 63Cu n(3,l) 54Cr n(2,O) 64Ni n( 1,O) 6*Ni a( 1,O) 63Cu n(2,O) 62Ni n(4,l) 53Cr n(l ,0) 5ov ~ ( 2 , 1 ) Remarks Zr K a X-ray interference Used for analysis Low intensity Copper interference Used for analysis Used for analysis Low intensity Low intensity Used for analysis Used for analysis Low intensity Iron interference Used for analysis Low intensity Low intensity Used for analysis the K X-rays of gold and platinum are not only measurable, but are also sufficiently well separated to allow them to be used for analysis and the possibility exists that the base metals may have low energy gamma-rays that are measurable within the efficiency range of the detector. The gold standards used contain minor amounts of copper, silver and platinum in a gold matrix.A typical spectrum obtained from the standard DEGULOR-M (DM) is shown in Fig. 4 and the identities of the marked peaks are given in Table 7 . The results given in Table 8 for the determination of silver can be taken as an example of similar results obtained for the other determinations; the latter are summarised in Table 9.50 100 150 200 250 300 350 400 450 500 550 600 650 Channel number Fig. 5 . Prompt photon spectrum from the proton bombardment of a geological ore ( E , = 4.5 MeV). The photons are identified in Table 10 I 1000 1200 1400 1600 1800 2000 Channel number Fig. 6 . Prompt photon spectrum from the proton bombardment of SRM 1154 Steel ( E , = 4.5 MeV). The photons are identified in Table 13 Analysis of geological ores The determination of rare-earth elements is becoming increas- ingly important in the characterisation of specimens, and in general, rapid and sensitive analytical methods for these elements are limited. The proposed method offers both these advantages and this is demonstrated by the determination of some rare-earth elements in geological ores.ANALYST, MARCH 1989, VOL.114 285 Table 14. Data for the determination of manganese in steels using 4.5-MeV protons and the 125.8-keV 55Mn p(1.0) gamma-ray Error Count for Mn content, Mean count/ 1% Mnl Mn found, No. of Absolute, Relative, Sample % mtm UC-1 UC-' % m/m analyses % mlm Yo SRM1152 . . . . 1.17 13 229 11 309 1.152 4 -0.018 -1.5 SRM1185 . . . . 1.22 14 173 11 617 1.234 4 +0.014 +1.1 SRM1155 . . . . 1.63 18 768 11 514 1.634 4 +0.004 +0.3 SRM1154 . . . . 1.74 19 998 11 493 1.742 4 +0.002 +0.1 No. of analyses: 16 Mean count (for 1% Mn): 11 483 i 128 Relative standard error: 1.1% Root mean square error: 0.0058% mlm Table 15. Summary of results for the analysis of SRM steels with protons of 4.5 MeV using an intrinsic germanium detect01 Element v . .. . Cr . . . . Mn . . . . Ni . . . . c u . . . . Mo . . . . Concentration range, YO m!m 0.04-0.06 17.0G20.00 1.2G1.80 10.0c-13 .OO 0.05-0.6 0.3-0.6 Yield for 1% of element/ Photon used counts UC-1 90-keV50Vp(2,1). 8572 378-keV 53Cr n( 1,O) 920 126-keV "Mn p(1,O) 11 483 67-keV 61Ni p( 1,O) 368 115-keVhSCu n(1,O) 7380 19.786-keV KB X-ray 16 383 Relative Root mean standard square error/ error, YO ug g-' 4.7 1140 1.9 873 1.1 58 3.6 780 2.6 390 3.1 65 The samples used in this investigation were carbonatites consisting mainly of CaC03 and containing various amounts of MgC03 and FeC03 with trace amounts of rare-earth elements. A typical spectrum obtained from a carbonatite specimen after a 10-min bombardment with protons of 4.5 MeV is shown in Fig.5 and the identities of the numbered peaks are given in Table 10. The high resolution of the intrinsic germanium detector allows the cluster of K X-ray photopeaks from La, Ce and Nd to be separated sufficiently for possible analytical use. The presence of prompt gamma- rays from Ca and Mn makes the method useful for multi- elemental analysis and can enhance further the characterisa- tion of the specimens. The results for the determination of lanthanum given in Table 11 indicate a sensitivity of 69 pg g-1 with an acceptable precision of 2.8%. Similar results were obtained for Mn, Sr, Ce and Nd as shown in Table 12. Analysis of steel alloys Steels represent a good system on which the method can be tested because most steels contain a variety of minor components in a matrix consisting largely of iron. The importance of high resolution spectrometry has been demonstrated pre~iously12~13 in the analysis of steel using alpha-induced prompt photons.The spectrum obtained from the bombardment of SRM 1154 Steel with protons of 4.5 MeV is shown in Fig. 6. The identity and origin of the numbered peaks in the spectrum are listed in Table 13 together with the possible interferences on the photopeaks suitable for use in analysis. Although Zr and Nb were not certified, their presence can clearly be seen by their K X-rays. The Mo KP X-ray was separated sufficiently from the Zr and Nb X-rays for it to be used for analysis. This is particularly significant in the determination of molybdenum because the yields of its prompt gamma-rays are relatively low14 for analytical pur- poses even when measured at higher energies with Ge(Li) detectors.The appearance of the 127-keV gamma-rays of 57Fe n(3,2), although interfering in the determination of manganese by its 126-keV 55Mn p(1,O) gamma-ray, is not significant because the high resolution of the detector used clearly distinguishes the photopeaks of these gamma-rays. Vanadium, Cr, Ni and Cu, although determined in this investigation, also provide a number of intense gamma-rays14 using Ge(Li) detectors, which may be more suitable for the determination of these elements. The present study may therefore serve as a complementary technique to the latter. Typical results for the determination of manganese in SRM steels containing 1-2% mim of Mn are given in Table 14.Similar determinations were carried out for V , Cr, Ni, Cu and Mo and these results are summarised in Table 15. Conclusion A survey of low energy prompt photons, induced by protons, on all the non-gaseous elements and the variation of their yields with bombarding energy formed an essential reference for the evaluation of the proposed technique. By considering the calculated sensitivities, the conditions of the analysis could be selected and the extent of interferences from other elements, under the selected experimental conditions, could then be predicted. The proposed technique was applied to the analysis of cements, gold alloys, geological material and steels. For cements, the known high yields of the 110- and 197-keV fluorine gamma-rays proved to be useful for the determination of this element.The simultaneous use of prompt X-rays and low energy gamma-rays made it possible to analyse gold alloys and geological ores. Transition metals were determined through the spectrometry of their low energy prompt gamma- rays. The analyses were shown to be rapid, multi-elemental and to have a relative precision of better than 3% for most elements. The sensitivity and accuracy are acceptable for major and minor components and in some instances for trace elements. The staff of the van de Graaff Group of the National Accelerator Centre are thanked for their co-operation and assistance. Financial assistance from the Council for Scientific and Industrial Research is gratefully acknowledged. This work forms part of a Masters thesis to be submitted (by L. G . L.) to the University of Cape Town and is published with permission.286 References 1. 2. 3. 4. 5 , 6. 7. 8. Peisach, M., and Gihwala, D., J. Radioanal. Chem., 1981, 61, 37. Gihwala, D.. and Peisach. M., Anal. Chim. Acra, 1982, 136, 311. Gihwala, D.. and Peisach, M., .Vucl. Instrum. Methods, 1982, 193. 371. Olivier, C.. Gihwala. D.. Peisach, M . , Pineda, D. A , , and Pienaar, H . S., J . Radioanal. Chem.. 1983, 76, 241. Olivier. C.. Peisach, M.. Horland. H. J . . and de Wet. B. S.. J . Radioanal. JVucl. Chem., 1986. 106. 107. Shroy, R. E . , Kraner. H. W., Jones, K. W., Jacobson, J . S., and Heller, L. I., ,Vucl. Instrum. Methods, 1978, 149. 313. Gihwala. D., and Peisach. M.. 5’. Afr. J . Chem.. 1985, 38, 181. Giles. I. S., and Peisach. M., J . Radioanal. Chem., 1976, 32; 105. ANALYST, MARCH 1989, VOL. 114 9. 10. 11. 12. 13. 14. Lackay, L. G . , Thesis, University of Cape Town, in prepara- tion. Peisach, M., J . Radioanal. Chem., 1974, 12, 251. Currie, L. A , , Anal. Chem., 1968, 40. 586. Gihwala, D., and Peisach, M., Radiochem. Radioanal. Letr., 1979, 40, 285. Gihwala, D., and Peisach, M., J . Radioanal. Chem., 1980, 5 5 . 163. Gihwala, D., Doctorate Thesis, University of Cape Town. 1982. Paper 8102843E Received July I4th, I988 Accepted August 26th, 1988

ISSN:0003-2654    

DOI:10.1039/AN9891400279

出版商:RSC    

年代:1989

数据来源: RSC

摘要:

ANALYST, MARCH 1989, VOL. 114 287 Proton-induced Spallation at 600 MeV* Rolf Michelt Zentraleinrichtung fur Strahlenschutz, Universitat Hannover, Hannover, FRG Beate Dittrich, Ulrich Herpers, Frank Peiffert and Thomas Schiffmann Abteilung Nuklearchemie, Universitat zu Koln, Cologne, FRG Peter Cloth, Peter Dragovitsch and Detlef Filges lnstitut fur Reaktorentwicklung, Kernforschungsanlage Julich, Julich, FRG In the course of a systematic investigation of proton-induced reactions up to p-energies of 3000 MeV, the target elements 0, Mg, Al, Si, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Y, Zr, Rh, Ba, Lu and Au were irradiated with 600-MeV protons a t the CERN synchrocyclotron. A consistent set of more than 200 thin-target cross-sections for the production of radionuclides and stable He, Ne and Xe isotopes has so far been measured.Here, 199 cross-sections for the production of radionuclides are presented. On the basis of the new data the quality of existing semi-empirical equations for the calculation of spallation cross-sections is discussed. In a more physical approach, the production of residual nuclides in the course of the intranuclear cascade was calculated using Monte Carlo techniques and compared with experimental cross-sections. Keywords: Proton-induced spallation reactions; integral production cross-sections; semi-empirical equations; intranuclear cascade calculations Integral reaction cross-sections for the production of residual nuclei by medium- and high-energy protons are the basic data for a large number of applications in science and technology.They are needed for the optimisation of isotope production, for the design and operation of high-energy accelerators, for the interpretation of reaction products of cosmic ray particles with terrestrial and extraterrestrial matter and for tests of nuclear reaction models. Our knowledge about medium- and high-energy production cross-sections is still not satisfactory, being incomplete with respect to the coverage of target - element combinations and to particle energies. Data for long-lived radionuclides, stable nuclides and nuclides far from stability are substantially missing. Moreover, the existing data (see references 1-5 for compilations) exhibit a considerable lack of accuracy manifested by discrepancies of up to an order of magnitude. Finally, the theoretical interpretation is mainly restricted to the development of semi-empirical models.In particular, the transition from pre-equilibrium to spallation reactions and special reaction types such as fragmentation and cluster emission are not well understood. Therefore, a project was initiated to investigate proton- induced spallation reactions at energies between 600 and 3000 MeV using the CERN synchrocyclotron and the Saturne synchrotron at LNS/Saclay . Together with previous work from our group up to 200 MeV,6.7 and related studies, these data will allow for a systematic survey of the production of residual nuclides ranging from compound nucleus reactions in statistical equilibrium over pre-equilibrium reactions to high- energy reactions such as spallation and fragmentation. Here, the results of 600-MeV irradiation experiments at CERN are reported. More than 200 thin-target cross-sections were determined for the target elements 0, Mg, Al, Si, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Y, Zr, Rh, Ba, Lu and Au.The residual nuclides investigated cover radionuclides with half-lives of between 15 h and 1.6 X 1 0 h years and also stable He, Ne and Xe isotopes. The techniques applied were gamma-ray spec- trometry and conventional and accelerator mass spec- trometry (AMS). * Presented at the 2nd International Conference on Nuclear and t Present address: Cksellschaft fur Reaktorsichcrheit, Cologne, Radiochemistry, Brighton, UK, 11-15 July, 1988. FRG. The cross-sections measured represent a consistent data set well suited for tests of nuclear reaction models.The new data allow us to evaluate critically earlier measurements and to test the limits of applicability of the semi-empirical equations for the calculation of spallation cross-sections by Rudstams and Silberberg and co-workers.Y,10 In order to allow for a more physical interpretation, an intranuclear cascade approach on the basis of Monte Carlo techniques was used. The calcula- tions were performed by the HETC/KFA-2 code of the HERMES code system.11 The theoretical results for the target element Fe were compared with the experimental data and the extension of the present models with respect to cluster emission are discussed. Experimental Cylindrical stacks of pure element foils and of selected chemical compounds were irradiated with 600-MeV protons at the CERN synchrocyclotron.The pure element foils were supplied by Goodfellow Metals UK, and covered the follow- ing elements, the numbers in parentheses indicating the specified percentage purity: Mg (99.9), A1 (99.9999), Si (99.999), Ti (99.6+), V (99.8), Cr (99.9), Fe (99.99+), Co (99.99+), Ni (99.99+), Cu (99.99+), Y (99.9), Zr (99.8+), Rh (99.9) and Au (99.9). For the investigation of Mn foils, an Mn - Ni alloy (88 + 12) was used. For the irradiation of Lu foils, an Lu - A1 alloy, supplied by Reactor Experiments, USA, with a Lu content of 5.5% and a specified purity of 99.8% and also Lu203 powder (99.9%) were used; the latter was preferred because of its higher Lu content. Barium was irradiated in the form of two types of Ba glass, supplied by Schott, Mainz, FRG, having Ba contents of 40.13% and 44.2%, respectively.In order to investigate the target element oxygen, high-purity quartz discs of Suprasil quality were used, supplied by Heraeus, Hanau, FRG. The foils were between 0.05 and 0.2 mm thick, except for Ba glass (3.5 mm), quartz (1 mm) and Si (1 mm). For irradiation, three foils of each target element were placed together in order to minimise recoil contamination and loss of the foil in the middle. Between different target elements catcher foils of 0.05-mm A1 (99.999%) or 0.05-mm Au (99.9%) were inserted, to avoid both recoil and cross- contamination. The total stacks were enclosed in A1 stack containers having a diameter of 10 mm. Two irradiations were288 ANALYST, MARCH 1989, VOL.114 performed in air using a defocused beam with diameters of 2 and 4 cm and a beam current of 2.5 x 10-6 A for 8.16 and 12.69 h, respectively. The constancy of the beam current was checked by the continuous observation of a Health Physics monitor located in the accelerator hall and by intermittent Faraday cup measurements. After irradiation the stacks were transported to Cologne and the foils were immediately unpacked and measured repeatedly by gamma-ray spectrometry over several months starting 24 h after irradiation. After sufficient cooling they were further processed chemically for determination of long-lived radionuclides by accelerator mass spectrometry or analysed by conventional mass spectrometry for stable rare gas isotopes. The absolute calibration of the gamma-ray spectrometers was carried out using calibrated radionuclide sources having an accuracy of better than 2%.The nuclear data used for calculating the cross-sections were the same as in our previous work.h.7 Half-lives were taken from the “Chart of the Nuclides”12 and gamma-ray energies and absolute intensities from the gamma-ray compilation of Erdtmann and Soyka.13 Further details of the gamma-ray spectrometric measure- ments, a description of the chemical separatlon schemes used for the determination of 1oBe and 2”Al and details of the AMS procedures can be found elsewhere. 1416 Flux measurement was performed by investigating the production of 22Na from aluminium. A cross-section of 16.0 mb was assumed for the reaction 27Al(p,3p3n)*ZNa at 600 MeV, according to the recommendation (“valeurs adoptkes”) of Tobailem and de Lassus St.Genies.17 This flux monitoring by 22Na from A1 is consistent with that by other monitor reactions such as 27Al(p,3~n)2~”a or 27Al(p. lop1 ln)7Be when using the “valeurs adoptees” from the compilations and evaluations of Tobailem and de Lassus St. Genies.17 This is demonstrated by comparing the cross-section ratios calculated from the latter work with those obtained in this work. The ratios of the “valeurs adoptees” are 1.45 for 22Na - ”Na, 0.45 for 7Be - ’“a, and 3.22 for 22Na - 7Be. They compare excellently with the values of 1.42,0.43 and 3.30, respectively, obtained in this work. The proton fluxes in the individual target foils were determined by measuring the production of ”Na in the next lying A1 foils.The proton fluxes derived by this method were about 1 X 1012 and 0.15 x 1012 cm-2 s-1 for the first and second irradiation experiments, respectively. There was no evidence for contributions of secondary particles, in particular of secondary neutrons, which was checked by looking for 59Fe from Co and for fWo from Co in the foils of the stacks. Moreover, Al, Fe and Co monitor foils were distributed around the stacks in the accelerator hall in order to check for a possible background of secondary particles. The measurements of these foils only resulted in upper limits for the neutron and proton background in the hall, which were more than three orders of magnitude lower than the actual primary proton fluxes. 14 Results More than 200 individual cross-sections for the production of radionuclides and stable rare gas isotopes have been measured so far.The measurements and evaluations are not yet finalised for the production of Kr isotopes from Y and Zr and for some of the long-lived nuclides to be measured by AMS. In this paper we present a total of 199 cross-sections for the production of radionuclides; those for 3He, 4He, 20Ne, 21Ne and 22Ne from Al, Mg and Si will be published soon@ and those for the production of stable Xe isotopes from Ba will be presented in the context of model calculations for the production of Xe by solar and galactic protons in extraterres- trial matter.19 The experimental data are presented in Tables 1-6. The reaction types investigated cover (p,xn) to (p140pxn), the Table 1.Cross-sections (mb) for the production of radionuclides from oxygen, aluminium, magnesium and silicon by 600-MeV protons. The eirors given here only consider the individual errors of the cross- section measurements (see text). If absolute errors are required 6% has to be quadratically added Target element Product Oxygen Magnesium Aluminium Silicon 7Be 11.3 f 0.5 6.43 k 0.28 4.88 k 0.10 5.39 k 0.22 “’Be 1.52 f 0.12* 1.03 k 0.07“ 1.03 f 0.07* 0.63 k 0.20 ”Na - 31.5 f 1.3 16.0t 19.6 k 0.6 - 7.94 k 0.38 11.3 k 0.3 5.15 i 0.19 13.0 f 0.8’ 2hA1 * Published already by Theis.lh t Monitor cross-section. - - - Table 2. Cross-sections (mb) for the production of radionuclides from titanium, vanadium, manganese and chromium by 600-MeV protons. The errors given here only consider the individual errors of the cross-section measurements (see text).If absolute error3 are required 6% has to be quadratically added Target element Product 7B e loBe 22Na 24Na 42K 47Ca 46SC 47% 4XSc 44Ti 4 w 48Cr ”Cr sZMn 54Mn 43 K 44Scni Titanium 1.95 k 0.09 0.85 k 0.05 0.90 f 0.04 1.37 ? 0.09 9.01 k 0.48 3.68 t 0.24 0.26 t 0.01 7.21 & 0.29 32.0 k 1.2 28.4 k 1.1 2.97 t 0.14 1.33 k 0.10 1.90 k 0.09 - Vanadium 1.72 k 0.08 0.56 i 0.03 1.10 f 0.06 2.46 t 0.15 0.36 t 0.04 8.81 k 0.36 23.3 f 0.9 15.4 k 0.9 4.88 f 0.21 13.4 IfI 0.3 0.047 k 0.002 2.58 k 0.11 - __ Chromium - 2.48 k 0.20 - 6.18 f 0.59 0.22 i 0.03 48.1 k 2.8 20.0k 1.1 3.60 k 0.22 77.4 k 4.3 2.11 & 0.13 180.0 k 9.9 1.38 2 0.12 61 .0 f 0.23 - - Manganese 1.68 & 0.12 3.82 k 0.21) - - 14.2 k 0.5 - - 15.7 k 0.8 0.10 k 0.08 41.7 k 1.6 6.21 k 0.25 Table 3. Cross-sections (mb) for the production of radionuclides from iron, cobalt, nickel and copper by 600-MeV protons. The errors given here only consider the individual errors of the cross-section measure- ments (see text). If absolute errors are required 6% has to be quadratically added Target element Iron 2.01 k 0.09 0.58 k 0.02* 0.40 ? 0.03 0.49 f 0.02* 0.95 t 0.05 8.4 iz 0.3 9.44 k 0.37 2.89 k 0.11 22.7 k 0.8 0.62 k 0.1 46.3 k 1.7 11.5 IfI 0.6 39.8 & 1.5 - - 1.36 t 0.06 0.23 t 0.01 0.036 2 0.007 Cobalt 1.41 f 0.07 0.18 t 0.02 1.06 k 0.05 6.87 f 0.27 8.86 f 0.34 3.34 t 0.13 0.93 f 0.07 16.1 k 0.9 0.37 k 0.02 37.9 k 0.2 11.2 k 0.5 30.8 f 1.2 0.87 t 0.11 8.97 k 0.35 28.3 k l . i 59.8 k 2.4 - - - - 0.22 f 0.02 * Published already by Theis.Ih Nickel Copper 2.63 k 0.12 0.46 k 0.02* - 0.39 f 0.03 0.36 f 0.02* - 1.61 f 0.08 0.136 f 0.017 - 0.507 i 0.027 - 3.85 i 0.20 - 2.33 k 0.09 5.03 k 0.19 5.5 f 0.2 - 22.7 t 0.9 1.86 k 0.10 42.8 f 1.6 15.8 k 0.6 16.4 k 0.7 0.30 f 0.02 38.4 f 1.5 75.6 f 2.8 23.4 k 0.9 2.31 t 0.22 2.25 k 0.19 23.7 f 0.9 - - 1 1 .1 kO.4 27.7 f 1.0 9.35 t 0.35 23.1 f 0.9 1.7 f 0.2 10.3 2 0.4 27.7 k 1.0 34.4 k 1.4 11.9 k 0.5 1.00 k 0.04 - - -ANALYST, MARCH 1989, VOL. 114 289 Table 4. Cross-sections (mb) for the production of radionuclides from yttrium and zirconium by 600-MeV protons. The, errors given here only consider the individual errors of the cross-section measurements (see text). If absolute errors are required 6% has to be quadratically added Production from yttrium Production from zirconium 1.95 i 0.20 8.2 f 0.5 12.2 k 0.7 11 .5 k 0.6 7.58 k 0.36 34.9 t 1.3 67.5 f 3.5 16.3 f 0.7 57.0 k 3.7 60.9 t 2.3 88.3 k 3.4 5.65 t 0.41 3.23 k 0.18 5.22 f 0.57 7.51 k 0.35 5.15 t 0.22 28.0k 1.1 55.2 t 2.9 9.5 t 0.4 102.0 t 9.7 66.9 f 2.5 51.1 k 2.7 19.4 k 1.5 44.8 k 2.4 58.6 k 2.2 2.42 t 0.15 1.00 t 0.01 Table 5.Cross-sections (mb) for the production of radionuclides from rhodium by 600-MeV protons. The errors given here only consider the individual errors of the cross-section measurements (see text). If absolute errors are required 6% has to be quadratically added Reaction Io3Rh( p. 14p2 1 n)hYGe Io7Rh(p, 12p 17n)7'Se "liRh(p ,9p 1 1 n)84Rb IoiRh (p ,7p 1 OnyY loiRh(p ,6plOn)"Zr IoiRh( p,6p3n)Y5Zr Io3Rh( p ,5p7n)"Nbm lo3Rh( p ,3p6n)Y5Tcrn IoiRh(p.2p5n)Y7Ru ""Rh(p.3n) 1olRh 1c'7Rh(p.2n)l(12Rh Cross- section Reaction 9.46 k 1.02 I()?Rh(p,13p17n)74As 6.69 k 0.25 IO?Rh(p,9pl2n)"'Rb 2.1 f 0.1 1*3Rh(p,8plln)85Sr 41.3 k 1.6 1[)3Rh(p,7p6n)Y]Y 38.2 k 1.5 10'Rh(p,6p9n)*YZr 0.20 k 0.02 l[)3Rh(p,Sp9n)Y11Nb 2.19 k 0.09 l('?Rh(p,Sp4n)Y5Nb 3.75 -t 0.15 lo3Rh(p.3pSn)"Tc 36.2 t 1.4 lo3Rh(p,Sn)YYRh 11.2 k 0.4 103Rh(p.3n)lo1Rhm 39.5 i 2.1 lo7Rh(p,2n)1('2Rhm Cross- section 0.94 k 0.07 23.4 k 1.2 51.5 k 3.2 20.8 k 1.7 41.8 k 1.6 38.0 * 3.5 1.18 i 0.06 23.2 k 1.2 7.86 k 0.31 38.6 k 2.12 18.2 t 0.7 Table 6.Cross-sections (nib) for the production of radionuclides from barium. lutetium and gold by 600-MeV protons. The errors given here only consider the individual errors of the cross-section measurements (see text).If absolute errors are required 6% has to be quadratically Production from Lu Production from Au added Production from Ba 2.56 t 0.12 11.2 k 0.4 11 .o k 2.5 46.2 k 2.5 18.0 t 0.8 13.7 k 0.5 37.8 k 1.8 30.1 i 1.2 28.4 t 4.2 61.2 k 2.3 52.3 t 3.8 43.9 k 7.0 25.0 f 2.9 87.1 k 4.1 30.4 k 3.9 8.26 k 0.46 0.61 i 0.05 1.36 5 0.11 0.89 k 0.04 20.7 k 0.9 27.3 k 1.2 65.6 t 2.4 24.2 k 1.0 42.6 k 2.5 60.3 k 3.0 69.4 f 3.1 77.0 t 3.2 differences in mass numbers between targets (AT) and products (Ap) ranging from 0 to 109. The new data provide a comprehensive survey of integral reaction data for light and medium target elements (AT < 103) in the "spallation" region, i. e., A-42 d AP d A I , and on fragmentation products such as 7Be and loBe.For heavy target elements most of the observed products are still target-like, except for Au, where even products of sequential fission are observed. Generally, the observed relative mass losses, ( A , - Ap)/AT, go up to 0.93. Here it should be pointed out that the regions of different reaction regimes cannot be distinguished unambiguously and that the transition from reactions with target-like products to spallation and further to fragmentation is smooth. This is one of the major problems in the theoretical interpretation in both the semi-empirical and the intranuclear cascade approaches (see below). With respect to the errors affecting the determination of production cross-sections, various sources have to be distin- guished. Some of them are general ones that affect all data in our experiments in the same way, and others that individually affect each reaction.In view of the severe problems of accuracy exhibited by a survey on earlier data, here all sources of error will be discussed in some detail. The general sources of error are as follows: 1. Uncertainties in the determination of the number of target atoms in the beam and their homogeneity over the target area. In this experiment with measurements of the target areas of the circular foils and comparisons with the actual weights of the target this error was estimated to be 2%. 2. Impurities in the target materials causing interfering nuclear reactions. Owing to the high purity of the target materials used, this influence is negligible. It has to be considered, however, for the determination of 7Be and '*Be from oxygen in the quartz targets.The errors resulting from the correction for the production of these two nuclides from Si are included in the errors quoted in Table 1. The same is true for the cross-sections for production of radionuclides from Mn, where the contribution of the second alloying component Ni had to be subtracted. Also this leads to an enhancement of the errors, which is accounted for in the errors given in Table 2. In those instances where the errors became unacceptably high no cross-sections for the respective products are re- ported. 3. Fluctuations in the beam intensity during irradiation. The constancy of the beam was monitored continuously and no indication of beam intensity fluctuations was observed.As the shortest half-life measured is 14.96 h (2jNa) and the longest irradiation lasted for 12.69 h, this influence was considered to be negligible. 4. Interference by reactions of secondary particles. As pointed out above, the secondary particle fluxes were more than three orders of magnitude lower than the primary fluxes, hence making this source of error negligible. 5. Recoil loss and recoil contamination. By analysing the foils in the middle of each "elementary unit" of three identical foils, these errors cancel out. 6. Error in the monitor cross-section. This error was not considered here and was taken to be zero. The considerations above point to the consistency at least for the three often used monitor reactions from A1 being better than 3% when using the "valeurs adoptkes" given by Tobailem and de Lassus St.Genies. 17 7. Uncertainty in the measurement of the monitoring radionuclide, here 22Na. The relative precision of this determi- nation was checked by multiple measurements and can be regarded as being better than 1%. 8. Error in the absolute calibration of gamma-ray spectro- meters. From the comparison of different calibrated radio- nuclide sets the error of absolute calibration can be taken to be 5%. It is larger than the error of relative efficiency calibration (see below). In the AMS measurements this source of error must not be considered. The errors especially affecting the AMS measurements were discussed in detail by Theis'" and are included in the errors given for the production of "'Be and 26A1.Unless stated otherwise, these general errors are not considered in the data given in Tables 1-6. They have to be quadratically added according to the laws of error propagation if absolute errors are required. The errors given here only involve the individual errors listed below: 1. Reproducibility of photopeak integration. Up to 15 measurements were used for the determination of each nuclide, hence allowing this source of error to be properly accounted for.290 2. Statistical counting error. This source of uncertainty is included by calculating the errors of individual photopeaks. The error in the background subtraction was taken into account. 3. Dead time and pile-up losses in gamma-ray spec- trometry. By varying the distance between the sources and the detector, the dead time was always kept below 10%, where the automatic dead-time correction of the measuring devices can be assumed to be correct.4. Self-absorption of gamma-rays in the sample. As only gamma energies in excess of 120 keV were used for evaluation and as the individual target thicknesses (except for Ba) were below 200 mg cm-2, this source of error was neglected. However, also for the Ba glass it was calculated to be negligible for the products determined.15 5 . Relative efficiency calibration of gamma-ray spec- trometers. From the quality of the radionuclide sources used this error is taken to be 2%. 6. Nuclear decay data. The error in the nuclear half-lives was not considered. or those for absolute gamma-ray intensi- ties. However, it should be pointed out that the errors given in Tables 1-6 also account for inconsistencies between different gamma lines.A detailed listing of the gamma-ray energies actually used can be found elsewhere.14.15 Owing to the large amount of new data presented here it is not possible to discuss for all the reactions the agreement with the partially existing earlier determinations. This has been done in detail elsewhere. I+lh From these comparisons one can conclude that for those reactions which were frequently measured in the past, e.g., spallation reactions from Fe, experimental data varying by up to an order of magnitude can be found at 600 MeV, but also for other energies.6.7 This is certainly insufficient for testing the quality of models of spallation reactions and also is unacceptable for most applica- tions.Even worse, for a large number of reactions dealt with in this study no or only a few data are available. This study has provided for the first time a consistent set of data covering wide ranges of target and product masses, which together with our earlier work on lower p-energiesh.7 and with the present experiments performed at Saclay will allow for comprehensive and consistent tests of nuclear reaction theories in the transition from pre-equilibrium to spallation nuclear reac- tions. On the basis of the currently available data this will now be done for the classical semi-empirical equations and for Monte Carlo calculations of the intranuclear cascade. ANALYST, MARCH 1989, VOL. 114 which is determined by equal evaporation probabilities of protons and neutrons. The proposed distribution functions on an isobar were Gaussian or nearly Gaussian.From the various parameters describing the exponential decrease in isobaric yields and the width and shape of the charge dispersion distribution, the most probable atomic numbers on the isobars were then taken as free parameters and their values were determined by fitting to experimental data, hence strongly depending on the quality of the experimental determinations. Today, there are several versions of these equations in use. Rudstams gave two semi-empirical equations, called the CDMD (charge dispersion mass dispersion) equation and the CDMD-(; (charge dispersion mass dispersion-Gaussian) equation. According to Rudstam, these equations should be applicable to p- and n-induced reactions on medium-mass targets and should be valid for mass losses around 0.5. He mentioned in particular that these equations are not applic- able to very light products.However, in applying these equations these restrictions were often forgotten and cross- sections were (inadequately) calculated without critically evaluating the limits of applicability. This can lead to extreme problems because for light fragmentation products such as 7Be the Rudstam equations underestimate the experimental data by up to four orders of magnitude in the energy range discussed here. Based on the work of Rudstam,g Silberberg and co-work- ers‘j31() later developed a model that took into account the evaporation of complex particles ( A d 4).This led to an equation that, according to the authors, is valid for target elements with 9 < AT S 209 and for products with 6 d A, S 200. It should be mentioned, however, that this equation also is not applicable to very small mass differences between the target and the product. Recently, a further semi-empirical approach was reported by Summerer21 which, however, will not be dealt with here. Here, rather the equations of Rudstam and by Silberberg and co-workers will be discussed with respect to the new exp er i me n t a1 data. Theoretical cr oss- se c t i o n s we re cal cu 1 ate d using the three equations in the form of the code SPALL by Routti and Sandberg.2’ As mentioned above, Rudstam’s equations must not be applied to very light products, because of the underestimation by several orders of magnitude.This is demonstrated in Fig. 1 for the CDMD equation, where the ratios of calculated to experimental cross-sections are plotted as a function of relative mass loss. However, also near a mass loss of 0.5 deviations of up to an order of magnitude are seen. In Fig. 2 this is shown for the CDMD-G equation, omitting the light products. As can be seen, the calculated data are good within a factor of two for many data. However. discrepancies of up to Discussion According to the “two-step” model proposed by Serber,“) high-energy reactions can be described by a fast intranuclear cascade of nucleon - nucleon interactions followed by a slow evaporation phase during which the highly excited residual nucleus of the first step de-excitates and the final product is formed.This simplified picture is able to describe the gross features of spallation reactions, but is hardly capable of dealing with all the different possible and actually occurring processes. In contrast to heavy ion reactions for nucleon- and light nuclei-induced reactions, few theoretical efforts have been made in the past and a more detailed theoretical interpretation is needed urgently. In particular for the caiculation of integral production cross-sections the theoretical treatment is still today mostly restricted to the use of semi-empirical equations that were first derived by RudstamX in 1966. The basic assumption of this semi-empirical model and its successorsY,10 is the strict applicability of Serber’s model. This leads to an exponential decrease in the isobaric yield with increasing difference between A.r and Ap.It is then assumed that the charge dispersion of the residual nuclides is determined during the evaporation phase. The resulting charge dispersion on an isobar is symmetrical around a most probable atomic number d 2 0.1 I- - C 0 @ 10-3 t 0 . - 0 0.25 0.5 0.75 I hA/A (target) Fig. 1. Ratio of cross-sections calculated by the semi-empirical CDMD equation of RudstamX to experimental values from this workANALYST, MARCH 1989, VOL. 114 29 1 10 I 1 v - m C I 0 0.1 I I I 0 0.25 0.5 0.75 1 bA/A (target) Fig. 2. Ratio of cross-sections calculated by the semi-empirical CDMD-G equation of Rudstam* to experimental values from this work c 7 1 0 0 0 0.1 ‘ 0 I 0 0.25 0.5 0.75 1 GAIA (target) Fig.3. Ratio of cross-sections calculated by the semi-empirical equations of Silberberg and co-workers’-10 to experimental values from this work an order of magnitude may occur. Hence it will be of particular interest to look for the improvements obtainable by using the equation of Silberberg and co-workers which, according to the authors, has a much wider range of applicability . The results with Silberberg and co-workers’ equation do not show such extreme failures as those given by Rudstam’s equations for the light products. There are no deviations by more than a factor of ten. As shown in Fig. 3, however, the ratios of calculated to experimental cross-sections still show a broad scatter which is not systematic in either the relative mass loss or in the mass difference between target and product.A more detailed analysis shows that the mean of the squares of the relative deviation of experimental and calculated cross- sections is 4.4, i.e., the equation is misleading by a factor of 2.1 on average. The distribution around a value of 1 of the ratio of the calculated to experimental data is fairly symmetric and no significant tendency for systematic over- or under-estimation is seen. This is expressed by the mean of the relative deviations, which is only 0.093. It cannot be decided from this analysis whether the large mean deviation is a general limitation of the semi-empirical approach or whether it is an expression of the poor quality of the former experimental database. In this context it will be of interest to see an analogous analysis for higher energies up to 3000 MeV which are currently being investigated.Considering the not yet well defined energy range of the spallation regime637 and the transition from pre-equilibrium to spallation, 600 MeV might be considered still to be a relatively low energy as the excitation functions still show a considerable energy depen- dence for many nuclides and the plateau cross-sections have not yet been reached. This can only be answered in the future. There is, however, another more physical way of calculating the distribution of residual nuclides, namely by following the intranuclear cascade via Monte Carlo techniques. Unfortu- nately, this method is handicapped by extremely long com- puter times when calculating residual nuclides with low production cross-sections with sufficient statistical accuracy.In this work, we report on calculations that were made for the target element Fe with natural isotopic composition for proton-induced reactions with energies between 150 MeV and 5 GeV. Further extensions of these calculations are planned. For these Monte Carlo calculations the high-energy trans- port code HETCKFA-2 originally introduced at ORNL23J4 was used within the newly developed HERMES code system. 11 Several modifications of HETC have been made and documented in detail .25,26 HERMES (High Energy Radiation Monte Carlo Elaborate System) is a system of Monte Carlo computer codes that allow the treatment of the different physical phenomena to be considered in computer simulations of radiation transport and interaction problems.For the calculation of spallation cross-sections for high-energy nucleons the “thin-target” set-up of HETCIKFA-2 was used to elaborate the history of residual nuclides, i. e., particles emitted from the struck nucleus are not allowed to react further with any other nucleus. The HETC code makes use of an intranuclear cascade - evaporation (INCE) model including inelastic collisions of protons, neutrons and charged and neutral pions and also evaporation of p, n, d, t, 3He and 4He, and describes the knock-on phase of the intranuclear cascade and the evaporation in the framework of the statistical theory according to Weisskopf.27 For the calculations the incident particles are selected uniformly over the geometric cross-section of the nucleus. A three-region step function is chosen to approximate the continuous Hofstadter density distribution of the nucleons in the nucleus.The nucleon momenta-in the nucleus are Fermi distributed and are normalised individually in each region. The potential energies are given as the sum of zero tempera- ture energies of the nucleons in each region plus the binding energy of the most loosely bound nucleon. Input data for the intranuclear cascade calculations are elastic and inelastic nucleon - nucleon cross-sections and also cross-sections for pion - nucleon elastic scattering, for charge exchange and for pion production. In all interactions the exclusion principle is taken into account. At low-energy cut-off of the intranuclear cascade, informa- tion about the mass, charge and excitation energy of the residual nucleus is transferred to the evaporation calculations according to the statistical theory of nuclear reactions.27 To this end, complete equilibration of the excitation energy between the emission of the last cascade nucleon and the first evaporation is assumed.Also, a re-equilibration of excitation energy between successive particle evaporations is assumed. An emission in the pre-equilibrium state is not allowed for. The history of the emitted particles (type, energy, momen- tum) and of the residual nuclei (mass, charge, recoil energy) is stored after each calculated reaction. After performing a pre-defined number of reactions, these data can be analysed by the module SIM of HETCKFA-2.This results in isobaric and isotopic yields of residual nuclides and in the yields of pre-defined particular nuclides per incident source particle. By multiplying these yields by the geometric cross-section of the target nuclide, absolute production cross-sections are derived, which are non-cumulative as they are not corrected for (3-decay. Hence the production of short-lived isobars must also be calculated in order to obtain results comparable to the experimental conditions.292 100 11 E , C .s 10 a, v, ANALYST, MARCH 1989, VOL. 114 - - 0.1 ’ L 1 10 100 1000 10 000 EnergylMeV Fig. 4. from natural iron ( X ) compared with HETC calculations (0) Experimental excitation function for the production of ”Cr 10 100 1000 En e rg ylMeV 10 000 Fig. 5 . Experimental excitation function for the production of 48V from natural iron ( X ) compared with HETC calculations (0) The quality of the calculations depends strongly on the number of calculated reactions in the Monte Carlo run.For the calculations in this work 200000 reactions of source particles were followed for each energy point, resulting in sufficient statistics for many nuclides. Some problems still exist, however, for extremely low cross-sections. Moreover, in the present calculations, owing to a lack of storage space, not all products from Fe could be calculated, so that only for a relatively small number of products can a comparison between theory and experiment be made. Further limitations in our calculations are due to the models used in HETC. It does not account for the emission of complex particles in the intra- nuclear cascade, it does not include pre-equilibrium decay and is not yet able to describe fragmentation.For a comparison of the HETC calculations with the experimental data, as examples in Figs. 4 and 5 the excitation functions for the production of W r and 4 W from iron are presented. Here for experimental cross-sections below 1 GeV only data from our work are shown; above 1 GeV exclusively the data of Raisbeck and Yiou2* were taken. This was done because for both reactions all existing data show severe scatter, on the basis of which only a general agreement of the calculated excitation functions with experiments can be stated. Comparing them, however, with a consistent data set allows a more detailed judgement.For the reaction Fe(p,3pxn)Wr the shape of the excitation function is fairly well described, but the theory underestimates the experimen- tal data by up to a factor of two. For the reaction Fe(p,4pxn)48V the picture looks different. Whereas the Table 7. Comparison of experimental 600-MeV cross-sections (mb) with theoretical values based on HETC calculations Cross-section Ratio, HETCi Reaction Experimental HETC experimental Fe(p,xn)Wo . . Fe(p,2pxn)QMn Fe( p ,2pxn)”Mn Fe( p ,3pxn)48Cr Fe(p,3pxn)”Cr Fe(p,4pxn)48V . . Fe(p, 14pxn)*6AI Fe(p, 16pxn)ZzNa , . 1.36k 0.06 . . 11.5 5 0 . 6 . . 39.8 -t 1.5 . . 0.62kO.1 . . 46.3 & 1.7 . . 22.7F0.8 . . 0.49 -t 0.02 . . 0.4Ok0.03 2.55 29.4 28.2 1.2 29.8 28.0 1 .58 0.34 1.88 2.56 0.71 1.95 0.64 1.23 3.22 0.85 experimental data above 600 MeV are well reproduced, the calculated cross-sections tend to overestimate the low-energy part of the excitation function.It is a general observation that discrepancies between theory and experiment increase in the low-energy region where pre-equilibrium emission and evap- oration of nucleons and light complex particles become important. The inclusion of pre-equilibrium models in the HETC calculation methods would certainly improve this situation. A severe problem occurs when calculating the production of fragmentation products such as 7Be. In these instances the theory fails and underestimates the experimental data by several orders of magnitude. However, this is reasonable, because the nuclear models used do not allow for emission of light particles with A > 4.Hence emission of 7Be as fragments is impossible. The production cross-sections calculated here have to be understood as being for that part of the production which is as residual nuclides rather than as ejectiles. The experimental data, in particular also for the production of 7Be from Lu, indicate the importance of these fragmentation processes. They will have to be considered in further extensions of HETC. In general, the energy dependence of the deviations between the HETC results and the experimental data allows one to evaluate the energy regions of applicability of the intranuclear cascade - evaporation model. For 600-MeV pro- tons the HETC calculations agree with the experimental data within a factor of three (Table 7). However, this also holds for products fairly near the target and a complete misinterpreta- tion is only seen for the fragmentation products.However, as the discrepancies between theory and experiment show systematic trends with energy and as they concentrate on particular reaction types, it is to be hoped that the necessary improvements in the HETC code can be made. Our new and consistent set of experimental cross-sections will provide a benchmark for testing such improvements. The authors are grateful to the CERN PSCC for making available the irradiations and to Drs. B. Allardyce and J. W. M. Tuyn and to the SC and Health Physics staff for their kind cooperation and for the essential support they provided before and during the irradiations. Funding was received from the Deutsche Forschungsgemeinschaft , Bonn, FRG.This support is gratefully acknowledged. References 1. 2. McGowan, F. K., and Milner, W. T., At. Data Ni-lcl. Data Tables, 1976, 18, 1. Burrows, T. W., and Dempsey, P . , “The Bibliography of Integral Charged Particle Nuclear Data,” Archival Edition, Parts 1 and 2, BNL-NCS-50640, Fourth Edition, National Technical Information Service, US Department of Commerce, Springfield, VA, 1980. Burrows, T. W., and Wyant, G., “The Bibliography of Integral Charged Particle Nuclear Data,” BNL-NCS-50640, Fourth Edition, Suppl. 1, National Technical Information Service, US Department of Commerce, Springfield, VA, 1981. 3.ANALYST, MARCH 1989, VOL. 114 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. Holden, N. E . , and Burrows, T.W., “The Bibliography of Integral Charged Particle Nuclear Data,” BNL-NCS-50640, Fourth Edition, Suppl. 2, National Technical Information Service, US Department of Commerce, Springfield, VA, 1982. Holden, N. E . , Ramavataram, S . , and Dunford, C. I>., “Integral Charged Particle Nuclear Data Bibliography,” BNL- NCS-5 1771, National Technical Information Service, US Department of Commerce, Springfield, VA, 1985. Michel, R . , and Stuck, R . , J . Geophys. Res., 1984,89B, Suppl., 673. Michel, R., Peiffer, F., and Stuck, R . , Nucl. Phys., 1985, A441, 617. Rudstam, G . , 2. Naturforsch., TeilA, 1966, 21, 1027. Silberberg, R . , and Tsao, C. H . , Astrophys. J., Suppl., 1973, 25, 315 and 335. Silberberg, R . , Tsao, C. H . , and Shapiro, M. M., in Shen. B. S.P., and Merker, M., Editors, “Spallation Nuclear Reactions and Their Applications,” Reidel, Dordrecht, 1976, Cloth, P . , Filges, D., Neef, R. D . , Sterzenbach, G . , Reul, C . , Armstrong, T . W., and Colborn, B. L., “HERMES-High Energy Radiation Monte Carlo Elaborate System,” Juel-2203, Kernforschungsanlage Julich, Julich, FRG, 1988. Seelmann-Eggebert, W., Pfennig, G . , Munzel, H., and e v e - Nebenius, H . , “Chart of the Nuclides,” Fifth Edition, Komunalschriften-Verlag Jehle, Munich, 1981. Erdtmann, G., and Soyka, W., “The Gamma Rays of the Radionuclides,” Verlag Chemie, Weinheim, 1979. Dragovitsch, P . , Thesis, Universitat zu K d n , 1987, Peiffer, F., Thesis, Universitat zu Kiiln, 1986. Theis, S . , Thesis, Universitat zu Koln, 1986. Tobailem, J . , and de Lassus St. Genies, C. H . , Additif No. 2 5 la CEA-N-1466(1), 1975, CEA-N-1466(4), 1977, and CEA-N- 1466(5), 1981, Service de Documentation, CEN Saclay, Gif- sur-Yvette, France. p. 49. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 293 Michel, R . , Peiffer, F.. Theis, S., Begemann. F . , Weber, H., Signer, P . , Wieler, R., Cloth, P.. Dragovitsch, P., Filges, D., and Englert, P., Nucl. Instrum. Methods Phys. Res. B , submitted for publication, 1988. Prescher, K., Peiffer, F., Stuck, R . , Michel, R . , Rao, M. N., and Methew, J . , in preparation. Serber, R . , Phys. Rev., 1947, 72, 1114. Summerer, K., GS1-Nachrichten, 1987, 7, 9. Routti, J . T . , and Sandberg, J . V., Comput. Phys. Commun., 1980, 21, 119. Armstrong, T. W., and Chandler, K. C., Nucl. Sci. Eng., 1972, 49, 110. Chandler, K. C., and Armstrong, T. W., “HETC, Monte Carlo High Energy Nucleon Meson Transport Code,” ORNL-4744, National Technical Information Service, W S Department of Commerce, Springfield, VA, 1972. Cloth, P., Filges, D., Sterzenbach, G . , Armstrong, T. W., and Colburn, B. L., “The KFA Version of the High Energy Transport Code HETC and the Generalized Evaluation Code SIMPEL,” Juel-Spez-196, Kernforschungsanlage Julich, Julich, FRG, 1983. Armstrong, T. W., Cloth, P . , Colborn, B. L., Filges, D., and Sterzenbach, G., in Schriber, S. O., Editor, “Proceedings of the International Collaboration on Advanced Neutron Sources, Chalk River, 1983, ICANS VII,” AECL-8488, Chalk River Nuclear Laboratory Report, Chalk River, Canada, 1984, p. 205. Weisskopf, V., Phys. Rev., 1940, 57, 295. Raisbeck, G. M., and Yiou, F., in “Proceedings o f the 14th International Cosmic Ray Conference, Munich, 1975 ,” Max Planck Institut fur Extraterrestrische Physik, Munich. 1975, Volume 2, p. 495. Paper 81028396 Received July I4th, 1988 Accepted September I9th, 1988

ISSN:0003-2654    

DOI:10.1039/AN9891400287

出版商:RSC    

年代:1989

数据来源: RSC

摘要:

ANALYST, MARCH 1989, VOL. 114 295 Production of Cosmogenic Nuclides in Meteoroids: Simulation Experiments and Modelling* Beate Dittrich, Ulrich Herpers and Thomas Schiffmann Abteilung Nuklearchemie, Universitat zu Koln, Cologne, FRG Rolf Michelt Zentraleinrichtung fur Strahlenschutz, Universitat Hannover, Hannover, FRG Peter Cloth, Peter Dragovitsch and Detlef Filges lnstitut fur Reaktorentwicklung, Kernforschungsanlage Julich, Julich, FRG Jorg Beer and Willy Wolfli lnstitut fur Mittelenergieph ysik der ETH Zurich, Zurich, Switzerland The production of cosmogenic nuclides in meteoroids by galactic protons was simulated in a series of thick-target experiments a t the 600-MeV proton beam of the CERN synchrocyclotron. Using a special technique, isotropic irradiation of three artificial meteoroids made out of diorite and gabbro with radii of 5,15 and 25 cm was performed.Results obtained for the production of 2*Na, 44Ti and 1OBe from Ti and Fe are reported. The experimental data were compared with model calculations based on Monte Carlo calculations of the high-energy transport of primary and secondary particles and on experimental and theoretical thin-target excitation functions. A method for translating the results obtained for the monoenergetic 600-MeV irradiations to real irradiation in space by a complex p-spectrum is presented. Keywords; Meteorites; galactic cosmic ray protons; thick-target simulation experiments; production rates; Monte Carlo modelling Cosmogenic nuclides are produced by the interaction of solar and galactic cosmic ray (CR) particles with matter.The abundances of these cosmogenic nuclides in extraterrestrial materials provide information about the irradiation history of cosmic dust, meteoroids, planetary and asteroidal surfaces and the history of the cosmic radiation itself. For reviews on these topics, see references 1-3. Accurate modelling of the depth- and size-dependent production rates of cosmogenic nuclides in extraterrestrial bodies provides the basis for the successful decipherment of this information. The present status of models for the description of the interactions of solar and galactic cosmic ray particles with matter can be characterised as follows: one is readily capable of calculating with adequate accuracy the depth- and size- dependent production of cosmogenic nuclides in extraterres- trial matter only for solar protons,d,j a-particles6 and primary galactic particles.7.8 However.an adequate description of the interaction of galactic cosmic ray (GCR) particles, taking properly into account the contribution of secondary nucleons to the production of cosmogenic nuclides, is still lacking for meteoroids and to a certain extent also for lunar surface materials. The situation is complicated by the fact that there exist two different approaches for describing the production of cos- mogenic nuclides by galactic cosmic ray particles: the results of which are often contradictory. The first is a "thin-target approach. '' Here the depth- and size-dependent production rates of a cosmogenic nuclide are calculated from depth- and size-dependent spectra of primary and secondary particles and from thin-target cross-sections of the underlying nuclear reactions.The fluxes of contributing particles are derived from experimental data on high energy interactions, from calculations of the intra- and inter-nuclear cascade and the related transport phenomena or from estimates of the multiplicities and interaction lengths of the various particle types. The second method is a "thick-target approach." Here the production depth profiles are derived from thick-target Presented at the 2nd Internationdl Conference on Nuclear and Radiochemistrq. Brighton. LK. 11-15 July, 1988 - To \\horn correspondence should be addressed irradiation experiments simulating the cosmic irradiation as closely as possible.As discussed in detail previously,9 neither approach was satisfactory, owing to a lack of a physical foundation for most thin-target models and to systematic experimental problems for nearly all earlier thick-target experiments. In order to improve this situation, a collaborative study was initiated in 1983 to examine as many aspects as possible of the production of cosmogenic nuclides in small meteoroids ( R 6 25 cm). In the course of experiment CERN SC 96, a series of thick-target experiments were performed at the CERN synchrocyclotron in order to simulate the production of cosmogenic nuclides in meteoroids by galactic protons. A proton energy of 600 MeV was used, which represents the energy of the maximum of the galactic p-spectrum. Using a special technique, isotropic irradiation of three artificial meteoroids made out of diorite and gabbro with radii of 5, 15 and 25 cm was carried out.thus avoiding the problems that affected earlier thick-target experiments. Individual pure element foils, degassed meteoritic material and special chemical compounds were filled into bores of the artificial meteoroids in order to determine the elemental production rates of all relevant cosmogenic nuclides for all target elements to be considered in extraterrestrial matter. To unify this thick-target approach with a thin-target approach and to demonstrate the equivalent of both types of modelling, the experimental data were compared with thin-target calculations based on Monte Carlo calculations of the high-energy transport of primary and secondary particles and on experimental and theoretical thin-target excitation functions.These model calculations allow the various production modes of cosmogenic nuclides to be distinguished. Up to now, depth profiles for more than 100 product nuclides have been measured covering short- and long-lived radionuclides and stable rare gas isotopes. A detailed descrip- tion of the first experiment with an artificial meteoroid with R = 5 cm has been given previously9~~(~ and later a comprehen- sive survey of all these experiments was prepared. covering all experimental data obtained up to the end of 1986, with the corresponding theoretical depth profiles." It also contains a bibliography of all the results published up to then. A detailed296 description of the second and third irradiation experiments is currently being prepared.12 However, there are still some data missing. e.g..for stable Ar. Kr and Xe isotopes and some long-lived radionuclides. Here, we report for the first time the results obtained for medium- and long-lived radionuclides such as 22Na, JJTi and "'Be from Ti and Fe target elements in artificial meteoroids with radii between 5 and 25 cm. Gamma- and X-ray spectrometry and accelerator mass spectrometry were used for the measurements. The data are discussed in the context of thin-target calculations as mentioned above and a method for translating the results obtained for the monoenergetic 600- MeV p-irradiations to the real irradiation conditions in space by a complex p-spectrum is presented.The limitations of the present model and further data requirements are discussed. ANALYST, MARCH 1989, VOL. 114 line of "Ti was disregarded because of spectroscopic interfer- ence with the 77.1-keV gamma-ray line of 21qPo from the natural background. The half-lives used were taken from the "Chart of the Nuclides"l4 and the gamma-ray energies and intensities from the compilation of Erdtmann and Soyka.15 Absolutely calibrated Ge(Li) and thin intrinsic Ge detectors were applied to gamma- and X-ray spectrometry, respec- tively. For 44Ti the measurements had to be corrected for self-absorption of the low-energy photons. However, these corrections were only about 5%. For the determination of ")Be, chemical separation from Ti and Fe foils was performed following procedures described previously in detai1.16,17 The particular advantage of these schemes is that they allow one to separate '"Be, 26AI.' T I , 41Ca and. for Fe, j3Mn for subsequent determination by AMS or NAA from only one target foil. The l(JBe/YBe ratios were measured with the 6-MV tandem van de Graaff accelerator at ETH Zurich. See reference 18 for a description of AMS techniques and further experimental details. From the measured activities and stable nuclide abun- dances, production rates were calculated using the above p-doses. Throughout this work the production rates (PRs) are given in units of s-1 g-1, normalised to a primary proton flux of 1 cm-2 s-1. Thin-target modelling of the production rates of all reaction products measured in the meteoroid models was carried out using Monte Carlo calculations of depth- and size-dependent spectra of primary and secondary nucleons with the high- energy transport (HET) code HETCiKFA 2.19 Using basic nuclear data the transport of primary particles and the production and transport of secondary particles (protons, neutrons. pions) were calculated.For secondary neutrons reaching energies below 15 MeV the KFA version of multi-group Monte Carlo transport code MORSE-CG was used. To this end the particles were analysed inside shells with a thickness of 1.0 cm. For all shells, the spectra of primary protons and secondary protons and neutrons were evaluated. As the space grid of 1 cm caused relatively large uncertainties in the centre of the larger meteoroid models. because of the small fractions of these volumes relative to the total volume.for the 15- and 25-cm spheres the centre spectra for r = 12-5 cm and for Y = 0-6 cm, respectively, were combined. Thus by giving away some space resolution, good statistical precision was obtained for all locations. For protons, 22 energy groups between 600 and 10 MeV were chosen. For neutrons, 113 groups were used. Thirteen of these groups were between 600 and 15 MeV. Below 15 MeV the MORSE-CG neutron transport analysis2(' with the EPR coupled neutron - gamma library.21 with 100 neutron groups (10-4 eV-14 MeV) and 21 gamma groups (10-2-14.0 MeV) was used, as derived from ENDF/B-IV.22 A detailed discus- sion of the fluxes and spectra obtained by these calculations can be found elsewhere.11-13 In principle, it is possible to calculate the abundance of residual nuclides by Monte Carlo techniques directly.In practice, however, this has not been feasible so far because the calculation of reliable data for reaction products with low yields requires unacceptably long computing times. There- fore, in general, an evaluated set of thin-target excitation functions for p-induced reactions (see references 9 and 13 for details) and a set of calculated functions for n-induced reactions23 was used to calculate production rates from the primary and secondary nucleon spectra. Thus. for "'Ti exclusively. the experimental thin-target cross-sections measured by our group3,*4 were used for p-induced reactions, and for n-induced reactions excitation functions were calcu- lated applying the code ALICE LIVERMORE 8225 and subsequent versions.26 For the production of 22Na and 1"Be from Ti and Fe exclusively, experimental cross-sections from various sources1~.16.24.27-~~ were used for p-induced reactions.Experimental Procedures and Calculation Methods Three spherical meteoroid models with radii of 5, 15 and 25 cm were made of diorite and gabbro. These materials are well suited as models for stony meteoroids because of their low water contents of less than 10-3 g g-1 and high densities of 3.0 g cm-3. However. the chemical compositions of diorite and gabbro" show some distinct differences from that of chon- dritic material. A higher Fe content in chondritic materials results in mean Z and A values of 11.3 and 23.1, respectively, for L-chondrites and of 11.9 and 24.4 for H-chondrites, whereas diorite and gabbro both have a mean Z of 10.4 and a mean A of 21.1.These differences must be accounted for when translating the results to cosmic irradiation conditions. The stony matrix of the meteoroid models itself was not intended as a target material for the investigation of produc- tion rates; it was meant only to serve for the development of the secondary cascade. The targets to be studied, which covered more than 20 elements and degassed meteoritic materials, were rather placed inside A1 containers in central bores. These containers probed up to nine different depths in the meteoroid models (see reference 13 for details). The irradiations were performed at the 600-MeV external p-beam of the CERN synchrocyclotron.Homogeneous irradiation of the model meteoroids was achieved by a complex movement of the target spheres, which consisted of the superposition of two orthogonal translational movements and two rotations around perpendicular axes. Each of the irradiations lasted for about 12 h. The fluxes of primary protons through the gabbro spheres and the homogeneity of the irradiations were measured via the "Na activities in various A1 monitor foils in front of the targets. The doses of primary protons measured were (4.82 i 0.02) X 1015 cm-2 ( R = 5 cm), (5.93 f 0.03) x 10'4 cm-Z(R = 15 cm) and (2.17 i 0.004) x 1014 cm-2 ( R = 25 cm). Thus. from the numbers of primary particles the irradiations were equivalent to cosmic ray exposure times exceeding 1 x 106 years even for the R = 25 cm meteoroid model. In order to check the background of secondary particles near the model meteoroids, numerous Fe and Co monitor foils were located in different positions around the target assembly in the experiment hall.The analysis of these foils resulted in upper limits for the background fluxes of 5.0 x 108 cm-2 s-1 for the secondary protons, 3.6 x 108 cm-2 s-1 for the fast secondary neutrons and 3.4 X 108 cm-2 s-1 for the thermal neutrons in all three experiments. A detailed analysis showed that the interference from these background fluxes was negligible in all instances. With regard to the particular nuclides described in this work, measurements were made by gamma- and X-ray spectrometry and by accelerator mass spectrometry (AMS). Sodium-22 (ti = 2.602 years) was determined via its 1274.50- keV gamma-line.having an absolute intensity of 0.9994. For 14Ti (ti = 47.3 years) the low-energy gamma-ray line of 67.8 keV was used, assuming an intensity of 0.877. The 78.4-keVANALYST, MARCH 1989, VOL. 114 Unfortunately, the code ALICE LIVERMORE in its present form does not allow the calculation of integral excitation functions for these two product nuclides from either target element because of the excessive mass differences between the targets and products. Therefore, the same excitation functions had to be adopted for p- and n-induced reactions. In this way theoretical production rates were derived describing the isotropic 600-MeV irradiation of the three model meteoroids. The comparison of measured and calculated production rate depth profiles allows, on the one hand, validation of the HET calculations.On the other hand. it offers the opportunity to investigate the depth and size dependence of the different primary and secondary contributions to a reaction product. Such comparisons will be performed in the next section. 0.08 - 0.07 0.06 ," 0.05 - 0.04 ' 0.03 0 0 297 - - - - - - Results of Terrestrial Simulation Experiments Generally. short- and medium-lived radionuclides have been investigated more thoroughly in the past than the long-lived radioactive or stable cosmogenic nuclides. This has resulted in a drastic difference in the availability of thin-target cross- section data. Based on the knowledge of the production of radionuclides. accumulated in the past in nuclear physics and technology studies.these nuclides are much better suited for a detailed analysis of the occurrences in a thick-target experi- ment and for the validation of HET calculations. On the other hand, the long-lived and stable nuclides are the more important cosmogenic nuclides with respect to unravelling the irradiation history of meteorites. Depth profiles for the production of about 100 radio- nuclides have so far been measured in the context of the CERN SC96 experiment, allowing a detailed survey of the various production modes of residual nuclides and for a critical discussion of the quality of Monte Carlo modelling.11 The phenomenological characterics of the depth- and size-depen- dence can be described as follows. The depth profiles for the production of so-called low- energy products with small mass differences between the target and product increase from the surface to the centre.The centre production rates for such products increase strongly with increasing radii, up to a factor of 3.5 in the radius range investigated. There are target - product combinations for which the centre production shows the highest values at R = 25 cm and the trend for the data indicates a further increasl whereas for others the highest values are obtained for R 0.02 0.01 0.3 1 - ..................................... ............. ... - .... ... . . . ... .- . - . _ . A 1 1 L .......................... . . . ...... ...... ......... ... ... . . . . :" ... 0 0 5 , ..: , -.-.-._._, .-. -.-.-. - ' - 0 5 10 15 0 -15 -10 - 5 rcm Fig.1. Experimental and theoretical production rates of from Ti in an artificial meteoroid with a radius of 15 cm. All production rates are normalised to a flux of incoming particles of 1 cm-'s-I. The open s\mbols with error bars represent the experimental data. For the theoretical depth profiles different production modes are distin- wished. The full line gives the total production, which is the sum of production b) primary protons (dashed line), by secondary protons (dotted line) and by secondary neutrons (dashed-dotted line) 15 cm. As the mass differences increase the gradient of the production rates inside an artificial meteoroid becomes smaller and the centre production rates show a slower increase with increasing radius. The maximum production rates are now observed at a radius of 15 cm.In contrast to these low- and medium-energy products, high-energy products with mass differences greater than 20-30 show depth profiles that decrease strongly from the surface to the centre. Also, the centre production rates decrease as a function of radius. However, a simple distinction of high- and low-energy products does not adequately describe the occur- rences in the artificial meteoroids. There are intermediate shapes showing just minor maxima or minima at the centre and even double-humped depth profiles. It is interesting that the profiles of 7Be from Mg, Al, Si and Fe are all constant within the limits of statistical error and also the dependences of centre production rates on radius do not fit into the general trends.11.12 Similar constant depth profiles have also been seen for loBe from Mg, A1 and Si.1(' The nuclear reactions most sensitive to depth and radius in the artificial meteoroids are low-energy neutron capture.exhibited by the production of 6(Co from Co. Cobalt-60. being produced by neutron capture. only gives direct evidence for the build-up of low-energy neutron fields. The centre produc- tion rates increase by a factor of 100 between radii of 5 and 25 cm. Whereas in the 5-cm artificial meteoroid no significant depth dependence can be seen, in the 15- and 25-cm spheres gradients of 30 and 400%, respectively, are found." Generally, the actual depth profiles and production rates are influenced by a complex balance of shapes of the underlying excitation functions and of the depth, size and energy dependences of primary and secondary nucleon fields. A particular target - product combination can be understood and quantitatively described only by taking into account all these factors.as is done by Monte Carlo transport calcula- tions. Among the radionuclides measured there are a number of classical n-monitor reactions such as *9Co(n.p)59Fe and 58Ni(n,p)58Co and also p-monitor reactions of the (p,xn) type. the products of which are exclusively or to an overwhelming extent produced by either protons or neutrons, respectively. Hence they allow for unambiguous monitoring of the respec- tive nucleon fields. The results of the HET calculations describe well the measured production rates. 11 However, these reactions are sensitive to relatively low energies ( E < 20 MeV) of protons and neutrons.For products with higher Q values a superposition of n- and p-induced reactions always has to be accounted for. However, also for such reactions in most instances the HET calculations in combination with our thin-target cross-section data base reproduce the experimental 0.1 0.09 c - 'I298 0.12 0.1 0.08 0.06 0.04 0.02 0 - ANALYST. MARCH 1989, VOL. 114 - 7 - _ _ - r- - _ _ - - - - - - - - - - - - - -__ - - - - - . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .-.r . . I . _ . _ _ - L 0.028 4 x 10-3 .......................................... ,_._. -.- - .-.- .-. ....... ...... ... ... - ... 7 , - , -. L. Y) 0.016 1 0.02 0.01- - ......................................-.-.- ,:,"' -.-'-' -. -. -' r i m Fig. 3. Experimental and theoretical production rates of 22Na from Fe in a n artificial meteoroid with a radius of 15 cm. All production rates are normalised to a flux of incoming particles of 1 c m - 2 s - ' . See Fig. 1 for the explanation of the symbols and lines data to within 20%. For many target - product combinations this agreement is even better than 5-10%. These general trends are also observed for the production rates measured in this work. From the measured depth profiles AJTi has to be regarded as a low- or medium-energy product from Ti. Whereas for the artificial meteoroid with R = 5 cm the measured production rates are constant to within 10%. for the R = 15 cm sphere an increase by about 20% from the surface to the centre is seen (Fig. 1) and for R = 25 cm this increase even amounts to 40%.The centre production rates have their maximum at R = 15 cm, the absolute values for R = 5 and 25 cm being 0.17 x and 0.22 x 10-4 g-1 s-1. respectively. As shown in Fig. 1. the agreement between the experimental and calculated depth profiles is still not satisfac- tory. The calculations underestimate the experimental data by 20% in the centre, whereas at the surface agreement to within j0i, is obtained. Secondary protons and neutrons make up more than half of the theoretical production rates and the contribution of secondary neutrons is less prominent than observed for other low- or medium-energy products.11 The importance of secondary particles decreases when considering the production of 22Na from Ti.Primary protons make up 75"i0 of the production in the artificial meteoroid with R = 15 cm (Fig. 2), clearly illustrating the high-energy character of the production of 22Na from Ti. The experimental centre production rates decrease monotonously with increas- ing radii. being 0.13 x 0.075 X and 0.035 X g-* s-1 for radii of 5 . 15 and 25 cm. respectively. The depth profiles for 5- and 15-cm radii are constant within the limits of error. whereas for the 25-cm sphere a 40% decrease from the surface to the centre was observed. There are similar problems as reported above for the production of 7Be from Fe. i.e., that theory predicts a decrease from the surface to the centre and that it does not explain the observed constancy over the entire sphere.It has to be kept in mind, however. that thin-target cross-sections for the production of 22Na from Ti are incompletely known. On the basis of the available data the importance of the observed discrepancy cannot be over- estimated. For the production of 2Wa from Fe more cross-section measurements are available from earlier work. Although there is still no perfect agreement between the calculated and experimental data. the scatter of the experimental data is compatible with the theoretical depth profile (Fig. 3). In spite of the fact that even higher energies are necessary to produce '?a from Fe than from Ti, the production by primary protons still makes up only 75% for the artificial meteoroid with R = 15 cm. secondary protons and neutrons contributing signifi- cantly but to different extents.The centre production rates are P I I .................................. ........ 0.02 U .... ... ... ._._.-. -.-.-.-.-. n r m -15 -10 -5 0 5 10 15 0.08 0.07 1 Fig. 4. Experimental and theoretical production rates of "'Be from Ti in artificial meteoroids with radii of 5. 15 and 25 cm. All production rates are normalised to a flux of incoming particles of 1 cm - 2 s- I , See Fig. 1 for the explanation of the symbols and lines 0.027 x 10-4, 0.017 x 10-4 and 0.013 x g-1 s-1 for R = 5 , 15 and 25 cm, respectively. The depth profiles are constant at R = 5 cm and decrease by about 20?6 from the surface to the centre for each of the remaining two artificial meteoroids. The production of *()Be from Ti and Fe is of particular interest, as such simple classifications as high- and low-energy products do not hold for this nuclide.In the energy region investigated "'Be is mainly a product of fragmentation. Therefore, simple threshold considerations do not hold. In Figs. 4 and 5 a survey of the depth profiles of the production of 1"Be from both elements is given for all three meteoroid models. For the R = 5-cm sphere in both instances nearly 90'10 of the production is due to primary protons. Theory and experiment are in fair agreement. both depth profiles being constant within the limits of error. With increasing radius in both instances the influence of secondary particles increases. The depth profiles for the production from Ti in the R = 15- and 25-cm spheres are also almost constant. The calculations slightly underestimate the R = 15-cm data.the predicted slight decrease from the surface to the centre not necessarily being in contradiction with the measurements. For R = 25 cm. however. the experimental and theoretical shapes are differ- ent. The agreement is still acceptable. particularly when considering the experimental errors and the coarse assump- tions about the thin-target excitation functions. For "'Be from Fe there is a slight indication in the experimental data of a double-humped profile, whereas for R = 25 cm a decrease from the centre to the surface is seen. Theory predicts in both instances a decrease from the surface to the centre. For both depth profiles the agreement obtained is not yet satisfactory. From the comparison of experimental and theoretical data presented in this paper, it can be stated that the contributionANALYST, MARCH 1989, VOL.114 0.02 299 1 -,-, - -,-. - .-.-._ .-.-.-. 0.02 0.03 1 ....................................... .... . _ . - . - . -._. 0.01 !J, I- ................................ 5 10-7, ..... ..... n -.I-. . ' '-.I-.-.- 8 -6 - 4 - 2 0 2 4 6 0.05 1 T r . _ _ _--r I 0.04 - 7 . I i 0.035 0.03 0.025 0.015 1 5 x 10-3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... - ._ . - . -, - . - . - -, - 0 -30 -20 -10 0 10 20 30 ricm Fig. 5. Experimental and theoretical production rates of '('Be from Fe in artificial meteroids with radii of 5. 15 and 25 cm. All production rates are normalised to a flux of incoming particles of 1 cm-2 5 - 1 , See Fig. 1 for the explanation of the symbols and lines of secondary particles to the production of residual nuclides also for the medium- and high-energy products must not be neglected when modelling the interactions of galactic cosmic ray protons even in small meteoroids.Secondary protons and neutrons contribute with different absolute amount and depth dependences, so that a uniform treatment of both particle types in the respective models is inadequate. The results obtained for products with "well known" reaction data as shomn in reference 11 strongly emphasise that thin-target calculations based on HET nucleon spectra and thin-target cross-sections are readily capable of describing the depth and size dependences of the production of residual nuclei in thick stony targets. at least at primary p-energies of 600 MeV.However, a necessary condition for an adequate description of production depth profiles is the availability of reliable excitation functions. For the production rates presented here the respective database is still insufficient, resulting in some lack of agreement between calculated and experimental data. Further work will have to be carried out in the future. Translation to Cosmic Irradiation Conditions For several reasons the production rates measured in the simulation experiments cannot be directly compared with real irradiation conditions in space. In the laboratory experiments. non-chondritic material was irradiated with monoenergetic 600-MeV protons, whereas real meteoroids are irradiated with a mixture of protons (about goo/,) and or-particles (about 10%)32 with complex continuous spectra.Even considering only the GCR protons, the production of secondaries depends strongly on the energy of primaries and on the mass number of the targets. Fraser and Bartholomew33 have shown that the production of neutrons by high-energy protons from target 0.16 1 I r 0.12 - ," 0.1 j I- ,_;- , elements with A < 210 increases linearily with increasing proton energy and with increasing mass number of the targets. Although the composition of the artificial meteoroids does not completely match that of ordinary chondrites, the influence of this difference will be marginal, as the mean mass numbers of gabbro and diorite, on the one hand, and H-chondrites on the other, differ by less than 15%.Hence the differences between the energies of the primary particles in the terrestrial simulation experiments and in the GCR irradiation in space will have a dominant influence on the production rates. On the basis of the present knowledge of the quality of HET calculations, the differences between 600-MeV and real GCR irradiations can be derived from such calculations. For this purpose. primary and secondary nucleon spectra were calcu- lated for incident protons with a GCR energy spectrum. irradiating meteoroid models made of gabbro with radii of 5. 10, 15 and 25 cm. As the input the arithmetic mean of the GCR p-spectra at 1965 solar minimum and 1969 solar maximum in the parameterised f o r m $\en h \ C;i\t3_rnoli and La134 was used, as in our previc'~ cosmogenic nuclides in meteoroids b> prinit~i-;~ &,iiactic pro- tons.' Also with these "GCR" spectra, theoretical production rates were calculated.They will be used to translate our findings for the terrestrial simulation experiments to cosmic irradiation conditions. As an example, in Fig. 6 the depth profile calculated for the production of "'Be from Fe by galactic protons is shown. Whereas the irradiation with 600-MeV protons (Fig. 5 ) resulted in a theoretical production decreasing by more than a factor of two from the surface to the centre, the "GCR" irradiation produces an essentially flat production profile, decreasing to the centre by only 10%. This indicates that in a monoenergetic irradiation the influences of high production thresholds and of strong gradients in the excitation functions are more important than for real GCR protons.For the "GCR" depth profile, moreover. the contributions of secondary particles are much more pronoun- ced than for 600-MeV proton irradiations. because of the increase of multiplicities by higher primary energies. The larger mean primary energy in a GCR spectrum together with these enlarged multiplicities for the production of primaries result in centre production rates in an R = 25-cm sphere 9.25 times higher than those calculated for the 600-MeV irradia- tion. This finding is in accordance with a previous finding. Already in the first simulation experiment with a 5-cm diorite sphere9 it turned out that the results could not be directly transferred to cosmic irradiation conditions.Hence the observed production rates of 21Ne in degassed material from the chondrite Ladder Creek were considerably lower than the300 ANALYST, MARCH 1989, VOL. 114 The authors are grateful to the CERN PSCC for making available the irradiations and to Drs. B. Allardyce and J. W. M. Tuyn and the SC and Health Physics staff for their kind cooperation and for the essential support they provided before and during the irradiations. Our thanks are due to R. Berndt, W. Pyschny and all the members of the mechanical workshop of the Abteilung Nuklearchemie, Universitat zu Koln, who constructed the irradiation machines. This work was supported in part by the Deutsche Forschungsgemein- schaft, Bonn. FRG. Table 1. Comparison of calculated production rates (10-6 g-* s-I) in the centre of gabbro spheres with a radius of 25 cm, irradiated with 600-MeV protons and with protons having a realistic GCR energy spectrum.The production rates are normalised to a flux of primary particles of 1 cm-* s-1 600-MeV Product Target energetic p-spectrum 600 ,MeV ‘“Be Ti 4.09 15.3 3.74 ZZNa Ti 2.68 13.1 4.89 aTi Ti 18.1 21.4 1.18 ‘“Be Fe 1.47 13.6 9.25 2’Na Fe 1.08 15.3 14.2 mono- GCR Ratio GCR/ canonical 21Ne production rate for chondrites of 0.31 x 10-8 cm3 (STP) g-1 per 106 years.35 In view of the new calculations this effect can easily be understood. Unfortunately, the translation from terrestrial simulation experiments to cosmic irradiation conditions cannot be achieved by a simple scaling factor. As shown in Table 1, the ratios of the centre production rates produced by a GCR spectrum to those of a 600-MeV irradiation range from 1.18 to 14.2 for the produc- tion rates dealt with here.In spite of the terrestrial simulation experiments not being directly comparable to cosmic irradiation conditions, they provide an experimental basis for the necessary validation of a thin-target model. Currently, such a model for the production of cosmogenic nuclides in meteoroids is being developed on the basis of experimental experience and the calculation methods described here. To do this with reasonable con- fidence, further validation of our calculation method by thick-target experiments at other energies is necessary. Also, the influence of target chemistry on the internuclear cascade and on particle transport has to be studied.Further, the measurements and calculations have to be extended to other cosmochemically important nuclides and to improving the thin-target cross-section database. Conclusions The production depth profiles of 44Ti from Ti and of 1OBe and 22Na from Ti and Fe have been measured in three artificial meteoroids with radii of 5 , 15 and 25 cm, isotropically irradiated with 600-MeV protons in order to simulate the GCR p-irradiation of meteoroids in space. HET calculations in combination with thin-target produc- tion cross-sections quantitatively describe the observed pro- duction depth profiles. The agreement between experiment and calculations is excellent provided that accurate excitation functions are available.11 Only reasonable agreement was obtained for the production rates measured in this work, indicating the necessity for further improvements to the thin-target cross-sections of the underlying nuclear reactions.The investigation showed that secondary particles are the dominant nuclear reacting specimen in meteoroids for low- energy products, and even for medium- and high-energy products they must not be neglected. Moreover, the different contributions of secondary protons and neutrons have to be accounted for. It turns out that a complete simulation of the GCR p-irradiation by a single monoenergetic experiment is not possible. This is particularly due to the strong dependence on the energy of multiplicities for the production of secondary protons and neutrons. On the basis of the validation of the HET calculations, which is provided by experiments carried out so far, it is possible to translate the present results to the actual irradia- tion of meteoroids in space by the complex GCR spectrum.1. 2. 3. 4 5 . 6. 7. 8. 9. 10. 11 12. 13. 14. 15, 16. 17. 18. 19. 20, 21. References Geiss. J.. Oeschger, H . , and Schwarz. U . , Space Sci. Rev., 1962, 1, 197. Lal, D.. Space Sci. Rev., 1972, 14, 3. Reedy, R . C . , Arnold, J . R . , and Lal, D . , Annu. Rev. Nucl. Part. Sci., 1983, 33, 505. Michel. R . , and Brinkmann, G . , J . Radioanal. Chem., 1980, 59, 467. Michel, R . , Brinkmann, G . , and Stuck, R . , Earth Planet. Sci. Lett., 1982, 59, 33; erratum, 1983, 64, 174. Michel, R . , Brinkmann. G . , and Stuck, R.: in Boeckhoff. K. H . , Editor, “Nuclear Data for Science and Technology, Proceedings of an International Conference, September 6-10, 1982, Antwerp,” Reidel, Dordrecht, 1983, p.952. Michel, R . , and Stuck. R . , J . Geophys. Res., 1984,89B, Suppl., 673. Reedy, R . C . , Lunar Planet. Sci., 1986, 17, 695. Michel, R . , Dragovitsch, P . , Englert, P., Pfeiffer, F., Stuck, R . ; Theis, S., Begemann, F.. Weber, H., Signer, P., Wieler, R . , Filges, D . , and Cloth, P., Nucl. Instrum. Methods Phys. Res., 1985. B16, 61. Englert, P . , Theis, S., Michel, R . , Tuniz, C., Moniot, R . K., Vajda, S.. Kruse, T. H . , Pal, D . K., and Herzog, G . F., Nucl. Instrum. Methods Phys. Res., 1984, 85, 415. Aylmer. D . . Begemann, F.; Cloth, P . , Dragovitsch. P.. Englert, P.. Filges, D . , Herpers, U . , Herzog, G . F.. Jermai- kan, A , , Klein, J ., Kruse, T. H . . Michel, R . , Moniot. R. K., Middleton, R . . Peiffer, F.: Signer, P., Stuck, R.. Theis, S., Tuniz, C., Vadja, S., Weber. H.. and Wieler, R . , “Monte Carlo Modelling and Comparison With Experiments of Nuclide Production in Thick Stony Targets Isotropically Irradiated With 600-MeV Protons,” Juel-2130, Kernforschungsanlage Julich, Julich, FRG, 1987. Michel, R.. Peiffer, F., Theis, S., Begemann, F.. Weber. H . . Signer, P . , Wieler, R . , Cloth, P., Dragovitsch, P . , Filges, D . , Englert, P.. Nucl. Instrum. Methods Phys. Res. B . submitted for publication, 1988. Dragovitsch. P . , Thesis, Universitat zu Koln, 1987. Seelmann-Eggebert, W., Pfennig. G . , Munzel, H . , and Kleve- Nebenius, H . , “Chart of the Nuclides,” Fifth Edition.Komu- nalschriften-Verlag Jehle, Munich, 1981. Erdtmann, G . , and Soyka, W.. “The Gamma Rays of the Radionuclides,” Verlag Chemie, Weinheim, 1979. Theis, S., Thesis, Universitat zu Koln, 1986. Vogt, S., and Herpers, L., Fresenius Z . Anal. Chem., 1988, 331, 186. Graf. T., Vogt, S., Bonani, G . , Herpers, U.. Signer, P., Suter, M., Wieler, R . , and Wolfli, W., Nucl. Instrum. Methods Phys. Res., 1987, B29, 262. Cloth, P., Filges, D . , Sterzenbach, G . , Reul, C.. Armstrong, T . W., Colburn, B. L.. Anders, B., and Bruckmann. H . . “HERMES-High Energy Radiation Monte Carlo Elaborate System.” Juel-2203. Kernforschungsanlage Julich. Julich. FRG, 1988. Emmett, M. B . , “The MORSE Monte Carlo Radiation Transport Code System” ORNL-4972, National Technical Information Service, US Department of Commerce, Spring- field, VA.1975. Barish, J . , Editor, “EPR Data Package: EPR Coupled 100- group Neutron 21-group Gamma Ray Cross Sections for EPR Neutronics,” RSIC Data Library Collection DLC-37. ORNL- TM-5249, National Technical Information Service, US Depart- ment of Commerce, Springfield, VA, 1979.ANALYST, MARCH 1989, VOL. 114 22. 23. 21. 25. 26. 27. 28. Kinsey, R . , “ENDF-102, Data Formats and Procedures for the Evaluated Nuclear Data File ENDF,” National Nuclear Data Center Brookhaven. BNL-NCS-50496, National Technical Information Service, US Department of Commerce, Spring- field, VA, 1979. 30. Michel, R . , and Peiffer, F., “Theoretical Predictions of Integral 31. Excitation Functions for n-induced Reactions in the Energy 32. Range from 2 to 190 MeV for the Target Elements Mg, Al, Si, 33. Ca. Ti, V, Cr, Mn, Fe, Co, Ni. Cu, Y, Zr, Te. W, Ba, Lu. Au,” KC-NSIG-86 (unpublished), Institut fur Kernchemie, Univer- sitat zu Koln. Koln. FRG, 1986. Blann, M.. and Bisplinghoff. J . , ALICE LIVERMORE 82, UCID-19614, Lawrence Livermore Laboratory, Livermore, CA, 1982. Blann. M . . ALICE LIVERMORE 87, personal communica- tion, 1987. Brodzinski, R . L . , Rancitelli. L. A , . Cooper, J. A , , and Wogman, N. A . , Phys. Rev. C , 1971, 4. 1250. Brodzinski. R . L.. Rancitelli, L. A , . Cooper. J . A , , and Wogman. N. A . , Phys. Rev. C. 1971, 4, 1257. 29. Schiffmann. T., Diplomarbeit. Universitat zu Koln. 1987. 34. 35. 301 Raisbeck, G . M., and Yiou, F., in “Proceedings of the 14th International Cosmic Ray Conference. Munich, 1975,” Max Planck Institut fur Extraterrestrische Physik, Munich, 1975, Volume 2, p. 495. Perron. C . , Phys. Rev. C, 1976. 14, 1108. Dedieu, J . , Thesis, Universitk de Bordeaux, 1979. Simpson, J . A , . Annu. Rev. Nucl. Part. Sci.. 1983, 33, 323. Fraser, J . S., and Bartholomew. G . A , , in Michaudon. A . , Cierjacks, S., and Chrien, R . E., Editors, “Neutron Sources (Neutron Physics and Nuclear Data in Science and Technology, 2);’ Pergamon Press, Oxford, 1983, p. 217. Castagnoli, G . , and Lal, D . , Radiocarbon, 1980. 22, 133. Nishizumi, K . , Regnier. S.. and Marti. K . , Earth Planet. Sci. Lett., 1980. 50, 156. Paper 81028381 Received July 14th, 1588 Accepted September 19th, 1588

ISSN:0003-2654    

DOI:10.1039/AN9891400295

出版商:RSC    

年代:1989

数据来源: RSC