OCTOBER, 1967 THE ANALYST Gamma-activation Analysis Vol. 92, No. 1099 A Review* BY C. A. BAKER (Analytical Sciences Division, A .E.R.E., Havwell, Bevks.) SUMMARY OF CONTENTS Introduction Types of gamma activation Nuclear considerations for (y,n) reactions Sources of high energy photons Electron accelerators Cyclic machines Linear machines Characterisation of the radiations Standards Determination of light elements, oxygen, nitrogen, carbon and fluorine The use of decay curves and energy thresholds The use of spectrometry and other instrumental methods Destructive analysis by rapid chemical separations Determination of heavier elements of atomic numbers equal to, or greater than, 10 Conclusions ACTIVATION analysis has now become firmly established as an analytical method of wide applicability.The high cross-section of many elements for thermal neutrons has enabled sensitive methods to be devised for trace-element analysis and fast-neutron activation by using 14-MeV neutrons produced by accelerators has proved useful for the measurement of major and minor components in many matrices. Charged-particle activations have also been applied to particular problems where the limited depth o€ penetration is no disadvantage and can even be turned to advantage in the analysis of surfaces. All activation analysis is free from the hazards of reagent blanks, and neutron activation has the additional advantage of homogeneous irradiation and the facility of etching the surface of a sample between activation and processing in order to remove any contamination. In recent years gamma-activation analysis has been exploited, particularly because of the possibility of doing intact analysis on ultra-pure materials for the light elements carbon, nitrogen and oxygen whose nuclear properties are unf avourable for neutron activation.Because of their penetrating nature gamma photons share with neutrons the advantages of homogeneous activation and post-irradiation etching that are of particular value when the metallurgical properties of high purity materials are being investigated. The more general applicability of gamma activation to the analysis of traces, minor and major components in a wide range of materials is now appreciated, and this review is concerned with the whole field of activation analysis with gamma photons.TYPES OF GAMMA ACTIVATION Most of the useful analytical applications of gamma activation require photons with energies in excess of 7 MeV, but, for the sake of completeness, the reactions observed with less energetic photons will be discussed briefly. ( y , ~ ) ACTIVATION : MOSSBAUER EFFECT- This is a very limited form of photon activation in which y-photons of a few hundred keV are emitted by metastable states and can be re-absorbed by other nuclei of the same isotope. There are no analytical methods based on this effect at the present time. For details see Summaries in advertisement pages. 601 Nature of the product nuclides * Reprints of this paper will be available shortly.602 BAKER : GAMMA-ACTIVATION ANALYSIS [AnaZyst, Vol. 92 ( 7 , ~ ’ ) ACTIVATION- Many stable nuclides, particularly of the heavier elements, can exist in a short-lived metastable state that can be produced by the interaction of a y-ray of sufficient energy, for example a reaction of the type 77Se (y,y’) 77mSe, the excess of energy being emitted as a photon 7‘.The cross-sections for such reactions are of the order of microbarns. Kaminishi,l Lukens, Otvos and Wagner2 tabulate isomers from selenium to mercury that can be made in this way with photons from 3 and 6-MeV linear accelerators. The degree of activation increases sharply with increasing energy of the incident gamma rays but, even under the most favourable conditions of 6-MeV energy and 1-mA beam current, the limit of detection is calculated to be one microgram. In the least favourable cases the limit may be several milligrams.(y,n) ACTIVATION IN WHICH NEUTRONS ARE DETECTED- There are two (y,n) reactions with threshold energies lower than the energies available from nuclear disintegrations (Goldstein3). The reaction 9Be (y,n) 8Be (threshold energy E,, 1.67 MeV) has been used to assay for beryllium with Z4Sb (half-life 60 days) and ssY (half-life 87 days) as sources of y-photons. Beryllium concentrations of only 60 p.p.m. can be determined with quite modest y-sources (300 mC) . Deuterium has been determined by the reaction 2D (y,n) lH (Eth 2.23 MeV) with thorium-228 (half-life 1.9 years) and sodium-24 (half-life 15 hours) as y-sources down to 0.02 per cent. v/v of deuterium. (y,n) REACTIONS IN WHICH THE PRODUCT NUCLEUS IS MEASURED- Most of the useful analytical work falls into this category.Basically it is only necessary to irradiate the sample with y-photons of sufficient energy (10 to 25MeV), measure the activity induced in the sample and compare this with a standard in a way that is entirely analogous to thermal neutron activation. (yJ2n) (7,~) (y,pn) J (7,an) REACTIONS- These reactions occur when more energetic photons are used than are necessary for the (y,ii) reactions. They are best regarded as unwanted interferences and avoided if possible. NUCLEAR CONSIDERATIONS FOR (y,n) REACTIONS Most nuclei will react with y-photons of a suitable energy to undergo a (7,n) reaction. The relation between the cross-section for the reaction and energy of the photon will be of the form shown in Fig. 1.Fig. 1. Energy dependence of cross-section for a ( y , n) reaction E, is the threshold energy and E,,,, is the energy of the so-called “giant resonance.” The values of E, and Em,,. vary from nuclide to nuclide, but in general Eth is 10 to 20 MeV for light nuclei, falling steadily to 7 to 10 MeV for heavy nuclei. Similarly, the value ofOctober, 19671 BAKER : GAMMA-ACTIVATION ANALYSIS 603 Ern=. falls with increasing atomic number from 20 to 25 MeV down to about 15 MeV. The cross-section at the peak varies from a few millibarns to a few hundred millibarns ( K ~ c h , ~ Sulin5 and Strauch6). SOURCES OF HIGH ENERGY PHOTONS- An intense source of y-photons is required having an energy in excess of the threshold and preferably in excess of Em,,. No known nuclear disintegration provides such energetic photons, but they can be produced by stopping particles of the required energy (“brems- strahlung” production).The most efficient conversion of energy into bremsstrahlung radiation occurs when a light particle is stopped in a heavy absorber, and this is realised physically by causing a beam of energetic electrons to impinge on a target or “radiator” of platinum, gold or tungsten. Most of the energy of the electron beam is dissipated as heat in the radiator, which may need to be water-cooled, but some energy is radiated in the forward direction as a beam of gamma photons. The energy distribution of photons in the brems- strahlung from a mono-energetic electron beam of energy E will be as shown in Fig. 2. The degree of activation induced by such a photon flux will be proportional to the sum of the products of flux and cross-section in each energy element in the region of overlap.The value of this function increases rapidly with energy, approximately as a 20th power near the threshold and as an 8th power in the region of maximum cross-section. Fig. 2. Energy distribution in a bremsstrahlung ELECTRON ACCELERATORS The two major classes are cyclic and linear. CYCLIC NACHINES- The betatron is a cyclic accelerator in which a pulsed beam of electrons is injected into an evacuated torus located between the poles of a powerful electro-magnet. The magnetic field is initially small but is increased over a period of about 5 milliseconds. This serves both to accelerate the electrons and to constrain them in a circular orbit in the torus.About loG revolutions are made, during which time the electrons are accelerated to perhaps 40 MeV, but the peak energy can be varied at will by varying the current flowing in the magnet. At the peak of each cycle of the power supply, the orbiting electrons are deflected on to a small internal platinum electrode and the bremsstrahlung is generated from that point. The mean current incident on the target is of the order of 10-8 amperes. The bremsstrahlung is radiated tangentially from the torus and the samples are placed at a convenient position in the beam. A considerable increase in intensity over a restricted area has been achieved by Chepel’, Viting and Chapyzhnikov,7 who describe the use of a hollow probe in the torus itself, so that the samples can be introduced at a point immediately in contact with the platinum electrode.A 100-fold increase in flux was observed compared to the best flux available outside the torus, but the sample size was restricted to about 1 mm square.604 BAKER GAMMA-ACTIVATION ANALYSIS [Analyst, Vol. 92 There is one report of the use of a much more powerful cyclic accelerator for gamma- activation analysis. Voigt and Abu-Sumra* used a synchrotron at 70 MeV, but no details are given of the machine or radiator. LINEAR MACHINES- The linear electron accelerator is by far the most useful source of high energy photons for gamma-activation analysis. Electrons are injected into an evacuated tube that consists of several sections of shaped waveguide, each one driven by a klystron amplifier.The electrons are accelerated on the crest of a travelling wave and reach energies of as high as 40 MeV, depending on the size of the machine, and mean electron currents of up to 50 pA have been used, although the heat dissipation can present some difficulties with such currents. The electron beam usually leaves the evacuated volume through a thin steel window and impinges on an air or water-cooled radiator of a heavy metal, such as platinum; the sample can be placed immediately behind the radiator. NATURE OF THE PRODUCT NUCLIDES Most of the products of (y,n) reactions are short-lived and even if a long-lived isotope is formed the cost of operation of the accelerator would normally make it uneconomical to irradiate for a period long enough to obtain a high sensitivity.Furthermore, neutron-deficient isotopes usually decay by positron emission, frequently without any accompanying gamma rays except for the 0-51-MeV annihilation radiation. CHARACTERISATION OF THE RADIATIONS- The most exploited techniques in activation analysis are chemical characterisation through radiochemical separations and gamma spectrometry. These two methods are limited in application to the products of gamma-activation analysis because of the short half-lives and paucity of gamma rays, respectively, but they have been used and supplemented with others as follows. Decay-carve resolzdion-This can range from the simple expedient of allowing the short- lived activities to decay and counting a long-lived tail, to refined computer analysis of up to six components.Energy dependence of cross-sections-By varying the energy of the bremsstrahlung the relative activation of different elements can be altered; in particular, the activation falls to zero for incident energies below the threshold energy. and y-spectrometry-/I-Spectrometers have been used to resolve the energy spectrum of positrons into its components, and several of the product nuclides, particularly those having high atomic weights, have significant contributions from y-decay modes so that y-spectrometry can be used. ChemicaZ se9arations-The separation of activities of individual elements by chemical separation is a very powerful technique. In many cases this has required that a procedure be developed or modified to make it rapid in relation to the half-life of the isotope required.STANDARDS- An essential feature of all activation analysis is the provision of an accurate standard. The gamma fluxes that are available are restricted in area and are heterogeneous, and great care must be taken in the irradiation of standards. Various workers have monitored the flux and energy by instrumental means; measured the degree of activation of sample and standard in separate irradiations by using a flux monitor; or compared sample and standard in the same radiation. DETERMINATION OF THE LIGHT ELEMENTS, OXYGEN, NITROGEN, CARBON AND FLUORINE The pioneer work was carried out in 1954 by Basile, Hur6, L6vi3que and Schuh1,g who determined oxygen in stearol, stearic acid and benzoic acid with a betatron to induce the reaction l60 (y,n) 150.Samples of 2.5 g containing several hundred milligrams of oxygen were irradiated for 4 minutes in a position 20 cm from the radiator. The energy was 18.6 MeV to avoid activation of the carbon, and a quartz monitor was included with each irradiation. By using this model system with the energy regulated so that only one element was activated,October, 19671 BAKER GAMMA-ACTIVATION ANALYSIS 605 it was shown that the initial count-rate of the induced activity was proportional to the oxygen content. The feasibility of intact determination of oxygen in beryllium oxide and alumina was also demonstrated. Beryllium is not activated, and aluminium-26 (half-life 6.5 seconds) can be allowed to decay before counting commences.THE USE OF DECAY CURVES AND ENERGY THRESHOLDS- A great deal of the work reported in the literature relies on the resolution of decay curves into components of half-lives 2, 10 and 20 minutes for the determination of oxygen, nitrogen and carbon, respectively. In addition, Englemann and others have used the energy dependence of the degree of activation with great effect. OXYGEN- AlbertlO and Albert, Engelmann, May and Petitll have reported determinations of oxygen, together with nitrogen and carbon, in beryllium, with the linear electron accelerator at Saclay as a source of photons. The accelerator was operated at 28 to 30 MeV and 50 PA, and the radiator consisted of two platinum discs that were cooled by water passing between them. Oxygen was determined by the reaction 1 6 0 (y,n) 150, the 2-minute component of.the decay curve being attributed to this nuclide. A thorough discussion of the determination of oxygen in this way is contained in three similar papers by Engelmann et aZ.,12713J4 who used the same linear electron accelerator as Albert for the source of photons. A pneumatic transfer system carried specimens from the irradiating position to the counters in 25 seconds. The counters were automatically operated and consisted of two opposed sodium iodide crystals equipped with coincidence circuits so that only 0.51 1-MeV annihilation radiation was detected. Methods are described for the non-destructive charac- isation of the positron activities by using decay-curve resolution and the energy dependence of the activation cross-section.After an irradiation of 5 minutes the component of the decay curve having a half-life of about 2 minutes was attributed by Engelmann to oxygen-15 (half-life 2.1 minutes) from the reaction l60 (y,n) 150 and phosphorus-30 (half-life 2.6 minutes) from the reactions 31P (y,n) 30P and 32S (y,pn) 30P. However, the thresholds of the three reactions are different and by carrying out three radiations, at 28, 18 (below the threshold for sulphur) and 14 MeV when the phosphorus alone is activated, it was possible to identify the individual contributors. Alternatively, the irradiation could be carried out in nominal energies of 27, 20 and 16 MeV, which were not strictly related to the threshold energies but nevertheless gave large differences in the ratio of specific activities for the three reactions.Specimens of “Mylar” film were included with each sample in order to monitor the current and energy of the machine during the irradiations. NITROGEN- AlbertlO suggested that nitrogen could be determined from the reaction 14N (y,n) 13N, and attributed the 10-minute component of the decay curve to this product. The resolution of the 10-minute component was uncertain because of the presence of 2-minute oxygen-15 and 20-minute carbon-1 1, and AlbertlO recommended that the nitrogen should be separated chemically by the Kjeldahl method, for which yields of 70 per cent. were claimed. Alternatively, the irradiation was carried out at 15 MeV, i.e., below the threshold of the interfering reactions. This method carried the penalty of a 500-fold loss in sensitivity.In America, Beard, Johnson and Bradshaw15 have applied gamma-activation analysis to the problem of quality control in ultra-pure beryllium manufacture. The individual activities were recognised by resolving the positron decay curve into three components attributed to oxygen, a mixture of the triad nitrogen, copper and iron, and carbon. Standards were irradiated separately and the beam currents monitored. The linear electron accelerator at Stanford was used and sensitivities of less than 1 pg were claimed. Enge1mannl2 also attributed the 10-minute component to a mixture of nitrogen-13, copper-62 and iron-53, but pointed out that the threshold energies and relative activation at different energies were not favourable for the application of the same methods that he used to discriminate between oxygen-15 and phosphorus-30.606 BAKER GAMMA-ACTIVATION ANALYSIS [Analyst, Vol.92 Engelmann12 therefore determined the copper and iron by some other method and made suitable corrections. The results obtained agreed well with similar irradiations in which pure nitrogen-13 was separated by the Kjeldahl method. CARBON AND-FLUORINE- Carbon is determined by the reaction I2C (y,n) llC, and all workers attribute the 20-minute component of the decay curve to this product. AlbertlO and Albert, Engelmann, May and Petitll pointed out that it is necessary to separate the active carbon in a chemically pure form for samples of aluminium, iron and zirconium in order to eliminate the interfering activity that is induced in the matrix.This was achieved by combusion of the sample in oxygen, then trapping out and counting the carbon dioxide formed. Engelmann12 assessed fluorine by measurements on the 112-minute component due to the reaction 19F (y,n) lSF. Fluorine is subject to interference from the reaction 23Na (y,ncc) 18F, but this can be reduced to negligible proportions by suitably reducing the energy of the irradiation. For materials of the highest purity, when the activities do not interfere with each other, sensitivities of the order of 0.01 to 0.1 pg have been achieved. THE USE OF SPECTROMETRY AND OTHER INSTRUMENTAL METHODS By using the principle of the “pocket” or “well” in the betatron torus to produce a high flux over a small area,’ oxygen has been determined in rubbers by P-spectrometry.l6 This was done by irradiating at 18 MeV, below the threshold energy for carbon activation, but the sensitivity was poor.By operating at 25 MeV a large increase in specific activity was obtained but the carbon activity interfered. To overcome this the positron emission from the thin samples was counted with a plastic scintillator, and pulse height analysis was used to discriminate against the counts from the decay of llC pf 0.99 MeV but retain those from 150 p+ 1.68 MeV. The counter was very inefficient but the over-all sensitivity of the method was increased by a factor of 20 compared with the 18-MeV irradiation. Separate irradiations of boron trioxide were used as standards and solutions of dibutyl phthalate were used to calibrate the method, Carbon has been determined in a piece of steel of archaeological interest by gamma activation and intact gamma spectrometry.A 70-MeV synchrotron was used to provide the bremsstrahlung. The analysis was complicated by the reaction 54Fe (y,pn) 52mMn (half-life 21.3 minutes) in the iron matrix which interfered with the measurement of carbon-11 (half-life 20-3 minutes). Fortunately the manganese-52m has a gamma ray at 1-435 MeV and by measurement of the gamma spectrum induced in pure iron it was possible to correct for this interference (Voigt and Abu Sumras). DESTRUCTIVE ANALYSIS BY RAPID CHEMICAL SEPARATIONS Baker, Pratchett and Williams1’ have used the linear electron accelerator at Hanvell as a source of y-photons to develop gamma-activation analysis methods for carbon and oxygen in materials that cannot be counted without separation of the required nuclide.RAPID SEPARATIONS OF RADIOCHEMICALLY PURE OXYGEN- The determination of oxygen in matrices, which themselves contribute a large short-lived activity ( e g . , iron and molybdenum), requires a chemical separation that is specific, of good chemical yield and very rapid. Baker has achieved this by using an inert-gas fusion in an iron bath at 2000” C, and subsequent purification of the oxygen activity by passing the gas stream through two copper oxide furnaces, one at 600” C and the other at 1100” C. The first furnace has been shown to oxidise the carbon monoxide without loss of radio-oxygen while the second furnace retained more than 99 per cent. of oxygen-15 without contamination by carbon-11 or other activities.The chemical separation was carried out in 4 minutes and the limit of detection was better than 1 pg. Engelmann has recently suggested a similar separation procedure in which gases evolved by fusion in a carbon arc furnace were absorbed on soda lime within the sodium iodide crystal assembly.October, 19671 BAKER : GAMMA-ACTIVATION ANALYSIS 607 CHEMICAL SEPARATION OF CARBON ACTIVITY- Baker has also determined carbon in low alloy and stainless steels by combustion in oxygen and counting the carbon dioxide evolved. Samples were irradiated, together with two standards of graphite of similar area to the sample placed in front of and behind the sample. The mean specific activity of the standards was compared with the activity found in the sample.Excellent agreement with published carbon contents of standard steels was achieved over the range 1 per cent. to 10 p.p.m. and the limit of detection was better than 0.1 pg. The technique has been applied to ultra-pure molybdenum. Ceramic materials such as magnesium oxide and calcium oxide have also been analysed by Baker for carbon by using a flux of borax at 1250" C in a stream of oxygen to dissolve the sample and remove the carbon as carbon dioxide. The high oxygen content of these materials resulted in a nuclear interference, 1 6 0 (y,na) W , that could only be eliminated by irradiating with an energy below the threshold for this reaction, 25.8 MeV. In spite of this the limit of detection was still better than 1 pg.The separation of individual radio-nuclides has the advantage that sources can be counted in a gross gamma well-crystal counter of high efficiency, rather than y - y coincidence counters used by many workers which are necessary to eliminate counts from gamma-emitters in the sample. DETERMINATION OF HEAVIER ELEMENTS, ATOMIC NUMBER EQUAL TO, OR GREATER THAN, 10 Engelmann mentions the possibility of determining zirconium and other elements, and this field is covered fully by Schweikert and Albertls The elements silicon, titanium, chrom- ium, manganese, iron, cobalt, nickel, copper, zinc, arsenic, strontium, zirconium, niobium, molybdenum, silver, cadmium, antimony, hafnium, tantalum, tungsten, platinum, thallium, lead and bismuth were divided into two groups according to the degree of activation induced by 27-MeV gamma photons in a 2-hour irradiation.Methods were indicated for the deter- mination of those elements for which the method is sensitive by decay-curve resolution, energy dependence of activation and gamma spectrometry for non-destructive characterisation of the induced activities. Specific examples were given of analyses that were superior to neutron-activation analysis for the following reasons: the nature of the photon-activation product is more suitable by virtue of sensitivity, gamma spectrum or half-life ; interference is eliminated; and shielding is eliminated in a high cross-section matrix, such as boron, tungsten, manganese, cadmium and hafnium. These advantages occur repeatedly in the following examples.Zirconium can be determined in hafnium with a sensitivity of 1 p.p.m. by irradiating for a short time and immediately counting the 0-59-MeV gamma ray from zirconium-89m (half-life 4.4 minutes) with a y spectrometer.ls Japanese workers report the determination of as little as 10 p.p.m. of zirconium in 100-mg samples of hafnium by y-irradiation at 20 MeV for 1 hour. After 4 days' decay the ratio of the peak heights in the gamma spectrum of zirconium-89 at 0.915 MeV and hafnium-175 at 0.34 MeV was determined. The advantage of using gamma activation rather than neutron activation is that the matrix does not have an enormous cross-section for the activating radiation and does not shield the impurity.lg Titanium has been determined in iron at the p.p.m. level by using the reaction 46Ti (y,n) 45Ti (half-life 3-1 hours).The titanium activity was separated chemically by extraction of the 8-hydroxyquinolinate in the presence of EDTA. The 3-hour half-life titanium-45 was more convenient in this respect than 5.8-minute titanium-51 from neutron activation.ls An interesting application concerns the determination of nickel in copper, which is limited in neutron activation by the reaction 65Cu (n,p) 65Ni, which generates nickel activity from the matrix. Gamma activation produced the nickel isotope 58Ni (y,n) 57Ni (half-life 36 hours), which is not subject to a nuclear interference. A dimethylglyoxime precipitation was used to separate the nickel activity.ls Chepel' and Skemarov20 have measured the chlorine as well as the fluorine content in halogen-containing hydrocarbon polymers.Specimens of PVC, Teflon and mixed polymers, 6 mm in diameter and 1.7 mm thick, were irradiated at 18 MeV with a betatron to produce the reactions 19F (y,n) 18F and 35Cl (y,n) 34Cl without production of interference from carbon-11.608 BAKER : GAMMA-ACTIVATION ANALYSIS [Analyst, Vol. 92 The 0.51-MeV annihilation gamma rays were counted and the decay curve resolved into its two components of half-lives 32 minutes and 1.7 hours. Pellets of lithium fluoride and sodium chloride were activated in separate irradiations to serve as standards; thin silver plates either side of each sample and standard served as beam monitors (silver-106, half-life 24 minutes). More recent work from Russia by Sulin5 and Berezin, Vitozhents, Sulin and Shornikov2l includes a comprehensive survey of the possible use of gamma-activation analysis for the analysis of rocks and minerals. The scope of the papers is limited to intact analysis by using varied energy of activating rays, decay-curve resolution, and gamma spectrometry to characterise the individual components.A useful compilation of nuclear data for most of the elements of the periodic table was included, together with estimates of the sensitivities likely to be achieved (100 to 1000 p.p.m.) in favourable cases with a betatron. Examples were given of successful analyses of model systems for the light elements, oxygen, nitrogen, carbon, phosphorus and sulphur and also for zinc, copper, zirconium by using synthetic mixtures with sample weights of up to 500 g.Mulvey, Cardarelli and MeyerZ2 used the reaction 12'1 (yln) 1261 (half-life 13.3 days) to determine iodine in biological material. Thermal neutron activation of biological materials usually results in high levels of activity due to sodium-24 and chlorine-38 and post-irradiation separation is necessary to isolate 25-minute iodine-128. By using gamma irradiation the analysis was performed directly on 0.9 per cent. sodium chloride media. After 3 days' decay period the only significant activity remaining was iodine-126 which could be measured by y-spectrometry. A limit of detection of 1 pg of iodine was claimed for a 90-minute irradiation at 22 MeV and 250pA. CONCLUSIONS Gamma-activation analysis has become established as a powerful technique for the determination of trace levels of carbon, nitrogen and oxygen in pure materials.It has also been applied to the determination of many other elements at trace levels. In general, the technique should be considered as a complement to neutron activation to be used when the nuclear properties of the impurity or the matrix favour the use of this method, for example, sensitivity, half-life, decay mode, interferences or shielding effect of the matrix. Except in the case of model systems, where the identification of the induced activity is un- equivocal, the characterisation, separation and measurement of activities is likely to be more difficult than with neutron-activation products. Two lines of research would appear to be profitable in this field. NOK-DESTRUCTIVE ANALYSIS- By irradiating at various energies for various times and following the decay of both annihilation radiation and the y-spectrum many results can be collected from which the composition of the sample can be deduced.Modern computers offer the best chance of extracting the maximum amount of information from these results. DESTRUCTIVE ANALYSIS- isotopes. separations could prove very useful in this field. Rapid dissolution procedures and chemical separations are required for the short-lived High temperature chemistry in various molten salt or metal baths, and gas-phase Appendix The sensitivity of any activation analysis method is very much dependent on the conditions of irradiation and counting. In Table I sensitivities have been calculated for ideal systems free from interference, for the following set of conditions which are realistic when using a large linear accelerator.Accelerator parameters : 5-pA, 30-MeV electron beam. 1 -mm tungsten radiator. Sample position 10 cm behind radiator. One half-life or 1 hour, whichever is the shorter. INTRINSIC SENSITIVITY OF GAMMA-ACTIVATION ANALYSIS Irradiation time :October, 19671 BAKER GAMMA-ACTIVATION ANALYSIS 609 Counting time : Counter: Commence immediately after irradiation, count for 1 half-life or 1 hour, whichever is the shorter. Well-type sodium iodide crystal with discriminator set to 100 keV. Limit of detection is defined as that weight of an element that will give a count equal to twice the standard deviation on the background count, which is 120 counts per minute.If a chemical separation is required before counting, the sensitivity must be halved for each half-life delay. Half-lives of less than 10 minutes are italicised. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Nuclide ‘2C 14N 1 6 0 lgF 20Ne 23Na 24Mn 2iA1 28Si 31P 32s 39K 40Ca 46Ti 50Cr “Mn 54Fe 58Ni 63Cu 75As igBr 97Rb 86Sr 99Y gOZr 93Nb 9 2 M ~ lo3Rh 107Ag 1131n 12%b 181Ta 197Au 35c1 59c0 642n 1 2 7 1 Half -life 20.3 minutes 10.05 minutes 2.1 minutes 1.8 7 hours 18 seconds 2.6 years 12 seconds 6.5 seconds 4.2 seconds 2-6 minutes 2-6 seconds 32.4 minutes 7.7 miflutes 0.9 second 3.1 hours 42 minutes 278 days 8.4 minutes 9 hours 37 hours 9-8 minutes 38 minutes 18 days 18.7 days 70 minutes 6.4 minutes 105 days 4.4 minutes 10.2 days 15.5 minutes 24 minutes 14-5 minutes 3.5 minutes 13 days 8.1 hours 9.5 hours 210 days TABLE I SENSITIV~TIES Limit of detection, pg 0.01 0.01 0.04 0.05 0.3 0.3 0.2 0.2 0.05 0.3 0.08 2.0 1.0 0.1 0.1 0.2 0.02 0.2 0.001 0.01 2.0 0.1 3.0 0.03 7 0.01 0.02 0.2 0.02 2.0 0.02 0.4 0.01 0-02 200 20 10 Limit reported J 0.01 pg in Be 0.1 pg chemical separation 1 p.p.m.in Be if pure J 1 p.p.m. in Be if pure -[ 1 pg chemical separation 0.1 pg in Be - - 1 p.p.m. chemical separation - - 1 p.p.m. chemical separation - - 0.1 pg in Hf - - 0.1 pg in Bi - - 1 pg in biological material - REFERENCES Kaminishi, K., J a p . J . A p p l . Phys., 1963, 2, 399. Lukens, H. R., jun., Otvos, J. W., and Wagner, C. D., I n t . J . Appl. Radiat. Isotopes, 1961, 11, 30. Goldstein, G., Analyt. Cham., 1963, 35, 1620. Koch, R. C., “Activation Analysis Handbook,” Academic Press, New York and London, 1960.Sulin, V. V., “Soviet Advances in Nuclear Geophysics,” Consultants Bureau, New York, 1965, Strauch, K., A . Rev. Nucl. Sci., 1953, 2, 105. Chepel’, L. V., Viting, B. I., and Chapyzhnikov, B. A, Instrums Exp. Tech., Pittsburgh, 1962,2,240. Voigt, A. F., Abu-Sumra, A., in “Proceedings of the International Conference on Modern Trends Basile, R., Hur6, J., L6v&que, P., and Schuhl, A., C.R. Hebd. Seanc. Acad. Sci., Paris, 1954,239,422. Albert, Ph., in “Proceedings of the International Conference on Modern Trends in Activation p. 129. in Activation Analysis,” College Station, Texas A and M University, 1965. Analysis,” College Station, Texas A and M University, 1961.610 BAKER 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. Albert, Ph., Engelmann, Ch., May, S., and Petit, J., C.R. Hebd. SBanc.Acad. Sci., Paris, 1962,254, Engelmann, Ch., Commissariat h 1’Energie A tomique, CEA-R-2559, Centre d’Etudes Nucldaires, -, in “Proceedings of the Symposium on Radiochemical Methods of Analysis, Salzburg, 1964,” Engelmann, Ch., Gosset, J., and Loeuillet, M., Bull. Soc. Chim. Fr., 1965, No. 2, 544. Beard, D., Johnson, R. H., and Hradshaw, W. G., Nucleonics, 1959, 17, No. 7, 90. Chepel’, L. V., Chapyzhnikov, B. A., andViting, B. I., J . Analyt. Chem. USSR, 1963, 18, No. 7, 749. Baker, C. A., Pratchett, A. G., and Williams, D. R., U.K. Atomic Energy Research Establishwent Schweikert, E., and Albert, Ph., in “Proceedings of the Symposium on Radiochemical Methods Oka, Y., and Kato, T., J . Chew. SOC. Japan, 1965, 86, 835. Chepel’, L. V., and Skemarov, F. V., Dokl. Akad. Nauk. SSR., 1964, 158, 976. Berezin, A. K., Vitozhents, G. Ch., Sulin, V. V., and Shornikov, S. I., in “Proceedings of the Symposium on Radiochemical Methods of Analysis, Salzburg, 1964,” I.A.E.A., Vienna, 1964, p. 361. Mulvey, P. F., jun., Cardarelli, J. A., and Meyer, R. A., “Symposium of Radioactive Sample Measurement Technique in Medicine and Biology,” I.A.E.A., Vienna, 1965. Received M a y 15th, 1967 119. Saclay, July, 1964. I.A.E.A., Vienna, 1964, p. 361. Report, AERE-R5363, H.M. Stationery Office, London, 1967. of Analysis, Salzburg, 1964,” I.A.E.A., Vienna, 1964, p. 323.