Scribd Logo
Upload

Welcome to Scribd!

  • 12 views

    Correction Factor

    Uploaded by

    Lucas

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd

    Correction Factor

    Uploaded by

    Lucas
    12 views14 pages

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Download as pdf
    12 views14 pages

    Correction Factor

    Uploaded by

    Lucas

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Download as pdf
    You are on page 1of 14
    ‘Ancient TL vol 16 No.2 1998 37 Dose-rate conversion factors: update Grzegorz Adamiec' and Martin Aitken? "Research Laboratory for Archaeology, Oxford University, 6 Keble Road, Oxford OX1 3QJ, UK. E-mail: grzegorz.adamiec@rlaha.ox.ac.uk Le Garret, 63 930 Augerolles, Puy-de-Déme, France. E-mail: martin.aitken@linacre.ox.ac.uk (Received 6 June 1998 : in final form 1” september 1998) Abstract : Dose-rate conversion factors relevant to luminescence and electron spin resonance dating have been derived from values for the energy carried by radiations emitted during nuclear transformations given in the current ENSDF (Evaluated Nuclear Structure Data File). For beta and gamma radiation the factors are a few percent lower than previously used. For the effective alpha dase-rate it is mare appropriate to use an approach based on article ranges and resultant values are given. Introduction In trapped charge dating evaluation of dose-rate is of equal importance to that of palaeodose. In most approaches the dose~ate is derived from measurement of radioelement concentration (or activity) by means of conversion factors, and because these factors are outside the ken of the dating specialist there is a tendency to take them for granted. as writen in stone. In fact they are based on nuclear dats tables and these are in an on-going process of refinement. The tables are dauntingly complex and fortunately for the dating communities @ formulation is available in which values for the radiation components of the energy release are given specifically from the dosimetry point of view, The overall tables are known as the Evaluated Nuclear Structure Data File (ENSDF) and the dosimetry- orientated formulation known as the Medical Interal Radiation Dose (MIRD) format. These are available on the Brookhaven National Laboratory Web site (hupii/wwwsnnde.bni-gov.) and the Intemational Atomic Energy Authority Web site ¢http:/www- nds.iaea.or.at); except for Table 7 and part of Table 8, the values of the present paper are based on MIRD downloaded on 22nd June 1998. Details about ENSDF has been given by Martin & Blickert-Toft (1970) and by Burrows (1998). Earlier use of ENSDF was made by Nambi & Aitken (1986) through the dosimetry-orientated format published by the International Commission on Radiation Protection as ICRP-38. Comparison with the values of Nambi & Aitken (1986) is summarized in Tables 4 and 8; since ICRP-38 was based on ENSDF as of 1978 it is not surprizing that there are differences. Prior use of nuclear data tables has been rade by Aitken (1974), Aitken & Bowman (1975), Bell (1976, 1977, and 1979), Carriveau & Troka (1978). Subsequent to Nambi & Aitken (1986), re- assessments have been made by Lititzis & Kokkoris, (1992) and by Ogoh et al. (1993), as discussed in Appendix A, ‘The paper is primarily written with luminescence dating in mind but it is largely applicable to dating by electron spin resonance as well. The data Tables 1,2 & 3 show the energy emission values and halflives obtained from the Brookhaven Web site mentioned above for the three radioactive series; Table 4 gives comparison between these data and those presented by Nambi & Aitken (1986). As wil be seen there are appreciable decreases inthe totals for beta radiation and gamma radiation from the Th- 232. series, the principal contributor to these decreases being Ac-228. It is not easy to pinpoint particular measurements that are respensible for decreases; this is because the Values are usually based on comprehensive schemes of nuclear energy levels incorporating a number of relevant measurements rather than on single direet measurements. In Table 5 the emission values have been converted into doserate, This ison the usial infinite matrix assumption thatthe dose-rate is equal to the rate of energy emission per unit mass, implying that there is homogeneity both in radioactive content and 38 in absorption coefficient. The value given for “full series’ is for the case of radioactive equilibrium, ic. the disintegration rate of each daughter is equal to that of the parent, except where modified by branching; the value labelled ‘pre-Rn’ corresponds to 100% escape of radon from the two principal series. This and other types of disequilibrium have been discussed by Krbetschek eral. (1994) and Olle er al. (1996), among others. Data for potassium and rubidium are given in Table 6. For potassium, the value for gamma radiation is unchanged from that of Nambi & Aitken (1986) but the value for beta radiation is lower by 4%. For rubidium the beta value is lower by a factor of 1.26; however, except forthe internal dose-rate in coarse grains of potassium feldspar, the change is ‘unimportant. The concentration ratio of 200:1 between potassium and rubidium used in Table 6 is an arbitrary choice that is within the range of ratios, encountered in samples. ‘The alpha particle contribution ‘The luminescence induced by alpha particles per gray of deposited energy is dependent on particle energy, decreasing as the particle energy decreases; this is in contrast t0 the case for the lightly-ionizing radiations, beta and gamma, for which the effectiveness is independent of energy. Thus in the A-value system, developed by Zimmerman (1971), it is necessary to know the precise energy of the alpha particles used for measuring k (the ratio of alpha effectiveness compared to that of beta or gamma radiation, effectiveness being the luminescence induced per ray) and also to make allowance for that fact thatthe alpha particles received during burial have a spectrum that spreads from zero to 8.8 MeV. To do this Zimmerman (1971) introduced K(effective) which for quartz he calculated, on the basis of range-energy data then available, as being less than &(3.7 MeV) by factor of 0.86 forthe thorium series and 0.80 for the It was implicit in the results of Zimmerman (1971) that, to a first approximation, the luminescence induced per unit length of alpha particle track is independent of particle energy. This was confirmed by Bowman (1976) using the alpha beam from a Van de Graaff generator and in respect of ESR by Lyons & Brennan (1989). This approximation is the basis of the three track-length approaches: the a- value system (Aitken & Bowman 1975), the b-value system (Bowman & Huntley 1984), and the commidirectional flux system (Valladas & Valladas 1982). Because the alpha count-rate from a thick layer of sample is proportional to the length of track Ancient Tvl 16 No.2 1998 generated in the sample by its constituent thorium and. uranium series this count-rate can be used to give direct determination of the alpha dose-rate during burial. Also required is measurement, for the sample concemed, of the {uminescence induced by unit length of track from an artificial source. By comparing this luminescence with that induced per ‘gray of beta (or gamma) radiation, alpha count-rate can be converted to effective alpha dose-rate (see Appendix B). It is ‘effective’ in the sense of being appropriate for use in the usual age equation in which the palacodose is expressed in terms of grays of beta (or gamma) radiation Although the track-length approaches are conceptually more difficult than the k-value system, the latter has the basic disadvantage that there is a need to have an accurate knowledge of the particle energy used in measurement of sample sensitivity. It is not just a matter of energy of the particles emerging from the alpha source; strictly, allowance should also bbe made for the degradation in energy as the particles penetrate the sample. Hence the preferred values for the effective alpha contribution are those based on track-length as given in Table 7. Concluding remarks Although the 8% decrease in the gamma dose-rate from the thorium series is substantial, there is considerable dilution when an actual context is considered, as illustrated in the example of Table 8. For fine-grain dating the total dose-rate is 2.11 according to the present paper compared to 2.18 according to Nambi and Aitken (1986), a decrease of 3%. For coarse-grain dating the values are 1.60 and 1.65 respectively, a decrease of 3% as well. For contexts in which there is dominance by thorium or potassium, the decreases will be greater. A historical survey of some published conversion factors is given in Appendix A. It will be seen that, for the specimen context of Table 8 at any rate, the totals of the present paper are a few percent Tower than all previous ones. However there is no ‘case for diluting the present factors by averaging because the earlier assessments are essentially based ‘on the progenitors of the present ENSDF and these are now superceded. It is reassuring t0 note that, at any rate for contexts not too dissimilar 10 the specimen one considered, no dramatic revision is needed in respect of ages based on earlier conversion factors: the highest fine-grain totals, those of Aitken (1974) and Bell (1979), are some 4% higher than that for the present paper; the highest coarse-grain total, that of Aitken (1974) is some 6% higher. “Ancient TL vol 16 No.2 1998, OF course, there are other influences on dose-rate accuracy besides the basic conversion factors, such as related to moisture content, grain- size, radioactive disequilibrium etc. These tend to introduce an uncertainty that is dominant and so the shifts consequent on changing to the factors of the present paper will usually be contained within error limits. Nevertheless, the case for utilization of these factors is just as strong as the case for using state-of- the-art refinements in palacodose determination. To do otherwise runs the risk of drifting into a situation where there will be two ages for a sample, one based ‘on ‘set in stone” factors and another based on the best available data from nuclear physics. Acknowledgements ‘We are sincerely grateful to Dr. T. W. Burrows of the Brookhaven National Laboratory for his patient explanation to us of various features of ENSDF. In addition, all of us in trapped charge dating owe an immense debt of gratitude to the multitude of nuclear physicists on whose work ENSDF is based. G.A. would like to express his gratitude to the Research Laboratory for Archaeology and the History of Art and to the Overseas Research Students ‘Awards Scheme for supporting his work. Finally, we are appreciative of a number of valuable suggestions made by the Reviewer. 39 40 Ancient TL vo 16 No. 1998 Table 1: Energy release in the Th-232 decay series Isotope Haltlife | Alpha | Beta fTh-232. 14,05 Ga 4.00|_— [Ra-228 575a)__ = 0.01 0.001 JAc-228 615h) 0.413] 0.85. Th-228 191d 331 0.019] 0.003] Ra-224 3.664 5.51] 0.002] 0.010] Rn-220 55.6 s 6.28] 0.001] [Po-216 0.145 s| 6.77] = <5x10"| Pb-212 10.6h[ _- 0.173] 0.144 Bi-212 60.6 m| 214 0.502] 0.103] Po-212 (0.641) 0:299 ms 3.63] : FTI-208 (0.359) 3.05 m[ 0.209] 1.205 Total 35.7 1.33 2.39] [Pre-Rn total 149] 0.444) 0.868} Notes for Table 1 1. Energies are given in MeV and represent the energy emitted per disintegration. 2. Non-SI units used in half-lives: 1a=1 year; Id=1 day; { hel hour; 1 m=1 minute 3, Branching ratios are shown in parenthesis against the radioelements in the branches; associated values given for energy release are after adjustment for branching. Note that the branching also affects the energy release of the radioelement at which the bifurcation occurs; thus the value given for the alpha release by Bi-212 is 35.9% of the full energy —- because TI-208 is formed by alpha emission from Bi-212. 4, Internal conversion and Auger electrons are included with the beta component (which is the average beta energy rather than the maximum). X-rays and annihilation radiation are included with the gamma component. Alpha recoil cenergies have not been included; this is on the basis that the contribution to luminescence will be negligible on account of the resultaat ionization density being even higher than for the alpha particles themselves; the total recoil energy does not exceed 2% of the alpha total for any series. Neutrinos have been ignored because of their very low absorption in matter. 5. A dash indicates that no radiations of that type are listed in MIRD. 6. At-216 has been omitted since its contribution to the total energy is insignificant. Ancien T. vol 16 No 1998 al Table 2: Energy release in the U-238 decay series. Isotope Half-life | Alpha Beta ‘Gamma, [U-238, 4.468 Gal 4.19] 0.007] 0.0011] [Th-234 zante 0.060] 0.0093] [Pa-234m_ 1.17 ml = 0.818] 0.0161] |Pa-234 (0.0016) 6.75 hl = 0,001 0.0023} lU-234 2.46 kal 4.68] 0.012] 0.0015] [Th-230 75.4 kal 458 0.013] 0.0014] [Ra-226 1600 al 477|___0.0038| 0.0074] [Rn-222 3.82 dl [549] ae 0.0004] |Po-218 3.11 ml 6.00] = = [Pb-214 26.8 m| : 0.294] 0.2521 Bi-214 199m] <0.005| 0.652| 1.4814] [Po-214 164 ms| 7.68] <5x10"|___0.0001 [Pb-210 22.3 al 0.033 0.0047] [Bi-210 5.01 dl <0.005| 0.389] - [Po-210 138.4 | Saif

    You might also like