Use of Sugars and Hair for ESR Emergency Dosimetry

A. TRIVEDI’ and C.L. GREENSTOCK*

‘Dosimetric Research Branch and *Radiation Biology Branch AECL Research

Chalk River, Ontario KOJ 1JO

ESR spectrometry of sugars and biological samples is being evaluated for emergency personnel dosimetry. Sugars are near tissue-equivalent, universally available in pure form and produce a simple, reproducible, low background signal. Of the sugars tested, sucrose and dextrose are the most sensitive and the ESR signals are proportional to X- or y-ray doses over the range of 0.5-10 Gy. There is little dependence on radiation energy or dose-rate, and the ESR signals remain stable for long periods post-irradiation. Human hair samples show considerable variability and signal complexity creating difficulties in dose assessment.

KEYWORDS: ESR dosimetry; sugars; human hair; low dose; dose-rate; energy response.

INTRODUCTION

An effective dosimeter for measuring accidental radiation exposure should ideally be sensitive, reproducible and practical, (Miiller and Streffer, 1991) as well as capable of measuring mixed fields and different energies, dose rates and radiation qualities. Since most applications involve retrospective analyses under emergency situations (Balonov eC al., 1989; Ramalho ef al., 1988), samples should be tissue-equivalent, readily available and impervious to adverse environmental conditions. Most solid state devices have limitations based on the above criteria, and there is no universally accepted method of retrospective biological dosimetry. Conventional methods of dosimetry include calorimetry, thermo- or lyo-luminescence, radiography, spectrophotometry, and ESR (McLaughlin et al., 1989). Amino acids, particularly alanine (Regulla and Deffner, 1982), and sugars (Nakajima, 1988; Greenstock et al., 1991) are close to tissue-equivalent and represent good, non-invasive bio-organic surrogates to living tissue. In addition, sugars are cheap, stable, universally available in pure form, and provide simple, low-background, dose-dependent ESR signals associated with radiation-induced free radicals. The nondestructive nature of ESR detection also allows one to study trapped species in biological samples (Ikeya et al., 1984) such as bone (Caracelli et al., 1986), teeth (Saito, 1979), hair, nails and dry skin, but the complexity and post-irradiation instability of some signals, inhomogeneity and metabolic activity of the samples and the problem of individual variability, considerably complicate their quantitative analysis (Pass and Dust, 1982; Pass and Aldrich, 1985).

This ESR study of the radiation-induced free radicals in sugars is aimed at evaluating the practical application of this technique to emergency personal dosimetry (Nakajima, 1988, 1989; Nakajima and Otsuki, 1990). In this regard, experiments have been carried out to estimate the minimum detectable dose limit, the nature of the dose- and dose-rate responses, energy dependence and post-irradiation signal stability, as well as practical considerations of sample handling, signal reproducibility and calibration against an internal reference.

86 ESR chimetry and applutiorv

EXPERIMENTAL

The ESR measurements were performed on a Varian E-109 spectrometer operating at X-band microwave frequencies. A single, multi-purpose cavity (Varian E231) operating in the TE,, mode was used at ambient temperature. Samples (200 mg) were placed in 4 mm ID quartz tubes containing a reference sample of MnCI, (7.7 x 10F6 mol L-‘) sealed in a capillary tube. The microwave power was 10 mW and the field modulation at 100 kHz was 0.4 mT, producing a single over- modulated peak in irradiated sugar samples. The magnetic field was set at 0.336 T and the spectra were obtained in 8 min. over a scan range of 100 mT. The recorder output was interfaced to an IBM PC computer using EP-Ware software (Scientific Software, Bloomington, IL) for data acquisition and analysis. Relative and absolute signal intensities are calculated as signal heights (peak-to-peak) or integrated areas under the double integral curves. Error bars show the overall precision obtained from averaging results from experiments done in triplicate. Daily machine calibration was performed using a reference sample of strong pitch, and showed an overall reproducibility of + 10% _

Radiation sources used were a 6oCo Gammacell 220 (AECL Research) at a dose rate of 1.5 Gy min-’ and a variable energy Phillips X-ray generator (50 to 2.50 kV) using variable dose-rates from 3 to 15 mGy min-‘. Primary dose calibrations were made using lithium fluoride (LiF) thermoluminescent dosimetry (TLD).

RESULTS AND DISCUSSION

Samples (200 mg) of granulated sugar (sucrose) were irradiated to different doses, added to quartz sample tubes (4 mm 1D) in the presence or absence of a reference sample (a 1 mm ID sealed quartz tube containing 7.7 pmol L-’ of MnC12) (Nakajima, 1988) and the tubes placed in the ESR cavity. A typical result showing the ESR spectra of unirradiated and irradiated glucose superimposed on the six-line Mn”+ reference signals are shown in Figs. la and lb. The single-line signal of irradiated sugar, which represents the sum of all individual signals from C-centered sugar radicals, appears under high field modulation as a single broad line in the center of the Mn’+ spectrum. This si

9, nal,

after subtraction from the unirradiated sugar control (essentially zero), is normalized to the Mn- signal, in order to reduce errors associated with any variability in machine response. The intensity of this corrected, normalized signal, measured either as the peak-to-peak height or the area under the curve, is directly proportional to absorbed dose as determined by independent dose calibration using TLD of selected LiF chips. Using this method, the dose response has been determined for a variety of mono- and di-saccharides and other sugars (Fig. 2). All samples exhibit very low background levels, and essentially the same ESR signal (g-2.003) upon irradiation. The most sensitive responses were consistently obtained with samples of sucrose and dextrose (glucose), and for this reason these sugars were selected for all subsequent experiments. Figure 3 shows the ESR results of dose-response experiments plotting corrected, normalized signal intensity as a function of dose (60Co y-rays) for irradiated dextrose and sucrose. As found in Fig. 2, the shape of the ESR spectrum of irradiated sugar does not change with dose. Sucrose is generally a more sensitive indicator of exposure than dextrose, but for both sugars the free radical signal increases linearly with dose above - 1 Gy. ESR can be used to estimate doses to sugars below 1 Gy, but the data are less accurate and follow a sigmoidal dose-response. Special efforts involving signal-averaging and slow, multiple scanning are necessary to produce good data in the low dose-range, and for practical purposes, the present lower limit of detection is -0.2-0.5 Gy. To increase the reproducibility and reliability, efforts must be taken to control variables that influence machine response, measure replicate samples, and monitor the ESR instrument on a d:zily basis by routinely calibrating signal intensity against a reference standard of strong pitch.

In the case of an emergency, a person can be exposed not only to different doses of radiation, but also to radiation of different dose rates and of different qualities or LETS. In order to test the dose rate effect in ESR dosimetry, sugar samples were irradiated with doses over a range of dose rates from 3 to 15 mGy min-‘.

6oCo y-rays to four different total As can be seen from Fig. 4, the signal

response is essentially independent of dose rate over the measured range. To test the energy response, sugar samples were irradiated with X-rays from 50 to 200 kV and the results compared

ESR dosimetry and applications 87

with 13’Cs and %o y-irradiations to the same total absorbed dose. TLD of LiF chips was used as the operational method of estimating dose, converting from primary ion chamber exposure measurements to those for absorbed dose. Each TLD measurement was corrected for the difference between the effective “Z” of the ESR sample and that of the LiF dosimeter using the ratios of their mass energy absorption coefficients at each energy.

Fig. 1. ESR spectra of a) unirradiated and b) irradiated glucose (1 Gy) superimposed on a Mn2+ reference sample (7.7 pm01 L-l). The sugar radical signal (central single line with g value of -2.003), the intensity of which is proportional to absorbed dose, appears between the third and fourth lines of the six-line Mn2+ spectrum.

z400

2ooo

1.600

1 .zw

0.800

0.400

O.OW

-O.XhNF-- Rib

Fig. 2. A chart showing the relative radiosensitivity of different sugar samples under equivalent conditions (a fixed dose of 5.4 Gy. 300 mg samples in a standard 4 mm ID quartz ESR tube).

Data in Fig. 5 suggest a slight dependence on radiation energy (and therefore LET) of the ESR signal for a given absorbed dose, especially when comparing 200 kV X-rays with 137Cs and %o y rays at 667 and 1250 keV, respectively.

88 ESR dosimetry and applications

The overall variation in energy response represents approximately &20% about the mean, which is not a major source of error when considering the uncertainties associated with nuclear accidents or other emergency situations requiring retrospective analyses under less than optimum conditions.

Another area of investigation focused on post-exposure signal stability. In the case of retrospective analysis it is important to search not only for a convenient, sensitive reproducible indicator of absorbed dose, but one that retains its signal indelibly for long periods so that it acts as an integrating life-time dosimeter and is suitable for determining past exposures even at long periods of time after the exposure.

5

c

a0 2 6 6 10

ABSORBED DOSE (Gyl

Fig. 3. A plot of ESR signal intensity against absorbed dose for y-irradiated dextrose (0) and sucrose (*), showing a linear dose response from 1 to 10 Gy.

6, I

0 000 0 004 0 008 0 012 0 016 0020

DOSE RATE IGy mm ')

Fig. 4. A plot of the dose-rate dependence of the ESR signals from dextrose y-irradiated to 1 Gy (o), 3 Gy (m), 5.4 Gy (O), and 10 Gy (0).

Fig. 5. A plot of ESR signal intensity for dextrose samples exposed to 5.4 Gy as a function of photon energy for X-rays (A) (So-200 kV), and y rays from 13’Cs (v) (667 keV) and 6oCo ( +) (1250 keV).

ESR dosimetry and applications 89

Samples of irradiated sugars were stored for up to six weeks after irradiation and monitored on a daily basis (Fig. 6). This is possible with ESR dosimetry in that, unlike TLD, X-ray film, ion chambers and neutron track-etch systems, samples can be repeatedly measured without erasing or disturbing the original radiation signal. As the data in Fig. 6 show storing samples under ambient conditions leads to a slow, predictable fading (TIA - 40-50 days) which may be quantitated and used to extrapolate back to estimate the initial exposure dose provided the decay rate and its dependence on dose and environmental factors is reproducible. Keeping samples cool, dry and dark may increase the post-exposure stability, so that samples many years old may be used to reveal useful results. This is not true of hematological, cytological, genetic or biochemical indicators of dose (Miiller and Streffer, 1991).

DAYS AFTER IRRADIATION

Fig. 6. A plot of ESR signal intensity of dextrose samples exposed to 1 Gy (o), 3 Gy (w), 5.4 Gy (O), and 7 Gy (0) as a function of post-irradiation time.

2

1,s

1.6

1.7

1.6

1.5

1.4

13

12

1.1

R&bus Intenslly per 1

Gram (kb Unts) 0.6

06 cT~==-w O7

06

0.5

04

0.3

02

01

0

BRUNETTE D)(m LTBROWN GREY RED BLONDE TmdHr

Em k9 ml zolm Gzi 4ofnm m 70mm 1440 ml”

Fig. 7. Timedependent ESR signals per unit weight in irradiated hair samples. Unirradiated controls are shown together with results for six different hair colors for a single 3 Gy dose.

Using a high modulation frequency, a clear ESR signal is observed in human hair samples after irradiation as previously noted by Brady et al. (1968). However, the use of hair for ESR dosimetry is complicated by a high background signal, particularly in dark hair, and a time-dependent variation in the post-irradiation response. Figure 7 shows the ESR signals obtained before and after a radiation

90 ESR dosimetry and applications

dose of 3 Gy for six different hair colors. In all cases, the ESR signals have been normalized to a unit mass (1 g) of hair, but it is clear that in unirradiated samples the background signals vary from very small in the case of blonde hair to values equivalent to several Gy for dark brown hair. A further complication arises from the time-dependent changes in the ESR spectrum of the irradiated samples. The results shown in Fig. 7 were analyzed manually before the computerized analysis system was installed, and are considered tentative. They indicate that without any pretreatment of the samples, consideration of cutting artifacts (Chandra and Symons, 1987) or attempts to differentiate between unirradiated and irradiated signals, a dose of 3 Gy can be readily detected in all but the light brown hair samples. In blonde hair which appears to be the most useful bio-indicator, the radiation response was maximal 20 minutes after irradiation; in red hair it occurred after approximately one hour; and for grey hair and brown hair the response was higher after 24 hours. Clearly a more detailed investigation, taking into account some of the confounding factors, is necessary before ESR dosimetry of human samples such as hair, teeth or nails can be fully realized as a useful, practical indicator of biological dose for emergency or life-time exposures.

REFERENCES

Balonov M.I., Keirim-Markus I.B., Margulis U.Y. and Osanov D.P. (1989) Methods for retrospective determination of absorbed doses in the human body resulting from external and internal exposure. In: Medical Aspects of the Chernobyl Accident. International Atomic Energy Agency, Vienna, 1989. IAEA TecDoc #516, pp 203-215.

Brady J.M., Aarestad N.O. and Swartz H.M. (1968) In vivo dosimetry by electron spin resonance spectroscopy. Health Phys., 15, 43-47.

Caracelli I., Terrile M.C. and Mascarenhas S. (1986) Electron spin resonance dosimetric properties of bone. Health Phys., SO, 259-263.

Chandra H. and Symons M.C.R. (1987) Sulphur radicals formed by altering ar-keratin. Nuture, Lond., 328, 833-834.

Dosimetry for Radiation Processing (1989) (W.L. McLaughlin, A.W. Boyd, K.H. Chadwick, J.C. McDonald and A. Miller, eds.) Taylor and Francis, London.

Greenstock C.L., Trivedi A., Kolios M., Mehta S. and Bonnot I. (1991) Using bio-organic substances for emergency dosimetry. In: Radiation Research: A lbentieth-Century Perspective Vol. 1, (J.D. Chapman, W.C. Dewey and G.F. Vhitmore, eds.) p 319, Academic Press.

Ikeya M., Mikajima J. and Okajima S. (1984) ESR dosimetry for atomic bomb survivors using shell buttons and tooth enamel. Jpn. J. Appl. Phys., 23, L697-L699.

Miiller W.-U. and Streffer C. (1991) Biological indicators for radiation damage. Znt. J. Radiat. Biol., 59, 863-873.

Nakajima T. (1982) The use of organic substances as emergency dosimeters. Znt. J. Appl. Radiat. Zsot., 33, 1077-1084.

Nakajima T. (1988) Sugar as an emergency populace dosimeter for radiation accidents. Health. Phys., 55, 951-955.

Nakajima T. (1989) Possibility of retrospective dosimetry for persons accidentally exposed to ionizing radiation using electron spin resonance of sugar and mother-of-pearl. Br. .I. Rudiol., 62, 148-153.

Nakajima T. and Otsuki T. (1990) Dosimetry for radiation emergencies: Radiation-induced free radicals in sugar of various countries and the effect of pulverising in the ESR signal. Appl. Radiat. Zsot., 41, 359-365.

Pass B. and Dust J. (1982) Cumulative dose as measured by ESR in dental enamel. Dent. Res., 61, 195.

Pass B. and Aldrich J. (1985) Dental enamel as an in vivo dosimeter. Med. Phys., 12, 305-307. Ramalho A.T., Nascimento A.C.H. and Natarajan A.T. (1988) Dose assessments by cytogenetic

analysis in the Goiania (Brazil) radiation accident. Radiat. Prot. Dosim., 25, 95-100. Regulla D.F. and Deffner U. (1982) Dosimetry by ESR spectroscopy of alanine. Znt. 1. Appl. Radiut.

zsot., 33, 1101-l 104. Saito K. (1979) Study of an asymmetric ESR signal in X-irradiated tooth enamel. CaZg Tissue. Zu.,

29, 95-99.