The New Operational Quantity Ambient Dose in Environmental Radiation Monitoring

Featured Application

The paper indicates that current environmental dosimeters must be modified for responding appropriately to low-energy photons when the new operational quantity Ambient Dose $H^{*}$ is introduced.

Abstract

In ICRU Report 95, the International Commission on Radiation Units and Measurements (ICRU) has proposed jointly with the International Commission on Radiological Protection (ICRP) new operational quantities for external radiation. The quantity for environmental monitoring is ambient dose $H^{*}$. This paper analyses how present ambient dosimeters and monitors for photons would respond to the new operational quantity. The results show that passive environmental dosimeters designed for determining ambient dose equivalent and capable of registering photons to energies as low as 10 keV show a strong overestimate of ambient dose in the energy interval from 15 keV to approximately 50 keV. Active dosimeters exhibit a low-energy cut-off and are not affected by the overestimation. In spectrometer-type ambient monitors, the operational quantity can be calculated by multiplying the unfolded count-rate spectrum with the conversion coefficient for $H^{*}$.

1. Introduction

The International Commission for Radiological Protection (ICRP) introduced in 1990 the effective dose E as the protection quantity for whole-body exposure at low doses and low dose rates [1,2]. Effective dose is defined as a weighted average of organ and tissue doses $D_{\mathrm{T},\mathrm{R}}$ over all radiation types R and all sensitive tissues T,

$$E=\sum _T{w}_{\mathrm{T}}\sum _R{w}_{\mathrm{R}}{D}_{\mathrm{T},\mathrm{R}}$$

(1)

In this equation, $w_{\mathrm{R}}$ and $w_{\mathrm{T}}$ are age- and sex-averaged radiation and tissue weighting factors, taking into account the relative effectiveness of radiation types and the radiation sensitivity of organs. Before the introduction of effective dose as a protection quantity, the radiation effectiveness was described by a quality factor $Q\left(L\right)$, depending on linear energy transfer L. Linear energy transfer indicates how much energy the particle loses per unit path length in its interactions with matter. In other contexts, L is also called stopping power. It is a purely physical quantity, whereas radiation weighting factors $w_{\mathrm{R}}$ are based on the experimentally determined relative biological effectiveness of radiation [2].

Radiation protection legislation in most countries demands that the effective dose of workers is determined, optimised and limited. Due to its definition over an extended body, effective dose cannot be measured. The values of E are calculated with the help of Monte Carlo radiation transport codes in anthropomorphic phantoms defined in ICRP Publication 110 [3]. Reference conversion coefficients between physical radiation field quantities fluence $\Phi$ (the number of particles passing a unit surface) or kerma in air $K_a$ (the kinetic energy initially released during the interaction of a neutral particle with matter, here air) and effective dose are published in ICRP Publication 116 [4]. The conversion coefficients are calculated for broad, unidirectional and monoenergetic radiation fields exposing the whole anthropomorphic phantom from standard directions: anterior–posterior (AP), posterior–anterior (PA), left and right lateral (LLAT, RLAT), rotational (ROT), and isotropic (ISO).

Practical radiation protection is based on measured values of potential radiation detriment, and therefore the International Commission on Radiation Units and Measurements (ICRU) has introduced operational quantities which are defined in a point and shall provide slightly conservative estimates of the protection quantity [5]. The operational quantity for the measurement of ambient radiation at workplaces and in the environment is ambient dose equivalent $H^{*}$(10). It is defined in the so-called ICRU-sphere, a 30 cm diameter sphere made from ICRU 4-component tissue, immersed in an expanded and aligned radiation field. In this idealised radiation field, all radiation vectors come from a single direction, and the field is homogeneous over the sphere. The ambient dose equivalent is defined as the absorbed dose in 10 mm depth of the sphere on the axis of the incoming radiation field, multiplied by the quality factor $Q\left(L\right)$ introduced above. Historically, the present operational quantities, among them ambient dose equivalent $H^{*}$(10), were defined by ICRU to depend on the $Q\left(L\right)$-formalism previous to the ICRP changing to the $w_{\mathrm{R}}$-formalism. Conversion coefficients between physical radiation field quantities (fluence, kerma in air) and $H^{*}$(10) are found in ICRP Publication 74 [6] and ICRU Report 57 [7]. It is customary that the value of the operational quantity is used as a surrogate for the value of the effective dose in the optimisation of radiation protection if the measured values are well below the dose limits.

The conceptual difference in the definitions of the protection quantity E and the operational quantity $H^{*}$(10) makes the system of radiation protection dosimetry unnecessarily complicated. Furthermore, a significant overestimation of the protection quantity E by the operational quantity $H^{*}$(10) is observed for photons for energies of less than 50 keV (Figure 1).

To overcome these difficulties, ICRU has proposed new operational quantities in its Report 95 [8]. The new operational quantities, personal dose $H_p$ and ambient dose $H^{*}$, are calculated with the same formalism as effective dose, using the numerical anthropomorphic phantoms and the weighting factors $w_{\mathrm{R}}$ and $w_{\mathrm{T}}$. The ambient dose $H^{*}$ is defined for a given particle energy $E_p$ as the maximum of effective dose over the incident radiation directions (DIR = AP, PA, LLAT, RLAT, ROT, ISO) as published in ICRP Publication 116 [4],

$$H^{*}\left(E_p\right)=\underset{\mathrm{DIR}}{max}\left(E(\mathrm{DIR},E_p)\right)$$

(2)

Due to the close relationship with the protection quantity E, the new operational quantity $H^{*}$ is numerically very close to it. At low photon energies, $H^{*}$ is in effect identical to E(AP), while it becomes identical to E(PA) from about 7 MeV (Figure 2). At even higher energies, it assumes the values of E(ISO).

Figure 3 shows the difference between the operational quantities $H^{*}$(10) and $H^{*}$ for photons. At energies above 200 keV, the new operational quantity is approximately 15% lower than the old one. At energies below 50 keV, the difference between the conversion coefficients is a factor of five for mono-energetic radiations. In practice for X-rays, the spectral-averaged conversion coefficients differ by about a factor of approximately two [9]. A similar behaviour can be observed for the ambient dose (equivalent) conversion coefficients for gamma rays from radionuclides, where the average is taken over all emitted photons [10].

Present ambient dosimeters and radiation monitors are calibrated in operational quantity $H^{*}$(10), and their design is optimised to represent this quantity as closely as possible. This paper analyses how present ambient dosimeters and photon monitors would respond to the operational quantity $H^{*}$, and it draws conclusions for the design of future dosimeters and monitors which are tailored to the new quantity. Previous findings on ambient photon dosimeter response to $H^{*}$ are published in [11,12], albeit for a more restricted range of energies.

2. Materials and Methods

The response R of a dosimeter or monitor is defined as the quotient of the value indicated by the dosimeter M and the conventional true value of the measured quantity C,

$$R={\displaystyle \frac{M}{C}}$$

(3)

The response of a detector is the inverse of its calibration factor. In the case of an ambient dosimeter measuring the ambient dose equivalent $H^{*}$(10) in a radiation field, this can be written

$$R_{\mathrm{ICRU}57}={\displaystyle \frac{M}{H^{*}\left(10\right)}}$$

(4)

The energy-dependent response to photons of an ambient dosimeter or monitor is determined with reference radiations from International Standard ISO 4037-3:2019 [13]. The standard describes radionuclide sources (${}^{137}\mathrm{Cs}$, ${}^{60}\mathrm{Co}$) and X-ray spectra. It is customary to use the spectra from the “narrow” N-series for performance testing and calibration of ambient dosimeters and monitors.

The change in the operational quantity to $H^{*}$ implies a change in the conventional true value in a given radiation field from $H^{*}$(10) to $H^{*}$, while the indication of the dosimeter depending on the physical characteristics of the radiation field will remain unchanged

$$R_{\mathrm{ICRU}95}={\displaystyle \frac{M}{H^{*}}}$$

(5)

Expanding the last expression with $H^{*}$(10) and introducing the dose conversion coefficients $h^{*}$ and $h^{*}$(10) yields

$$R_{\mathrm{ICRU}95}={\displaystyle \frac{M}{H^{*}\left(10\right)}}{\displaystyle \frac{H^{*}\left(10\right)}{H^{*}}}=R_{\mathrm{ICRU}57}{\displaystyle \frac{H^{*}\left(10\right)}{H^{*}}}=R_{\mathrm{ICRU}57}{\displaystyle \frac{h^{*}\left(10\right)}{h^{*}}}$$

(6)

The dose conversion coefficients relate the operational quantity to a physical field quantity, for photon radiation usually kerma in air, $K_a$. For the X-ray spectra, the conversion coefficients are averaged over the radiation spectrum. Dose conversion coefficients $h^{*}$(10) and $h^{*}$ are given for monoenergetic radiations in [7,8] and for radionuclides and X-ray spectra in [9,13], respectively.

Published data for the energy-dependent response of environmental dosimeters and monitors to photons is taken from scientific publications and manufacturer’s data. Data are usually represented graphically as a response function in which the response to one of the spectra from the ISO-4037-3 narrow series is marked with an entry at the average energy of the spectrum. Table 1 lists the survey instruments and dosimeters for which response functions were obtained, and their sources.

The plots of response functions published in the references were digitised with the PlotDigitizer program [14]. Here, the user marks data points on a 2-dimensional plot, and the program converts the geometric coordinates of the data point into its physical coordinates, determined by the scales of ordinate and abscissa. The digitised energy values on the ordinate are identified with the closest mean energy of X-spectra in the N-series, with the photon energy of ${}^{137}\mathrm{Cs}$ (662 keV) or the average photon energy of ${}^{60}\mathrm{Co}$ (1250 keV). With this, the energy values have no uncertainty. The values on the abscissa can be determined with a precision of about 2% on a linear scale and 5% on a logarithmic scale. The estimates obtained for the spectral response to ambient dose equivalent $H^{*}$(10) were converted with the help of Equation (6) to response values for the new quantity ambient dose $H^{*}$.

Table 1. Data sources for energy-dependent response functions for ambient dosimeters and monitors.

Ambient Dosimeter or Survey Instrument	Reference
Automess 6150 AD6 Geiger–Müller survey monitor	[15]
Centronics IG5-A20 2 MPa Ar ionisation chamber	[16]
Seibersdorf environmental dosimeter	[17]
BEOSL environmental dosimeter	[18]
2 cm sphere	[19]

Table 1. Data sources for energy-dependent response functions for ambient dosimeters and monitors.

Ambient Dosimeter or Survey Instrument	Reference
Automess 6150 AD6 Geiger–Müller survey monitor	[15]
Centronics IG5-A20 2 MPa Ar ionisation chamber	[16]
Seibersdorf environmental dosimeter	[17]
BEOSL environmental dosimeter	[18]
2 cm sphere	[19]

3. Results

3.1. Active Dosimeters

The two active dosimeters (or radiation monitors) analysed are from the families of Geiger–Müller (GM) counters and ionisation chambers, two widely used monitor types in environmental surveillance networks. Figure 4 shows the energy-dependent response function for the old and the new operational quantity for one GM-counter [15] and one ionisation chamber [16], which are chosen to represent their types.

The Automess 6150 AD GM-counter [15] is a hand-held survey instrument. The energy-dependent response of the 6150 AD for $H^{*}$ and $H^{*}$(10) (Figure 4a) is, as anticipated, approximately 15% higher in the new operational quantity over the whole energy range between 50 keV and 1250 keV. This difference in response can be corrected by an adjustment of the calibration factor of the instrument. The 6150 AD is not capable of registering photons with an energy lower than 50 keV due to its rugged metal casing designed for fieldwork, and it is therefore not affected by the overestimate of $H^{*}$ by $H^{*}$(10) below this energy.

Similar conclusions can be drawn for the high-pressure ionisation chamber Centronics IG5-A20 [16] (Figure 4b). The solid steel envelope, required to contain the argon gas under 2 MPa pressure, reduces the response to photons with energies below 80 keV so that a reasonable measurement is no longer possible. For the energies from 100 keV on, the difference between the indications of $H^{*}$ and $H^{*}$(10) is, as expected, 15 %. In radiation fields where such energies dominate, the chamber can be recalibrated to respond correctly to the new operational quantity $H^{*}$.

3.2. Passive Environmental Dosimeters

Passive environmental dosimeters are designed for long-term surveillance tasks in situations where one does not expect an excess of the dose limit for the public to occur. They are deployed, for example, at the site boundaries of nuclear or accelerator facilities, where they serve mainly to record the good performance of radiation protection programmes. The dosimeters consist of thermoluminescent (TL) or optically stimulated luminescent (OSL) detectors in a housing providing protection against influences from the weather and filtering to adjust the energy response of the detector. The energy-dependent response functions of three passive environmental dosimeters for $H^{*}$(10) and $H^{*}$ are shown in Figure 5.

A commercially available model is the $H^{*}$(10) TL Area dosimeter Seibersdorf [17], designed and marketed by the eponymous Austrian research centre. It is based on four TLD detectors in a “Harshaw” card, placed in a 6 cm by 6 cm cylindrical PMMA moderator covered by a thin aluminium shell. The BEOSL, an OSL dosimeter using beryllium as sensitive material [18], is marketed in a slightly different variant by Mirion Dosimetrieservice (AWST) in Munich, Germany. A Swedish publication [19] refers to a design for which the author could not find a commercial source.

The three dosimeter types are optimised for measuring ambient dose equivalent, $H^{*}$(10). This is achieved by covering the sensitive elements with a material layer which is equivalent to 10 mm of tissue. Since the dosimeters are relatively small, they cannot have the same response characteristics as the 30 cm diameter ICRU sphere. This is corrected by the choice of materials (PMMA with a density exceeding 1 g/cm³, different metallic shielding layers), leading to the response curves in red in Figure 5a–c. When applying Equation (6) to obtain the $H^{*}$-response, one can observe for all three passive dosimeters at energies above 100 keV the 15% difference between the old and the new quantities. At the lowest energies for which a response value was reported, the difference is the highest. For the Seibersdorf dosimeter (Figure 5a), it assumes a factor of more than 5 at a mean photon energy of 15 keV. The two other designs in Figure 5b,c do not report reference values at such low energies, but the trend to a strong overestimation is also visible. At energies above 100 keV, the response curves for the old and new operational quantities are proportional to each other with a difference of 15%, which can be taken into account by a recalibration if the radiation field contains only high-energy photon emitters.

3.3. Spectrometer-Type Environmental Dosimeters

Photon spectrometers based on solid-state detectors deliver an energy-dependent pulse-height spectrum of the incoming radiation. Such detectors are frequently used for measuring terrestrial radiation or in airborne gamma ray surveys, which could be deployed after a radiological accident. HPGe spectrometers were originally used for these tasks [20,21]. They have an unrivalled energy resolution, but they need to be cooled by liquid nitrogen. Their sensitivity to mechanical shocks makes them unsuited for airborne measurements. They are now complemented by organic scintillators or scintillator crystals, which can be operated at room temperature [22,23,24]. Scintillators are coupled to photomultipliers to derive an electrical signal which can be amplified, filtered and recorded. The pulse height spectrum of a solid-state spectrometer does not represent the incoming photon spectrum; it is modified by the detector response function, based on the interaction mechanisms of photons in the detector. Only a material- and energy-dependent fraction of photons is registered as a photo peak at the initial energy; the rest is scattered and registered as a Compton background at lower energy. The response function of spectrometers can reliably be simulated with Monte Carlo radiation transport methods [24], allowing the original energy spectrum of the photon field to be recovered. From this, the operational quantities can be calculated by

$$\begin{array}{cc}\hfill H^{*}\left(10\right)=& {\int }_{E_p}{\displaystyle \frac{d{E}_p}{d\varphi }}{h}_{\varphi ,E_p}^{*}\left(10\right)d\varphi ,\hfill \end{array}$$

(7)

$$\begin{array}{cc}\hfill H^{*}=& {\int }_{E_p}{\displaystyle \frac{d{E}_p}{d\varphi }}{h}_{\varphi ,E_p}^{*}d\varphi \hfill \end{array}$$

(8)

The integration extends over the full energy spectrum. This shows that once the energy spectrum is recovered from the detector pulse height spectrum, the calculation of either operational quantity reduces to a simple calculation. Any other energy-dependent quantity can be calculated in a similar fashion, for example, kerma in air $K_a$.

4. Discussion

The introduction of a new operational quantity for environmental dosimetry, $H^{*}$, by ICRU in its Report 95 [8] implies that dosimeters optimised for the present quantity $H^{*}$(10) will deliver too high values of the new quantity if they are not adapted. Depending on the detector type in the dosimeter, there are different routes to adapt their response to ambient dose $H^{*}$.

For active photon dosimeters, with a low-energy cut-off energy situated at around 50 keV, recalibration of the dosimeter is possible because the two quantities, ambient dose $H^{*}$ and ambient dose equivalent $H^{*}$(10), are proportional to each other in the energy range from approximately 100 keV until the highest energies of interest for environmental measurements. Reduction in the sensitivity, or analogously, increase of the calibration factor by approximately 15%, is sufficient for the dosimeter or monitor to indicate $H^{*}$ correctly.

Passive environmental dosimeters designed for determining ambient dose equivalent and capable of registering photon energies as low as 15 keV show a strong overestimation of ambient dose in the energy interval from 15 keV to approximately 50 keV. Their energy-dependent response can be modified by adding different filtration to their casing, for example metallic filters of different material and thickness. Today, possible filter configurations can be designed on the desktop with the help of Monte Carlo simulations [25] before testing a physical prototype in the calibration laboratory. Adding metallic filtration will help correct the overestimation of $H^{*}$ at low energies, but it will inevitably also lower the sensitivity of the dosimeter. A balance has to be found between sufficient sensitivity for small environmental ambient doses, often only a few mSv in a year, and an optimally corrected energy response.

A less onerous modification of the energy-dependent response concerns spectrometer-based environmental dosimeters. Once the count rate spectrum is unfolded to the true photon spectrum, a multiplication with the energy-dependent fluence-to-ambient dose conversion coefficients $h^{*}\left(E\right)$ is sufficient to obtain a measurement of $H^{*}$. There are now various types of detectors with spectrometric capabilities on the market which can be operated at ambient temperature: scintillators such as LaBr₃ [22,23] and pixel-type detectors, fully integrated with detection, amplification and digitisation on a single chip, such as the DECAL [26] or the TIMEPIX 4 detectors [27]. If the cost of such detectors becomes comparable to that of ionisation chambers, they may become useful complementary environmental dosimeters, delivering additional information in the form of an approximate, measured fluence spectrum of the ambient radiation field.