New beta primary standard (BPS2) and revised ISO 6980 at PTB

In this work, firstly, the main characteristics of PTB’s new beta primary standard (BPS2) are presented in comparison to its previous one (BPS1), i.e. to the Böhm chamber. Secondly, the main changes of the revised ISO standard series on beta reference radiation fields, ISO 6980, are described. Finally, the application of the revised ISO 6980 to both PTB’s new and old beta primary standard is implemented. It turns out that both devices, BPS2 and BPS1, agree on average to about ± 1 % with each other, i.e. well within their one sigma statistical uncertainties of approximately 3 % for 147Pm and 1.5 % to 1.8 % for 85Kr, 90Sr/90Y and 106Ru/106Rh radiation qualities. Therefore, as of mid-2023, PTB moved from the 2004–2006 version of ISO 6980 to the revised version from 2022 and from the old beta primary standard, BPS1 (i.e. the Böhm chamber) to the new one, BPS2.


Introduction
Primary dosimetry for beta radiation protection has a long history at the Physikalisch-Technische Bundesanstalt (PTB) which is closely connected to the scientific work of Böhm [1][2][3][4][5][6][7][8][9][10][11][12], which found one of its highlights in the development of an extrapolation chamber for primary beta dosimetry: the Böhm chamber [13]. This has been and still is commercially available since the late 1980s [14] and is in use as the primary standard for beta dosimetry at PTB as well as many other national metrology institutes. Accordingly, major parts of the standard series relating to beta dosimetry published in 2004 and 2006 by the International Organization for Standardization (ISO), ISO 6980 [15][16][17], are based on the many works of Böhm cited above. Also, like the Böhm chamber, ISO 6980 has been used by PTB as well as many other national metrology institutes Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. [13,[18][19][20][21][22][23]. Since the 2004/2006 publication of ISO 6980 (and partly before) further investigations were undertaken at PTB regarding several topics: • a beta secondary standard irradiation facility BSS 2 providing quality assured irradiations in terms of the operational quantities H p (0.07), H p (3), H ′ (0.07) and H ′ (3) in reference radiation fields according to ISO 6980 was continuously improved [24]: it provides three radiation sources: 147 Pm, 85 Kr and 90 Sr/ 90 Y with beta mean energies of 0.07 MeV, 0.25 MeV and 0.80 MeV, respectively; • a 106 Rh/ 106 Ru source with a beta mean energy of 1.2 MeV and several more extensions to the BSS 2 were described [25]; • the BSS 2 has been and still is commercially available since the late 1970s [26]; • the depth dose curves of the BSS 2 radiation fields were determined and proved to be universal for all BSS 2 radiation sources of the same radionuclide [27,28]; • the photon contamination of the BSS 2 radiation fields was determined [29]; • the simulation of the radiation fields of the BSS 2 was performed and the radiation fields were provided in ASCII and binary format [30]; • several correction factors for the different quantities and phantoms used in beta dosimetry were determined by Monte Carlo particle transport simulations and published [31]; these correction factors are ready for implementation in ISO 6980-3; • two energy-reduced radiation fields based on the 90 Sr/ 90 Y source of the BSS 2 were described in order to fill the energy gap between the mean energy of the 85 Kr and 90 Sr/ 90 Y source [32]; • a new beta primary standard was built at PTB [33] and finally, • several correction factors for primary beta dosimetry according to ISO 6980-2 were re-determined by means of Monte Carlo particle transport simulations [33,34].
In this work, three main topics are described: • the main characteristics of PTB's new beta primary standard, • the main changes of the revised ISO 6980 standards, and • the application of the revised ISO 6980 to both PTB's new and old beta primary standard.

Reasons for a new primary standard
The main reasons for developing a new beta primary standard (BPS2) at PTB were twofold: Firstly, as outlined above, the Böhm chamber (BPS1) has been and is still in use at several national metrology institutes worldwide since the late 1980s. Thus, past comparison measurements suffer from the possible risk of systematic issues with the chamber that were not revealed as long as nearly all the comparison participants used the same type of instrument to determine the reference value of the dose rate, i.e. a Böhm chamber. Therefore, another chamber design (BPS2) using different dimensions and different materials compared to the ones used for the Böhm chamber should be used.
Secondly, the measurements necessary for a source calibration at PTB are only semi-automatic: • the extrapolation curve measurement, i.e. the measurement of the ionization current depending on the chamber depth which is varied from 250 µm up to 2500 µm in steps of 125 µm, resulting in 19 different chamber depths, is carried out automatically by means of a software controlled motor mounted to PTB's Böhm chamber [35]; for information: these measurements are performed with an absorber of approximately 0.05 mm polyethylene terephthalate (PET) in front of the chamber resulting, together with the entrance foil of the chamber, in approximately 0.07 mm tissue equivalent depth; • however, the measurement of the depth dose curve, i.e. the measurement of the ionization current depending on the depth in a tissue equivalent phantom which is varied from nearly zero depth, i.e. without any absorber in front of the chamber, up to more than 20 mm tissue depth (resulting in the complete absorption of the beta radiation) using up to 24 different absorbers placed in front of the chamber, is done manually, i.e. about every 25 min another absorber needs to be placed in front of the chamber by hand-this should be automated; for information: these measurements are performed at a fixed chamber depth of 1000 µm; using the rather thick absorbers results in the measurement of the photon contribution to the dose rate (x-ray, gamma ray and bremsstrahlung emissions from the source).
Therefore, a new measuring facility was constructed. Figure 1 shows the different components of PTB's new beta primary measuring facility: it consists of a new beta primary standard extrapolation chamber (BPS2) embedded in a slab phantom. A motor is attached to the chamber to automatically adjust the chamber depth. Furthermore, a magazine is incorporated in the design, containing 24 absorbers of different thicknesses (between zero and 20 mm) and a mechanism to automatically position the absorbers in front of the chamber. Details of the chamber's geometry, dimensions and material are given in table 1 as well as figure 2. In contrast to the Böhm chamber, BPS1, (left and center), both electrodes of the new chamber, BPS2, (right), i.e. the collecting electrode at the back of the chamber as well as the front electrode, are covered with layers of very thin (sputtered) Aluminum on the inner side of the chamber while graphite was used in the Böhm chamber (BPS1). This has an advantage over the Böhm chamber, as the latter's contacts consist of graphite that has been sprayed on and therefore does not have a well-defined thickness. Furthermore, the collecting electrode of the new chamber is contacted using electrically conductive polyether ether ketone (PEEK ELS), see table 1 and figure 2. Behind the active volume is a 2 mm thick isolating layer of PEEK and located behind that is an electrically conductive 19 mm layer of PEEK ELS to prevent betas penetrating, i.e. electrons, to form space charges behind the active volume, i.e. free electrons that diffuse through the material. Behind that is another 32 mm of PEEK, (figure 2 right side). This concept preventing space charges is not present in the Böhm chamber routinely sold (and routinely used at PTB, figure 2 left side) but is present in a few examples built by PTW on PTB's special request, (figure 2 middle section). The active collecting volume, 30 mm in diameter, is surrounded by a 15 mm wide guard ring to ensure a homogeneous electrical field inside the active volume. Located outside of this guard ring is another 10 mm wide ring divided into three segments-again, see figure 2 right side. The electrical capacity of these three segments can be measured independently to make sure the back and front electrodes are parallel to each other. Finally, another small difference is that in BPS1, a copper wire insulated by 2 mm polymethyl methacrylate (PMMA) is used for the electric contact of the  Further details for Monte Carlo transport simulations in the chamber were described earlier [33].

Overview of changes
As a consequence of the many updated and published data for both primary beta dosimetry as well as irradiations and calibrations of instruments, see section 1, the ISO working group for reference radiation fields, ISO TC85 SC2 WG2 [36], decided in 2019 to revise the ISO 6980 standard series. The new version was issued in October 2022 with the following main changes: for all three parts-1, -2 and -3: (1) inclusion of the quantities H p (3) and H ′ (3); (2) inclusion of 106 Ru/ 106 Rh series 1 sources; (3) inclusion of energy-reduced beta-particle fields based on 90 Sr/ 90 Y sources; (4) removal of 14  The revised versions of the standards are available from ISO [39][40][41]. They contain the changes listed above which are described in detail in the following subsection. (3) and H ′ (3). As the dose limit for the lens of the eye has significantly been reduced [42] it may be necessary to monitor the dose to the lens of the eye in radiation fields with a significant dose contribution due to beta radiation. Therefore, the corresponding operational quantities of H p (3) and H ′ (3) were introduced into the standard series. The implementation was achieved by introducing a better description of the depth dose curves in ISO 6980-2, see below, section 3.2.7, and corresponding conversion coefficients in ISO 6980-3. 106 Ru/ 106 Rh series 1 sources. Radiation fields with a beta mean energy above that of 90 Sr/ 90 Y (∼0.8 MeV) were added; corresponding data and conversion coefficients for 106 Ru/ 106 Rh sources with a beta mean energy of ∼1.2 MeV were added in all three parts of ISO 6980 for several more distances than before and also using a beamflattening filter.

Inclusion of energy-reduced beta-particle fields based on 90 Sr/ 90 Y sources.
Radiation fields with a beta mean energy between that of 85 Kr and 90 Sr/ 90 Y (∼0.25 MeV and ∼0.8 MeV) were added; corresponding data and conversion coefficients for 90 Sr/ 90 Y sources with an absorber of a few mm PMMA between the source and the reference point resulting in beta mean energies of ∼0.55 MeV were added in all three parts of ISO 6980.

Removal of 14 C sources.
The beta mean energy of 14 C sources is rather small (∼0.04 MeV) and, therefore, its radiation does not contribute to the dose to local skin, i.e. it does not penetrate 0.07 mm of tissue. Consequently, it was agreed by the corresponding ISO working group to remove all information regarding 14 C sources from the ISO 6980 standard series. [37] and its amendment AMD.1:2015 [38].

Reference to the terms and definitions in clause 3 of ISO 29661:2012
Since the original date of publication of the standard series (2004-2006) a general standard on reference radiation fields for radiation protection containing definitions and fundamental concepts was issued in 2012 [37] and amended in 2015 [38]. Consequently, the revised standard series ISO 6980 refer to that general standard and adopted its definitions and fundamental concepts instead of repeating them in the revised standard series ISO 6980.

Inclusion of correction factors for primary dosimetry
based on radiation transport simulations. Three of the correction factors to be applied for primary dosimetry using an extrapolation chamber were re-determined in the past by means of Monte Carlo radiation transport simulations [33,34]: • k ba : the correction factor for the difference in backscatter between tissue and the material of the collecting electrode and guard ring; • k pe : the correction factor for perturbation of the beta-particle flux density caused by the side walls of the extrapolation chamber and • k ih : the correction factor for the inhomogeneity of the absorbed dose rate inside the collecting volume; k ih comprises the former three factors * k ra : the correction for the radial non-uniformity of the beam, i.e. perpendicular to the beam axis; * k ac : the correction factor for attenuation of beta particles in the collecting volume and * k di : the Correction factor for the axial non-uniformity of the beta-particle field; Among these three, k pe shows the largest changes compared to the previous data set as the method used for its measurement at that time was an indirect one [3] as opposed to the direct determination via simulations used nowadays [33,34].
Correspondingly new values of these three correction factors were adopted in the revised ISO 6980-2.

Use of Chebyshev polynomials with twelve parameters instead of ordinary polynomials with three parameters for the description of transmission functions.
Transmission functions are used to determine the correction factor to account for variations in the attenuation and scattering of beta particles between the source and the collecting volume due to variations from reference conditions and for differences of the entrance window to a tissue-equivalent thickness of 0.07 mm, k abs . These transmission functions describing the depth dose curves were changed from ordinary polynomials of the second order (i.e. with three parameters) to Chebyshev polynomials with 12 parameters [27,28,33,34]. The new polynomials have the great benefit that the whole depth dose curve can be described from zero tissue equivalent depth up to several mm instead of only a small range of around 0.07 mm tissue depth. With this description of the transmission functions, the correction factor k abs can also be determined for 3 mm tissue depth, i.e. for the quantities H p (3) and H ′ (3), not only for 0.07 mm. Consequently, the new descriptions of the transmission functions were adopted in ISO 6980-2.

Inclusion of a correction factor for the source to cham-
ber distance at different phantom depth, k ph . For the measurement of the depth dose curves absorbers of different thickness are placed in front of the extrapolation chamber while the distance between the source and the front of the chamber is kept constant at the reference distance, y 0 . Thus, the distance between the source and the front of the absorber, which represents the phantom's surface, is smaller by the absorber thickness than y 0 . This is taken into account by the correction factor k ph [27,28,33,34]. This correction factor was added in ISO 6980-2.

Inclusion of a correction factor for the stopping power
ratio at different phantom depths, k Sta . As outlined above, the revised ISO 6980-2 contains depth dose curves up to several mm of tissue equivalent depth, see 3.2.1 and 3.2.7. With increasing tissue depth, the beta particles continuously lose energy on their way through the phantom resulting in a reduction of its beta mean energy. Consequently, the spectrum averaged ratio of the stopping powers of tissue and the absorber material were determined [27,28,33,34]. To account for this change, the correction factor k Sta was added in ISO 6980-2.

Inclusion of a correction factor for primary dosimetry to use the Spencer-Attix theory instead of the Bragg-Gray
theory, k SA . The Spencer-Attix cavity theory is considered to be more accurate than the Bragg-Gray one as it accounts for the variation in the response measured as a function of cavity dimension whereas the Bragg-Gray theory does not. Therefore, the Spencer-Attix stopping power ratios depending on the chamber depth have been calculated [43,44]. Corresponding correction factors were determined [33,34] and added in ISO 6980-2.

Inclusion of correction factors for the differentiation between different quantities at the same depth: Hp(0.07) vs. H ′ (0.07) and Hp(3) vs. H ′ (3).
An extrapolation chamber is rather large (several cm in diameter and thickness). Consequently, the dose measured is, using appropriate corrections which are described in ISO 6980-2, the dose in a slab at a certain depth, e.g. 0.07 mm or 3 mm. This dose represents the operational quantity H p (0.07) or H p (3), respectively, which are defined in a slab phantom (30 cm × 30 cm × 15 cm) made of International Commission on Radiation Units & Measurements (ICRU) four-element tissue [45,46]: H p (0.07) slab and H p (3) slab . However, the corresponding area quantities, H ′ (0.07) and H ′ (3), are defined in the ICRU sphere (30 cm in diameter), also made of ICRU fourelement tissue. Therefore, corresponding correction factors to account for the sphere instead of the slab were determined [31] and added in ISO 6980-3.

Inclusion of correction factors for the differentiation between different phantoms for the same quantity, such as the slab and rod or slab and cylinder phantom for the quant-
ities Hp(0.07) and Hp(3), respectively. As described above, extrapolation measurements yield the quantities H p (0.07) slab or H p (3) slab being defined in a slab phantom. However, quite often in beta dosimetry ring dosemeters are worn on a finger or, more seldom, eye lens dosemeters worn near the eye are used. Such extremity dosemeters are calibrated and irradiated on a rod or cylinder phantom [37,38], leading to H p (0.07) rod or H p (3) cyl , respectively. Corresponding correction factors to account for the rod or cylinder instead of the slab were determined [31] and also added in ISO 6980-3.

Outlook: further revision of ISO 6980
Shortly after the revised ISO 6980 was issued some minor inconsistencies and errors became obvious. Therefore, a new minor revision is planned for 2023. However, the values of all correction factors will remain unchanged and, consequently, not affect the results presented in this publication-which will, in turn, remain valid without any change once the newly revised ISO 6980 will be issued, possibly as soon as 2023.

Outlook: quantities according to ICRU 95
In 2020 the ICRU proposed a new set of operational quantities for radiation protection [47]. However, as these quantities have not yet been internationally adopted in basic safety standards on radiation protection, by bodies such as the International Atomic Energy Agency and the European Commission, they were not adopted in the revised standard series ISO 6980. Anyway, for possible future use, corresponding conversion coefficients have already been determined [48] and are ready for a possible future revision of ISO 6980.

Implementation of the revised ISO 6980 to Böhm chambers and to PTB's new primary standard
Both the previous and the revised version of ISO 6980 describe in their main text the methods to determine correction factors necessary for beta reference dosimetry. Mostly informative annexes provide values of correction factors-exemplarily for the Böhm chamber (BPS1) as this chamber type is in worldwide use. The values were taken from the literature: for the old ISO 6980 mainly from Böhm [13] and for the revised ISO 6980 mainly from Behrens [33,34]. Furthermore, Behrens [33,34] provides corresponding correction factors for PTB's new primary standard (BPS2). With these two sets of values, both determined according to the revised ISO 6980, a comparison of PTB's new and old beta primary standard was possible: BPS2 vs. BPS1. For this comparison, the following source types were used: 147 Pm, 85 Kr and 90 Sr/ 90 Y, from the BSS 2 [24][25][26] and 106 Ru/ 106 Rh, the latter mounted in a source holder of the BSS 2 [25]. For each type, the same source was subsequently calibrated using the old and the new beta primary standard measuring facility. For both beta primary standards the correction factors according to the revised ISO 6980 were applied, i.e. for BPS1 from the annexes in the revised ISO standard and for BPS2 from [33]. The revised ISO 6980 was used as its correction factors proved to be more suitable than those of the previous version-especially for 85 Kr as the corrected extrapolation curves are much better linear compared to those curves obtained using the correction factors from the previous version of ISO 6980 [33]. Furthermore, no set of correction factors determined in accordance with the previous version of ISO 6980 are available for PTB's new primary standard. Figure 3 shows an overview of the ratio of the dose rates measured with the BPS2 and the BPS1 depending on the beta mean energy. For some radionuclides and distances, more than one data point exists because more than one specimen of the sources was measured. Furthermore, for the same radionuclide, different distances and the optional presence of the beam flattening filter result in slightly different beta mean energies resulting in several data points close to each other. To identify the single data points more easily, the same ratios are presented in detail in figures 4-7 separately for the four radionuclides. The textboxes in the figures give the mean ratios for the four radionuclides together with their standard deviationsall of them agreeing quite well with unity, i.e. on average to about ± 1 %. Also, the single ratios fairly agree with unity, at maximum to about 2.5 %, see the ratios for 147 Pm in figure 3, within their one sigma uncertainties (k = 1) which are represented by the error bars. Only very few ratios are slightly outside of this range, i.e. the lowest point for 147 Pm at 20 cm distance with a beam flattening filter and the points for 85 Kr and 106 Ru/ 106 Rh at 50 cm distance with beam flattening filter. In conclusion, these results demonstrate, that the two beta primary standard measuring facilities of PTB, BPS1 and BPS2, provide comparable results. Therefore, it is concluded that both, PTB's new and old beta primary standard, BPS2 and BPS1, comply with each other and provide reliable results.

Impact of the change from the old to the revised ISO 6980 and from PTB's Böhm to its new primary chamber
In section 4, the correction factors according to the revised ISO 6980 were applied for both PTB's beta primary standards. In contrast, up to now, PTB used its Böhm chamber (BPS1) with the correction factors according to the old ISO 6980 while, in future, PTB will operate its new primary standard (BPS2) applying the revised ISO 6980. Therefore, figure 8 shows the ratio of the dose rates measured with the BPS2 (applying the revised ISO 6980, i.e. the correction factors according to [33]) and the BPS1 (applying the old ISO 6980). These ratios reflect the change of PTB's reported dose rates due to the implementation of PTB's new primary standard (BPS2) and the revised ISO 6980: on average up to 1.7 %, see 147 Pm ratio in the textbox in figure 8, with single values up to almost 5 %, again, see the ratios for 147 Pm in that figure.       In general, the ratios are slightly above unity except for Kr-85. To identify the reasons, figure 9 shows a comparison of resulting dose rates depending on those correction factors in ISO 6980 which are different in the revised and the old version. To correctly interpret this figure, it is important to notice that these correction factors, except the one for backscatter, k ba , depend on the chamber depth and that the dose rate of a source is deduced from the slope of the extrapolation curve (ionization current versus chamber depth). Therefore, a change of the correction factors depending on the chamber depth have a non-linear impact on the dose rate and can, therefore, only be determined for actual measurement data but not universally. Thus, figure 9 shows the ratios of resulting dose rates when applying the different versions of correction factors to the actual measurements mentioned above. It can be seen that the four correction factors have different influences on the different radionuclides: the corrections for backscatter, k ba , and inhomogeneity, k ih , increase the dose rate of 147 Pm sources each by about 1.5 % -2 % resulting in an overall increase of 2 % -4 % while the other nuclides are almost unaffected, especially by the correction for inhomogeneity. This can be well understood as the radiation fields from 147 Pm sources are rather inhomogeneous [32] which was, obviously, not sufficiently considered in the old version of ISO 6980. Furthermore, the correction for perturbation of the radiation in the chamber wall results in slightly larger dose rates especially for 90 Sr/ 90 Y and 106 Ru/ 106 Rh sources-by about 0.5 % up to 1.5 %. Together with a slightly increased backscatter factor, k ba , this results in approximately 2% larger dose rates for 90 Sr/ 90 Y and 106 Ru/ 106 Rh sources. For 85 Kr sources the different changes result only in less than 1 % dose rate increase-although its extrapolation curves are more linear than before (especially due to the change of k pe )-as mentioned above. This demonstrates that the method in ISO 6980-2 to determine the slope of the extrapolation curves based on a quadratic instead of a linear fit is quite robust and well justified. Finally, it can be seen that the introduction of the Spencer-Attix instead of the Bragg-Gray theory, realized by the correction factor k SA , only results in a nearly negligible increase of the dose rate, i.e. less than 1 % for all four radionuclides.
In section 4 it could be shown that PTB's old and new primary extrapolation chambers agree with each other within their one sigma statistical uncertainties; in section 5 it could be shown that the changes from the old to the revised ISO 6980 can be well understood and are well justified. Therefore, PTB moved from its old beta primary standard (BPS1, i.e. the Böhm chamber) to its new one (BPS2) in mid-2023, and from the 2004-2006 version of ISO 6980 to its revised version as of 2022. Of course, corresponding references will be given in PTB's future calibration certificates.

Conclusions
In this work, firstly, the main characteristics of PTB's new beta primary standard (BPS2) in comparison to the previous one (BPS1), i.e. to the Böhm chamber, secondly, the main changes of the revised ISO standard series on beta reference radiation fields, ISO 6980, and, thirdly, the application of the revised ISO 6980 to both are presented. It turns out that both devices, BPS2 and BPS1, agree on average to about ± 1 % with each other, i.e. well within their one sigma statistical uncertainties of approximately 3 % for 147 Pm and 1.5 % to 1.8 % for 85 Kr, 90 Sr/ 90 Y and 106 Ru/ 106 Rh radiation qualities, provided the correction factors according to the revised ISO 6980 series are applied to both. This is a significant improvement compared to the application of the previous edition of ISO 6980: when using this edition, the deviations are also only up to 1.7 % on average, but individual extreme values occur up to almost 5 %, compared to a maximum 2.5 % when using the updated series of standards. Therefore, in mid-2023, PTB moved from the 2004-2006 version of ISO 6980 to its revised version as of 2022 and from its old beta primary standard (BPS1, i.e. the Böhm chamber) to its new one (BPS2) resulting in slight increases of the dose rates from all source types covered in ISO 6980 except for 85 Kr. These increases have their origin in the move from the previous to the revised version of ISO 6980.