The information in this spreadsheet is copied directly from Appendix 1: Performance prerequisites and required information for inclusion on the AAPM’s Ionization Chamber Registry (Version 6)

References are in Appendix 1

Instructions:

Under columns C and E, indicate whether data were provided and whether or not it is acceptable (i.e., passes the specification) with a Y or N. Cell format will change automatically.

Under column D, indicate the value of a given specification and/or any comments.

Provided?

Acceptable?

For an ionization chamber to be added to the Registry, the following information must be provided by the manufacturer for publishing within the Registry. All physical dimensions must be in units of mm.

Useful voltage range over which linear chamber behavior is observed using Jaffé plots (1/M vs. 1/V) obtained in beams with differing dose per pulse, Dpp, values (at least three), which must be provided. Jaffé plots are also required for investigation of ion recombination behavior, discussed below. The recommended operating voltage and maximum bias voltage, Vmax, must be specified.

For an ionization chamber to be added to the Registry, it must meet the performance requirements listed below. Unless otherwise specified the characterization tests should be performed for a representative chamber with the manufacturer specified operating voltage in a cobalt-60 or linac beam.

Waterproof limitations: If specified as waterproof, demonstrate no impact on chamber behavior by measuring chamber response and leakage before and after being immersed in water for 16 hours. Leakage should not increase to more than 50 fA and response should not change by more than 0.2 %.

The chamber must demonstrate that accumulated absorbed dose dependence to 1 kGy is consistent with the results demonstrated above (trend within 0.3 % over 2 years) to verify long-term stability. For testing, higher doses can be applied and the result for 1 kGy be interpolated from the data. This testing may be done with the same chamber(s) used for stability testing and combined with results demonstrated above (trend within 0.3 % over 2 years) to verify long-term stability.

Preirradiation stabilization effects: less than a 0.5 % change in chamber response from beam-on after delivering 0.7 Gy. For this test, the ionization chamber response must be stable to within 0.1 % of the average response in a beam with a field size of at least 10 cm x 10 cm. The chamber should not have been irradiated for at least 72 hours before doing this test. The manufacturer must provide a plot showing ionization chamber signal vs. irradiation time without chamber preirradiation to demonstrate this behavior. An example of chamber stabilization is given in Figure 1 of Appendix 1.

The chamber must be shown to have its active volume openly communicate with the local environment, with the chamber response following the ideal gas law for temperature and pressure dependence. The chamber must show 90 % of its reduction in response within 10 seconds with a relative reduction in air pressure of 5 % to 10 %.

Effects on the chamber response due to variations in humidity must be less than 0.5 % and the leakage current7 must not increase to greater than 50 fA for relative humidity levels between 20 % and 70 %.

Initial ion recombination: Cinit should be less than 0.002 (ϒ/U < 0.6 in the notation of Bruggmoser et al.9) for TG-51 reference conditions.2 That is, the component of initial recombination should contribute less than 0.2 %. The difference in the initial recombination correction between opposite polarities should be less than 0.1 %.2

A full Monte Carlo simulation10 of any new chamber must be performed for a chamber to be included on the Registry. The EGSnrc code system11 is the most widely-benchmarked code for this type of simulation, but alternate codes such as PENELOPE12 or others may be used if suitable benchmark calculations are performed. Both of these code systems have been shown to pass the Fano test,13,14 albeit by careful selection of user parameters with PENELOPE. Any other code may be used for calculations of kQ factors, but in order for these factors to be included in this Registry, it must be shown that the Monte Carlo code can pass the Fano test within 0.1 %. To demonstrate an appropriate understanding of these types of simulations, the registrant must calculate kQ factors for at least one ionization chamber that employs a similar geometry to the applicant chamber and has previously published kQ factors. If chamber geometries for existing chamber types are not available one can create models using a document that describes generic ionization chamber geometries (WGICR Appendix 6). These results must be provided at the time of the Registry application. Statistical uncertainties on calculated kQ factors must be less than 0.1 %.

For photon beams, the source can be modelled using photon beam spectra across the entire energy range of interest for beam qualities with %dd(10)x between 63.0 % and 86.0%. For electron beams, kQ data must be provided for beam qualities with R50 between 1.7 cm and 8.7 cm. For electron beams with energies equal to or greater than 15 MeV (R50 equal to or greater than 6.5 cm) the source model must include effects from contaminant photons (i.e., by using a full accelerator model or a phase-space source to simulate the source) since contaminant electrons can impact calculated kQ factors by up to 0.5 % for high-energy beams.15,16

Experimental kQ values for 27 different ionization chamber models were published by McEwen5 for megavoltage photon beams using a direct comparison of reference chamber measurements against a primary standard water calorimeter and indirectly calibrating other ionization chambers against these calibrated reference chambers. This type of indirect kQ measurement is required for inclusion of an ionization chamber on the Registry.

The starting point is obtaining a cobalt-60 ND,wCo calibration from a primary or secondary standards laboratory for the chamber of interest. Ideally, the chamber of interest would also be calibrated in photon or electron linac beams and ND,wQ calibration coefficients provided by the primary or secondary laboratory. However, these calibrations may not be possible, especially for linac electron beams where absorbed dose primary standards are typically not available. If ND,w is available for the chamber of interest for both cobalt-60 beams and linac beams for the modality and energy range of interest, then kQ (normally provided with the calibration certificate if a linac calibration service is available) is simply

This approach is similar to that used to obtain kQ in the literature.17-21 A report on how the measurements were carried out, along with plots of kQ vs. beam quality specifier [%dd(10)x for MV photon beams or R50 for electron beams], and fits to the data of the forms provided in the Addenda to TG-512,3 must be provided. These measurements must be compared to those for similar chambers in the literature if possible.

Measured kQ factors must be provided for a set of five chambers of the same type. The relative variation (standard deviation) of measured kQ among the set of five chambers must be ≤0.5 %.

For a chamber to be added to the Registry, measured and Monte Carlo calculated kQ factors must agree within combined uncertainties.

NOTE: For photon beams, the typical level of agreement of measured and calculated kQ factors is 0.3 %.22