Gold-198
Au-198 · Gold, Z = 79, A = 198
At 9.06e+15 Bq/g — 2.45e+5 Ci/g — a gigabecquerel of Au-198 amounts to 110 ng, which is why activity rather than mass is how anyone states the quantity. Gold-198 decays by beta-minus emission with a half-life of 2.6941 days, falling to 16.51% of today's activity in a week and 6.8e-9% in three months.
An air kerma rate constant of 0.0545 mGy·m²/(GBq·h) is 1.4× less than Cs-137 and 5.6× less than Co-60, and 39 of 96 among the photon emitters carried here.
Halving the air kerma rate calls for 2.77 mm of lead, or 9.49 mm of steel where lead is unwelcome, and a factor of ten calls for 9.33 mm of lead — sheet thicknesses that a glovebox or a transport container can carry. Reaching 20 µSv/h from 1 GBq at a metre takes 4.02 mm of lead.
The beta endpoint is 1.37 MeV, mean 0.313 MeV over 3 branches; that endpoint comes from a branch of only 0.025%, while 98.99% of decays take the 0.961 MeV branch. Shielding is sized on the endpoint, dose on the mean. That endpoint stops in 5.11 mm of acrylic or 2.43 mm of glass. Of the beta energy, 0.22% turns into X-rays in acrylic and 2.76% in lead, worked out for the 0.961 MeV branch, the hardest in at least 1% of decays.
Au-198 activation foils and the retention of implanted gold grains
Au-198 is the beta-emitting product formed when stable gold captures a neutron, making an irradiated foil a record of neutron exposure. The resulting activity can be measured after the foil has been removed from the irradiation position. A germanium spectrometer separates its characteristic gamma peak from other activation products, and the measured activity is converted into neutron fluence using the target inventory and reaction cross section. This is a measurement of accumulated exposure; the foil is not an electronic monitor that displays changes while irradiation proceeds.
Gold activation also distinguishes components of a mixed radiation field. In a clinical accelerator study, researchers compared uncovered foils with foils enclosed in cadmium, which strongly absorbs thermal neutrons. The covered specimens predominantly registered the epithermal contribution, allowing the thermal component to be evaluated separately. Irradiation duration, decay before counting and neutron absorption within the foil enter the calculation. High-energy photons can make Au-196 in the same material, so identifying the activation product matters: activity in an irradiated gold sample cannot automatically be assigned entirely to Au-198.
In brachytherapy, Au-198 is instead deliberately implanted in small grains to irradiate a nearby tumour. Reports from oral cancer services describe placement beneath the mucosa, with the arrangement adapted to the lesion. The grains remain radioactive after implantation, and their positions affect the delivered dose. A study of hospitalised patients found displaced grains particularly in superficial tongue and lip lesions. An implant can therefore lose a source while the source itself remains intact; physical retention is a separate concern from radioactive decay.
The reported monitoring combined radiographs of the implant with radiation surveys of treatment and patient rooms. Dislodged grains were found on floors, while another was detected in the abdomen after being swallowed. Early loss can reduce irradiation near the original implant position and expose staff elsewhere. These observations give Au-198 a practical distinction from an activation foil: the clinical question includes where each grain has moved, not simply how much activity remains. Gold's chemical stability does not remove the consequences of an active grain entering an unintended location.
- LNHB / DDEP, Recommended nuclear data tables: Au-198
- Haddad Kh., Anjak O. and Yousef B., Neutron and high energy photon fluence estimation in CLINAC using gold activation foils, Reports of Practical Oncology and Radiotherapy 24(1) (2019), pp. 41–46. DOI: 10.1016/j.rpor.2018.08.009.
- Konishi et al., Investigating the displacement of radio-active sources during gold-198 grain brachytherapy for hospitalized oral cancer patients, Journal of Contemporary Brachytherapy (2022)
- IAEA, Manual for reactor produced radioisotopes, IAEA-TECDOC-1340 (2003), Gold-198, pp. 63–65
Half-life, specific activity, dose rate and beta energies
| Half-life | 2.6941 days (2.328e+5 s) |
| Decay mode | beta-minus decay |
| Specific activity | 9.06e+15 Bq/g (2.45e+5 Ci/g) |
| Air kerma rate constant Γ (δ = 20 keV) | 0.0545 mGy·m²/(GBq·h) |
| Dose rate, 1 GBq at 1 m | 0.0545 mGy/h |
| Dose rate, 1 Ci at 1 m | 2.02 mGy/h |
| Kerma-weighted mean photon energy | 416.5 keV |
| Beta endpoint / mean | 1.37 MeV / 0.313 MeV |
412 keV carries 98% of the dose rate
One further line below the 20 keV cutoff, at 11.78 keV with 1.19% emission probability, is excluded here and from Γ. Probability and dose share.
| Energy (keV) | Emission probability (%) | Share of dose rate (%) |
|---|---|---|
| 411.80 | 95.62 | 97.82 |
| 675.88 | 0.80512 | 1.34 |
| 1087.68 | 0.15892 | 0.40 |
| 70.82 | 1.36883 | 0.22 |
| 68.89 | 0.80747 | 0.13 |
| 80.32 | 0.46455 | 0.08 |
| 82.48 | 0.13611 | 0.02 |
2.77 mm of lead halves this spectrum
Solved numerically across all 7 lines, narrow beam. Why not one representative energy.
| Material | HVL (mm) | TVL (mm) | TVL / HVL |
|---|---|---|---|
| lead | 2.77 | 9.33 | 3.36 |
| tungsten | 1.96 | 6.58 | 3.36 |
| iron | 9.49 | 31.7 | 3.34 |
| copper | 8.34 | 27.9 | 3.34 |
| concrete | 31.2 | 104 | 3.33 |
| water | 66.2 | 220 | 3.33 |
| aluminum | 28.0 | 93.4 | 3.33 |
A single energy would give 3.32. Spectrum and attenuation ratios.
5.11 mm of acrylic stops the 1.37 MeV endpoint
Katz–Penfold fit to the 1.37 MeV endpoint — a stopping thickness rather than an attenuation length. Why the material matters more than the thickness.
| Absorber | Range for the endpoint (mm) |
|---|---|
| acrylic | 5.11 |
| aluminum | 2.25 |
| water | 6.08 |
| glass | 2.43 |
The conversion below uses the 0.961 MeV branch; the rarer 1.37 MeV branch sets only the range above.
| Shield material | Fraction of beta energy converted to X-rays |
|---|---|
| acrylic (Z ≈ 6.5) | 0.22% |
| lead (Z = 82) | 2.76% |
Activity over days and weeks
Ten half-lives is 26.9 days — for Au-198 itself, a storage problem rather than a disposal one, since those ten half-lives already take it below 0.1% of today's activity. The mean life 1/λ is 3.89 days.
| Elapsed | Fraction remaining |
|---|---|
| 1 half-life | 50.0 % |
| 2 half-lives | 25.0 % |
| 5 half-lives | 3.13 % |
| 10 half-lives | 0.0977 % |
| Time to fall to 10 % of today's activity | 8.95 days |
| Time to fall to 1 % | 17.9 days |
| Time to fall to 0.1 % | 26.8 days |
Limits of these dose rates
- 5.11 mm of acrylic for the 1.37 MeV endpoint: a fitted depth of penetration, not a skin dose.
- Air kerma, 1 µGy counted as 1 µSv — not H*(10).
- Model limits and internal dose.
Gamma, beta and decay calculators for Au-198
- Gamma dose rate and shielding
- Beta dose rate and shielding
- Decay and half-life
- Mass and activity
- Detection limits (MDA / MDC)
Other Gold nuclides: Au-198m
Computed from the IAEA Nuclear Data Section — Livechart API (ENSDF) and the NIST X-Ray Mass Attenuation Coefficients. Derivations and citations.