Radiation shielding data for use with F-18
Published: Jun 14, 2025
Source: Ionactive radiation protection resource
Prelim
There is much data online regarding radiation shielding of F-18 - and this spans several decades. More recently AI may play a role, but be very careful - we have tested AI on a number of radiation shielding scenarios, and in some cases it is pretty good, in other cases it fails spectacularly (including incorrectly "reading" tables and graphs from online resources). So we at Ionactive feel there is still a place for carefully curated resources, and this is one of them.
F-18 is an artificially produced (via a cyclotron) radioactive position emitter. Per decay a positron is emitted and this quickly annihilates with an electron which yields two 511 KeV photons. This decay feature can be used in the medical diagnostic practice of Position Emission Tomography (PET). Other positron emitters are also used in medicine such as C-11, N-13, Rb-82 and G-87 but will not be discussed further here.
An internet review of PET shielding will probably reveal a TVT (10th value thickness), also known as a TVL (10th value layer), of 16-17mm for F-18 with lead (a popular PET radiation shielding material). Ionactive also promotes this value here - F-18 (Fluorine-18) Radiation Safety Data. However, often in shielding design careful optimisation is required and the multiple use of a TVT (TVL) may lead to over design (with increased costs). This is particularly relevant in the design of PET rooms (e.g. PET scanner rooms, dose uptake areas, hot WC, hot waste rooms etc). Furthermore, whilst such resource can often be used to calculate shielding based on a point source, often additional factors are introduced, such as patient attenuation (for administered F-18) and the rapid decay of F-18 (T1/2 = 110 minutes).
Whilst this resource provides some relevant shielding data for F-18, it does not provide detailed design advice. It may be useful for architects during the early design process where different shielding materials may be considered (e.g. lead, or perhaps a mixture of lead and high density concrete blocks). The formal design process should always be undertaken with advice from a Radiation Protection Adviser (RPA).
Radiation shielding data for use with F-18
The following table is fairly self-explanatory - the provenance of data is discussed below with examples of use.
Lead (cm) | Transmission (T) | Concrete (cm) | Concrete (cm) | Concrete (cm) | Steel (cm) |
|---|---|---|---|---|---|
0.25 | 0.738 | 6 | 5 | 3 | 1 |
0.5 | 0.523 | 9 | 8 | 6 | 2.1 |
0.75 | 0.362 | 13 | 11 | 8 | 3 |
1 | 0.248 | 16 | 13 | 9 | 3.9 |
1.25 | 0.170 | 19 | 16 | 11 | 4.7 |
1.5 | 0.116 | 22 | 18 | 13 | 5.5 |
1.75 | 0.079 | 25 | 21 | 15 | 6.3 |
2 | 0.053 | 28 | 23 | 16 | 7 |
2.25 | 0.036 | 31 | 26 | 18 | 7.8 |
2.5 | 0.025 | 34 | 28 | 20 | 8.5 |
2.75 | 0.017 | 37 | 31 | 22 | 9.2 |
3 | 0.011 | 40 | 33 | 24 | 9.9 |
3.25 | 0.008 | 43 | 36 | 25 | 10.6 |
3.5 | 0.005 | 46 | 38 | 27 | 11.3 |
The above data has been partly derived from the document: AAPM Task Group 108: PET and PET/CT Shielding Requirements (and other literature searches). The formula, as written in the AAPM document (based on Archer et al) is incorrect, and will not yield the results above (or in the AAPM document tables), nor can they output data for other points not tabulated. We have corrected and rewritten them below as follows:
\[T = \left[ \left(1 + \frac{\beta}{\alpha} \right) e^{\alpha \gamma X} - \frac{\beta}{\alpha} \right]^{-\frac{1}{\gamma}}\]
The terms in the above expression have the following meaning:
- T = transmission
- X = thickness (cm)
- α = attenuation constant (material specific)
- β= empirical fitting parameter (material specific)
- γ = curve shape parameter (material specific)
α, β and γ in the above expression are essentially curve fitting parameters (these fit Monte Carlo modelling results, based on a number of points, into a continuous expression). The fitting parameters are material specific and are as follows.
Material | α (cm-1) | β (cm-1) | γ |
|---|---|---|---|
Lead | 1.543 | -0.4408 | 2.136 |
Concrete | 0.1539 | -0.1161 | 2.0752 |
Steel (iron) | 0.5704 | -0.3063 | 0.6326 |
If required, the above expression can be solved for X (cm), such that the thickness of a material can be obtained from the following expression (where T is known). \[X = \frac{1}{\alpha \gamma} \ln\left( \frac{T^{-\gamma} + \frac{\beta}{\alpha}}{1 + \frac{\beta}{\alpha}} \right)\]
If you wish to obtain concrete thickness (X), the output will be for 2.25 g/cm3. If you require a different density (eg. 1.9, 3.2 or some other density, g/cm3) then it is permissible to scale the results by density. Example: suppose you wish to use 3.2 g/cm3 concrete, and your result for X is 16 cm (at 2.25 g/cm3), then the higher density concrete requirement is 16 cm x (2.25/3.2) = 11.25 cm (in practical terms about 11 cm as shown in the table above).
[Ionactive comment: The above noted density scaling for concrete is valid because 511 KeV photons in concrete, a relatively low atomic number material, is via Compton scatter. Whilst the density of concrete may change, the overall composition remains similar and so the photon interactions are also similar. The same density scaling between say concrete and steel will not work (try it using the above table). Steel overall has a much higher effective atomic number (as well as physical density), so proportionally has a higher electron density than concrete, and so the attenuation of 511 keV photons by Compton scatter is even more effective. This combination of higher physical and electron density means steel is a more effective shield than concrete, for a given thickness (X), but density scaling between the two will under or over estimate the thickness of shielding required, (depending in which direction the scaling is done i.e. 'steel ⇔ concrete'). This is similar with lead and steel for 511 keV photons (try it in the table above). Whilst the physical densities are not that far apart (11.34 g/cm3 vs. 7.85 g/cm3), the electron density difference is considerable (Zlead = 82 vs Ziron = 26). Different buildup and scatter will also play their part. A this photon energy never try and scale lead and steel by density. Overall, unless you are sure of your methodology, don't density scale different shielding materials at any energy.]
How to use the F-18 radiation shielding resource
Here are some examples of how you might use the resource presented in this article.
I have transmission (attenuation), I need material thickness
Suppose you have a transmission requirement of 0.1, what F-18 radiation shielding choices do you have? Looking at the table and using the nearest T value (0.116) you will note the choices are:
- 1.5 cm lead
- 22 cm 1.9 g/cm3 concrete
- 18 cm 2.25 g /cm3 concrete
- 13 cm 3.2 g/cm3 concrete
- 5.3 cm steel
[Ionactive comment: If we take concrete (2.25 g/cm3) specifically, we can calculate the requirement for a T=0.1. This yields 19.4 cm. There are at least 3 internet sites which quote a TVL / TVT for F-18 with concrete at around 11cm. No density is given, but regardless of density this is way off and will lead to inadequate shielding.]
If you have a specific transmission requirement, use the "T=" expression to determine thickness (if you are wanting concrete make sure you know the density required and scale as noted above).
I have a lead shielding requirement, but I need to use lead and concrete
Suppose your F-18 shielding calculations show that you need 25 mm (2.5 cm) of lead. However, your architect has stated they can accommodate 10 mm lead, but structurally they would prefer the rest to be 2.5 g/cm3 concrete. So you have a hybrid shielding situation. Looking at the above table:
- You already have the 10mm lead.
- You need another 15mm lead equivalent, but this must be 2.5 g/cm3 concrete.
- From the table you note that 1.5 cm (15mm) lead is equivalent to 18 cm of concrete.
- Therefore, your hybrid shielding (original requirement being 2.5 cm lead) will comprise 10 mm lead and 18 cm of 2.25 g/cm3 concrete.
Suppose the builder then wants to use barytes bricks specified at 3.2 g/cm3. Using the table we can see that the requirement is then 13 cm of 3.2g/cm3 concrete blocks. Since barytes blocks are normally supplied in 100mm thickness, this might not be the best optimisation, since two sets of 100 mm barytes block will significantly exceed the shielding requirement!
A word of caution!
In many cases PET is combined with CT in modern PET/CT scanners. Whilst some sites might operate a combined scanner for PET/CT diagnostics, others may take advantage of the dual modality and split workloads (e.g. 25% PET/CT and 75% CT). Often workload (and occupancy) are built into (UK at least) shielding calculations, this dominating over instantaneous dose rate (IDR) alone.
If a combined PET/CT facility has shielding dominated by lead (say 2.5cm for example), then the impact of the CT side of the system on the shielding will be negligible. The CT will probably be specified up to 140 kV (with a range from 80-140 kV). The room shielding impact from the CT is by scattered x-rays rather than primary x-ray beam.
Suppose instead that the facility is made with 28 cm of 2.25g/cm3 concrete - we know this is broadly equivalent to 2.5 cm of lead for 511 KeV photons. The equivalence between lead / concrete does NOT hold for the 140 kV x-rays. The concrete noted above will be "good enough" but will be considerably less effective than the lead. However, since the x-rays are scattered, their photon energy will be considerably less than 140 KeV (most likely < 50 KeV) so the photo electric effect will dominate in the concrete.
Overall the shielding will work with either F-18, 140 kV (scattered) x-rays, or both (with lead, concrete or a hybrid of both). But be aware of how this equivalence breaks down at lower KeV photons. Suppose the PET/CT room uses 28cm concrete as noted earlier, and undergoes a change of use - perhaps to primary 140kV x-rays shot against the wall (for whatever reason). The fact that the wall may work adequately for 511 KeV photons from F-18, does not mean it will offer the expected shielding performance for 140 kV photons.
Handy F-18 shielding calculator
Try out the following calculator F-18 shielding calculator.
Note that this calculator should be used for educational purposes only. When using the layered material function, you may notice a slight discrepancy between the total transmission calculated, compared to calculating individual materials (and their transmission) . This is down to the different curve fitments of each of the shielding materials.
Aspects of the calculator are as follows:
- Concrete density (g/cm3) - change this if you know the density of the concrete you are using.
- Transmission - enter your desired transmission (0.1, 0.001 etc) and see the required thickness for lead, concrete and steel. Compare these with the const TVL, and note that these may over or underestimate the required thickness.
- Layered materials - Select thicknesses of a hybrid shield and see the total transmission.
