During target irradiation, a few atoms of the wanted radionuclide are produced within the bulk target material. The energy released in a nuclear reaction is large relative to the electron binding energies and the radionuclide is, therefore, usually ‘born’ almost naked with no or few orbit electrons. This ‘hot atom’ will undergo chemical reactions depending on the target composition. In a gas or liquid target, these hot atom reactions may even cause the activity to be lost in covalent bonds to the target holder material.
During irradiation, the target is also heated and its structure and composition may change. A pressed powder target may be sintered and become more ceramic, which makes it more difficult to dissolve. The target may melt and the radioactivity may diffuse in the target and even possibly evaporate. In designing a separation method, all of these factors have to be considered. Fast, efficient, and safe methods are required to separate the few picograms of radioactive product from the bulk target material, which is present in gram quantities.
The separation of the radionuclide already starts in the target, as demonstrated in the production of 11CO2. Carbon-11 is produced in a (p, α) reaction on nitrogen gas. To enable the production of CO2, some trace amounts of oxygen gas (0.1–0.5%) are added. However, at low beam currents, mainly CO will be formed since the target will not be heated. At high beam currents, the CO will be oxidized to the chemical form CO2. The separation is made by letting the target pass through a liquid nitrogen trap, which is simple and efficient. By adding hydrogen gas instead, the product will be CH4.
The skill in hot-atom chemistry is to obtain a suitable chemical form of the radioactive product, especially when working with gas and liquid targets. Solid targets are usually dissolved and chemically processed to obtain the wanted chemical form for separation.
The concept of specific activity (a), i.e., the activity per mass of a preparation, is essential in radiopharmacy. If 100% of the product contains radioactive atoms, often called the theoretical a, then the relationship between the activity (A) in becquerels and the number of radioactive atoms (N) is given by:
N = A / λ
Where λ is the decay constant (1/s), which can be calculated from the half-life (T1/2) in seconds as:
λ = ln(2) / T1/2
The specific activity (a) expressed as activity per number of radioactive atoms is then:
A/N = λ = ln(2) / T1/2
For a short-lived radionuclide, a will be relatively large compared to a long-lived isotope. For example, a for 11C (T1/2 = 20 min) is 1.5 × 108 times larger than for 14C (T1/2 = 5730 years).
The specific activity (a) expressed in this way is a theoretical value that is rarely obtained in practical work. When producing 11C, the target gas and target holder will contain stable carbon, which dilutes the radioactive carbon and competes in the labelling process afterwards. A more empirical way to define a is to divide the activity by the total mass of the element under consideration. This value for 11C will usually be a few thousand times lower than the theoretical value, while the production of 14C can come closer to the theoretical a.
In the labelling process, a is usually expressed as the activity per number of molecules (a sum of labelled and unlabelled molecules). Instead of using the number of atoms or molecules, it is common to use the mole concept by dividing N by Avogadro’s number (NA = 6.022 × 1023). A common unit for a is then gigabecquerels per micromole.
If the radioactive atoms are produced and separated from the target without any stable isotopes, the process is said to be ‘carrier-free’. If stable isotopes are introduced as contaminants in the target or in the separation procedure, the process is said to have ‘no carrier added’, i.e., no stable isotope is deliberately added. Both of these processes usually give a high final a. However, it may be necessary to use a target of the same element or it may be necessary to add extra mass of the same element in order for the separation process to work. In this case, a carrier is added deliberately, and a will usually be low.
It should be noted that a carrier does not necessarily need to be of the same element. When labelling a radiopharmaceutical with a chelator and metal ions, any ion fitting into the chelator will compete. For example, when labelling a peptide with 111In, the activity will usually be delivered as InCl3 in a weak acid. By sampling the activity with a stainless steel needle, Fe ions will be released and will probably completely ruin the labelling process by outnumbering the 111In atoms.