Chapter 13 from BASICS OF NUCLEAR PHYSICS AND OF RADIATION DETECTION AND MEASUREMENT - An open-access textbook for nuclear and radiochemistry students by Jukka Lehto

Radiation imaging in used to locate, and in many cases also to quantify, radionuclide or a radionuclide-bearing compound from solid material. There are two basic types of imaging techniques: planar imaging giving information of radionuclide distribution at two dimensions and tomography giving three-dimensional information. The latter technique is only briefly described at the end of the chapter. Imaging techniques are typically used in biological and medical applications to locate target molecules. To enable the location of these molecules they have been labelled with a radionuclide, typically a beta-emitting radionuclide in planar imaging and a gamma-emitting radionuclide in tomography. Radiation emitted by these radionuclides is then detected by autoradiography or using technique based on CCD camera filming in case of planar imaging and by an array of gamma detectors in case of tomography.

Autoradiography can be divided into two categories, film autoradiography and storage phosphor screen autoradiography. The prefix auto means that the source of radiation is within the sample unlike in other types of radiographies in which the sample is exposed to an external radiation source, such as X-rays. Autoradiography dates back to late 19^th century when Henri Bequerel discovered in 1896 that uranium salts produced an image on photographic plates (Figure XIII.1).

Figure XIII.1. Image of a uranium salt on a photographic plate (autoradiogram) determined by Henri Bequerel in 1896 (http://www.japanfocus.org/-elin_o_hara-lavick/3196/article.html).

In film autoradiography a film is apposed to a radionuclide-bearing sample. The sample should be flat and as smooth as possible, for example pressed plant or polished rock surface. The film consists of a 0.2 mm polymeric (polyester or cellulose acetate) support plate coated with an emulsion comprising fine silver halide (AgCl, AgI, AgBr) grains in gelatin. The outer surface facing the sample can have a very thin protective cover. Radiation, typically beta particles but also alpha particles, emitted from the sample pass the surface cover and ionize silver atoms in the emulsion layer, which is typically 10-20 µm thick. The released electrons travel in the emulsion and after losing their kinetic energy reduce Ag+ ions into metallic silver Ag forming a latent, invisible image of the radionuclide distribution on the sample. These latent metallic silver centers comprise only of a few silver atoms. When the film is developed in a reducing liquid, Ag⁺ ions around the latent silver metal centres reduce and the amount of metallic silver in the crystal increases by a factor of 10⁸-10¹⁰ making them visible either by eye (macro autoradiography) or by microscope (micro autoradiography).

Figure XIII.2. Principles of autoradiography (http://lifeofplant.blogspot.fi/2011/12/autoradiography. html).

The autoradiogram seen on a film gives a qualitative picture of the distribution of the target radionuclide, or the compound/material bearing the radionuclide, in the sample. Depending on the type and thickness of the sample and the type and energy of radiation the image represents the radionuclide distribution either on the surface of the sample or also in its bulk. If, for example, the sample is rock, the density of which is about 2-3 g/cm³ only radiation originating from the sample surface (alpha particles), or very close to it (beta particles), can be detected on the film due to self-absorption of radiation in the sample at higher sample depths. In the case of imaging a plant sample, having a much lower density, much larger fraction of the radiation comes from the inner part of the sample giving thus also information of radionuclide distribution in its bulk. From the sensitivity point of view autoradiography technique is best suited for tracers utilizing beta emitters of an intermediate energy, such as ¹⁴C (E_max = 156 keV) and ³⁵S (E_max = 167 keV) for which the energy is high enough to avoid self-absorption but low enough to avoid penetration of beta particles through the reactive gel layer.

An important parameter in autoradiography is the resolution, which means the ability of the system to differentiate two individual points in the sample. A typical resolution range is from 5 µm to 50 µm. The resolution is dependent on the following factors, in the order of importance:

1. Distance between the film and the sample. Closer contact to the sample can be obtained by using a fluid silver halide emulsion without the polymeric support, which improves resolution by 5-7 times at maximum.
2. Energy of radiation. The lower the beta energy the better the resolution due to a shorter range of emitted beta particles. The resolution with the low energy beta emitter ³H (E_max = 18 keV) is about ten times better than with the high energy beta emitter ³²P (E_max = 1710 keV). Resolution with the intermediate energy beta emitter ¹⁴C (E_max = 156 keV) is in between these two.
3. Thickness of the sample, the resolution being the better the thinner the sample is.

Figure XIII.3 shows an autoradiogram of a rock impregnated with polymethylmetacrylate labelled with ¹⁴C. The dark areas represent pores (the method is described in detail later in the chapter). In addition to the sample, also standards with known activities of the target nuclide are prepared and their autoradiograms are determined in an identical way as that of the sample. These standards are used to quantify the radionuclide distribution in the sample. The darkness or grey level distribution of the autoradiogram is measured with an optical densitometry measuring the absorption of exposed light at various points of the autoradiogram. The absorption values are converted to optical densities, which are compared point by point to those observed with standard samples and relative activity values can thus be determined at various points at 5-50 µm resolution. The autoradiogram can also be scanned and the grey level values at various points, pixels, are determined with a computer. Exposure times of autoradiographic films vary in a wide range up to weeks, mostly depending on the activity levels. Finding a suitable exposure time requires optimization and experience.

Figure XIII.3. An autoradiogram of the surface of a rock impregnated with polymethylmetacrylate labelled with ¹⁴C. Standard series with varying ¹⁴C activities are at the bottom.

In storage phosphor screen autoradiography, also known as digital autoradiography, the radiation emitted from the sample excites molecules in a phosphor screen apposed to the sample. The excitations are relaxed by scanning with a laser beam, the light emitted in de-excitation is detected and an image is created in a computer based on detected light intensities at all scanned points. The storage term in the name of the process means that the energy from the emitted radiation hitting phosphor molecules is stored in the phosphor crystal as excitation energy. Phosphor is a general name of compounds, which are able to emit light in de-excitation processes.

The phosphor screen, also known as an imaging plate, consists of a polymer support; polyester for example, over which there is a thin layer (150 μm) of phosphor compound bariumfluorobromide BaFBr doped with trace amounts of divalent Eu²⁺ which replace Ba²⁺ ions in the crystal. The crystal size of BaFBr:Eu²⁺ is very small, at about 5 µm. Since the typical oxidation state of europium is +III, Eu²⁺ is readily ionized to Eu3+ when a beta particle from the sample hits the phosphor molecules. The electrons originating from the ionization are trapped in barium vacancies resulting in the excitation of the BaFBr molecules. After exposure, the excitation points are located on points where the radionuclide was present in the sample. To make this “latent” image visible the excitations are relaxed by scanning the image plate with a laser beam and light intensity emitted in the de-excitations at all scanned points (pixels) are detected with a photomultiplier tube. Laser beam moves the trapped electrons to conduction band where they finally combine with Eu³⁺ ions to regain Eu²⁺ ions (Figure XIII.4). This process is called photostimulated luminescence (PSL). Typically the scanning resolution, pixel size, in digital autoradiography varies from 5 to 500 μm. After scanning the plate, it is erased from excitations by intensive light after which the plate can be reused.

Figure XIII.4. Detection process of beta radiation in a phosphor imaging plate.

Figure XIII.5. Left: structure of a phosphor screen. Right: scanning of the screen with laser beam and detection of the emitted light with photomultiplier tube.

Storage phosphor screen autoradiography has several advantages over film autoradiography. First, it has clearly higher sensitivity over film autoradiography, 50-100 times for ¹⁴C imaging, for example. This has a direct effect on exposure times, which are much shorter in case of phosphor screens. Another advantage is that the phosphor screens can be reused unlike films that are used only once (an advantage of film over the phosphor screen is that the film is a durable record of the results while in case of phosphor screen the data is only in an electronic form). Furthermore, an advantage of phosphor screen over film is that the grey level data can be directly digitized to computer while in case of film after development the film needs to be digitized for optical density calculation. Comparing the linearity of light intensity (PSL) response with respect to measured activity the phosphor screen is clearly better compared to film. The linear range in case of phosphor screen is four orders of magnitude while in case of film it is only two orders of magnitude.

Figure XIII.6. Optical densities and light intensity response (PSL) as a function of detected activity for film (black) and for phosphor screen (white).

Using CCD camera for two-dimensional on-line beta counting is still a more advanced method for imaging beta radiation from planar sources. In the apparatuses based on this technique, such as BetaImager or MicroImager from BiospaceLab, beta particles are transformed into light by scintillation process and the light photons are detected with a CCD camera. At its best modification this technique can offer ten times better spatial resolution compared to phosphor screens. Moreover, it gives real-time information on beta emissions from the studied surface. This also shortens the imaging time since one step compared to phosphor screen and two steps compared to film autoradiography can be avoided.

If a three-dimensional picture of the radionuclide distribution in a sample is needed one could cut thin slices of the sample, determine their autoradiograms and superimpose them to get the three-dimensional picture. This would, however, be very laborious and not suitable to determine distribution of a short-lived radionuclide, and particularly to distribution in a human body. For this purpose tomographic methods are the choice and they are widely used in the development and clinical use of radiopharmaceuticals. Depending on the type of radionuclide either single photon emission tomography (SPECT) or positron emission tomography (PET) are two choices. In the SPECT mode a radiopharmaceutical labelled with a gamma-emitting radionuclide, most typically ^99mTc, is injected into a body of a test animal or human. In PET mode the label is a positron emitter, most typically 18F. Thereafter the distribution of the radiopharmaceutical in the body is followed with a gamma camera in case of SPECT and with a PET camera in case of PET, both detecting gamma rays outside the body. Gamma camera comprise an array of collimated Na(I) detectors capable to separate gamma rays emitting from various parts of the body. PET camera makes use of two 511 keV gamma rays emitted in opposite directions in the annihilation of positron particles. PET camera consist of an array of Na(I) detectors in a ring. The target is positioned inside the ring and the camera detects pulses in coincidence mode, i.e. when two gamma rays hit detectors on opposite sides of the ring a pulse is registered while in case of only one gamma ray the pulse is rejected. Both SPECT and PET tomographies are powerful tools in medical imaging and they are increasingly used also in the preclinical development.

Figure XIII.7. Formation and detection of positron annihilation gamma rays (left) (http://www.cellsighttech.com/technology/pet.html) and scheme of PET camera (right) (http://www.lookfordiagnosis.com/mesh_info.php?term=Positron-Emission+Tomography&lang=1).

Applications of autoradiography can be divided into two categories: those where the actual study target is the radionuclide and those where radionuclides are tracers to study existence/distribution/ concentration etc. of other substances.

In environmental radioactivity studies it is a common way to identify and localize particles with higher than typical activities. These particles are present in the environment from fallouts from the nuclear weapons tests in the 1950' to 1970's and from the Chernobyl accident, as well as from releases from nuclear facilities. Particles can be found from air sampling filters and from soils and sediments. Figure XIII.8 shows a film autoradiogram of an air filter taken from a nuclear power plant during maintenance work. The points seen as dark spots in the autoradiogram represent individual particles or their agglomerates removed from the air by filtration (pore size typically about 0.2 μm); the darker the spots are the larger the particles and the higher is their activity. The activity of the largest particle in this autoradiogram was 25 Bq. Based on the information obtained from the autoradiogram larger particles can be localized and further also isolated with the aid of a microscope. The isolated particle can then be characterized with respect to elemental, radionuclide and isotopic composition using a variety of methods, such as scanning electron microscopy, gamma spectrometry, XANES/EXAFS spectroscopy, as well as alpha and mass spectroscopy.

Figure XIII.8. Autoradiogram of an air filter sample taken from a nuclear power plant during maintenance (http://www.stuk.fi/julkaisut_maaraykset/kirjasarja/fi_FI/kirjasarja2/). The diameter of the image is about 10 cm.

At the Laboratory of Radiochemistry, University of Helsinki, a unique method to determine rock porosities utilizing autoradiography has been developed. In this method a rock piece is heated in vacuum to remove water and air from its pores. Then the piece is impregnated with methylmetacrylate (MMA) monomer solution labelled with ¹⁴C. MMA having a lower viscosity than water fills nanometer scale pores in the rock. ¹⁴C-MMA within the rock is polymerized by irradiation or chemically to polymethytmetacrylate (¹⁴C-PMMA). Now autoradiograms can be produced from the sawn surfaces of the impregnated rock piece, to observe the distribution of ¹⁴C in the rock in two dimensions. This distribution corresponds to the distribution of ¹⁴C-PMMA in the rock, which in turn corresponds to porosity of the rock (Figure XIII.9). From the grey levels on the autoradiogram, porosities at various parts of the rock can be determined at micrometer scale. By superimposing the mineralogical composition of the rock surface one can conclude in which minerals the porosity can be found. Here, most of the porosity was found in the inter and intragranular space, in the dark minerals which are micas and in altered plagioclase grains.

Figure XIII.9. Photograph of a polished rock piece surface (left) and an autoradiogram from the same surface (right) after impregnating the rock with ¹⁴C labeled MMA and polymerizing it into ¹⁴C-PMMA.

In the development of a radiopharmaceutical the product needs to pass preclinical tests prior to human tests. An essential part of the preclinical tests are imaging studies to reveal distribution of the product into various organs. These imaging studies are carried out by animals, either with living animals or with specific organs/tissues of dead animals. Both autoradiography and PET/SPECT imaging are used in these studies. The autoradiography tests can be divided into in vivo and ex vivo tests. In the former an organ or tissue is equilibrated with a radiopharmaceutical-bearing solution and in the latter radiopharmaceutical is injected into a living animal. After desired contact time the animal is sacrificed and the distribution of the radiopharmaceutical in the body is determined by measuring radioactivity of various organs separated from the carcass. More detailed distribution can be observed by freezing the organ/tissue or the whole body, by taking thin slices with microtome and by making autoradiograms from the slices. An example of a series of slices taken from a rat’s brain incubated with a solution containing a 18F-labelled radiopharmaceutical ¹⁸F-CTF-FP is shown in Figure XIII.10. The autoradiograms show the spatial distribution of the ¹⁸F radioactivity (red indicates the highest levels, blue the lowest levels), with nonspecific uptake partly subtracted. STR indicates striatum; AMY, amygdala; HIP, hippocampus; LC, locus coeruleus; RAP, raphe nuclei; SN, substantia nigra; CTX, frontal cortex; and CERE, cerebellum.

Figure XIII.10. Autoradiograms of ex vivo rat brain sections at 15 min after injection of dopamine transporter (DAT) radioligand [¹⁸F]β-CFT-FP. The upper row depicts a control rat, and the lower row depicts a rat pretreated with the DAT inhibitor GBR12909 (Koivula, T. et al. Nucl. Med. Biol. 35 (2):177-183).

For the quality control of radiopharmaceutical products HPLC (high performance liquid chromatography) and TLC (thin layer chromatography) methods are used. The latter, TLC, utilizes autoradiography. In this method a drop of a radiopharmaceutical product is applied on a TLC plate and the chromatogram is developed with a proper mobile phase. The run separates chemically different products on the plate and their chemical nature can be determined by their position along the transfer track on the plate. Various compounds are separated into individual spots on the plate. Their relative radioactivity contents can be measured either by radioactivity scanning of the plate or by making an autoradiogram, film or phosphor screen, of the plate. Distribution of radioactivity on the TLC plate can then be seen, and quantified, from the darkness and the position of the identified spots.

Nuclear track methods are based on tracks created by charged particles (from H⁺ up) in solid state nuclear track detector (SSNTD) apposed to the sample emitting the particles. SSNTD can be used to locate particles with elevated alpha activity or fissile material and quantify their amounts by the number of detected tracks. SSNTDs are typically made of plastics. Also other detector materials, such as mica and glass, are used but they are not discussed here. The plastic detectors are made of cellulose nitrate, polycarbonate, polethyleneterephthalate and polyallyldiglycol carbonate, of which the latter has the best sensitivity, i.e. it can produce detectable tracks most effectively. Polyallyldiglycol carbonate is also known with a code name CR-39. It is also able to register tracks from alpha particles, which are not the case with polycarbonate detector. Plastic SSNT detectors are thin foils with thickness varying in the range of 100-1000 µm. The tracks created in the detector are so small, tens of nanometers, that they cannot be seen by eye. They can be detected directly with transmission electron microscopy (TEM) and their number can be counted by eye or computer programs developed for this purpose. Alternatively, the size of the tracks can be enlarged by etching, typically with 2-6M NaOH, which enables detection and counting of the tracks with an optical microscope. Etching is carried at a slightly elevated temperature (50-60 ºC) for about an hour. Furthermore, the tracks can be widened by electrical methods after chemical etching and detected by image analysis techniques. Figure XIII.11 presents a scanned Makrofol film for determination of radon content in the indoor air by counting the number of tracks on the film and magnified image of the tracks by optical microscope.

Figure XIII.11. a) Indoor radon monitor having a polycarbonate film detector b) Tracks due alpha particles from radon in polycarbonate film, magnification 40, the photographed area is about 1.3 by 1.0 mm (http://pages.csam.montclair.edu/~kowalski/cf/327squeeze.html).

There are a number applications of SSNTD methods, but here only two, alpha track analysis and fission track analysis, are briefly described. In environmental radioactivity research they are typically used to locate and quantify alpha-emitting radionuclides and fissile material in low concentrations in soil or sediment, for example. Figure XIII.12 shows an image of an alpha-emitting particle in sediment sample. Here, the active particle is mostly embedded among other, non-active material.

Figure XIII.12. An SSNTD image of an alpha-emitting particle in sediment (J. Jernström, M. Eriksson, J. Osán, G. Tamborini, S. Török, R. Simon, G. Falkenberg, A. Alsecz and M. Betti, Non-destructive characterisation of low radioactive particles from Irish Sea sediment by micro X-ray synchrotron radiation techniques: micro X-ray fluorescence (μ-XRF) and micro X-ray absorption near edge structure (μ-XANES) spectroscopy, J. Anal. At. Spectrom. 2004, 19, 1428-1433).

Figure XIII.13. Fission tracks on a polycarbonate SSNTD (http://barc.ernet.in/publications/nl/2005/200506-2.pdf).

Presence and amount of fissile materials, particularly of ²³⁵U and ²³⁹Pu, can be determined by fission track analysis. The sample with the apposed SSNTD is exposed to thermal neutron radiation resulting in fission events in the material. The fission fragments cause tracks in the detector and the amount of fissile material can be determined by taking into account the number of the tracks, neutron flux, exposure time and the reaction cross section. An example of fission tracks seen on a polycarbonate detector is shown in Figure XIII.13.