Chapter 11 from BASICS OF NUCLEAR PHYSICS AND OF RADIATION DETECTION AND MEASUREMENT - An open-access textbook for nuclear and radiochemistry students by Jukka Lehto

In an ordinary radiochemical laboratory the alpha-emitting radionuclides studied are those listed in Table XI.I. Of these Po, Ra, Th and U isotopes are naturally occurring radionuclides while Pu and Am isotopes are artificial transuranium nuclides. The natural alpha-emitting radionuclides belong to the decay series beginning from ²³⁸U, ²³⁵U and ²³²Th. The sources of the transuranium elements are the the nuclear weapons tests in the 1950' and 1960's and of the Chernobyl accident in 1986 as well as the nuclear waste, especially the spent nuclear fuel. Some of these radionuclides, such as ²³⁵U, ²²⁶Ra and ²⁴¹Am emit gamma radiation, which can in some cases be used for their measurement. The intensities and/or gamma ray energies are, however, typically so low that the gamma spectrometric measurement does not yield accurate results. Moreover, gamma spectrometry does not allow determination of isotopic composition of uranium, which is important information in many studies. Accurate measurements, enabling also determination of isotopic compositions, are obtained either by alpha spectrometry of by mass spectrometry. The former is discussed here in this chapter.

Table XI.I. Most typical alpha-emitting radionuclides studied in radiochemical laboratories.

Semiconductor detectors were discussed already in chapter IX where gamma spectrometry was described. In gamma spectrometry the detector material is germanium whereas in alpha spectrometry the material is silicon. The principal idea in both is the same. They are both diodes composing of an n-type Si/Ge, having an electron donor additive, such as phosphorus, P(V), and a p-type Si/Ge, having an electron acceptor additive, such as boron, B(III). When these are attached to each other a depletion zone is developed around the interface due to combination of electrons and holes on the interface. When a reverse bias voltage is applied across the crystal this depletion zone widens. These are transferred close to electrodes due to the voltage applied. The thickness of the zone is dependent on the voltage applied being typically only 40-60 V in silicon alpha detectors. For the detection of gamma rays the depletion zone needs to be thick, several centimeters, in order to absorb the readily penetrating gamma rays. This is accomplished by using a larger crystal made of very pure germanium and by using a high voltage up to 5000 V. In the case of alpha detection with Si-detectors the depletion zone should be very thin due to the short range on alpha particles in silicon, only 30 μm. Typically the depletion zone in silicon detectors used in alpha spectrometry is 100-200 μm. There are two types of silicon detectors in production (Figure XI.1): surface barrier detectors (SBB) and passivated ion-implanted detectors (PIPS) the latter being a more modern construction mode.

Figure XI.1. Production of silicon detectors for alpha spectrometry. Left: surface barrier detectors. Right: Passivated ion-implanted detector. (http://www.ortec-online.com/Products-Solutions/RadiationDetectors/silicon-charged-particle-detectors.aspx).

To produce a detector, the edges of a silicon wafer, with thickness less than 500 µm, are first insulated from each other to prevent continuous current across the wafer. In SBB detectors the insulation is done with epoxy resin and a ring mounted around the wafer. In PIPS detectors the surface of the wafer is first passivated by heating which results in the formation of about 50 nm thick non-conducting SiO₂ layer. This layer is removed from the middle of both sides of the wafer. To transform the silicon wafer into a diode the other side is treated with acceptor atoms producing p-type layer and the other side with donor atoms to produce n-type layer. In the SBB detectors this is accomplished by forming a thin, 100-200 nm, layer of Au on the other side (n-type) and a layer of Al on the other (p-type). In PIPS detectors this is done by ion-implantation technique by bombarding high energy atoms on the sides. The PIPS detectors have several advantages over SSB detectors:

• The surface layer is mechanically and chemically more resistant and can be cleaned with alcohol, for example. In SBB detectors the gold surface is very sensitive and cannot be touched at all.
• The “window”, the passive layer on the surface is somewhat thinner resulting in a better energy resolution.

The detectors are rather small in size (Figure XI.2.) Their diameters are only 2 to 4 centimeters and thickness less than 500 µm (about 1 cm including the metallic cover). Table XI.II. shows properties of alpha detectors available from Canberra. As seen from the table the resolution is very good, 20-40 keV. Resolution is here determined for ²⁴¹Am 5.486 MeV alpha particles. Thus the relative resolution is 4-8%. The resolution is better for the smallest detectors, being about two-times better for the smallest detector in Table XI.II compared to the largest. The background in alpha detectors is very low, only 4-16 counts per day being directly proportional to the surface area of the detector. The background is almost solely caused by cosmic radiation. It is evident that the counting efficiency is better for the larger crystals. Thus one needs to make a compromise with respect to resolution on the one hand and to counting efficiency on the other when selecting a detector. The selection depends naturally on what is needed, high resolution or high efficiency. When measuring alpha activities in environmental and biological samples the activity levels are typically very low requiring very long counting times. In this case a larger detector would be desirable. On the other hand, the background pulses increase with detector size and the overall performance is also dependent on the sample size in comparison with the detector size. In some cases highest possible resolution is a priority. For example, typically ²³⁹Pu and ²⁴⁰Pu activities cannot be measured individually from an alpha spectrum due to overlap of their alpha peaks. Using a high-resolution alpha detector one may distinguish these two nuclides by deconvolution technique separating overlapping peaks with a mathematical fitting process.

Figure XI.2. Canberra alpha detectors (http://www.canberra.com/products/detectors/pdf/ passivated_pips_C39313a.pdf).

Table XI.II. Properties of Canberra silicon detectors for alpha spectrometry (http://www.canberra.com/products/detectors/pdf/passivated_pips_C39313a.pdf).