GAMMA-RAY SCINTILLATION COUNTING

In a scintillation spectrometer, the measurement of absorbed radiation energy is accomplished by determining the intensity of each scintillation produced by radiation as it impinges on a phosphor, or fluor. The excited states produced by the deposition of the energy decay by emitting photons which can be observed and analyzed. The difference between fluorescence and phosphorescence is a function of the rate of decay of the excited state of the material: if it is rapid, 10^-8 sec or faster, the material is a fluor; if slower, it is a phosphor.

Scintillators are divided into two classes, based on their physical structure: organic and inorganic. In general, the inorganic scintillators are widely used for beta and alpha measurements, and are available in liquid and solid forms. They have faster response than the inorganics, but produce less light, and yield less photoelectric events due to the low Z of the material.

Inorganic materials are crystalline, transparent, and doped with an impurity. The most widely used is Sodium Iodide, doped with Thallium [NaI(Tl)]. It is linear in its response to various energy gammas. The function of the dopant is to shift the wavelength of the photon emitted by the excited molecule to a value which is not absorbed by the crystal. Typical Tl atom concentrations are about 0.1%.

In a solid-state sense, a gamma-ray interacting with the crystal moves electrons from the valence band (by ionization of the molecules) to the conduction band. The positively charged "hole" drifts to an activator (Tl) site, ionizing the activator. A conduction band electron then drops into the hole, forming an excited neutral activator, which then decays with the emission of a visible photon. Since this photon energy is less than the width of the " forbidden gap" between the conduction and valence bands, the crystal cannot absorb the photon, resulting in near-perfect transmission of the flash.

If accurate, quantitative measurements of gamma-ray energy are to be made, the intensity of the scintillation must be proportional to the energy of the energy of the ray. If the energy of the incident radiation is only partially absorbed by the crystal, with some part lost by scattering, a distortion of the data will result.

The energy of the gamma-ray is transferred to the sodium iodide crystal by three processes: the photoelectric effect, the Compton effect, and pair production (at high photon energies).

In a photoelectric interaction, the entire energy of the photon is transferred to an electron, which is called a photoelectron. It is a highly ionizing particle and has a path length of less than a millimeter in the crystal. The intensity of the scintillation produced along the path of this photoelectron is proportional to the incident photon energy.

A Compton electron acquires only part of the incident photon energy. The balance is carried by the scattered gamma to another site, often outside the crystal. If this scattered photon has a photoelectric collision before leaving the crystal, the entire energy of the incident photon is dissipated in the crystal, preserving the proportionality of total scintillation intensity to incident photon energy.

If the incident photon energy is in excess of 1.02 MeV, pair production also occurs. This process is always accompanied by annihilation radiation (two 0.51 MeV photons). One or both photons may leave the crystal.

We see that the scintillator should be as dense as possible, and as large as possible. Small crystals are satisfactory for low energy photons which have small mean free paths, but large crystals are required for satisfactory performance at higher photon energies.

The detector assembly of a scintillation counter contains the scintillating crystal, a light-transmitting coupling, and a photomultiplier tube, all encased in a light-tight container. The photomultiplier tube converts each scintillation in the crystal to an electrical pulse. If it is assumed that the efficiency of light collection by the photocathode is independent of the location of the pulse in the crystal, and that an electron is emitted from the photocathode for each photon striking it, then the output pulse-height is proportional to the scintillation intensity. Amplification by the photomultiplier tube is about 10⁹.

The picture shows the construction (idealized) of a typical solid-crystal scintillation detector commonly used for the measurement of gamma radiation. One gamma ray is shown entering the crystal, where it interacts and produces electrons which, in turn, generate photons in the visible range. One such photon is shown moving form the crystal, through a light-coupling system (designed to pass the light with minimum losses) onto a photocathode, where it ejects electrons from the material. The sketch shows one of these photons accelerated toward a dynode maintained at a positive potential relative to the photocathode. The energy gained by the electron results in the ejection of several electrons upon impact on the dynode (only two are show for clarity). The cascade produced...dynode to dynode...illustrates the amplification achieved. The high voltage connections required to maintain the dynodes at proper voltage is not shown. The entire structure, photomultiplier tube, crystal, and coupler, is encased in a light-tight structure to prevent ambient light from entering the system...In operation, even a small light leak will render the system useless.

A preamplifier transmits these small pulses to an amplifier where they are linearly amplified to produce output pulses typically ranging from 0 - 10 volts. The pulse-height analyzer sorts pulses according to their height (amplitude or voltage). Since pulse-height is proportional to the incident radiation energy, the analyzer is cataloging the photon energy deposited in the crystal. Typically, a lower discriminator level, E_L, is established and pulses with an amplitude less than E_L are rejected and do not appear at the analyzer output. An upper discriminator level E_U is also established, and pulses above this value will also be rejected. Thus, only those pulses in the window E between E_L and E_U pass through to the scaler or rate meter.

Setting the lower discriminator to its minimum (0%) setting, and the upper to its maximum (100%) will allow all pulses to be counted, and the set-up will behave in much the same manner as a Geiger device, ie: every event is counted. However, single channel scalers are often used with the window set to count only the energy area of interest. If, for example, it is desired to determined the uncollided gammas at some location, the window could be set to define a narrow domain about the source energy, rejecting the scattered gammas from the same source (e.g., if the source was ⁵⁴Mn which has a .835 MeV gamma, setting the window from .750 to 9.50 MeV would count the uncollided gammas and reject most of the scattered gamma component).

With patience, a single channel scaler may also be used as a spectrometer by making repeated counts of a sample at successively higher window settings (e.g., window from 0 to 1%, next from 1 to 2%, all the way up to 99 to 100% ). Plotting these counts versus energy will yield a differential spectrum.