Liquid scintillation counting in nuclear medicine

Abstract

Many of the radionuclides used in nuclear medicine can be measured by liquid scintillation (LS) counting, and the technique is the only practical approach to counting low-energy β emissions. This review is intended to be a brief exposition of the capabilities of LS counting and of some precautions that should be observed. The liquid scintillation process is basically simple: Electrons and/or photons from radioactive disintegrations are absorbed by a solvent, with the resulting energy being transferred to the π electrons of an organic scintillator. The scintillator emits fluorescence in its return to its least-excited ground state; the emitted photoelectron spectrum is proportionate to the energy of the photons emitted in the original radioactive disintegration. The photons are absorbed by the cathode of a photomultiplier tube, with the resulting pulse in voltage being amplified and resolved in nanoseconds before being recorded as a count. In modern LS counters, coincidence circuitry eliminates much of the background inherent in the counting system. Errors that must be avoided or assessed include quenching: impurity or chemical quenching that increases with the atomic number of the impurities and color quenching that distorts the photoelectron spectrum in inverse proportion to the emitted energy. Use of quench correction curves and methods used to determine the loss of efficiency or pulse height are necessary in LS counting if there is any significant variation in sample preparation. If there are two radionuclides to be counted in a sample, a quench correction must always be applied because the pulse height shifts to a lower energy region whenever quenching occurs. Thus the spillover or overlap of two isotope spectra will vary with the energy of the radionuclides and the degree and type of quenching. None of the three methods of determining the extent of quenching (internal standardization, sample channels ratios, or external standardization) is ideal for all situations or is by any means foolproof. For example, if the scintillator solution and sample are not mixed to near homogeneity, counts may be lost because of self-absorption. Excitation of the scintillation solvent is decreased and cannot adequately be comprehended by any of the quench correction techniques. Because most scintillation solvents will tolerate only minimal amounts of aqueous samples of the sort usually of interest in nuclear medicine, oxygen flask combustion (for 14 C or 3 H), hydrolysis, addition of surfactants (solubilizers), or thixotropic gels—all relatively expensive procedures—may have to be used for sample preparation. The cost of sample preparation is frequently the only significant unfavorable feature of LS counting when compared to gamma (Nalcrystal) scintillation counting.