One of the principal advantages of NAA is that it is nearly free of any matrix interference effects as the vast majority of samples are completely transparent to both the probe (the neutron) and the analytical signal (the gamma ray). Moreover, because NAA can most often be applied instrumentally (no need for sample digestion or dissolution), there is little if any opportunity for reagent or laboratory contamination.

Interferences can occur when different elements in the sample produce gamma rays of nearly the same energy. Usually this problem can be circumvented by choosing alternate gamma rays for these elements or by waiting for the shorter-lived nuclide to decay prior to counting. The reaction we have described above is simple neutron capture. Other interferences can occur if another type of nuclear reaction concurrently produces the radionuclide of interest.

The limit of detection for a particular element will depend upon the measured count rate (R) of the gamma ray being monitored and the background upon which that gamma-ray peak sits. From equation 1, it is clear that the measured count rate (R) for a given isotope can be increased by:

• increasing the detector efficiency (moving the sample closer to the detector)
• increasing the irradiation time (t_i)
• decreasing the decay time (t_d)

In the alternative, the sensitivity of the measurement can in many cases be improved by increasing the overall signal or total number of counts (R·t_m). This is accomplished by simply increasing the measurement time t_m.

We can improve the limit of detection for an element if we increase the ratio of the activity of the gamma ray of interest to the background through any combination of steps 1, 2 or 3. Conversely, changing the detector efficiency, irradiation time, or decay time to cause a lower peak-to-background ratio will worsen the detection limit. How this ratio changes is primarily a function of the activity of other isotopes produced in the neutron irradiation of the sample.

Another common limitation of instrumental NAA occurs when the activity generated by the irradiation of the bulk matrix is so large that it masks the signal of interest. Fortunately, in NAA the excitation and measurement steps are sufficiently separated in time to allow the analyst, when necessary, to employ time delays or chemical separations to improve significantly both selectivity and sensitivity.