Nuclear Structure Research

Neutron Gamma Discrimination

In some nuclear reactions free neutrons and gamma rays are produced. If the reaction that one is studying produces free neutrons, it may be useful to take data only when a neutron is present. This increases the chance that the data being taken is related to the relevant reaction and not an undesired one.

In order to acquire data only when neutrons are present (gate data acquisition on neutrons), one must first be able to detect neutrons reliably. Any particle that penetrates the casing surrounding a liquid scintillator being used as a neutron detector has a chance of causing an electrical pulse from the photo-multiplier at the back of the detector. In low energy nuclear reactions any charged particles produced will likely not penetrate the shielding of the detector. Therefor the two types of particles with which we are primarily concerned will be neutrons and gamma rays.

Fortunately for us, the electrical pulse emitted by the photo-multiplier tube (PMT) caused by a gamma ray is measurably shorter in duration than the pulse caused by a neutron. This difference in pulse duration allows us to discriminate between neutrons and gamma rays. We have two ways in in which we can set up a data acquisition gate on neutrons, either through hardware or software.

Hardware Gating

The following is a schematic diagram of the electronics that we used for hardware gating on neutrons.

When a signal comes from the Photo-Multiplier Tube (PMT) of the liquid scintillator detector it is fed into a pre-amplifier to increase its strength. The output of the pre-amplifier is then split into two channels, a fast channel (left) and a slow channel (right).

In the fast channel the Fast Filter Amp amplifies the signal again, quickly at the expense of preserving its shape. This amplified signal is fed into the Constant Fraction Discriminator (CFD). The CFD emits a pulse whenever the voltage of its input is above a user defined value. The pulse is fed to a Gate and Delay Generator which is only included in case the signal needs to be delayed to adjust the timing difference between the fast and slow channels. The signal starts a timer in the Time to Pulse Height Convertor (TAC).

In the slow channel the signal is fed into a Spectroscopy Amp which has more processing time than the Fast Filter Amp, but does a better job of preserving the shape of the signal. The preservation of the signal shape is important both because this is the signal that is sent to the ADC (Analog to Digital Convertor) as the energy signal and since we need to know the duration of the signal with some precision in order to discriminate between neutrons and gamma-rays. The Uni-Polar signal is sent to the ADC since the ADC can only interpret signals which are positive. The Bi-Polar signal is sent to the Timing Single Channel Analyzer (Timing SCA). The Timing SCA emits a pulse to stop the TAC when its input signal crosses zero voltage at the end of a pulse.

The TAC measures the time between its start and stop signals and emits a square pulse with amplitude proportional to that time through the TPHC channel. This pulse is fed into the ADC as the time signal. The SCA channel produces a pulse when the time measured by the TAC is between an upper and lower threshold set by the user. These thresholds can be set such that only signals with duration consistent with neutron events cause a signal on this channel. This signal is fed to a Gate and Delay Generator which makes a square signal on which the ADCs are gated. Thus energy and time data from the liquid scintillator should only be recorded for neutron events.

In order to record gamma-ray data only in coincidence with these neutron events one could simply gate acquisition using this same Gate signal.

List of component model numbers:Pre-Amp, Fast Filter Amp, Spectroscopy Amp, Delay Amp, Constant Fraction Discriminator, Timing Single Channel Analyzer, Gate and Delay Generator, Time to Pulse Height Convertor,

Software Gating

To accomplish neutron gating we need to record both energy and duration data for each event in the liquid scintillator. We need data analysis software that can sort and display events based on whether or not they are in coincidence with other particular events. To sort our data we used a program called SCANU, and to display it we used one called DAMM. We plotted the 2-D spectrum (Energy vs. Duration) of the liquid scintillator data. To get a better idea of which areas in the 2-D spectrum corresponded to neutrons and which to gamma-rays we compared spectra from a 152Eu gamma-ray source and a PuBe neutron source. There were gamma-rays in the spectra from both sources, but only the PuBe source spectrum had significant numbers of neutrons.

The picture above and to the left is a 2-D spectrum from a 152Eu source. The picture above and to the right is a 2-D spectrum from a PuBe source.

The peak in the 152Eu spectrum is at around 60 on the time (x) axis. The gamma-rays from the 152Eu source have signal duration that is centered at this channel. The left peak in the PuBe spectrum is at the same time coordinate which leads us to believe that it is due to gamma-rays as well. You can see from the spectra that the gamma-rays from the PuBe source extend to higher energies than those from the 152Eu source. The peak at 100 on the time axis of the PuBe spectrum is not present in the 152Eu spectrum. Since this peak is at longer duration than the gamma-ray peak, we believe that it is caused by the presence of the neutrons emitted by the PuBe source. We later used the positions of these two peaks to get a rough estimate of which sections of the 2-D spectrum corresponded to neutrons and which sections corresponded to gamma-rays.

After having studied these spectra we were able to identify neutrons and gamma-rays in our in-beam Energy vs. Duration spectrum (Below). To get better time resolution in our in-beam spectrum we delayed the start signal for the TAC. This spread out the spectrum along the time axis. This spreading shifted our gamma-ray and neutron peaks to channels 100 and 150 respectively.

When gating on the neutrons in this spectrum we chose to take events which were in

Last Updated July 27th, 2001.

Copyright © 2000-2001, Fred Letson and Josh Thompson