Radiation detection and measurement education is necessary for a wide variety of people, such as Nuclear Engineering students, radiation workers, radiotherpahy and others working with radiation. The equipments and laboratory setup needed for delivering this type of education are expensive and difficult to assemble due to the large variety of the type of experiments that are desirable to cover. For a basic radiation detection and measurement laboratory education, one needs to have gamma, beta, alpha and neutron sources, at least one detector to detect each type of radiation and other supplemental nuclear instrumentation to perform experiments. Although the radiation exposure due to these radioactive sources in an education laboratory is low, shielding and a private secured place is needed. Nearly two hundred thousand dollars is needed to construct such a basic laboratory described as above. In order to overcome these difficulties, a model of a basic Radiation Detection Laboratory has been created to provide a virtual environment for designing and simulating such experiments. This software, which is a GPL licensed, open source, free software, is an easy to use solution for Radiation Detection and Measurement education. Project team : Dagistan Sahin, Muzaffer Sena Sahin, Korcan Kayrin.


Radiation can be electromagnetic radiation such as gamma and X-rays, or particle radiation such as beta, positron, neutron or heavy charged particle, such as alpha and ions. Heavy charged particles mainly interact with matter due to Coulomb forces. These particles have a very short range in matter. Beta electrons and positrons also have short ranges compared to that of neutrons or gammas. However, neutrons and gammas are neutral particles, and have longer ranges of interaction through matter. This study considers only neutron and gamma interactions with matter.

There is a lot of work done to simulate these types of experiments using Monte Carlo methods. However, most of the avaliable codes are for experienced users and require long learning curves.  Furthermore, these specific nuclear codes used in these simulations are difficult to setup and edit without deep knowledge about radiation physics and an understanding of the codes. To achieve these difficulties experienced by the end user, an extendible, flexible and easy to use environment is created using a high level language JAVATM. A full analog Monte Carlo model is implemented for radiation sources, detectors and intervening materials such as shielding and collimators. The necessary nuclear instrumentation for experiments are also modeled both graphically and numerically to maintain a consistent and real-life feeling virtual environment. The analog Monte Carlo simulation for these detectors  uses available cross-section data libraries [1]. Gamma cross-section data from XCOM [2] , and neutron cross section data from ENDF [3] were used in RADlab. Beta and alpha interaction is implemented using stopping power data from SRIM software [4] and NIST database [5]. Available gamma detectors in the RADlab software are CdZnTe, NaI(Tl)3x3, NaI(Tl)5x5, NaI(Tl)7x7, NaI(Tl)2x2, NaI(Tl)2x2 Well Type, BGO, HPGe, Geiger Muller. Neutron detectors are BF3 1x7, BF3 3x7. Beta-Pips and Alpha-Pips detectors are modeled as beta and alpha detectors, respectively. The environment, in which the experiments are carried on, can be set to air, water or vacuum. For the environments, again the cross-section data from the mentioned databases are used.

 Moreover one needs some shielding material to measure its properties, such as the attenuation coefficient. An Aluminum shield is modeled as an example for this purpose. A useful collimator is modeled made out of pure lead . The basic nuclear spectroscopy instruments are also modeled in RADlab. These are high voltage supplies for the detectors, Preamplifier and Amplifier for signal filtering, Multi Channel Analyzer (MCA) and Oscilloscope to visualize the signals. Additionally, instruments such as Coincidence Unit, Delay Amplifier, Single Channel Analyzer (SCA), Random Pulse Generator and Counter are also modeled.  Some of the most popular radioactive sources are modeled using nuclear data from Lund/LBNL database [6].All of the sources are assumed to be isotropic point sources. The gamma sources 133Ba, 60Co, 137Cs, 22Na and a mixed source are available in the program. Two neutron sources modeled as Americium-Berillium and natural 252Cf. Alpha sources avaliable in RADlab are 241Am, 230Th and 148Gd. and beta sources are 137Cs and 204Tl.

Experiment Simulation Model

The program gives the flexibility to create experiments by choosing from variety of sources, detectors and instruments, using a simple experiment creation wizard. Once the user constructs the experiment, the environment in which the experiment would be performed can be selected, and changed durin the experiments. A cool feature is to set the environment to air, water or vacuum. 
After that the user starts an experiment, the analog Monte Carlo engine in RADlab starts the simulation by tracking the sources first.  RADlab uses Javalution api to enchance computation by using parallel computing. For every ten milliseconds, the sources in the environment decay via forced decay. The radioactive particles generated from the sources initially penetrate to the corresponding environment isotropically. While the particles travelling and interacting with the environment, the program checks whether they interact with a detector, material, shield or collimator. When a particle interacts with a detector and deposits some amount of energy, the detector generates a signal on its output. Then this signal is transferred to the instrument connected to the detector. When an instrument receives a signal it processes the signal and transmits to its output.  Therefore, the experiments are quite realistic in terms of random nature of radiation and the interactions with matter. Hence the radioactive sources decay randomly based on their decay constant, sum and escape peaks are generated in the spectrum.
The working principles of the nuclear instruments are modeled so that they behave similiar to the instruments found in a real radiation detection and measurement laboratory. When the final instrument is a MCA or Oscilloscope, user sees the output. Furthermore, the program can draw the particle paths while the simulation is running to give more visual inside of particle interactions. This feature is especially usefull for teaching purposes. Four type of experiment can be simulated using Radlab. These are gamma experiments. neutron experiments, alpha and beta experiments.

Gamma Experiments

Gamma particles interact with matter basically with three major ways. These are Photoelectric Effect, Compton Scattering, Coherent Scattering and Pair Production [1].

The dominant low energy photon process is the Photoelectric Effect. In photoelectric absorption the incoming gamma ray interacts with a nucleus of the material and disappears, transferring all of its energy to the absorber atom. If the kinetic energy of the gamma ray is larger than the binding energy of the electrons of the nucleus, it may cause emission of an electron with excess energy, i.e. the gamma-ray energy minus the binding energy. The atom is then left in excited state. The excited atom releases fluorescent X rays and Auger electrons and returns to its ground state. If this process happens in a detector, the whole energy of the gamma photon is assumed to be deposited in the detector.

Compton Scattering is an inelastic scattering of a photon from an electron in the shell of a nucleus. Compton scattering is the process by which a gamma-ray is scattered with an outer bound electron of the material. The gamma-ray after scattering has a new direction angle with respect to its original direction. This process is modeled using Klein-Nishina [7] sampling model for incoherent scattering.

Coherent Scattering is an elastic scattering of photons from atoms. This process is an elastic scattering event which occurs when a gamma-ray interacts coherently with electrons of a nucleus [18]. The gamma ray loses only an infinitesimally small amount of energy in this event [18]. This process can be neglected in most practical applications. This process has a very low probability for a wide energy range of photons. So this process is neglected in RADlab.

When a high energy photon interacts in the fields of a nucleus, it can annihilate and produce an electron-positron pair. This process is called Pair Production. When a gamma-ray energy is more than or equal to two times the rest mass energy of an electron, and the gamma ray travels within the vicinity of the nucleus, then the gamma ray may be converted into two particles. An electron and a positron are emitted. They share the excess kinetic energy of the gamma-ray above 1.02 MeV. After a very short time, the positron particle annihilates with an electron, and two gamma rays are emitted in the opposite directions to each other and both have 0.511 MeV kinetic energy. When this process happens within a detector in a RADlab simulation, the energy difference between gamma energy and 1.022MeV is assumed to be totally deposited in the detector, and two gammas are produced and released from the origin of event as it happens in the process, so that the single and double-escape peaks are generated in the gamma spectra.

All of the gamma sources, gamma detectors and instruments can be used to create various gamma experiments. The user can analyze the spectrum for different types of sources. And the user can use different types of detectors listed above to analyze the responses of these detectors via their spectrums. Moreover the user can use different sizes of the same detector to see the size effect. For example there are NaI5x5 and NaI7x7 detectors which are not likely to be found in basic laboratories. Using these detectors and the NaI3x3 detector the user can create an experiment in which one can see the effect of detector size on the spectrum and on the geometric efficiency.
For instance, using the Aluminum material an experiment to determine the attenuation coefficient of Aluminum can be constructed, which is a quantity that gives information about how much of the incoming photons are transferred through the material. In Figure-1, the aluminum attenuation coefficient experiment setup is constructed with a NaI2x2 detector, 137Cs source and relevant nuclear instrumentation.

Figure 1: Attenuation Coefficient Experiment

Coincidence Unit, SCA, Delay Amplifier and the other instruments allows the user to simulate gamma-gamma, gamma-beta coincidence experiments. Coincidence experiments are generally used to determine the absolute activity of a given radioactive source or to analyze nuclear decay properties. In the Figure 2, gamma-gamma coincidence experiment setup is shown for a 22 Na source with two NaI3x3 detectors and other instrumentation.

Figure2: Gamma-Gamma Coincidence Experiment

Neutronic Experiments

The interactions processes of neutrons with matter are fundamentally different from those for the interactions of photons. Whereas photons interact, more often, with the atomic electrons, neutrons interact essentially only with the atomic nucleus. These interactions can be grouped in two main mechanism, scattering and radiative capture [8]. Scattering of a neutron can be elastic or inelastic. If an elastic scattering occurs it is assumed to be isotropic, and the no energy is deposited within the detector. Inelastic scattering is not modeled hence low probability and practicality. In a radiative capture event the neutron is captured by the nucleus, and one or more gamma rays are emitted. This reaction is very important since it can cause the new nuclide to emit charged particles, which deposit their energy in the detector. Other neutron interactions are not needed for this simulation and they are ignored. 

Neutrons are neutral particles, and they are difficult to detect when they are highly energetic, since they would not interact with the detector. In Radiation Detection and Measurement education the most popular experiment is to investigate the slowing down properties of neutrons in water. The neutron detectors and sources can be used to create and simulate such experiments in Radlab. In the Figure 3, an AmBe source is used as e and a BF3 1x7 was  used as a neutron detector. The spectrum properties, such as the visible Wall Effect, shows how realistic the RADlab analog Monte Carlo engine performs the simulations.

Figure 3: Am-Be Source  BF3 detector experiment

Charged Particle (Alpha and Beta) Experiments

Because alpha particles are so massive, they are only slightly deflected when they interact with atomic electrons. They therefore move in more or less straight lines as they travel in a medium [8]. While travelling they lose their energy primarily by Coulomb field. Neglecting other interactions the energy loss of alpha particles can be modeled using stopping power data [4]. When an alpha particle penetrates into a detector it deposits energy using this assumption. The attenuation of beta rays in matter in some ways more complicated than alpha particles [8]. For performance reasons and simplicity the beta particles are modeled like alpha particles.

The major experiments about alpha sources are to determine their range in materials, which can be simulated using Radlab. For the beta sources the end point energy determination experiments can also simulated using Radlab. The Figure 4 shows the experiment for 230Th spectrum with alpha pips detector and the Figure 5 shows the spectrum of 137Cs with beta pips detector.

Figure 4: 230Th Alpha Spectrum

Figure 5: 137Cs Beta Spectrum

An easy to use simulation tool for radiation detection and measurement education is created. The software’s computer resource requirements are low so that with a modern personal home computer one can use this software efficiently. Furthermore in RADLAB one can setup experiments which may be impossible to construct in real life. For example in RADLAB one can make a gamma experiment easily in water, which will be very hard to construct in real life. Furthermore, RADLAB creates a safe environment for the student, since there is no real radiation risk and also no expensive instrument damage risk while learning radiation detection and measurement.

Future Work

Since the software is not developed with a professional team of developers there are some bugs to be fixed. However current release of the software is operational and usable. Neutron activation analysis is used to determine the concentrations, ingredients of a known or unknown material. This requires a nuclear reactor available to irradiate the sample and a radiation detection laboratory to analyze. Furthermore while during such an experiment the experimenter needs to predict the activity of the sample after irradiation. So the neutron activation analysis and spectroscopy module will be added to Radlab in order to enable activity, dose and spectrum prediction before neutron activation experiments.


[1] Book, Alex F Bielajew, Fundamentals of the Monte Carlo method for neutral and charged particle transport, The university of Michigan,September 17,2001
[2] XCOM Database, NIST National Institute of Standards and Technology, http://physics.nist.gov.
[3] ENDF Database, Experimental Nuclear Reaction Data, IAEA, http://www-nds.iaea.org/
[4] SRIM-2008 Sofware , http://www.srim.org/SRIM/SRIMLEGL.htm
[5] Stopping-Power and Range Tables for Electrons,Protons, and Helium Ions, http://physics.nist.gov/PhysRefData/star/Text/contents.html
[6] S.Y.F. Chu, L.P. Ekstrom and R.B. Firestone,Lund/LBNL Nuclear Data, nucleardata.nuclear.lu.se
[7] Ivan Lux, Ph. D.Laszlo Koblinger, Ph,Monte Carlo Transport Methods: Neutron and Photon Calculation, CRC Press Inc., 1991
[8] J. Kenneth Shultis, Richard E. Faw, Radiation Shielding,Prentice Hall PTR, New Jersey,1996