Experiments with a Tiny Radioisotope Neutron Source

July 20, 2008

By placing beryllium in intimate contact with an alpha-emitting radioisotope, neutrons are produced. At home, one can approach this well-known reaction by lining up the sealed americium sources from a quantity of old-fashioned ionization smoke detectors on a sheet of beryllium metal. The neutron yield is easily detected; see this link for more information. My own toy AmBe neutron source currently produces an estimated 1000 neutrons per second. That makes it more than three orders of magnitude weaker than my Farnsworth Fusor. But are there enough neutrons to perform some detectable nuclear reactions? As it turns out…yes.

Neutrons give rise to prompt gamma rays when they are captured by many nuclei. The following experiments involve the detection of high-energy (> 4.4 MeV) gamma signatures from neutron capture in chlorine, iron, and titanium, in the form of inexpensive and readily-available compounds mixed with water and placed near the 1000 n / s AmBe source. A gamma ray spectrum is collected with a very efficient scintillation detector (2×2″ BGO) over a period of hours.



For the chlorine experiment, a 40 lb. bag of rock salt was purchased at Lowe’s for about $4. A two-gallon pail was filled to the brim with salt, followed by successive additions of water and more salt to obtain the highest chlorine density possible. A piece of PVC pipe, capped on the bottom, enters the salt pail from the top. The neutron source fits down in this pipe so that it is nearly surrounded on all sides by salt. The lead-shielded BGO scintillator views the side of the pail at the level of the source.

The iron experiment is similar, except the medium in use is 10 lb. of ferrous oxide (FeO) purchased at New Mexico Clay here in Albuquerque. The oxide is divided among two PETE jars and water is added to wet the oxide, drive out air, and fill the jars. The neutron source is taped behind one of the jars and the assemblage is surrounded by UHMW-PE bricks to act as a neutron moderator / reflector.

Titanium was obtained in the form of titanium dioxide powder from New Mexico Clay. Enough dioxide (~8 lb) was loaded into a large cylindrical Rubbermaid food-storage container to fill the sonofabitch. Water was added to displace air and top off, and as before, PE bricks were arranged around the outside. In this instance the neutron source sits in an acrylic tube penetrating the TiO2 from the top.

Results follow below…



(Click any image for a larger version.)

The gamma spectra are dominated by a 4.44 MeV gamma energy from the the neutron source itself. When the 9Be(α,n)12C reaction occurs, the carbon residue is frequently left in its first excited state at 4.44 MeV, from which it de-excites by gamma emission. The peak is broad, but the poor resolution is due to Doppler broadening rather than the detector (the 12C nucleus de-excites very rapidly, while still hauling ass post-conception). This radiation effectively masks any (n,g) counts at lower energies. However, the high-energy region above 5 MeV is left wide open, and expected signature radiations from neutron capture in 35Cl, 56Fe, and 48Ti can be clearly seen in the respective experiments. In each case, I’ve marked the spectra with colored lines indicating expected radiation. Single-escape peaks are prominent at these energies, and are marked as “E1” and color-coded according to which photopeak they are associated with.

A few concluding points:

  • The AmBe source derived from smoke detectors took hours because of how weak it is. Similar results could probably be obtained in about a minute with a suitable Farnsworth Fusor. However, electronic neutron sources have the disadvantage of creating copious electrical noise that can interfere with the detector.
  • Water is added to each “target” material to act as a moderator. The (n,g) reactions producing these gamma rays are more probable for slow neutrons than for the fast neutrons directly from the AmBe source.
  • Other elements for which this high-energy prompt-gamma technique should be easy include Hg, Ni, Cr, and possibly N. A good data reference for prompt gamma neutron activation analysis (PGNAA) is the “Database of Prompt Gamma Rays from Slow Neutron Capture for Elemental Analysis,” an IAEA publication.


  1. How are you protecting yourself? I know that neutrons are difficult to shield and have always been afraid to attempt any neutron activiation experiments, since I have no way of measuring neutron flux. I’m not afraid of gamma, because at least I can detect its presence. I have plenty of beryllium and alpha sources, but am still concerned about neutron capture in my own body. Any advice would be appreciated.


  2. […] of antimony, indium, and aluminum is possible using a homemade neutron source containing ~5.6 millicuries of Am-241 in sealed smoke detector sources pressed against beryllium.  […]

  3. Thanks for sending me this link.  It is greattly appreciated.   Art Friedman, Ph.D

  4. Hi ! My only alpha source is an Am241 button from a smoke detector, with a activity of 1 microcurie. Could a such weak source, when used with beryllium, could activate some material in a way that it could be detectable with an average geiger counter or a 1cc CsI detector for gamma spectrometry ?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: