Algorithm to catch nuclear traffickers

A US research team has discovered that coupling commercially available spectral X-ray detectors with a specialised algorithm can improve the detection of uranium and plutonium in small, layered objects such as baggage. Their study has been published in the Journal of Applied Physics.

The US International Atomic Energy Agency (IAEA) claims that the greatest danger to nuclear security comes from terrorists acquiring sufficient quantities of plutonium or highly enriched uranium (HEU) to construct a crude nuclear explosive device - materials which may be trafficked in small quantities which are hard to detect.

“Current radiographic methods are limited in their ability to determine the presence of nuclear materials in containers or composite objects,” said the researchers. “A central problem is the inability to distinguish the attenuation pattern of high-density metals from those with a greater thickness of a less-dense material.”

The new approach enhances the powers of X-ray imaging and provides a new tool to impede nuclear trafficking. The study was conducted by a joint research team from the University of Texas at Austin (UT) and the Department of Energy’s Pacific Northwest National Laboratory (PNNL).

“We first had to develop a computational model for how X-rays move through materials and how they are detected so that we could predict what an image will look like once the radiation passed through an object,” said co-author Mark Deinert from UT. “With that in hand, we applied an ‘inverse algorithm’, varying the composition of the object until the predicted image matched the measured one. We also gave our algorithm additional details about density and other factors - a process called ‘regularisation’ - to adaptively enhance its ability to discriminate materials.”

This is a simulated inspection of a layered baggage-like object that contains a thin, shielded plutonium wedge (a), not to scale. A single energy radiograph is shown in (b) along with the material estimations from the adaptive inverse algorithm which show an equivalent of colour vision (by material) for a single-view radiograph using spectral X-ray detector data. Credit: M Deinert/UT Austin.

“We wanted to show that spectrally sensitive detectors can be used to discriminate plutonium and other high-atomic-number elements from multiple layers of other materials using a single-view radiograph,” said lead author Andrew Gilbert, a doctoral student of Deinert’s working at PNNL. “In simulated radiographs, we were able to detect the presence of plutonium with a mass resolution per unit area of at least 0.07 g/cm^-2; in other words, we can locate a sample of plutonium with a thickness of only 0.036 mm.”

Deinert said his team will next expand the concept to improve detection on a larger scale, with the authors stating they will look at adapting the algorithm for use with multiple X-ray spectra and detectors that cannot discriminate X-ray energy.

“This type of imaging offers less information as the K-edges and low-energy uniqueness of the attenuation coefficients cannot be directly observed,” they noted. “However, it allows inspection of larger and denser objects because X-ray sources can be used with energies beyond what is detectable by spectral radiography detectors.”