Mailing List

For all the latest news and features, sign up to receive our FREE updates by email:

Muon tomography: looking inside dangerous places

Posted on 19. October, 2015.

Bookmark and Share

Free Current Commentary article in Science Progress by Prof Chris Rhodes


In 1936, during a study of cosmic radiation at Caltech, Carl D. Anderson and Seth Neddermeyer detected negatively charged particles whose trajectories in a magnetic field curved less acutely than those of electrons but more acutely than for protons. On the assumption that its charge was the same as that on an electron, it was concluded that this new particle was heavier than an electron but lighter than a proton. The new particle was initially named as a mesotron, based on the Greek word meso-, meaning “mid‑”. Re‑dubbed as a muon, the particle’s existence was confirmed in 1937 by J.C. Street and E.C. Stevenson in a cloud chamber experiment.

Muons are formed when high energy protons from cosmic rays strike the nuclei of light elements in the Earth’s upper atmosphere. The process initially produces pions (and other short-lived particles, e.g. kaons), which decay over a distance of a few meters into muons, with accompanying muon neutrinos. The muon tends to continue in the motional direction of the proton that created it, and travels at a nearlight velocity. The muon decays on a microsecond timescale, and in the absence of relativistic effects would only travel a (half-survival) distance of 456 m2. However, according to the theory of special relativity, the effect of time dilation enables the muons to live long enough to reach the Earth’s surface. The effect may be viewed alternatively in terms of length contraction which, in the inertial frame of the muon, means that the critical distance that the particle must travel is shortened. Either way, it is the relativistic velocity of the muons which preserves them such that they can not only reach the Earth’s surface, but penetrate hundreds of metres into the ground. It is this highly penetrating property of muons, due to their very high momentum (typically 3 – 4 GeV/c), that allows them to be used for imaging much thicker samples than can be accessed using X‑rays. The muon flux at the Earth’s surface is of the order of 10,000 particles per square metre, per minute, meaning that every second one muon passes through an area about the size of a human hand.

Muon transmission imaging (muon radiography)

Cosmic ray muons have been used for imaging purposes since the 1950s, the pioneer being E.P. George who employed them to determine the depth of the ice burden above the Guthega–Munyang tunnel in Australia. In the next decade, Luis Alvarez, who won the Nobel Prize for Physics in 1968, for his work on the hydrogen bubble chamber, and also postulated correctly that the dinosaurs had become extinct due to an asteroid impact (not a massive volcanic eruption), used muon transmission imaging to look for hidden chambers in the Pyramid of Chephren in Giza. It is sometimes said that he found “nothing” but, more accurately, he was able to demonstrate that no such chambers were present, which is a definite result and of considerable significance. Indeed, the technique can be used to “screen” pyramids, in order that only those with apparently interesting internal features are chosen for more detailed physical exploration by archaeologists, the others being left alone. A more recent investigation is of “The Pyramid of the Sun” at Teotihuacan – “The City of the Gods” – near modern day Mexico City, discovered by the Aztecs in the fourteenth century (and many centuries after it was constructed). By volume, this is the third largest pyramid known on earth, and is 74 m in height, set on a square base with 225 m sides. Its exterior is covered with 3 million tonnes of volcanic rock, while the interior is a mound of earth. Muon detectors were placed under the centre of the pyramid, inside a tunnel that runs under its base. While the muons are deflected when they hit more dense materials, if a cavity is present, more of them will pass through to the other side of the pyramid. Thus a relative two-dimensional density map is created, from which it was concluded that, in contrast to the smaller and neighbouring Pyramid of the Moon, the Sun Pyramid contains no hidden chambers. However, the pyramid is less dense on one side than the other by 20%, which has led to some speculation that the structure could collapse, although this is disputed. A tree-covered mound, about 20 m in height, at an ancient Mayan site in Belize (“La Milpa”) is currently under investigation. The mound is termed “Structure 3”, and is one of four structures there of sufficient size that they might contain pyramids. Unlike the Chephren and Sun pyramids, which have tunnels running underneath or within them, into which muon detectors could be placed, Structure 3 has no such tunnel. Hence an alternative strategy was employed, with two solar-powered detectors placed in trenches on either side of the mound, each of which collects muons passing laterally through the pyramid. Thus, a stereographic 3‑dimensional image can be compiled of the internal structure of the pyramid.

Cosmic-ray muons have also been used to obtain images of magma chambers, in an attempt to predict when volcanoes will erupt. Tanaka and his co‑workers from the University of Tokyo have investigated the upper portion of the Asama volcano in Japan, and demonstrated that under the bottom of the crater is a region of low density rock. Such regions of low-density can be incorporated into computer simulations aimed to predict the development of forthcoming eruptions, and to identify which locations close to the volcano are the most dangerous. The same group developed a portable cosmic-ray muon telescope module, with which they were able to successfully image the density distribution of magma in the conduit of the Mt Iwodake volcano. With a 1-m2 detector placed at the foot of the volcano, a map was obtained which represented the differential absorption of cosmic-ray muons according to the varying thickness
and density beneath the crater floor. Thus, an image was obtained of the conduit itself and its structure, working on the assumption that the density anomaly is localised in the vent area. It was concluded11 that the apparent position of the magma head is in accord with the degassing model of rhyolitic systems that was advanced in 2002.
Muon radiography is able to determine internal structures to a resolution of around 10 m, which is better by a factor of 10 than the indirect methods that are typically used. Furthermore, the development of internal structures can be monitored as a function of time, so that any changes may be evidenced. The thicker is the rock, the longer is the time required to collect sufficient muons to yield an image of satisfactory quality. This is because the muon flux loses its “brightness” as it travels through the rock, and accordingly the time resolution can be weeks, months, or even years. 

The detectors used in work of this kind are dubbed “muon telescopes”, and are analogous to the film used to obtain X‑ray images. Near-horizontal muons are detectedfollowing their passage through the body of the volcano, as they emerge from the edifice of it. It is possible to determine the path of each single muon passing through the volcano, which provides a directional map of the muon absorption. To use the X‑ray image analogy, because more muons are absorbed by denser rocks, the map of muon intensities provides a “negative” of the rock densities within the body of the volcano. Although it is not yet possible to use such images to predict when a volcanic eruption will occur, they are usefully employed in computer simulations to ascertain the nature of a likely eruption.
Muons are also being investigated to image probably the most famous volcano in the world, Mount Vesuvius , which overlooks the city of Naples, and poses the greatest risk of any volcano in Europe. This is the Mu‑Ray project, which is funded by the Istituto Nazionale di Fisica Nucleare (Italian National Institute for Nuclear Physics) and the Istituto Nazionale di Geofisica e Vulcanologia (Italian National Institute for Geophysics and Volcanology), with contributions by the Italian Ministry of Education and Research (MIUR-PRIN), Fermilab (USA) and IN2P3Orsay (France), with support from the Provincia di Napoli (Province of Naples) and the Istituto Fondazione Banco di Napoli (Foundation of the Bank of Naples). To image Vesuvius is problematic, since the crater within its summit is 500 m wide and 300 m deep. This large scale means that the muons must penetrate practically 2 km of rock before they can reach the detector at the opposite side of the mountain, and it is only muons with very high energies, and travelling in an almost horizontal direction that can accomplish this. The number of muons reaching the detector is accordingly relatively low, so that it is difficult to obtain a good quality image. The detectors, made from strips of scintillator material, can be made to cover relatively large areas and are durable over long exposure times, even in the proximity of a volcano. It is intended to construct two separate 4 m2 telescope arrays, with which to count muons over a period of a year or more, depending on experience gained with the currently recording (since Spring 2013) 1 m2 prototype detector, and further funding being forthcoming.

Download the full article, free, from Science Progress, Volume 98, Number 3, September 2015, pp. 291-299.

Image: The crater of Mount Vesuvius, photographed from the air. The surrounding region is densely populated. The Mu-Ray project aims to use cosmic-ray muons to obtain an image of the interior of the volcano. Credit: Pastorius.