© European Physical Society, EDP Sciences, 2025

Muons are elementary particles similar to electrons but with a mass approximately 200 times greater. They are produced naturally in the upper atmosphere by the interaction of cosmic rays with atomic nuclei and can also be generated artificially in accelerators. Muons interact less with matter compared to other charged particles such as electrons or protons, allowing them to penetrate substantial thicknesses of dense materials such as rock, concrete, or metal. This high penetration power makes muons particularly valuable for non-invasive imaging applications, especially in contexts where conventional radiation (e.g., X-rays) is insufficient.

This property of muons, together with the fact that cosmogenic muons are naturally and freely produced and are easy to detect, forms the basis of “muography”. Its first documented application dates back to 1955, to measure the overburden of a tunnel in Australia. The second, and most famous, application was published in 1970, when L. Alvarez’s team used cosmic muons to search for hidden chambers inside Khafre’s pyramid in Egypt. More applications followed, especially in the 21^st century, and muography is nowadays a booming research area, with a steadily growing trend of publications.

Cosmic-ray muography

Two main muography techniques are used, as illustrated in Figure 1. Absorption muography relies on the energy loss of muons through ionization as they pass through matter. Muons lose energy via ionization and eventually decay, leaving no detectable signal. Muon trajectories are reconstructed using a tracker placed downstream of the object, and the deficit of muons with respect to the free sky is used to infer the amount of matter traversed. This technique is directly sensitive to the material’s density, allowing researchers to test different hypotheses about an object’s internal composition. Scattering muography instead exploits the angular deflection of muons caused by interactions with atomic nuclei. The average deviation depends strongly on the atomic number (Z, i.e. number of protons) of the material, enabling elemental discrimination. To measure this deviation, the muon path must be tracked both before and after crossing the object, requiring it to be “sandwiched” between two trackers.

FIG. 1 Scattering (top) and absorption (bottom) muography of a small statue. Study based on simulated data, to illustrate the different imaging quality achievable with the two methods for a relatively small object (80 cm height). Reproduced from: iScience 28, 112094 (2025)

Each method has distinct strengths and limitations. Absorption muography requires only one tracker and yields 2D density projections, which can be combined from multiple angles to form a 3D density map. However, it does not distinguish between material properties beyond density, and lacks sensitivity for small features. Scattering muography inherently provides 3D information and is sensitive to atomic composition due to the dependence of scattering angle on Z. Yet, it demands a dual-tracker setup and is unsuitable for human-sized statues unless the object can be placed between detectors or a complex detector installation is undertaken. Thus, it is best suited to relatively small objects.

From pyramids to domes: muography in cultural heritage studies

Muography has already proven effective in studies of large-scale cultural heritage. A landmark example is the ScanPyramids project, which used three detector types (plastic scintillators, micro-pattern gaseous detectors, and nuclear emulsions) to reveal a low-density anomaly in Khufu’s Great Pyramid in 2017. Additional data have been collected since then, leading to the announcement in 2023 of a corridor-like structure of about 9 m length with a transverse section of about 2 m by 2 m. Ground Penetrating Radar and ultrasonic tests reinforced the findings, leading to a successful visual confirmation via endoscope (Figure 2). Other recent successes of muography include the detection of potential structural hazards in the Xi’an city wall (China) – Figure 3 - the identification of unknown internal features and voids in the Svyato-Troitsky Danilov monastery (Russia), and evidence for unknown structures in an archaeological site in Naples (Italy), located 10 meters below street level, including one compatible with the hypothesis of a hidden, currently inaccessible, burial chamber.

FIG. 2 A hidden corridor inside Khufu’s Great Pyramid photographed by an endoscope. The existence of the corridor was first discovered using muography (Nat. Commun. 2023; 14:1144) and confirmed by an independent team using GPR and ultrasonic testing (NDT E Int. 2023; 139, 102809). Photo: Egyptian Ministry of Tourism and Antiquities.

FIG. 3 Low-density regions identified via 3D inversion of muography data in the Xi’an city walls. These include a known cavity (the transformer room) and an unexpected pattern near the northern surface of the rampart, hinting at a potential issue for its stability. Reproduced from J. Appl. Phys. 133, 014901 (2023).

The examples above all rely on absorption muography, but alternative approaches have also been proposed. One such proposal involves using scattering muography to detect iron chains within the masonry of Florence Cathedral’s dome, supported by a successful proof-of-concept on a mock-up wall. A newer approach, alignment-based muography, has been proposed for the Palazzo della Loggia in Brescia (Italy). It uses the fact that muons travel in straight lines in the absence of electromagnetic fields. It requires two muon telescopes: one fixed to a reference structure and the other on the element being monitored. Misalignment of reconstructed muon tracks indicates displacement, enabling long-term structural stability monitoring.

A research gap: muography for medium-sized cultural artifacts

In all the examples discussed so far, the muon detectors used are relatively large and not easily transportable. This is because large heritage structures significantly attenuate the cosmic muon flux, requiring detectors with a wide cross-sectional area (on the order of a square meter) to collect sufficient data within a reasonable time. However, muography remains under-explored for medium-sized cultural heritage objects, such as human-sized statues. These are too large for X-ray or microwave imaging, yet too small for standard absorption muography, which suffers from poor resolution at this scale. Their intermediate size presents a unique challenge, calling for tailored detector designs and advanced reconstruction methods to make muography viable for such artifacts. In this regime, scattering muography offers better spatial resolution as well as the advantage of material discrimination through sensitivity to atomic number, but its requirement for detectors on both sides of the object creates extra logistic challenges in many cultural heritage settings.

Muography may be useful for detecting hidden features or internal anomalies in heritage objects. It can also serve as a non-invasive moisture sensor: water infiltration alters the effective density of porous materials, leading to detectable variations in muon attenuation. The detection of internal cracks or defects in statues remains a challenging application of muography. However, ongoing research in civil engineering, using muography to inspect aging infrastructure, may drive developments that could benefit cultural heritage studies as well. Addressing this research gap will benefit from compact, portable muon detectors. In confined spaces like museums or churches, equipment must be small, easy to install, and possibly removable outside of public hours.

From the cosmos to accelerators

Cosmic-ray muography also has some fundamental limitations, the most blatant being that the muon flux is relatively small, of the order of 100 Hz/m²; i.e. roughly a muon per minute through the surface of a thumb. This means that very long exposure times - from several hours to several months, depending on the size and density of the object to be imaged - are needed in order to form a useful image. Moreover, cosmic muon direction and energy cannot be controlled. One can, however, overcome these drawbacks by using a muon beam produced by a particle accelerator. A few facilities in the world, e.g., CERN and PSI in Switzerland, ISIS in the UK, TRIUMF in Canada, FNAL in USA, and J-PARC in Japan, are able to produce muon beams with precisely controlled energy and high intensity. While mostly produced for the purposes of fundamental physics experiments, these muon beams can be used for cultural heritage studies. A technique called Muonic Atom X-ray Emission Spectroscopy (μ-XES) or Muon Induced X-ray Emission (MIXE) exploits beams of negative muons at sufficiently low energy to be stopped (through energy loss by ionization) inside the object of interest. Stopped negative muons form socalled “muonic atoms” by displacing an electron in one of the outer atomic shells, and these atomic systems relax to their lowest-energy state by emitting X-rays, which are in turn detected and whose spectra depend on the atomic number Z of the material. With respect to the conceptually similar PIXE and PIGE techniques, which rely on proton beams, MIXE has the advantage of probing deeper in the material, thanks to the large muon penetrating power. Moreover, activation of the material is negligible with respect to irradiation with neutrons or protons.

While this method is very promising for the analysis of relatively small objects (it has been used e.g. for the study of ancient Roman and Japanese coins), it suffers a fundamental limitation: the objects to be analyzed have to be transported to one of the few existing muon beam facilities in the world. Some very ambitious ideas have been proposed toward a transportable muon generator, but artificial muon production must necessarily start from high-energy collisions of other particles, which implies radiological hazards, like all methods based on an artificial particle source, because of the byproducts of the collisions.

Conclusions

Muography is a promising nondestructive subsurface imaging technique for cultural heritage analysis, thanks to the excellent penetration properties of muons. The inherent availability of cosmogenic muons led to the recent boom of cosmic-ray muography, making it a low-cost and portable alternative to other imaging methods, especially for large structures. The use of artificial muon beams from particle accelerators is becoming increasingly promising, and complementary, as they offer controlled flux, directionality, and energy, overcoming many limitations of cosmic-ray muography and enabling high-resolution imaging and elemental analysis of smaller cultural heritage objects. n

About the Author

Andrea Giammanco is a particle physicist of the Fonds National de la Recherche Scientifique (FNRS) and the director of the Centre for Cosmology, Particle Physics and Phenomenology (CP3), in Belgium. His research includes various applications of muography, from cultural heritage to volcanology.

References

All Figures

	FIG. 1 Scattering (top) and absorption (bottom) muography of a small statue. Study based on simulated data, to illustrate the different imaging quality achievable with the two methods for a relatively small object (80 cm height). Reproduced from: iScience 28, 112094 (2025)
In the text

	FIG. 2 A hidden corridor inside Khufu’s Great Pyramid photographed by an endoscope. The existence of the corridor was first discovered using muography (Nat. Commun. 2023; 14:1144) and confirmed by an independent team using GPR and ultrasonic testing (NDT E Int. 2023; 139, 102809). Photo: Egyptian Ministry of Tourism and Antiquities.
In the text

	FIG. 3 Low-density regions identified via 3D inversion of muography data in the Xi’an city walls. These include a known cavity (the transformer room) and an unexpected pattern near the northern surface of the rampart, hinting at a potential issue for its stability. Reproduced from J. Appl. Phys. 133, 014901 (2023).
In the text