Que Sont les Rayons X et Comment Sont-ils Utilisés?

X-rays are a form of electromagnetic radiation with wavelengths roughly between 0.01 and 10 nanometers, placing them between ultraviolet light and gamma rays on the energy spectrum. They were discovered by Wilhelm Conrad Röntgen in November 1895, and within weeks the scientific and public reaction was enormous, because these invisible rays could pass through solid objects and reveal the structures hidden inside them. That core property still drives nearly every use of X-rays today, from a chest radiograph in a hospital to the inspection of welds in a pipeline, the mapping of protein structures in a laboratory, and the study of distant black holes in astrophysics.

How Röntgen Found a “New Kind of Ray”

Röntgen was experimenting with cathode-ray tubes in his laboratory in Würzburg, Germany, when he noticed that a fluorescent screen across the room was glowing even though the tube was covered. After seven weeks of intensive investigation, he reported the discovery on December 28, 1895, in a communication titled “On a new kind of ray.”¹ He named them “X-rays” specifically to emphasize that their nature was unknown. The announcement triggered immediate worldwide excitement: within months, physicians were using X-rays to locate broken bones and foreign objects in patients.² It took years of further work by other physicists to establish that X-rays were electromagnetic waves, just like visible light but with far more energy per photon.

How X-Rays Are Produced

In a standard X-ray tube, electrons are accelerated from a heated filament toward a metal target, usually tungsten. When those fast-moving electrons slam into the target, they lose energy in two ways. Some collide with the target’s atoms and knock inner-shell electrons out of orbit; when outer-shell electrons drop down to fill the gap, the atom emits a characteristic X-ray photon at a specific energy. The other, more common mechanism is bremsstrahlung (German for “braking radiation”): the electron decelerates as it passes near an atomic nucleus, and the lost kinetic energy comes out as an X-ray photon. Bremsstrahlung produces a continuous spread of energies up to the maximum voltage applied to the tube. Modeling this process accurately involves tracking the depth, energy, and angular distribution of electrons as they penetrate the target, combined with the physics of the bremsstrahlung interaction itself.³

The practical takeaway is that by adjusting the tube voltage and current, operators can tune both the energy and the intensity of the X-ray beam. Higher voltage means more penetrating (harder) X-rays; higher current means more photons per second (a brighter beam). This tunability is what makes X-rays so versatile across different applications.

Why X-Rays Can See Through You

X-rays interact with matter in several ways, but two dominate in the medical energy range. In photoelectric absorption, a photon is completely absorbed by an atom, ejecting an inner electron. In Compton scattering, a photon bounces off a loosely bound electron, losing some energy and changing direction. The balance between these two processes depends heavily on the composition and density of the material the X-rays are passing through.

Photoelectric absorption rises dramatically with the atomic number of the material. Bone, rich in calcium (atomic number 20), absorbs far more X-rays than the surrounding soft tissue, which is mostly carbon, hydrogen, and oxygen (atomic numbers 6, 1, and 8). The photoelectric absorption cross-section scales roughly with the fourth power of atomic number, while Compton scattering scales only linearly with it.⁴ That steep difference is the whole reason a plain X-ray image works: bone blocks X-rays effectively, soft tissue lets most of them through, and air in the lungs lets nearly all of them through. The detector on the other side records these differences as varying shades of white and black.

Compton scatter, though, is a nuisance. It is the dominant source of attenuation in diagnostic imaging energy ranges, and scattered photons that reach the detector carry no useful spatial information. In computed tomography, Compton scatter can produce “cupping” artifacts and dark streaks between dense structures in the reconstructed image.⁵ In conventional radiography, scatter can reduce image contrast by as much as half.⁶ Anti-scatter grids, beam-shaping attenuators, and computational corrections are all used routinely to mitigate the problem.

Medical Radiography and Computed Tomography

The simplest medical X-ray application is a plain radiograph: a single exposure produces a two-dimensional projection image. Chest X-rays remain one of the most commonly ordered imaging studies in the world, useful for spotting pneumonia, heart enlargement, lung tumors, and even vertebral fractures. Radiologists reading routine chest or abdominal films can detect a substantial portion of moderate-to-severe vertebral compression fractures in the mid-thoracic and lumbar spine, particularly in older women, even when the fractures were not the reason the scan was ordered.⁷ A single radiograph is quick, inexpensive, and delivers a relatively low radiation dose.

Computed tomography takes things further by rotating the X-ray source and detector around the patient, collecting projections from many angles, and then using mathematical algorithms to reconstruct cross-sectional slices. Early CT scanners relied on a method called filtered back projection. Modern scanners increasingly use iterative reconstruction algorithms, which compare simulated projection data against the real measurements and refine the image through repeated cycles. These techniques account for noise and the geometry of the scanning system more accurately. Studies have shown that iterative reconstruction can reduce the radiation dose needed for a diagnostic-quality CT scan by roughly a quarter to three-quarters compared to the older approach, without sacrificing image quality.⁸ That is a meaningful improvement, because CT scans deliver considerably more radiation than plain radiographs, and reducing dose without losing diagnostic value is one of the field’s ongoing goals.

Biological Effects and DNA Damage

X-rays are ionizing radiation, meaning each photon carries enough energy to strip electrons from atoms and break chemical bonds. When X-rays pass through living cells, the most consequential damage occurs to DNA. A single photon can cause a simple break in one strand of the double helix, which the cell usually repairs efficiently. The greater concern is complex or clustered damage, where multiple lesions (breaks, base damage, crosslinks) occur close together on the DNA molecule. Accumulating evidence points to these clustered lesions as the key factor determining whether the biological outcome after exposure is manageable repair or something more harmful, such as mutations or cell death.⁹

The severity of biological effects depends on the dose, the dose rate, the type of radiation, and the type of tissue exposed. A dental X-ray delivers a tiny fraction of a millisievert, well within the range the body handles without measurable harm. A full-body CT scan delivers a considerably larger dose. At very high doses, like those sometimes encountered in radiation accidents, acute effects such as skin burns and suppression of blood cell production can follow. At the low doses typical of diagnostic imaging, the primary concern is a small statistical increase in lifetime cancer risk. This is the reason radiation protection exists.

Radiation Protection Principles

The guiding framework for radiation safety in medical imaging has been ALARA, an acronym for “as low as reasonably achievable,” introduced by the International Commission on Radiological Protection in 1977 and still the backbone of radioprotection worldwide.¹⁰ ALARA rests on three pillars. The justification principle says that an X-ray exam should only happen when the expected clinical benefit outweighs the radiation risk. The optimization principle means doing everything practical to minimize dose during the exam: collimating the beam tightly so it covers only the area of interest, using the shortest acquisition time that produces a diagnostic image, and shielding sensitive areas when possible. The dose limitation principle sets absolute upper bounds on exposure for workers and the public.

More recent refinements include ALADA (“as low as diagnostically acceptable”) and ALADAIP, which adds “indicative of pathology.” These newer frameworks push the emphasis further toward diagnostic purpose: the dose should be low enough to minimize harm, but not so low that the image quality drops below what a clinician needs to make a correct diagnosis. The point is that an underexposed, unreadable scan helps nobody and may lead to a repeat exposure, which is worse than getting the dose right the first time.

Industrial Inspection and Non-Destructive Testing

Outside medicine, one of the largest uses of X-rays is non-destructive testing (NDT). Manufacturers need to verify that welds, castings, and composite structures are free of internal cracks, voids, and inclusions, without cutting the part open. X-ray radiography and CT scanning can reveal these flaws by detecting differences in how much radiation passes through sound material versus a defect.

A recently developed technique goes a step further. By using a welding filler wire with a higher X-ray attenuation than the surrounding base material, the weld fusion zone itself becomes clearly visible in the X-ray image, allowing engineers to measure its geometry precisely after the weld is made.¹¹ This kind of quantitative characterization of the weld shape is valuable in aerospace, automotive, and pipeline applications where weld quality can be safety-critical.

The food industry also relies heavily on X-ray systems. Foreign materials in food products, such as metal fragments, glass shards, stones, and bone, are a persistent safety concern. Over the past two decades, foreign materials have accounted for roughly one in ten food recalls, with plastic fragments being the most common type of complaint.¹² Inline X-ray inspection systems on production lines can detect many of these contaminants in real time, scanning every package as it moves through the process. Compared to metal detectors, X-ray systems can also catch non-metallic contaminants like glass and stone, making them a more comprehensive screening tool.

Protein Crystallography and Synchrotrons

When X-rays hit a crystal, the regular arrangement of atoms acts like a three-dimensional diffraction grating, scattering the X-ray beam into a pattern of spots. By measuring the positions and intensities of those spots, researchers can work backward to determine how the atoms are arranged inside the crystal. This technique, X-ray crystallography, has been central to structural biology for decades. It revealed the double-helix structure of DNA, the architecture of thousands of proteins, and the shapes of drug molecules bound to their targets.

The brightest and most tunable X-ray beams come from synchrotrons, large ring-shaped particle accelerators where electrons traveling near the speed of light emit intense radiation as they are bent by magnets. Synchrotron beamlines have been used to determine over 70 percent of all macromolecular structures deposited in the Protein Data Bank, contributed by more than 13,000 different research groups.¹³ Despite major advances in synchrotron technology over the years, the median resolution of structures determined at these facilities has held steady at about 2 ångströms. This is not because the hardware has stagnated; rather, it reflects the intrinsic quality of the protein crystals scientists can grow. The beam is no longer the bottleneck.

One example of the kind of detail crystallography can reveal involves studying how flexible a protein is within its crystal. In work on calmodulin, researchers refined their crystallographic models using multiple conformers and saw clear improvements in how well the model matched the data when they allowed up to four copies of the molecule to adopt different positions.¹⁴ The linker region of calmodulin, known to be flexible, showed the most movement. This kind of analysis matters for drug design, because a protein’s flexibility determines which binding pockets open and close, and therefore which drug molecules can fit.

X-Rays in Astrophysics

The universe is full of X-ray sources, but our atmosphere absorbs virtually all cosmic X-rays before they reach the ground. That is fortunate for life on Earth but frustrating for astronomers, which is why X-ray astronomy relies entirely on space-based telescopes. Missions like Chandra and XMM-Newton have mapped the X-ray sky in extraordinary detail, revealing phenomena invisible at other wavelengths.

Many of the brightest cosmic X-ray sources are binary systems where matter from a normal star falls onto a compact object, either a neutron star or a black hole. As the infalling gas spirals inward, it heats to millions of degrees and radiates intensely in X-rays. A Chandra study of IC 10 X-1, a luminous X-ray binary in a small starburst galaxy near the Milky Way, confirmed that the source has an average X-ray luminosity of about 1.5 × 10³⁸ ergs per second and varies in brightness by a factor of roughly two over very short timescales. The X-ray source was found to be coincident with a Wolf-Rayet star, making it a candidate for a Wolf-Rayet/black hole binary, an exotic and astrophysically informative system.¹⁵ Studying such systems helps constrain models of stellar evolution, black hole formation, and the extreme physics of accretion.

Beyond individual sources, X-ray observations have revealed vast halos of superheated gas surrounding galaxy clusters, mapped the distribution of dark matter through gravitational lensing effects on background X-ray sources, and detected the afterglows of gamma-ray bursts. X-ray astronomy, in short, opens a window onto the most energetic processes in the cosmos.

Phase-Contrast Imaging for Soft Tissues

Conventional X-ray imaging relies on absorption differences, which is why it excels at showing bone but struggles with soft tissue. A liver and a kidney, for instance, absorb X-rays at nearly the same rate, making them hard to distinguish on a standard radiograph without injecting a contrast agent. Phase-contrast X-ray imaging sidesteps this limitation by measuring how the X-ray wave is bent (refracted) as it passes through tissue, rather than just how much of it is absorbed. Since different soft tissues refract X-rays by different amounts, this approach can produce detailed images of structures that would otherwise be invisible.

Phase-contrast techniques retain the excellent spatial resolution and tissue penetration of conventional X-ray methods while dramatically improving soft-tissue contrast.¹⁶ This makes them promising for applications in tissue engineering, where researchers need to evaluate scaffolds and growing tissues without destroying their samples. In oncology, synchrotron-based phase-contrast imaging using X-ray interferometers has shown the ability to depict fine morphological structures in soft-tissue tumors with high sensitivity and high spatial resolution.¹⁷ For now, most phase-contrast work relies on synchrotron facilities, which limits clinical availability. But compact laboratory-scale phase-contrast systems are under active development, and if they succeed, they could expand the diagnostic reach of X-rays into territory currently dominated by MRI.

Art Conservation and Cultural Heritage

X-rays have a long and productive relationship with art historians and conservators. Underneath the visible surface of a painting there may be earlier compositions, corrections by the artist, structural damage, or later restorations. Standard X-ray radiography can image the internal layers of a painting because pigments containing heavy elements like lead (in lead white) or mercury (in vermilion) absorb X-rays strongly, while lighter pigments and the canvas itself are more transparent.

X-ray radiography and infrared reflectography have been the standard methods for studying the subsurface structure of cultural heritage artifacts for decades. More recently, scanning X-ray fluorescence (XRF) has expanded the toolkit by mapping the distribution of specific chemical elements across a painting’s surface, which allows researchers to identify which pigments were used and where, layer by layer.¹⁸ These are entirely non-destructive analyses, meaning the artwork is never touched, cut, or sampled. Famous paintings by Rembrandt, Vermeer, and Van Gogh have all been studied this way, sometimes revealing hidden portraits or abandoned compositions beneath the finished work. For conservators deciding how to restore a damaged painting, knowing what lies underneath is not just interesting but practically essential to avoid destroying original material.

Museum and archaeology applications extend beyond paintings. X-ray CT scanning of Egyptian mummies, for instance, allows researchers to study skeletal pathology, dental health, and even the contents of sealed funerary jars without unwrapping or opening anything. The same principle applies to corroded archaeological metalwork: a CT scan can reveal the original shape of an object hidden under centuries of corrosion products, guiding the conservator’s cleaning strategy.