X-Ray Crystallography: Seeing Atoms Through Diffraction

Why Visible Light Cannot See Atoms

Any imaging method that relies on focusing waves into a picture is fundamentally limited by the wavelength of the wave it uses. As a rough rule, you cannot resolve details much finer than about half the wavelength of the radiation involved, because features smaller than that simply do not disturb the wave enough to leave a detectable trace. Visible light has a wavelength of roughly 400 to 700 nanometers, which is thousands of times larger than the spacing between atoms in a solid, typically a few tenths of a nanometer, or a few angstroms (1 angstrom equals 0.1 nanometers). Trying to resolve individual atoms with visible light is like trying to map the grooves of a vinyl record using ocean waves as your probe: the waves are so much larger than the features that they simply roll over them without registering any detail. X-rays, by contrast, have wavelengths on the order of 0.1 to a few angstroms, placing them in almost exactly the same size range as the spacing between atomic planes in a crystal. That match in scale is what makes X-rays, rather than any form of visible light, the natural tool for probing atomic structure.

Bragg's Law: The Geometry of Constructive Interference

A crystal is not a random jumble of atoms but a highly ordered, repeating lattice, which can be thought of as stacks of parallel atomic planes spaced a fixed distance apart. When a beam of X-rays strikes these planes, each plane reflects a small fraction of the beam, much like a partially silvered mirror. Most reflected waves cancel each other out through destructive interference, but at very specific angles the waves reflected from successive planes emerge perfectly in step, reinforcing one another into a strong, detectable beam. William Henry Bragg and his son William Lawrence Bragg worked out exactly when this constructive interference happens, giving us Bragg's Law: n times lambda equals 2 times d times sine of theta (written n λ = 2 d sin θ), where n is a positive integer called the diffraction order, lambda is the wavelength of the X-rays, d is the spacing between the parallel atomic planes, and theta is the angle of incidence measured from the plane itself, not from the normal to the plane. The extra distance traveled by the wave reflecting off the second, deeper plane compared to the first is exactly 2 times d times sine of theta, and constructive interference occurs precisely when that extra path length equals a whole number of wavelengths, n times lambda. At any other angle, the reflected waves drift out of phase and cancel, so a crystal only 'lights up' at a discrete, predictable set of angles rather than scattering X-rays smoothly in every direction.

A Worked Sense-Check

It helps to run one concrete number through the formula. Suppose an X-ray beam with wavelength 1.54 angstroms (a common value produced by copper-target X-ray tubes) strikes a set of crystal planes spaced 3.0 angstroms apart, and we want the first-order reflection, so n equals 1. Rearranging Bragg's Law, n times lambda equals 2 times d times sine of theta, to solve for sine of theta gives sine of theta equals (n times lambda) divided by (2 times d), which here is 1.54 divided by (2 times 3.0), or 1.54 divided by 6.0, which works out to about 0.257. Taking the inverse sine of 0.257 gives theta of approximately 15 degrees. So for this wavelength and plane spacing, a detector would register a bright first-order diffraction spot when the crystal (and detector) sit at roughly 15 degrees from the incident beam relative to those particular planes, and a second, weaker spot for n equals 2 at a larger angle, and so on. Every distinct family of atomic planes in the crystal, each with its own spacing d and orientation, produces its own set of allowed angles, which is why a real diffraction pattern from a crystal is a rich constellation of many spots rather than a single ring.

A Crystal as a Three-Dimensional Diffraction Grating

An ordinary diffraction grating is a surface ruled with evenly spaced lines that splits light into a fan of angles depending on wavelength. A crystal does the same trick, but in three dimensions at once: its atoms sit on a repeating lattice, and that lattice can be sliced into countless different families of parallel planes, each running in a different direction and each with its own characteristic spacing d. Because every one of those plane families satisfies Bragg's Law at its own particular angle, a single crystal illuminated from one direction produces not just one diffracted beam but an entire array of spots, each corresponding to reflection from a different set of planes. This is why X-ray crystallography works at all: the crystal's own periodic order acts as nature's own finely tuned diffraction grating, and the positions and intensities of the resulting spots encode detailed information about where the atoms actually sit. In practice, a single fixed orientation of the crystal only captures a slice of this information, because only planes that happen to satisfy Bragg's Law at that exact orientation produce a visible spot. Real experiments therefore rotate the crystal systematically through many angles, collecting a full diffraction pattern at each orientation, so that essentially every relevant family of planes gets its turn to satisfy the Bragg condition and register a spot.

From Spots to Structure: The Phase Problem and Landmark Discoveries

Collecting thousands of diffraction spots across many crystal orientations gives you the intensity of each spot, which tells you how strongly that family of planes scatters, but a detector can only record intensity, not the phase of the scattered wave, even though the phase is equally essential for reconstructing the electron-density map that shows where the atoms actually are. This missing information is famously called the phase problem, and solving it, whether through mathematical tricks like comparing crystals with and without added heavy atoms, or through modern computational methods that guess and refine trial structures, is the hardest part of any real structure determination. Once the phases are recovered, the intensities and phases together can be combined mathematically (via a Fourier transform) into a three-dimensional electron-density map, a kind of contoured fog showing where electrons, and therefore atoms, are statistically likely to be found; fitting a chemical model into that fog yields the final atomic structure. This approach has produced some of the most consequential discoveries in science. Rosalind Franklin's X-ray diffraction images of DNA fibers, most famously Photograph 51, captured the characteristic cross-shaped pattern that let James Watson and Francis Crick deduce the helical spacing and diameter needed to propose the double-helix structure of DNA in 1953. In the decades since, protein crystallography has determined the atomic structures of thousands of enzymes, receptors, and viruses, work that underpins modern structure-based drug design, where researchers use a protein's exact three-dimensional shape to engineer molecules that fit precisely into its active site.

Frequently asked questions

Try it live

Everything above runs in your browser — open X-Ray Crystallography: Seeing Atoms Through Diffraction and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.