Different parts of a particle scatter together
Send X-rays through a liquid containing nanoparticles. Most light continues forward; a small fraction leaves at a slight angle. Differences in electron density between a particle and its surroundings produce scattering. This density difference is called contrast, and a larger contrast generally makes the particles easier to detect.
Waves scattered from different parts of one particle meet at the detector. Along some directions their crests arrive together and reinforce. Along others a crest meets a trough and the intensity falls. This addition and cancellation is interference. Distances inside the particle determine the angles at which its bright and dark features appear.
Larger particles move features towards the beam
Imagine two spheres of the same kind, with every length in the second doubled. At a fixed small angle, the path difference between rays from corresponding parts is approximately doubled too. Reducing the angle by about half restores a similar interference condition. A larger particle therefore moves its main scattering features towards smaller angles.
Scattering work often replaces angle with a spatial-scale coordinate called . It depends on wavelength and the angle between incoming and outgoing directions. Smaller values are sensitive to the overall object; larger values progressively reveal finer features.
Very close to the beam, many compact particles have similar smoothly decreasing curves. Their initial fall measures a radius of gyration: how far scattering matter spreads from its centre. Particles with different details can share this overall scale. Features farther out help separate their shapes.
Moving a detector farther away spreads the same small angles over more pixels. Its finite area then misses some larger angles. Rings from a standard with known spacing help determine the distance and the position of the direct beam.
Random rods can also make a round halo
One fixed rod can scatter quite differently along different directions. In a liquid full of randomly oriented rods, those directional patterns add together. Averaging over orientations leaves a circular halo. Randomly oriented clusters can behave similarly.
Pixels at equal distance from the beam centre can be averaged to form a radial intensity curve. Sphere minima, the decay from finite rods and interference among subunits of a cluster produce different radial features. Sufficient angular coverage and photon counts can distinguish some of these candidates.
A few large particles can dominate
At very small angles, electrons within one particle scatter approximately in step. Their wave amplitudes add before the intensity is squared. For uniform spheres of equal contrast, doubling radius multiplies volume and electron count by eight. The forward intensity per particle can rise by sixty-four. A small number of large particles can therefore contribute most of a mixture’s low-angle signal. A size average based on counting particles may differ substantially from the size to which scattering is most sensitive.
Sources: [1]
The direct beam and the cell also matter
The direct beam is much brighter than the nearby scattering. A small beamstop intercepts it to protect the detector. A diode inside the stop can measure transmitted light. Surrounding pixels collect weak scattering, but the stop also removes the central angles that often carry the clearest information about overall size.
Cell windows, solvent and detector dark counts contribute too. Measurements of the sample, a matched empty cell and a closed-beam dark frame need corrections for exposure, flux and transmission. Subtracting two large signals can leave a noisy small difference. Occasionally negative corrected intensities are a natural consequence of counting fluctuations.
Read a size together with its angular range
When the centre is largely hidden, several sizes or shapes may follow the remaining tail equally well. Particle arrangements also affect scattering, and repeated scattering in a thick sample changes the pattern further. Dilution, a second detector distance or another measurement can help separate these effects.
A useful size result identifies its assumed shape, observed angular range and background treatment. The halo supplies clues from many particles together. Covering enough spatial scales and comparing possible explanations gradually narrows the structures compatible with those clues.