Isomer shift, quadrupole splitting and magnetic hyperfine splitting explained in terms of what they tell you about an iron site, plus the absorber preparation and source planning that decide whether the spectrum is worth fitting.
A group synthesises iron oxide nanoparticles, runs diffraction, and finds a pattern consistent with magnetite. Magnetometry gives a loop with almost no coercivity. Someone concludes the particles are not magnetic. They are, in fact, magnetic and small, and the two results only make sense together once you know which iron sites are present and how fast their moments are flipping. That is the question Mossbauer spectroscopy answers, and almost nothing else on a laboratory bench does.
The technique rests on recoil free emission and absorption of a gamma ray by a nucleus bound in a solid. For iron the relevant transition in iron-57 sits at about 14.4 keV, and the natural line width of that transition is extraordinarily narrow relative to its energy. Narrow enough that moving the source a few millimetres per second shifts the gamma energy by more than the line width, through the Doppler effect.
That is why a Mossbauer spectrometer is built around a velocity transducer rather than a monochromator. The source is driven back and forth on a linear motor, counts are accumulated against velocity, and the resulting spectrum is plotted with velocity in millimetres per second on the horizontal axis. The narrowness is the whole point: it makes the nucleus sensitive to interactions with its own electrons that are far too small to see any other way, and the three of them that matter are described below.
The nucleus has a slightly different radius in its excited and ground states, so the energy of the transition depends on the density of s electrons at the nucleus. Change the oxidation state, the spin state or the coordination of the iron and you change that density, and the whole spectrum moves along the velocity axis. That displacement is the isomer shift, quoted relative to a reference, conventionally alpha iron foil at room temperature.
In practice the isomer shift is how you separate ferrous from ferric iron, and high spin from low spin. In a mixed valence material it lets you assign each component of a fitted spectrum to a valence rather than guessing. For anyone working on iron oxides, spinels, clays, catalysts, steels or iron bearing glasses, that assignment is the first thing the spectrum gives you and it is quantitative in a way that a diffraction pattern is not.
The excited state of the iron-57 nucleus has a quadrupole moment, so it responds to an electric field gradient at the nuclear site. That gradient comes from the arrangement of charge around the atom: the ligands, the site symmetry, distortion of the coordination polyhedron, and the iron valence electrons themselves. Where the site is perfectly cubic there is no gradient and no splitting. Where it is distorted, the line splits into a doublet, and the separation of the two lines is the quadrupole splitting.
Read together, isomer shift and quadrupole splitting identify the site. Two ferric components with different quadrupole splittings are two chemically distinct ferric environments, for example tetrahedral and octahedral positions in a spinel, or iron in a surface layer against iron in the interior of a particle. The relative areas of those components give the relative populations, subject to the assumption that the recoil free fraction is similar for both, which is worth stating rather than assuming silently.
If the nucleus sits in a magnetic field, whether from magnetic ordering in the material or from an applied field, the nuclear levels split further and the spectrum becomes a six line pattern. The separation of those lines measures the magnetic hyperfine field at the nucleus, which is characteristic of the phase. Metallic iron, haematite, magnetite and goethite each give a recognisable sextet, and magnetite gives two overlapping sextets because it has iron on two distinct sites.
The point that most often gets missed concerns small particles. A magnetically ordered particle below a certain size has a moment that flips direction thermally, and if it flips faster than the nuclear precession the nucleus sees an average field of zero. The sextet collapses to a doublet, and the room temperature spectrum looks like a non-magnetic material. Cooling the sample slows the flipping until the sextet reappears, and the temperature at which that happens is related to particle size and anisotropy. So a doublet at room temperature and a sextet at low temperature is not a contradiction: it is a direct measurement of superparamagnetic relaxation, and it explains the flat magnetometry loop that started the confusion.
Be clear about what the technique cannot do. It is nucleus specific, so with an iron-57 source it sees iron and nothing else. Tin-119 and a handful of other isotopes are practical, but there is no general elemental survey here. It does not image, it has no spatial resolution in the ordinary arrangement, and it will not tell you how much iron is in the sample as a weight fraction. It tells you, with unusual authority, what the iron that is present is doing.
Sample preparation is where most first spectra go wrong. The absorber has to be thin enough that lines are not saturated and broadened, and thick enough to give a workable count rate in a sensible time. The controlling quantity is the amount of the resonant isotope per unit area, not the mass of powder, so an iron rich oxide and a dilute clay need very different absorber loadings.
The source is a consumable. Cobalt-57 in a rhodium matrix has a half life of roughly nine months, so a spectrometer that gave a good spectrum overnight when the source was fresh will need progressively longer runs, and eventually a replacement. Build that into the running cost rather than discovering it in year two. A sealed radioactive source also has to be imported and held under authorisation from the Bangladesh Atomic Energy Regulatory Authority, with the associated documentation, storage arrangements and disposal route for the spent source. That paperwork has a lead time measured in months, and it is the item that most often delays commissioning of an otherwise straightforward instrument.
If your material contains iron and the question is which iron, this is the measurement that answers it. If the question is how much iron, use something else and keep the spectrometer for the harder question.