Home / Courses / CSIR NET Physics / Practice / Diffraction and the structure factor

Diffraction and the structure factor

Solid state physics and devices · 87 questions and templates for CSIR NET Physics, in order from warm-up to top rank. Check each answer, then read the full solution.

Loading questions… (they need JavaScript; nothing you type is sent anywhere)

Written questions with solutions

The same questions as a list, so you can print them or read without the practice box. Templates (fresh numbers) appear only in the practice box above.

  1. MCQWarm-up

    The structure factor Fhkl of a crystal depends on
    1. only the lattice constant
    2. the kinds of atoms in the basis and their positions in the cell
    3. only the X-ray wavelength
    4. the number of unit cells in the crystal
    Answer and solution

    Answer: B

    Fhkl=∑jfje2πi(hxj+kyj+lzj), a sum over the atoms of one cell: their form factors fj (which kind of atom) and their fractional positions (xj,yj,zj). The lattice fixes where the peaks are; F fixes how strong each is. The number of cells changes the overall height and sharpness of the peaks, not F.

    Why people get this wrong: The wavelength enters only indirectly, through the fall of fj with sin⁡θ/λ.

  2. MCQWarm-up

    A diffraction experiment measures the integrated intensity of each reflection. This gives directly
    1. |Fhkl|2, but not the phase of Fhkl
    2. the full complex Fhkl, phase included
    3. only the sign of Fhkl
    4. the electron density at each atom directly
    Answer and solution

    Answer: A

    The intensity is proportional to FF∗=|Fhkl|2, so the phase of F is lost. Since the electron density is the Fourier sum ρ(r→)=1V∑Fhkle−2πi(hx+ky+lz), it cannot be rebuilt from intensities alone: this is the phase problem of crystallography, solved by direct methods, heavy atoms or anomalous scattering.

    Why people get this wrong: Detectors count photons (energy), never field amplitudes with their phase.

  3. MCQWarm-up

    The X-ray atomic form factor of a K+ ion (Z=19) in the forward direction (θ=0) is
    1. 19
    2. 20
    3. 9.5
    4. 18
    Answer and solution

    Answer: D

    X-rays scatter from electrons. At θ=0 all the electrons scatter in phase, so f(0) equals the number of electrons. K+ has lost one electron: f(0)=19−1=18. (This is why K+ and Cl−, both with 18 electrons, scatter almost alike.)

    Why people get this wrong: Using Z=19 ignores the charge of the ion.

  4. MCQWarm-up

    With Δk→=k→′−k→ the change of wave vector on scattering, which of these is one of the three Laue equations?
    1. a→1×Δk→=2πh
    2. |Δk→|=2π/|a→1|
    3. a→1⋅Δk→=2πh, with h an integer
    4. a→1⋅k→=2πh
    Answer and solution

    Answer: C

    Waves scattered from points one lattice vector apart differ in phase by Δk→⋅a→i. All cells scatter in phase when a→i⋅Δk→=2π× integer for i=1,2,3. Together the three equations say Δk→=G→hkl.

    Why people get this wrong: The condition is on the change Δk→, not on the incident k→ itself.

  5. MCQWarm-up

    In the Ewald construction for elastic scattering of wavelength λ, the sphere
    1. has radius 2π/λ and passes through the origin of the reciprocal lattice
    2. has radius λ and is centred on the origin
    3. has radius 4π/λ and is centred on the origin
    4. has radius 2π/dhkl for the strongest reflection
    Answer and solution

    Answer: A

    Draw the incident k→ (length 2π/λ) so that it ends at the reciprocal-lattice origin. A sphere of radius |k→| about its starting point passes through the origin. Every reciprocal lattice point on this sphere gives k→′=k→+G→ with |k→′|=|k→|: a diffracted beam.

    Why people get this wrong: The radius 4π/λ belongs to the limiting sphere (all orientations), centred on the origin.

  6. NATWarm-up

    The radius of the Ewald sphere for Cu Kα X-rays (λ=1.54 Å) is ______ Å−1 (to three decimal places).
    Answer and solution

    Answer: 4.08 Å⁻¹ (±0.004)

    |k→|=2π/λ=6.2832/1.54=4.080 Å−1.

    Why people get this wrong: Using 1/λ=0.649 Å−1 mixes the crystallographers' convention with the 2π convention for G→.

  7. MCQWarm-up

    Sodium chloride has the rock-salt structure (4 Na+ and 4 Cl− ions in the conventional cube). Na+ has 10 electrons and Cl− has 18. The X-ray structure factor F000 of the conventional cell is
    1. 28
    2. 56
    3. 112
    4. 224
    Answer and solution

    Answer: C

    At hkl=000 every phase factor is 1 and each form factor equals its number of electrons, so F000=∑jfj(0)=4×10+4×18=112: the number of electrons in the cell.

    Why people get this wrong: 28 is one formula unit; the conventional cube holds four.

  8. MCQStandard

    The X-ray atomic form factor f(q→) of an atom is, apart from constants,
    1. the square of the electron density at the nucleus
    2. the Fourier transform of the electron density ρ(r→) of the atom
    3. the Laplace transform of the potential energy of the atom
    4. the number of electrons divided by sin⁡θ
    Answer and solution

    Answer: B

    Each volume element ρ(r→)d3r scatters with a phase factor eiq→⋅r→, so f(q→)=∫ρ(r→)eiq→⋅r→d3r with q=4πsin⁡θ/λ. At q=0 it is the number of electrons; it falls as q grows because the cloud has a finite size. The structure factor is the same idea applied to a whole cell.

    Why people get this wrong: The fall with angle is not 1/sin⁡θ; it is set by the size of the cloud (a wider cloud falls faster).

  9. MCQStandard

    KCl has the rock-salt structure, yet its X-ray powder pattern looks like that of a simple cubic crystal of lattice constant a/2. The reason is that
    1. KCl really has a simple cubic Bravais lattice
    2. its lattice constant is too large for the odd reflections to fit
    3. K+ and Cl− both have 18 electrons, so their form factors are nearly equal and the all-odd reflections almost cancel
    4. the Cl− ions do not scatter X-rays
    Answer and solution

    Answer: C

    For rock salt, all-odd reflections have F=4(f+−f−) and all-even ones F=4(f++f−). K+ and Cl− are isoelectronic, so f+≈f− and the all-odd lines nearly vanish. What remains (all even) is what an sc crystal of identical atoms with spacing a/2 would give. The lattice is still fcc.

    Why people get this wrong: Missing lines here are a near-accident of form factors, not a change of Bravais lattice. In NaCl (f+≠f−) the odd lines are clearly seen.

  10. MCQStandard

    For the diamond structure of identical atoms (form factor f taken the same for both reflections), the ratio |F111|2/|F220|2 is
    1. 1
    2. 2
    3. 0
    4. 1/2
    Answer and solution

    Answer: D

    Diamond is fcc with a basis at 0 and (14,14,14), so for unmixed indices F=4f[1+eiπ(h+k+l)/2]. (111): 1+e3πi/2=1−i, |F|2=16f2⋅2=32f2. (220): 1+e2πi=2, |F|2=64f2. Ratio =1/2.

    Why people get this wrong: Forgetting that 1−i has modulus 2, not 1 or 0.

  11. MCQStandard

    Friedel's law |Fhkl|=|Fh‾k‾l‾| (no anomalous scattering) means that a diffraction pattern
    1. shows a centre of symmetry even when the crystal has none
    2. always has 4-fold symmetry
    3. can tell left- and right-handed crystals apart
    4. contains only even orders of reflection
    Answer and solution

    Answer: A

    With real form factors, Fh‾k‾l‾=∑fje−2πi(…)=Fhkl∗, so the two intensities are equal. The pattern therefore always looks centrosymmetric, whether or not the crystal is. Only anomalous scattering near an absorption edge (complex f) breaks this and reveals handedness.

    Why people get this wrong: Concluding the crystal must have an inversion centre because its pattern does.

  12. MSQStandard

    Which statements comparing thermal-neutron and X-ray diffraction are correct?
    1. Neutron scattering lengths do not grow steadily with atomic number.
    2. Neutrons locate hydrogen (or deuterium) atoms far better than X-rays.
    3. The nuclear scattering length of an atom falls off with angle as strongly as an X-ray form factor.
    4. Neutrons can reveal the ordering of atomic magnetic moments.
    Answer and solution

    Answer: A, B, D

    Neutrons scatter from nuclei through the strong force, with lengths b that vary irregularly from isotope to isotope (even negative, as for H). Light atoms such as H/D are therefore visible next to heavy ones. The nucleus is tiny compared with the wavelength, so b does not fall with angle (unlike f, which reflects the size of the electron cloud). The neutron's magnetic moment also scatters from unpaired electron spins, which shows magnetic order.

    Why people get this wrong: The magnetic part of neutron scattering does fall with angle (it comes from the electron cloud); the nuclear part does not.

  13. NATStandard

    Atoms in a crystal vibrate with a mean-square displacement ⟨u2⟩=0.010 Å2 (all three directions together). Using I=I0exp⁡(−G2⟨u2⟩/3), the fraction of the static intensity left in the reflection from planes of spacing 1.0 Å is ______ (to three decimal places).
    Answer and solution

    Answer: 0.877 (±0.003)

    G=2π/d=6.283 Å−1, G2=39.48 Å−2. Exponent =39.48×0.010/3=0.1316. I/I0=e−0.1316=0.877. The peak keeps its sharpness; 12% of its intensity moves into a diffuse background.

    Why people get this wrong: Using ⟨u2⟩ without the 1/3 (that form needs the one-direction value ⟨ux2⟩).

  14. MCQStandard

    For elastic scattering through an angle 2θ with wavelength λ, the magnitude of the scattering vector Δk→ is
    1. 2πsin⁡θ/λ
    2. 4πcos⁡θ/λ
    3. 4πsin⁡2θ/λ
    4. 4πsin⁡θ/λ
    Answer and solution

    Answer: D

    k→ and k→′ both have length k=2π/λ and make an angle 2θ. Their difference is the base of an isosceles triangle: |Δk→|=2ksin⁡θ=4πsin⁡θ/λ. Setting it equal to |G→|=2πn/d gives Bragg's law 2dsin⁡θ=nλ.

    Why people get this wrong: Using the full scattering angle 2θ inside the sine.

  15. NATStandard

    A beam of wavelength 1.5 Å falls perpendicular to a long row of identical atoms spaced 5.0 Å apart. The first-order diffracted beams lie on a cone whose half-angle, measured from the row, is ______ degrees (to one decimal place).
    Answer and solution

    Answer: 72.5 degree (±0.1)

    For a single row only one Laue equation applies: a(cos⁡α−cos⁡α0)=hλ, with α the angle to the row. At normal incidence cos⁡α0=0, so cos⁡α=λ/a=0.30 and α=72.5∘. Every direction at that angle to the row works: a cone. A 3D crystal needs three such cones to meet, which is why a fixed crystal in a monochromatic beam rarely diffracts.

    Why people get this wrong: Taking the angle from the normal (17.5∘) instead of from the row.

  16. MCQStandard

    In a powder pattern the multiplicity factor phkl multiplies the intensity of a line because
    1. higher orders of reflection are stronger
    2. all planes of the form {hkl} have the same spacing and diffract into the same ring
    3. the Lorentz factor is larger at low angles
    4. each crystallite reflects p times before the beam leaves
    Answer and solution

    Answer: B

    In a cubic crystal, for example, the 6 orientations (100),(010),…,(001‾) all have spacing a and so all add to the same cone of angle 2θ. Randomly oriented grains put each of them in reflecting position equally often, so the ring is p times stronger: p=6 for {100}, 12 for {110}, 8 for {111}.

    Why people get this wrong: Multiplicity counts symmetry-equivalent planes, not orders of reflection.

  17. MSQStandard

    Which statements about the structure factor Fhkl are correct?
    1. Fhkl is in general a complex number.
    2. Fhkl is (up to the cell volume) the Fourier coefficient of the electron density for G→hkl.
    3. Moving the origin of the cell changes |Fhkl|.
    4. Fhkl=0 means the lattice has no planes with those indices.
    Answer and solution

    Answer: A, B

    (1) True: a sum of phase factors is complex unless the cell is centrosymmetric about the origin. (2) True: ρ(r→)=1V∑G→FG→e−iG→⋅r→. (3) False: shifting the origin by r→0 multiplies F by e−iG→⋅r→0, changing only its phase. (4) False: the planes exist; the waves from the different basis atoms cancel.

    Why people get this wrong: Statement 4 is the classic slip: "absent reflection" means destructive interference within the basis, not missing planes.

  18. MSQStandard

    Which statements about the Ewald sphere construction are correct?
    1. Its radius is 2π/λ.
    2. A reflection occurs whenever a reciprocal lattice point lies on the sphere.
    3. For a fixed crystal in a monochromatic beam, usually few or no points lie on it.
    4. Shorter wavelengths give a smaller sphere.
    Answer and solution

    Answer: A, B, C

    The radius is |k→|=2π/λ, so a shorter wavelength gives a larger sphere (statement 4 false). A point on the sphere means |k→+G→|=|k→|: the Laue condition. Reciprocal lattice points are a discrete set, so a sphere through the origin in a general orientation misses them almost always. That is why experiments rotate the crystal, use a powder, or use white radiation (a range of spheres: the Laue method).

    Why people get this wrong: Inverse relation: λ small, sphere big.

  19. MCQStandard

    The same reflection (hkl) of a crystal is measured first with Cu Kα (1.54 Å) and then with Mo Kα (0.71 Å) X-rays. Ignoring absorption edges, the atomic form factors that enter Fhkl are
    1. larger with Mo, because θ is smaller
    2. smaller with Mo, because the wavelength is shorter
    3. the same, because sin⁡θ/λ=1/2dhkl is fixed by the reflection
    4. the same only if the crystal is cubic
    Answer and solution

    Answer: C

    f depends on the scattering vector q=4πsin⁡θ/λ. Bragg's law gives sin⁡θ/λ=1/2dhkl for every wavelength, so a given reflection always has the same q. With Mo the reflection appears at a smaller angle, but f, and so |Fhkl|, is unchanged.

    Why people get this wrong: f depends on sin⁡θ/λ, not on θ alone.

  20. MCQExam level

    An hcp crystal of identical atoms has atoms at (0,0,0) and (13,23,12) in the hexagonal cell. Which reflection is absent?
    1. (100)
    2. (001)
    3. (101)
    4. (002)
    Answer and solution

    Answer: B

    F=f[1+e2πi((h+2k)/3+l/2)]. It vanishes when the phase is an odd multiple of π: h+2k=3n and l odd. (001): h+2k=0, l=1: F=f(1+eiπ)=0. The others: (100) phase 2π/3 (|F|2=f2), (101) phase 5π/3 (3f2), (002) phase 2π (4f2).

    Why people get this wrong: Physically, the B-layer half way up scatters exactly out of phase with the A-layers for (001), so the basal reflection appears first as (002).

  21. NATExam level

    For the hcp structure of identical atoms (atoms at (0,0,0) and (13,23,12)), |F101|2 in units of f2 is ______.
    Answer and solution

    Answer: 3

    Phase of the second atom: 2π(13+12)=5π3. |1+eiϕ|2=2+2cos⁡ϕ=2+2cos⁡300∘=2+1=3.

    Why people get this wrong: Squaring the real part only, or using h+k instead of h+2k (the second atom is at 13,23).

  22. NATExam level

    Ordered Cu3Au has Au at the cube corners and Cu at the face centres. Taking fAu=79 and fCu=29 for both lines (forward values, a rough approximation), the ratio of a superlattice to a fundamental structure-factor intensity, |F100|2/|F111|2, is ______ (to three decimal places).
    Answer and solution

    Answer: 0.091 (±0.002)

    Sites (0,0,0) Au and (0,12,12),(12,0,12),(12,12,0) Cu. Unmixed (111): all phases 1, F=fAu+3fCu=166. Mixed (100): the phase factor is e2πix, which is +1 for Au and for the Cu at (0,12,12), and −1 for the other two Cu. So F=fAu+fCu−2fCu=fAu−fCu=50. Ratio =2500/27556=0.091.

    Why people get this wrong: Expecting (100) to be absent as in fcc: it is absent only when all four sites hold the same (or the same average) atom.

  23. MCQExam level

    Cu3Au is heated above its order-disorder temperature, where Au and Cu occupy the four fcc sites at random. Its powder pattern then
    1. loses the mixed-index (superlattice) lines and shows only the fcc lines
    2. gains extra lines at half-integer indices
    3. keeps every line with unchanged intensity
    4. loses the unmixed-index lines and keeps the mixed ones
    Answer and solution

    Answer: A

    With random occupation each site scatters like an average atom f‾=14fAu+34fCu. All four sites are then equivalent: an fcc lattice of identical (average) atoms, for which mixed indices give F=0. The fundamental lines remain with F=4f‾=fAu+3fCu, unchanged.

    Why people get this wrong: Disorder does not wipe out the fcc lines; it only removes the extra lines that order created.

  24. NATExam level

    In fluorite CaF2, Ca sits on the fcc sites and F at (14,14,14) and (34,34,34) from each. Taking forward values fCa=18 (Ca2+) and fF=10 (F−), the magnitude |F200| is ______.
    Answer and solution

    Answer: 8

    For unmixed indices the fcc sum gives a factor 4: F=4[fCa+fF(eiπ(h+k+l)/2+e3iπ(h+k+l)/2)]. For (200), h+k+l=2: eiπ+e3iπ=−2. F=4(18−20)=−8, so |F200|=8. The two F ions sit half way between Ca planes for this reflection and almost cancel the Ca.

    Why people get this wrong: Adding the F contributions with a plus sign gives 4(18+20)=152, which is the (400) or (220) value.

  25. MCQExam level

    Cubic perovskite SrTiO3 has Sr at (0,0,0), Ti at (12,12,12) and O at (12,12,0), (12,0,12), (0,12,12). The structure factor F110 is
    1. fSr−fTi−fO
    2. fSr−fTi+3fO
    3. fSr+fTi+3fO
    4. fSr+fTi−fO
    Answer and solution

    Answer: D

    Phases e2πi(x+y) for (110): Sr e0=1; Ti e2πi=1; O at (12,12,0): e2πi=1; O at (12,0,12): eiπ=−1; O at (0,12,12): eiπ=−1. Sum: fSr+fTi+fO−2fO=fSr+fTi−fO.

    Why people get this wrong: The other options are F100, F111 and F200 in turn: each oxygen must be phased separately.

  26. NATExam level

    X-rays of wavelength 1.54 Å are scattered elastically through 2θ=90∘. The magnitude of the scattering vector is ______ Å−1 (to two decimal places).
    Answer and solution

    Answer: 5.77 Å⁻¹ (±0.01)

    q=4πsin⁡θ/λ=12.566×sin⁡45∘/1.54=12.566×0.7071/1.54=5.77 Å−1. A Bragg peak there would mean a reciprocal vector of this length, i.e. d=2π/q=1.089 Å.

    Why people get this wrong: Putting sin⁡90∘ (the full scattering angle) into the formula gives 8.16 Å−1.

  27. MSQExam level

    A 1D crystal of period a has atom A at x=0 and atom B at x=a/2 in each cell. Which statements about its reflections h (with G=2πh/a) are correct?
    1. Odd h have |Fh|=|fA−fB|.
    2. If fA=fB, odd h vanish, because the true period is then a/2.
    3. Even h have intensity proportional to fA2+fB2.
    4. The positions of the peaks depend on the ratio fA/fB.
    Answer and solution

    Answer: A, B

    Fh=fA+fBeiπh=fA+(−1)hfB. Odd: fA−fB; even: fA+fB, so the even intensity is (fA+fB)2, not fA2+fB2. If the atoms are identical, the chain repeats every a/2 and its reciprocal lattice has spacing 4π/a: the odd h were never true reciprocal vectors. Peak positions come from the lattice only; the basis sets their heights.

    Why people get this wrong: Adding intensities (fA2+fB2) instead of amplitudes: the atoms scatter coherently.

  28. NATExam level

    For an unpolarised X-ray beam, the polarisation factor of a reflection is (1+cos2⁡2θ)/2. Its value at a scattering angle 2θ=60∘ is ______ (to three decimal places).
    Answer and solution

    Answer: 0.625 (±0.002)

    The incident field splits into a component perpendicular to the scattering plane (scattered fully) and one in the plane (scattered with factor cos⁡2θ in amplitude). Averaging the intensities: (1+cos2⁡2θ)/2=(1+0.25)/2=0.625. At 2θ=90∘ the factor falls to 0.5.

    Why people get this wrong: Using θ instead of 2θ: the scattering angle between the beams is 2θ.

  29. MCQExam level

    In a transmission electron microscope at 200 kV (λ≈0.025 Å), a thin single crystal shows a whole grid of diffraction spots at once, without being rotated. The main reason is that
    1. electrons are scattered more weakly than X-rays
    2. the Ewald sphere is then so large that near the origin it is almost a plane, cutting a whole layer of reciprocal lattice points
    3. electrons have no wave properties at that energy
    4. the crystal contains many grains in every orientation
    Answer and solution

    Answer: B

    The Ewald radius 2π/λ≈250 Å−1 is about a hundred times the reciprocal lattice spacing (∼2 Å−1), so near the origin the sphere is nearly flat. A thin specimen also stretches each reciprocal point into a rod along the beam. Together they let the flat sphere cross a whole plane of points: the spot pattern is a picture of a reciprocal-lattice section.

    Why people get this wrong: Electrons actually scatter far more strongly than X-rays (which is why samples must be thin).

  30. MCQExam level

    A crystal orders antiferromagnetically so that the magnetic cell is twice the chemical cell along each axis. Below the Néel temperature its neutron diffraction pattern shows
    1. no change, because neutrons do not see magnetic moments
    2. only a uniform rise in background
    3. a shift of all nuclear peaks to lower angles
    4. extra peaks at positions that are half-integer in the chemical-cell indices
    Answer and solution

    Answer: D

    Doubling the real-space period halves the reciprocal-lattice spacing for the magnetic scattering, so new reciprocal points appear half way between the chemical ones: peaks at (121212) and similar in chemical units (as in MnO). X-rays, which barely see spin, show no such peaks. Their intensity grows as the sublattice magnetisation squared below TN.

    Why people get this wrong: Ferromagnetic order (same period) adds intensity on top of existing peaks; antiferromagnetic order adds new peaks.

← All topics · Timed mock test →