Introduction to Crystallography: Lecture 9 — Fourier Transforms

Scientific concepts, discoveries, and nature phenomena

Fourier transforms in crystallography

• Goal: Determine electron density inside a crystal (a periodic function) from diffraction data.
• Core idea: Electron density can be expanded as a Fourier series (sum of sine/cosine terms or complex exponentials).

Fourier synthesis vs. Fourier analysis

• Fourier synthesis: Add many waves (sine/cosine components) to reconstruct a periodic function.
• Fourier analysis: Decompose a periodic function into its component waves.

Fourier series completeness

• A complete set of reflections corresponds to a complete Fourier series.
• In practice, only a finite number of reflections are measured.

---

Structure factors, diffraction, and electron-density reconstruction

• The structure factor (F_{hkl}) is interpreted as the amplitude (magnitude) of an electron-density wave in the crystallographic direction ((hkl)).
• The diffraction pattern can be understood as a Fourier transform of the crystal (electron density).
• The inverse Fourier transform applied to measured structure factors allows reconstruction of the original electron density.

Key relationship (integral form)

• The structure factor can be written as an integral over electron density with a complex exponential phase factor (the lecture uses the crystallographic form involving (Hx + Ky + Lz)).
• Electron density is obtained from the inverse transform of the structure factors.

Real space vs. reciprocal space

• The lecture emphasizes that real-space periodicity becomes discrete spots in reciprocal space.

---

Practical limitations: truncation errors and finite reflections

• Truncation errors: Using only a limited number of reflections produces artifacts in reconstructed electron density.
  • Can lead to negative regions in electron-density maps (physically meaningless, but mathematically caused by an incomplete Fourier series).
• Resolution dependence on maximum scattering:
  • Measuring diffraction to larger (d^*) (or equivalently smaller (d)-spacing) improves electron-density map quality and helps identify atom positions more reliably.
• Rule of thumb:
  • Small molecules can often be solved with a finite (and relatively modest) number of reflections if resolution/quality are good.
  • Proteins require different strategies.

---

Example phenomena: wavelength differences and interference

• Different (hkl) reflections correspond to waves with different effective periodicities in the electron-density modulation.
  • Therefore, simple intuition about “same-wavelength x-ray interference” doesn’t transfer directly.

1D toy example logic

• Summing 1st/2nd/3rd “waves” gives a periodic function reminiscent of an atomic arrangement, but incomplete sums distort it.
• Adding progressively more Fourier terms improves the correspondence between peaks and atom positions.

---

Convolution and instrumental effects (envelope functions)

Convolution concept in reciprocal space

The measured diffraction pattern is influenced by relationships between:

• Unit cell contents transform (molecular/unit-cell transform),
• Fringe function / lattice periodicity (generating discrete reciprocal-lattice points),
• Instrument setup (slits), producing an envelope function that affects measured intensity distribution and resolution.

Slit width effects

• Wider slits broaden the beam’s angular spread.
• This changes the envelope function and therefore alters the effective resolution.

---

Analogy: diffraction from holes (Fourier optics intuition)

• Single hole / single object: Produces a characteristic diffraction (ring) pattern tied to its Fourier transform.
• Two holes: Superimposes a fringe function (interference term depending on the separation vector).
  • Can yield cancellation or reinforcement depending on phase.
• Extension to crystals:
  • Treat atoms/electron density as a periodic array analogous to a periodic set of holes.
  • Crystal diffraction arises from:
    • unit-cell (molecular) transform × lattice fringe (reciprocal lattice modulation).

---

Atomic orbitals and scattering factors

• Atomic electron distribution → atomic scattering factor:
  • The Fourier transform of atomic orbitals/electron density yields the atomic scattering factor (f).
• Qualitatively, different orbitals (e.g., 1s, 2s, 3s, p orbitals) have different spatial extents, producing different scattering-factor decay patterns with scattering angle.

---

Protein vs. small-molecule crystallography

Why proteins need special strategies

Proteins often cannot reach small-molecule-like resolution for reasons including:

• fewer total unit cells effectively contribute,
• internal motions/disorder (thermal + conformational flexibility) reduce phase coherence,
• flexible regions (loose termini/loops) may be poorly defined or absent.

Protein strategy: use prior knowledge

• Combine imperfect electron density with known amino acid propensity for α-helices and β-sheets to infer structure at lower resolution.

Small-molecule strategy

• With good crystals and high resolution, electron density may allow direct atom assignment.

---

Methodologies / workflow outlined

Data-to-density workflow (conceptual)

1. Measure many diffraction reflections → obtain structure factors (F_{hkl}).
2. Use the inverse Fourier transform of (F_{hkl}) to reconstruct the electron density map.

Practical considerations

• Ensure adequate resolution (the extent of measured reciprocal-space data).
• Recognize and mitigate truncation errors from incomplete reflection sets.
• Account for instrument envelope functions (e.g., slit settings).
• Apply correct scaling using (f_{000}) (the lecture emphasizes the importance of knowing/setting the unit-cell electron count).

---

Researchers or sources featured

No specific researchers, institutions, or external published sources were explicitly named in the provided subtitles.