Spectroscopy & Structure Determination

Reading a molecule's structure from its spectra, and knowing when you are wrong

About this course

How spectra determine molecular structure — selection rules and resolution, IR, Raman and UV-visible spectroscopy, mass spectrometry, one- and two-dimensional NMR, X-ray diffraction and absolute configuration, and how assignments fail.

What you will be able to do

  • Use selection rules to predict which transitions appear in IR, Raman and UV-visible spectra
  • Assign a molecular formula from exact mass and interpret fragmentation patterns
  • Interpret chemical shift, coupling and integration in NMR spectra
  • Assemble a structure from IR, MS, one- and two-dimensional NMR data together
  • Evaluate an assignment for impurities, stereochemical ambiguity and the risk of error

What this course covers

  • The crystal as a coherent scattering instrument
  • The phase problem
  • X-ray structure as a refined model, not a photograph
  • Data-to-parameter ratio and its consequences
  • What a crystal structure does and does not settle
  • The price of admission: crystallization as a phase transition
  • Crystallography's place in the analytical workflow
  • kT as the spectroscopic ruler
  • NMR sensitivity and the population excess
  • Selection rules and the transition moment
  • Infrared vs. Raman activity and mutual exclusion
  • Quantifying 'forbidden': molar absorptivity scale
  • Mechanisms that relax selection rules
  • The inverse problem and overdetermination
  • NMR signal weakness: the 1-in-25,000 problem
  • Magnet strength, cryoprobes, and tube size as sensitivity tools
  • Field uniformity and shimming requirements
  • Chemical shift and the parts-per-million scale
  • Chemical shift regions and the electron-density principle
  • Electronegativity correlation for chemical shifts
  • Ring-current model and aromatic deshielding
  • Integration as exact atom count and its T₁ condition
  • Harmonic oscillator model of bond vibration
  • Deuterium isotope test of the harmonic model
  • Localization condition for group frequencies
  • Carbonyl frequency as a bond-stiffness measurement
  • Overdetermination: combining multiple spectral constraints
  • Failure modes: hydrogen bonding, Fermi resonance, coupling
  • Transferability vs. specificity as opposite properties
  • Mass spectrometry and resolving power continuity
  • Nuclear binding energy and exact atomic masses
  • Nominal vs exact mass and resolving-power cost
  • Exact mass collapse to molecular formula and degrees of unsaturation
  • The nitrogen rule
  • Isotope patterns as direct structural readouts
  • Limits of exact mass at high molecular weight
  • Symmetry theorem: all scalar observables blind to chirality
  • Friedel's law and diffraction blindness
  • General principle: chiral probe required to break symmetry
  • Anomalous scattering and the Bijvoet signal size
  • The Flack parameter and its uncertainty
  • Relative vs. absolute configuration; historical context
  • Chemical routes: Mosher's method and chiral chromatography
  • Pharmaceutical consequence: the racemate arithmetic
  • Raman signal strength and the fourth-power law
  • The laser-wavelength trade-off in instrument design
  • The depolarization ratio as a symmetry measurement
  • Infrared vs. Raman complementarity
  • Water transparency and biological applications of Raman
  • Resonance Raman: selectivity and its price
  • Practical instrument choice and closing synthesis
  • Three sources of spectral line broadening
  • Comparing techniques against their natural-width floor
  • Natural linewidth as a Fourier theorem
  • Intensity–resolution trade-off and the role of field strength in NMR
  • Resolving power as a cross-technique comparison
  • Exchange broadening and measuring activation barriers
  • Every linewidth as a positive measurement
  • Stability inequality: why fragmentation is non-random
  • Electron ionization energy and reproducibility
  • Fragmentation as a rate problem; molecular ion optionality
  • Alpha cleavage, Stevenson's rule, and the thermodynamics-as-kinetics proxy
  • Even-electron rule and fragmentation cascade shape
  • Exact mass of neutral losses and the McLafferty rearrangement
  • Library matching: value, limits, and verification checklist
  • Coupling constants vs. chemical shifts: field dependence
  • Spin-spin splitting: the N+1 rule and Pascal's Triangle intensities
  • Non-equivalent neighbors and equivalent-proton splitting rules
  • Diagnostic coupling constant values and their structural meaning
  • The Karplus relation: dihedral angle from coupling constant
  • Structural ambiguity of the Karplus inverse: a non-instrumental limit
  • Second-order (AB) spectra and the geometric-mean correction
  • Mixture assumption underlying all spectroscopic techniques
  • How mixtures corrupt each spectroscopic technique
  • Chromatographic resolution as a Z-score and its cost
  • Peak capacity and the statistical ceiling on clean peaks
  • What a chromatogram is not: area percent vs. composition
  • Synthesis: the discipline of qualified quantitation
  • UV-Vis as a quantitative, not structural, tool
  • Optimum absorbance and the precision floor
  • Asymmetric error penalty and practical consequences
  • Woodward–Fieser rules: accurate, transferable, and empty
  • Free-electron (particle-in-a-box) model for polyenes
  • Bond alternation as the cause of polyene saturation
  • What UV-Vis is actually for: quantitative uses
  • Inverse detection sensitivity gain
  • Cost and resolution of the second frequency axis
  • COSY: proton–proton correlation
  • HSQC: one-bond carbon–proton correlation
  • HMBC: multi-bond C–H correlation to quaternary carbons
  • Limits of cross-peak interpretation
  • NOESY and the inverse sixth-power distance relationship
  • Evidence ranking: the likelihood-ratio ledger
  • Confirmation bias as multiplicative inflation
  • The double-count error in evidence tables
  • Resolving conflicting observations by ledger rank
  • Error compounding through inference chains
  • The disciplined structure-determination procedure
  • Anomalous observations as the most informative data
  • Lecture 11 recomputed: one arithmetic identity
  • What establishes a structure: the ledger used
  • The square penalty: what a wrong line costs
  • Correlated evidence: the double-count mechanism
  • Overturning a published claim: one hard line suffices
  • The procedural defense: a five-minute protocol
  • Course synthesis: engineering vs mathematics as the master question
  • Calibration gap: stated confidence vs. observed error rate
  • Distribution of misassignment types
  • Dangerous vs. safe structural classes
  • Hexacyclinol case study: shape of a misassignment story
  • Computed carbon shifts as a discriminator: the central quantitative result
  • Computed shifts as comparative, not absolute, measurement
  • Requirements for a safe structure: journal policy and method limits

Syllabus

  1. Spectra and Their Limits

    1. L1 · What a Spectrum Is: Selection Rules and What They Forbid
    2. L2 · Resolution, Linewidth, and the Limit You Cannot Buy Past

  2. Vibrational and Electronic Spectroscopy

    1. L3 · Infrared: Group Frequencies and Why They Transfer
    2. L4 · Raman: the Complement, and Why It Sees What IR Cannot
    3. L5 · UV-Visible: Chromophores, and the Spectrum That Is Not a Fingerprint

  3. Mass Spectrometry

    1. L6 · Mass Spectrometry: Exact Mass and the Formula It Fixes
    2. L7 · Fragmentation: Reading a Molecule From Its Pieces

  4. NMR and Combined Assignment

    1. L8 · NMR I: Chemical Shift as a Measurement of Electron Density
    2. L9 · NMR II: Coupling, and the Geometry It Reports
    3. L10 · NMR III: Relaxation, Integration, and What the Peak Areas Mean
    4. L11 · Two-Dimensional NMR: Correlations, Not Peaks
    5. L12 · Putting It Together: Structure From Four Spectra

  5. Diffraction, Chirality and Wrong Structures

    1. L13 · X-Ray Diffraction: the Structure You Can See, and Its Price
    2. L14 · Chirality and Absolute Configuration: the Hardest Question
    3. L15 · Mixtures, Impurities, and the Spectrum of Something Else
    4. L16 · When the Structure Is Wrong: Retracted Assignments, and the Course as One Argument