📖 Core Concepts
CHEM303: Physical Chemistry II: Core Concepts
Personal study notes for Skipper. Quantum Mechanics, Spectroscopy, and Kinetics.
0. What Is This Subject Really About?
Until now, Skipper, you have mostly used molecular structure as a set of rules. In CHEM201/202 you learned that a carbonyl shows up near 1700 cm−1, that an aldehyde proton sits near δ 9–10, and that conjugation pushes absorption toward the red. In CHEM301 you learned to count states with partition functions, but the energy levels you counted were handed to you. This course explains where those levels come from, and it shows how every spectrum you have ever read is a direct readout of them.
The course has one central thread:
Matter confined in space has quantized energy. Light can move a molecule between those levels only when energy and symmetry both allow it. The rate at which molecules move between states, by light or by collision, is set by how many pathways exist and how high the barriers are.
The three parts of the course title follow from this thread:
- Quantum mechanics gives you the levels. You solve (or approximate) the Schrödinger equation for the system of interest.
- Spectroscopy probes the levels. Each region of the electromagnetic spectrum matches a type of motion: radio waves for nuclear spin flips, microwaves for rotation, infrared for vibration, UV-visible for electronic rearrangement. Selection rules, which come from symmetry, decide which transitions are allowed.
- Kinetics tells you how fast molecules move between states. Transition state theory turns the statistical mechanics of CHEM301 into a rate constant. Marcus theory does the same for electron transfer, and Michaelis–Menten kinetics does it for enzymes.
Your earlier courses feed directly into this one. MATH301 (linear algebra) supplies the main language: operators are linear maps, observables are Hermitian operators, measurements are eigenvalues, and group representations are matrices. MATH102 (differential equations) supplies the solution techniques, especially separation of variables and series solutions. CHEM301 supplies the Boltzmann distribution, which you will need for line intensities, NMR sensitivity, and the Eyring equation. Atkins, de Paula, and Keeler organize their treatment the same way, moving from model systems to spectroscopy to dynamics [Atkins, de Paula & Keeler, 2018]. For deeper formal treatment, especially of operators and group theory, use Atkins and Friedman [Atkins & Friedman, 2011].
1. The Postulates: Wavefunctions, Operators, and the Born Interpretation
Quantum mechanics is not derived from classical mechanics. It is postulated, and the postulates are justified because their predictions match experiment. The working set is:
- State. The state of a system is completely described by a wavefunction ψ(r, t). Everything that can be known is encoded in ψ.
- Observables. Every observable quantity corresponds to a linear Hermitian operator Ω̂. Position becomes "multiply by x," and momentum becomes −iħ ∂/∂x.
- Measurement. A measurement of Ω can return only an eigenvalue ω of Ω̂, where Ω̂ψ = ωψ.
- Probabilities. If ψ is expanded in the eigenfunctions of Ω̂ (ψ = Σ cnφn), the probability of obtaining ωn is |cn|². The average over many measurements is the expectation value ⟨Ω⟩ = ∫ψ*Ω̂ψ dτ / ∫ψ*ψ dτ.
- Dynamics. ψ evolves according to the time-dependent Schrödinger equation, iħ ∂ψ/∂t = Ĥψ.
Time-independent Schrödinger equation (stationary states):
Ĥψ = Eψ, Ĥ = −(ħ²/2m)∇² + V(r)
Momentum operator: p̂ₓ = −iħ ∂/∂x
Commutator: [x̂, p̂ₓ] = iħ → Δx·Δpₓ ≥ ħ/2
The Born interpretation. Max Born proposed that |ψ|² is a probability density [Born, 1926]. The probability of finding the particle in a volume element dτ is |ψ|²dτ. This one statement imposes strict conditions on acceptable wavefunctions. They must be single-valued, continuous, have continuous first derivatives wherever V is finite, and be square-integrable so that they can be normalized. These boundary conditions are what produce quantization. Quantization is not an extra assumption added to the theory. It follows because only certain energies give wavefunctions that behave properly at the edges.
An analogy. Think of a guitar string clamped at both ends. Any wiggle is conceivable, but only those with a whole number of half-wavelengths can persist as standing waves. The clamps (the boundary conditions) select the allowed frequencies. In quantum mechanics, the potential and the requirement that ψ be normalizable act as the clamps.
The link to linear algebra. Hermitian operators have three properties from MATH301 that you will rely on constantly:
- Their eigenvalues are real. This is why measured quantities are real numbers.
- Eigenfunctions with different eigenvalues are orthogonal.
- Their eigenfunctions form a complete set, so any state can be expanded in them.
Two operators that commute share a common set of eigenfunctions, which means both observables can be sharp at the same time. Two that do not commute, like x̂ and p̂, obey an uncertainty relation. The uncertainty principle is therefore not about clumsy measuring instruments. It is a statement about the structure of the operators.
Something surprising. In a stationary state, |ψ|² does not change in time, even though ψ itself oscillates as e−iEt/ħ. Nothing observable changes. A "stationary state" really is static, and an isolated atom in an eigenstate does not radiate. Radiation, and therefore spectroscopy, requires a superposition of states whose interference term oscillates at the frequency (E₂ − E₁)/h. This is the deep origin of the Bohr frequency condition ΔE = hν.
2. Three Model Systems: Translation, Vibration, Rotation
Nearly all molecular spectroscopy rests on three exactly solvable problems. Learn each one as a pattern of energy levels whose spacing reveals a physical property.
2a. Particle in a Box: Confinement Creates Quantization
Place a particle of mass m in a one-dimensional region of length L with infinitely high walls. Setting ψ(0) = ψ(L) = 0 gives:
ψₙ(x) = (2/L)^½ sin(nπx/L), n = 1, 2, 3, ... Eₙ = n²h² / (8mL²) Spacing: Eₙ₊₁ − Eₙ = (2n + 1)h² / (8mL²)
Three lessons come from this result:
- Zero-point energy. n = 0 is not allowed, because it would make ψ zero everywhere. A confined particle can never be at rest, which is consistent with the uncertainty principle.
- Size matters as 1/L². A bigger box gives more closely spaced levels. In the macroscopic limit the levels form a continuum, which is the correspondence principle at work.
- Mass matters as 1/m. Electrons, being light, show large quantization effects. Nuclei show much smaller ones.
Application to conjugated π systems (the free-electron model). Treat the N π electrons of a linear polyene as particles in a box whose length spans the conjugated chain. Fill the levels two electrons at a time. The HOMO is n = N/2 and the LUMO is n = N/2 + 1, so:
ΔE(HOMO→LUMO) = (N + 1)h² / (8mₑL²) → λ = 8mₑcL² / [h(N + 1)]
As the chain grows, L² rises faster than N + 1, so λ shifts red. This is exactly the conjugation trend you memorized in CHEM202: β-carotene, with 11 conjugated double bonds, is orange.
The model is crude. For short polyenes it overestimates λ badly. With L ≈ 1 nm, hexatriene is predicted near 450 nm, but it actually absorbs near 260 nm. The reason is real bond-length alternation, which the flat-bottomed box ignores. The model works much better for cyanine dyes, where resonance equalizes the bond lengths, and that is where Hans Kuhn applied it with real success [Kuhn, 1949]. The general lesson is to trust a model's trends more than its absolute numbers, and to understand why it fails.
2b. Harmonic Oscillator: The Model for Vibration
Near the bottom of any potential well, V(x) ≈ ½kx². This is just a Taylor expansion, which you know from MATH102. For a diatomic molecule with reduced mass μ = m₁m₂/(m₁ + m₂):
E_v = (v + ½)ħω, ω = (k/μ)^½, v = 0, 1, 2, ... ψ_v(x) = N_v H_v(y) e^(−y²/2), y = x/α, α = (ħ²/μk)^¼ (H_v = Hermite polynomials) Zero-point energy: E₀ = ½ħω
The levels are evenly spaced. This is the defining feature of the harmonic oscillator. Unlike the box, every transition between adjacent levels has the same energy, ħω.
Selection rules. For an infrared transition to occur, two conditions must hold:
- Gross selection rule. The electric dipole moment must change during the vibration, (∂μ/∂x)₀ ≠ 0. Homonuclear diatomics such as N₂ and O₂ are therefore IR-inactive, which is fortunate for anyone measuring IR spectra in air.
- Specific selection rule. Δv = ±1. This follows because x connects Hv only to Hv±1, a recursion property of the Hermite polynomials.
Anharmonicity. Real bonds break, so the potential flattens at large extension. The Morse potential captures this:
V(x) = hcD̃ₑ [1 − e^(−a x)]² G̃(v) = (v + ½)ν̃ − (v + ½)² xₑν̃ (wavenumber terms)
The levels converge as v rises, overtones (Δv = ±2, ±3) become weakly allowed, and a Birge–Sponer extrapolation of the converging spacings estimates the dissociation energy. Note that the spectroscopic dissociation energy D₀ = Dₑ − zero-point energy. Bonds dissociate from v = 0, not from the bottom of the well.
Isotope effects. Because ω ∝ μ−½, replacing H with D lowers a stretching frequency by about a factor of √2. A C–H stretch near 2900 cm−1 becomes a C–D stretch near 2100 cm−1. The zero-point energy drops too. Since the C–D bond starts lower in the well, it needs more energy to reach the transition state. This is the quantum origin of the kinetic isotope effect you met in CHEM201.
2c. Rigid Rotor: The Model for Rotation
For a diatomic molecule rotating with fixed bond length R, the moment of inertia is I = μR². The angular wavefunctions are the spherical harmonics YJ,M(θ, φ):
E_J = (ħ²/2I) J(J + 1), J = 0, 1, 2, ... Degeneracy: g_J = 2J + 1 (M_J = −J, ..., +J) In wavenumbers: F̃(J) = B̃ J(J + 1), B̃ = ħ / (4πcI) Pure rotational selection rules: permanent dipole needed; ΔJ = ±1 Line positions: ν̃ = 2B̃(J + 1) → equally spaced lines, separation 2B̃
The spacing between lines gives B̃, B̃ gives I, and I gives R. Microwave spectroscopy measures bond lengths to roughly a thousandth of an ångström, which makes it one of the most precise structural methods in chemistry.
Line intensities come from CHEM301. The population of level J is proportional to (2J + 1)e−EJ/kT, so the intensities rise and then fall. The most populated level is Jmax ≈ (kT/2hcB̃)½ − ½. You can therefore read the temperature directly from a rotational spectrum.
A beautiful unification. The rigid rotor's angular wavefunctions are exactly the angular parts of hydrogen's orbitals. Rotation of a molecule and the angular motion of an electron are the same mathematics, because both describe motion on a sphere.
3. The Hydrogen Atom and Many-Electron Atoms
3a. Hydrogen: The Only Exact Atom
The Coulomb potential V = −e²/(4πε₀r) is spherically symmetric, so the Schrödinger equation separates in spherical coordinates:
ψ_{n,l,mₗ}(r, θ, φ) = R_{n,l}(r) · Y_{l,mₗ}(θ, φ)
Radial part: R_{n,l}(r) ∝ ρ^l L_{n+l}^{2l+1}(ρ) e^(−ρ/2) (associated Laguerre polynomials)
Angular part: Y_{l,mₗ} ∝ P_l^{|mₗ|}(cos θ) e^(i mₗ φ) (associated Legendre polynomials)
Eₙ = −μe⁴ / (32π²ε₀²ħ²n²) = −hcR_H / n²
n = 1, 2, 3, ...; l = 0, ..., n−1; mₗ = −l, ..., +l
Degeneracy (ignoring spin): n²
Radial nodes: n − l − 1 Angular nodes: l
Each quantum number arises from a boundary condition:
- mₗ comes from requiring ψ to be single-valued as φ goes from 0 to 2π.
- l comes from requiring the Legendre equation to have solutions that stay finite at the poles.
- n comes from requiring the radial function to vanish as r → ∞, which truncates the Laguerre series.
This is the guitar-string principle in three dimensions.
The radial distribution function. P(r) = r²R(r)² gives the probability of finding the electron in a thin shell at radius r. For the 1s orbital, ψ² is largest at the nucleus, yet P(r) peaks at the Bohr radius a₀ = 52.9 pm. These two statements do not contradict each other. A shell's volume grows as r², so a large shell of modest density can hold more probability than a tiny region of high density near the nucleus.
The "accidental" degeneracy. In hydrogen, 2s and 2p have the same energy, because the energy depends only on n. This reflects a hidden symmetry of the pure 1/r potential. Add a second electron and the degeneracy breaks: penetrating s orbitals feel more nuclear charge and drop below p. That is the origin of the Aufbau ordering you learned in CHEM101.
3b. Spin, the Pauli Principle, and Slater Determinants
Electrons carry an intrinsic angular momentum, spin, with s = ½ and mₛ = ±½. Spin does not appear in the non-relativistic Schrödinger equation. It comes from relativistic quantum mechanics (the Dirac equation), and in this course we simply attach it as a label.
The Pauli principle, in its fundamental form [Pauli, 1925]: the total wavefunction must be antisymmetric under exchange of any two electrons.
Ψ(1, 2) = −Ψ(2, 1)
"No two electrons may share all four quantum numbers" is a consequence of this rule, not the rule itself. A Slater determinant builds antisymmetry in automatically:
1 | φ₁(1) φ₂(1) ... φ_N(1) |
Ψ = ——————— | φ₁(2) φ₂(2) ... φ_N(2) |
√(N!) | ... |
| φ₁(N) φ₂(N) ... φ_N(N) |
Swapping two electrons swaps two rows, which flips the sign of the determinant. Putting two electrons in the same spin-orbital makes two columns identical, which makes the determinant zero. The Pauli exclusion principle is therefore a property of determinants, which you studied in MATH301.
Exchange and Hund's rule. Antisymmetry makes electrons of the same spin avoid each other: the wavefunction vanishes when two same-spin electrons sit at the same point. This "Fermi hole" lowers their mutual repulsion. That lowering is the exchange energy, and it is why triplets usually lie below the corresponding singlets (Hund's first rule). It is also why phosphorescence, from a triplet, appears at longer wavelength than fluorescence, from a singlet. This is a quantum effect with no classical counterpart.
3c. The Hartree–Fock Approximation
Electron–electron repulsion makes the many-electron Schrödinger equation impossible to solve exactly. Hartree–Fock (HF) theory assumes the wavefunction is a single Slater determinant. Each electron then moves in the average field of all the others:
f̂ φᵢ = εᵢ φᵢ f̂ = ĥ_core + Σⱼ (2Ĵⱼ − K̂ⱼ) (Ĵ: Coulomb operator, K̂: exchange operator)
The catch is that f̂ depends on the orbitals you are trying to find. You guess orbitals, build f̂, solve for new orbitals, and repeat until nothing changes. That procedure is the self-consistent field (SCF).
HF typically recovers more than 99% of the total energy. The missing part, the correlation energy, is small but chemically decisive, because it is often comparable to reaction energies. Correlated methods such as configuration interaction, coupled cluster, and perturbation theory exist to recover it. Density functional theory approaches the problem differently, approximating exchange and correlation together through a functional of the electron density.
4. Molecular Orbital Theory: Where Chemical Bonds Come From
H₂⁺ is the simplest molecule: one electron and two protons. Within the Born–Oppenheimer approximation, the nuclei are clamped because they are heavy and slow. We then write the molecular orbital as a linear combination of atomic orbitals (LCAO):
ψ± = (1sₐ ± 1s_b) / [2(1 ± S)]^½ S = ∫1sₐ 1s_b dτ (overlap integral) α = Hₐₐ = ∫1sₐ Ĥ 1sₐ dτ (Coulomb integral) β = Hₐ_b = ∫1sₐ Ĥ 1s_b dτ (resonance integral, negative) E± = (α ± β) / (1 ± S)
Why does bonding happen? The common textbook answer is "electron density accumulates between the nuclei." That answer is correct but incomplete. The deeper analysis, which Atkins and Friedman discuss, involves a subtle balance between potential and kinetic energy [Atkins & Friedman, 2011]. Delocalizing the electron over two centers lowers its kinetic energy, much as a bigger box lowers the energy in Section 2a, and that effect is central. Bonding is partly the particle in a box again: give the electron more room and its energy drops.
The antibonding asymmetry. Because the denominators are 1 + S and 1 − S, the antibonding orbital is raised more than the bonding orbital is lowered. This is why He₂, with two electrons in each orbital, is unbound: the destabilization outweighs the stabilization. It is also why filled–filled orbital interactions are repulsive, which underlies steric repulsion.
H₂ and beyond. Place two electrons with opposite spins in σg(1s). Bond order is defined as ½(bonding electrons − antibonding electrons). For second-row diatomics, s–p mixing places σg(2p) above πu(2p) for B₂ through N₂, and below it for O₂ and F₂. O₂ has two unpaired electrons in degenerate πg* orbitals, so it is paramagnetic with a 3Σg− ground state. Simple Lewis structures cannot explain this, which made it one of MO theory's early triumphs.
The variation principle. For any trial function, the expectation value of the energy is at least the true ground-state energy E₀. When you minimize the energy with respect to the LCAO coefficients, you obtain the secular equations, a generalized eigenvalue problem (H − ES)c = 0. MO theory is linear algebra applied to chemistry. Hückel theory [Hückel, 1931] is the simplest version: S = 0, β is applied only between neighbors, and the result is the 4n + 2 aromaticity rule from CHEM202.
5. Symmetry and Group Theory: The Grammar of Selection Rules
Group theory is the most powerful shortcut in this course. Here is the idea. If an integral's integrand is odd overall under any symmetry operation of the molecule, the integral is exactly zero. You do not have to evaluate it. You only need to know how each piece of the integrand transforms.
Point groups. The symmetry operations of a molecule are E, Cₙ, σ, i, and Sₙ. Together they form a mathematical group, which has closure, identity, inverses, and associativity. Some examples:
- H₂O belongs to C₂v.
- NH₃ belongs to C₃v.
- Benzene belongs to D₆h.
- CH₄ belongs to Td.
Representations and characters. Each symmetry operation can be written as a matrix acting on some basis, such as atomic displacements or orbitals. The trace of that matrix is its character χ(R). A representation that can be block-diagonalized into smaller pieces is reducible. The smallest pieces are the irreducible representations (irreps), which are listed in the character table. The reduction formula gives the number of times irrep i appears in a reducible representation Γ:
nᵢ = (1/h) Σ_classes g_c · χ_Γ(R) · χᵢ(R) h = order of the group, g_c = number of operations in the class
This formula works because the characters of irreps are orthogonal vectors, a result directly analogous to the orthogonality of eigenvectors in MATH301.
The master selection rule. An integral ∫ψf* Ô ψi dτ can be nonzero only if the direct product Γ(ψf) ⊗ Γ(Ô) ⊗ Γ(ψi) contains the totally symmetric irrep (A₁, A₁g, and so on). For specific spectroscopies:
- IR-active vibrations transform like x, y, or z, the components of the dipole operator.
- Raman-active vibrations transform like x², xy, and the other quadratic functions, the components of the polarizability.
- Rule of mutual exclusion. In a centrosymmetric molecule, no mode is both IR- and Raman-active, because x is u (antisymmetric under inversion) while x² is g (symmetric).
Worked examples. Water has 3N − 6 = 3 vibrations, which reduce to 2A₁ + B₂ in C₂v. (The B label is B₁ or B₂ depending on which plane you call σv. Conventions differ between books, so check yours.) All three modes are both IR- and Raman-active.
CO₂ is linear, with 3N − 5 = 4 modes, and belongs to D∞h:
- The symmetric stretch (Σg+) is Raman-only. Your eye can see why: the dipole stays zero throughout the motion.
- The antisymmetric stretch (Σu+, about 2349 cm−1) and the degenerate bend (Πu, about 667 cm−1) are IR-only.
The bend near 667 cm−1 absorbs thermal infrared emitted by Earth, which is the core physics of CO₂ as a greenhouse gas.
6. Spectroscopy as One Idea: The Transition Dipole Moment
All absorption spectroscopy obeys the same two requirements:
Resonance: ΔE = hν (Bohr frequency condition) Intensity: I ∝ |μ_fi|², μ_fi = ⟨ψ_f | μ̂ | ψ_i⟩ (transition dipole moment)
Light's oscillating electric field couples to the molecule's dipole operator. Time-dependent perturbation theory then gives Fermi's golden rule: the transition rate is proportional to |μfi|² times the density of available states. Every selection rule in this course is a statement about when μfi = 0.
6a. Rotational (Microwave) Spectroscopy
The key facts are the rigid-rotor results from Section 2c. Real molecules stretch as they spin faster, which adds a centrifugal distortion term: F̃(J) = B̃J(J + 1) − D̃J²(J + 1)². For polyatomic molecules, you classify the rotor by its principal moments of inertia as linear, symmetric top, spherical top, or asymmetric top. Spherical tops such as CH₄ have no permanent dipole and therefore no pure rotational spectrum. Combining data from several isotopologues, by measuring B̃ for 12C and 13C versions for example, lets you solve for the complete molecular geometry.
6b. Vibrational Spectroscopy: IR and Raman
A nonlinear molecule with N atoms has 3N − 6 normal modes. In each normal mode, every atom moves at the same frequency and in phase. Mathematically, the normal modes are the eigenvectors of the mass-weighted Hessian matrix, so finding them means diagonalizing a matrix (MATH301 again). In normal-mode coordinates, the coupled vibrations separate into 3N − 6 independent harmonic oscillators.
Group frequencies. Normal modes involve the whole molecule, yet a carbonyl still absorbs near 1700 cm−1 in nearly any molecule. This works because a C=O stretch differs so much in force constant and mass from its surroundings that it barely mixes with other motions. The localization is approximate, which explains why group frequencies shift with environment. Those shifts carry information, and you will see this when we reach the amide I band.
Raman spectroscopy is inelastic light scattering. Light induces a dipole μind = αE. If the vibration modulates the polarizability (∂α/∂Q ≠ 0), the scattered light carries sidebands at ν₀ ± νvib:
- Stokes lines, at lower frequency, are stronger.
- Anti-Stokes lines, at higher frequency, are weaker because they require molecules that start in excited vibrational levels.
The anti-Stokes/Stokes intensity ratio therefore acts as a Boltzmann thermometer. Rotational Raman obeys ΔJ = 0, ±2, so even homonuclear diatomics like N₂ show rotational Raman spectra.
Rovibrational structure. In gas-phase IR, each vibrational band splits into a P branch (ΔJ = −1) and an R branch (ΔJ = +1), with a gap where the Q branch (ΔJ = 0) would be, because Q is forbidden for most diatomics. In HCl, the P and R branches also show doubled lines from the 35Cl/37Cl isotopes. One spectrum thus reveals the bond length, the force constant, and the isotopic composition.
7. Electronic Spectroscopy and Photochemistry
7a. Franck–Condon Principle
Electrons move much faster than nuclei. An electronic transition happens on a timescale of about 10−15 s, while a vibration takes about 10−13 s. The transition is therefore vertical on a potential energy diagram: the nuclei have no time to move during the jump [Franck, 1926; Condon, 1926]. In the Born–Oppenheimer approximation, the transition dipole factors into an electronic part and a vibrational overlap:
μ_fi ≈ μ_el · ⟨χ_v' | χ_v⟩ Franck–Condon factor: |⟨χ_v' | χ_v⟩|²
If the excited state has a longer equilibrium bond length, which is typical because an electron has been promoted into an antibonding orbital, the strongest vibronic line will not be 0→0. The intensity spreads over a progression of vibrational lines. This explains the shape of absorption bands.
Beer–Lambert revisited. In CHEM302 you used A = εcl as an empirical law. Now you can see what ε means. The integrated absorption coefficient is proportional to |μfi|², and the oscillator strength f = (8π²mₑν/3he²)|μfi|² measures how "allowed" a transition is.
- Allowed π→π* transitions have ε of roughly 10⁴–10⁵ M−1cm−1.
- Symmetry-forbidden n→π* transitions have ε of roughly 10–100.
- Spin-forbidden singlet→triplet absorption is weaker still.
A molar absorptivity is a quantum mechanical matrix element in disguise.
7b. The Jablonski Diagram and Quantum Yield
After a molecule absorbs a photon into an excited singlet state (Sn), the following processes compete:
- Vibrational relaxation and internal conversion (about 10−12 s) bring it rapidly to the lowest vibrational level of S₁. Kasha's rule states that emission almost always occurs from the lowest excited state of a given multiplicity [Kasha, 1950].
- Fluorescence (S₁ → S₀, spin-allowed, about 10−9 s) is red-shifted from absorption. This is the Stokes shift, caused by relaxation in both the excited and ground states.
- Intersystem crossing (S₁ → T₁) is spin-forbidden but enabled by spin–orbit coupling, which is strong in molecules with heavy atoms.
- Phosphorescence (T₁ → S₀) is slow, lasting microseconds to seconds, because it is spin-forbidden. It is also lower in energy than fluorescence, because of the exchange energy from Section 3b.
Fluorescence quantum yield: Φ_F = k_r / (k_r + k_IC + k_ISC + k_q[Q]) Observed lifetime: τ = 1 / (k_r + k_IC + k_ISC + k_q[Q]) Stern–Volmer quenching: Φ₀/Φ = 1 + k_q τ₀ [Q]
The quantum yield is a branching ratio, the same idea as product ratios from competing first-order pathways in CHEM102.
7c. Förster Resonance Energy Transfer (FRET)
An excited donor can transfer its energy without emitting a photon. The transfer happens through dipole–dipole coupling to a nearby acceptor whose absorption overlaps the donor's emission [Förster, 1948]:
E = 1 / [1 + (r/R₀)⁶] R₀⁶ ∝ κ² Φ_D J / n⁴ κ² = orientation factor (2/3 for isotropic averaging), J = spectral overlap integral, n = refractive index, R₀ = Förster distance (typically ~2–8 nm)
The r−6 dependence makes the efficiency switch from about 1 to about 0 over a narrow distance range around R₀. That steepness is what makes FRET such a sensitive ruler. Stryer and Haugland demonstrated this in 1967 using polyproline spacers of defined length, calling FRET "a spectroscopic ruler" [Stryer & Haugland, 1967]. The r−6 law itself comes from dipole–dipole physics: the coupling energy scales as r−3, and the golden-rule rate depends on the coupling squared.
8. NMR Spectroscopy: Quantum Mechanics You Can Hold in Your Hand
NMR is where this course meets the organic chemistry labs you have already taken. The syllabus's 1951 Journal of Chemical Physics paper, "Nuclear Magnetic Resonance Spectra and Chemical Structure" (DOI: 10.1063/1.1699910), comes from the brief period in the early 1950s when physicists realized that resonance frequencies depended on chemical environment. That discovery turned a nuclear physics technique into chemistry's most important structural tool [J. Chem. Phys., 1951]. (The course listing does not give the author, so look up the DOI and record the attribution correctly in your notes rather than guessing.)
8a. Nuclear Zeeman Effect and the Larmor Frequency
Nuclear magnetic moment: μ = γħI Energy in field B₀: E = −γħ m_I B₀, m_I = −I, ..., +I For I = ½ (¹H, ¹³C, ¹⁵N, ¹⁹F, ³¹P): ΔE = γħB₀ → ν_L = γB₀ / 2π (Larmor frequency) ¹H: γ/2π ≈ 42.58 MHz T⁻¹ → 14.1 T magnet ≈ 600 MHz spectrometer
The sensitivity problem. At 600 MHz and 298 K, ΔE/kT ≈ 1 × 10−4. The fractional excess population in the lower spin state is only about 5 × 10−5, roughly 1 extra spin in every 20,000. NMR detects only that tiny surplus. This is why chemists pay so much for stronger magnets: the signal grows steeply with field strength, because both the population difference and the detected frequency increase.
8b. Chemical Shift
Electrons circulating around a nucleus, driven by B₀, produce a small opposing field. The nucleus therefore experiences Bloc = (1 − σ)B₀, where σ is the shielding constant. To make the scale independent of field strength, we report:
δ = (ν − ν_ref) / ν_ref × 10⁶ (ppm; reference TMS)
Electron-withdrawing groups deshield, which gives higher δ. Aromatic ring currents deshield protons outside the ring and shield protons above it. This explains why benzene's protons appear near 7.3 ppm and why protons positioned over aromatic rings in folded proteins can appear unusually upfield.
8c. J-Coupling
Spin–spin coupling between nuclei is transmitted through the bonding electrons, mainly by the Fermi contact interaction. The splitting is measured in Hz and is independent of B₀. Chemical shifts in Hz, by contrast, scale with B₀. This difference is the experimental test for telling a coupling from two distinct chemical shifts: run the sample at two field strengths and see which separation changes.
For first-order spectra (Δν ≫ J), n equivalent neighbors produce n + 1 lines with binomial intensities. Three-bond couplings depend on the dihedral angle φ through the Karplus relation [Karplus, 1959]:
³J(φ) = A cos²φ + B cos φ + C
Coupling constants therefore encode conformation. Keep this in mind for the peptide section.
8d. Relaxation, FT-NMR, and 2D NMR
- T₁ (longitudinal, spin–lattice relaxation) is the recovery of z-magnetization toward Boltzmann equilibrium. It requires energy exchange with the surroundings, driven by fluctuating local fields at the Larmor frequency.
- T₂ (transverse, spin–spin relaxation) is the loss of phase coherence in the xy-plane. T₂ is at most 2T₁, and in practice T₂ ≤ T₁. The natural linewidth is Δν½ = 1/(πT₂), so faster dephasing produces broader lines. Large, slowly tumbling molecules have short T₂ values, which is why NMR of large proteins is difficult.
- FT-NMR. A short, intense radiofrequency pulse tips the magnetization so that all frequencies are excited at once. The receiver records the free induction decay (FID), and a Fourier transform converts it into a spectrum. This multiplex approach, the Fellgett advantage, enables signal averaging and made 13C NMR routine. You studied Fourier series in MATH102, and here is their payoff.
- 2D NMR. A second time variable (t₁) is incremented across many experiments. After a two-dimensional Fourier transform, off-diagonal cross peaks show which nuclei are connected. In COSY, the connection is through-bond J-coupling. In NOESY, it is through-space dipolar cross-relaxation (the nuclear Overhauser effect, which scales as r−6, the same law as FRET). In HSQC, it is a direct heteronuclear bond.
Your BIOL302 physiology course mentioned MRI. MRI contrast is T₁ and T₂ contrast between tissues, so this section is also the physics behind every clinical MRI scan.
9. Advanced Chemical Kinetics: Why Rates Are What They Are
9a. Transition State Theory and the Eyring Equation
In CHEM102 you wrote k = A e−Eₐ/RT and treated A as an empirical constant. Transition state theory (TST) gives that prefactor physical meaning [Eyring, 1935]. TST assumes that reactants are in quasi-equilibrium with an activated complex at the saddle point of the potential energy surface, and that the complex crosses to products at a rate set by one special vibrational mode. That mode has turned into translation along the reaction coordinate.
k = κ (k_B T / h) K‡ = κ (k_B T / h) exp(−ΔG‡/RT) = κ (k_B T / h) exp(ΔS‡/R) exp(−ΔH‡/RT) k_B T / h ≈ 6.2 × 10¹² s⁻¹ at 298 K (a universal frequency) κ = transmission coefficient (≤ 1; fraction of crossings that don't recross) Eyring plot: ln(k/T) = −ΔH‡/(RT) + ln(k_B/h) + ΔS‡/R Relation to Arrhenius (unimolecular, solution): Eₐ = ΔH‡ + RT
Where kBT/h comes from. This is where CHEM301 pays off. K‡ is written using partition functions. The vibrational partition function for the reaction-coordinate mode, in the limit of low frequency, is kBT/hν. When you multiply by the crossing frequency ν, the ν cancels and leaves a frequency that is the same for every reaction. That cancellation is what the Eyring equation rests on.
Activation entropy carries mechanistic information:
- A large negative ΔS‡ indicates an ordered, associative transition state, such as a bimolecular reaction or a cyclic arrangement like a Diels–Alder reaction.
- A positive ΔS‡ suggests dissociation or loosening in the transition state, or the release of ordered solvent.
The Arrhenius A-factor told you very little. ΔS‡ tells you about the structure of the transition state.
9b. Marcus Theory of Electron Transfer
Electron transfer is fast, but the surrounding nuclei (bond lengths and solvent dipoles) must first reorganize so that the donor and acceptor states have equal energy. This is the Franck–Condon principle again, now applied to a reaction. Marcus modeled the reactant and product free energies as intersecting parabolas along a collective solvent coordinate [Marcus, 1956]:
ΔG‡ = (λ + ΔG°)² / (4λ)
λ = reorganization energy (energy to distort reactants to product geometry
without transferring the electron)
The result that surprised chemists. As ΔG° becomes more negative, the rate first increases. It peaks when −ΔG° = λ, where the reaction is barrierless. Beyond that point the rate decreases, even though the reaction is more favorable. This is the Marcus inverted region. It contradicts the intuition that more downhill means faster, and it was controversial for decades until Miller, Calcaterra, and Closs confirmed it experimentally in 1984 [Miller, Calcaterra & Closs, 1984]. Marcus received the 1992 Nobel Prize. The inverted region is one reason photosynthetic reaction centers avoid wasteful charge recombination: the recombination reaction is so exergonic that it falls into the inverted region and is slow.
9c. Diffusion-Controlled Reactions
In solution, two reactants must first meet. If every encounter leads to reaction, the rate is limited by diffusion [Smoluchowski, 1917]:
k_d = 4π R* D N_A (R* = reaction distance, D = D_A + D_B) With Stokes–Einstein (D = k_B T / 6πηa) for similar-sized species: k_d ≈ 8RT / (3η) Water, 298 K (η ≈ 0.89 mPa·s): k_d ≈ 7 × 10⁹ M⁻¹ s⁻¹
This is the speed limit for bimolecular reactions in water, and it matters a great deal in the next section.
10. Enzyme Kinetics Through the Lens of TST
The mechanism and steady-state treatment are as follows [Michaelis & Menten, 1913]:
E + S ⇌ ES → E + P (k₁, k₋₁, k₂ = k_cat) Steady state d[ES]/dt = 0 gives: v = V_max [S] / (K_M + [S]), V_max = k_cat [E]₀, K_M = (k₋₁ + k_cat) / k₁
The TST interpretation treats each kinetic parameter as an Eyring barrier:
- kcat = (kBT/h) exp(−ΔG‡cat/RT). It reflects the barrier from ES to the transition state ES‡.
- kcat/KM, the specificity constant, reflects the barrier from free E + S to ES‡. It is the second-order rate constant at low [S], and it is the right quantity for comparing how well an enzyme handles competing substrates.
- Transition-state stabilization. Pauling's insight was that an enzyme binds the transition state much more tightly than the substrate. Catalysis comes from that differential binding. Rate enhancements reach 10¹⁷ or more over uncatalyzed reactions, and transition-state analogs make extraordinarily potent inhibitors because of this principle.
- "Perfect" enzymes. Triose phosphate isomerase and a few others have kcat/KM ≈ 10⁸–10⁹ M−1s−1. That is close to the diffusion limit from Section 9c. For these enzymes, evolution has made chemistry faster than the encounter rate, so encounter is now the slow step [Albery & Knowles, 1976].
Inhibition mechanisms. Define α = 1 + [I]/KI for inhibitor binding to free enzyme and α′ = 1 + [I]/K′I for binding to ES.
General (mixed): v = V_max [S] / (α K_M + α′ [S]) Competitive (binds E only, α′ = 1): apparent K_M ↑ (×α), V_max unchanged Uncompetitive (binds ES only, α = 1): K_M and V_max both ↓ (÷α′) Noncompetitive (α = α′): V_max ↓, K_M unchanged
On a Lineweaver–Burk plot (1/v versus 1/[S]), the three cases look different:
- Competitive inhibition gives lines that intersect on the y-axis.
- Uncompetitive inhibition gives parallel lines.
- Noncompetitive inhibition gives lines that intersect on the x-axis.
Use the double-reciprocal plot for diagnosis only. For fitting parameters, prefer nonlinear regression, as you learned in MATH201, because the reciprocal transformation distorts the errors.
11. Connections to Peptide Chemistry
Almost every concept in this course becomes a tool for studying peptides and proteins. This is where your chemistry and biology training come together.
- The peptide bond as an MO problem. The amide nitrogen lone pair delocalizes into the C=O π* orbital. This gives the C–N bond partial double-bond character, a planar amide, and a rotational barrier of roughly 80–90 kJ/mol. The resulting planarity is why a protein backbone has only two main flexible dihedrals per residue, φ and ψ, which is the basis of the Ramachandran plot. MO theory dictates protein architecture.
- Vibrational spectroscopy of secondary structure. The amide I band, mostly C=O stretch near 1600–1700 cm−1, is a group frequency perturbed by hydrogen bonding and by coupling between neighboring amides. α-Helices absorb near 1650–1658 cm−1, and β-sheets absorb lower, near 1620–1640 cm−1 [Krimm & Bandekar, 1986]. Amide II (N–H bend combined with C–N stretch) appears near 1550 cm−1. When you exchange amide N–H for N–D in D₂O, the reduced-mass shift from Section 2b appears directly in the spectrum.
- UV-Vis. The peptide backbone has a π→π* transition near 190 nm and an n→π* transition near 220 nm. The aromatic side chains (Trp, Tyr, and to a lesser extent Phe) produce the familiar A₂₈₀ that you use for protein concentration via Beer–Lambert. Circular dichroism of the backbone transitions reports helix and sheet content, and it relies on the same transition dipole formalism, extended to chiral molecules.
- Fluorescence and FRET. Tryptophan fluorescence is extremely sensitive to its surroundings. It blue-shifts when buried, which makes it a built-in folding probe. FRET between dye pairs attached to a peptide measures end-to-end distances, folding transitions, and conformational dynamics, down to the level of single molecules.
- NMR structure determination [Wüthrich, 1986]. Each NMR concept from Section 8 maps to a structural measurement:
- The 1H–15N HSQC spectrum is a protein's fingerprint, with roughly one cross peak per backbone amide.
- NOESY cross-peak intensities, which scale as r−6, give inter-proton distances up to about 5 Å.
- The Karplus relation applied to 3J(HN–Hα) gives φ angles: about 4 Hz indicates helix, and about 8–10 Hz indicates β-strand.
- Hα chemical shifts systematically indicate secondary structure.
- T₁, T₂, and heteronuclear NOE measurements report backbone motion on picosecond-to-nanosecond timescales.
- Peptide kinetics through TST. Two examples show how the Eyring equation applies:
- Uncatalyzed hydrolysis of a peptide bond at neutral pH has a half-life of hundreds of years at room temperature [Radzicka & Wolfenden, 1996]. Plug that into Eyring and you get a ΔG‡ near 100 kJ/mol, which also tells you how much transition-state stabilization a protease must provide.
- Prolyl cis/trans isomerization is slow, on the scale of seconds to minutes, because of the amide rotational barrier. It is often the rate-limiting step in protein folding, and peptidyl-prolyl isomerases exist specifically to catalyze it.
- Electron transfer through proteins. In respiratory chain proteins and photosynthetic reaction centers, Marcus theory governs electron tunneling through peptide frameworks. Rates fall off exponentially with distance, and the protein environment tunes λ.
- Computational peptide chemistry. Hartree–Fock and its correlated successors parameterize the force fields used in molecular dynamics simulations of peptides. The harmonic bond terms in those force fields are the harmonic oscillator from Section 2b.
12. Common Misconceptions to Watch Out For
- "|ψ|² is where the electron is smeared out." No. |ψ|² is a probability density for the outcome of a position measurement. A single measurement always finds the whole electron at one point.
- "Zero-point energy is energy the molecule could release if cooled enough." It cannot be extracted. It is the lowest possible energy of the confined system, and it persists at 0 K.
- "The uncertainty principle is about measurement disturbance." Its formal statement concerns the spread of a quantum state for non-commuting operators. Even an ideal measurement cannot remove it.
- "Electrons in orbitals orbit the nucleus." Stationary states have time-independent probability densities. Nothing circulates in the classical sense. The 1s orbital has zero orbital angular momentum, which no classical orbit could have.
- "The 1s electron is most likely found at the nucleus" versus "at a₀." Both statements are correct about different quantities. The probability density ψ² peaks at r = 0. The radial probability P(r) = r²R² peaks at a₀. Always ask which quantity is being discussed.
- "Orbitals are physical objects." Hartree–Fock orbitals are mathematical constructions inside an approximation. Any unitary rotation of the occupied orbitals (canonical versus localized, for example) gives the same total wavefunction. Only the total electron density and the energy are observable.
- "Bonding and antibonding effects are symmetric." Antibonding orbitals are destabilized more than bonding orbitals are stabilized, because of the 1 ± S denominators.
- "IR-inactive means the vibration doesn't exist." N₂ vibrates perfectly well. It simply has no changing dipole to couple to light. Raman spectroscopy sees it.
- "Forbidden transitions never happen." Selection rules come from approximations, such as the electric dipole limit, the harmonic potential, and pure spin states. Spin–orbit coupling, vibronic coupling, and anharmonicity all relax them. "Forbidden" usually means "weak."
- "A bigger J splitting means a stronger magnet." J is field-independent in Hz. Chemical shift differences in Hz grow with B₀. This is why higher fields simplify second-order multiplets.
- "T₂ relaxation releases energy to the lattice." Pure T₂ processes are dephasing and need no energy exchange. Energy exchange is the T₁ process.
- "The transition state is an intermediate you could isolate." It is a saddle point with a lifetime of roughly one vibrational period, and it has no well of its own. Intermediates sit in local minima.
- "Eₐ = ΔH‡." They differ by RT for unimolecular reactions in solution, and by different amounts in other cases. The difference is small but real when you analyze data precisely.
- "More exergonic means faster." Marcus showed that beyond −ΔG° = λ, rates fall. Rate–driving force relationships are not always monotonic.
- "KM is the dissociation constant of ES." That is only true when kcat ≪ k₋₁, the original Michaelis–Menten rapid-equilibrium assumption. In general, KM = (k₋₁ + kcat)/k₁, which is a kinetic constant rather than a thermodynamic one.
- "FRET is emission by the donor and reabsorption by the acceptor." Förster transfer is nonradiative dipole–dipole coupling, and no photon is exchanged. Its r−6 distance law is different from the geometry of trivial reabsorption.
Closing Thought
Skipper, here is the theme to carry through the semester. The same few ideas keep returning in new forms:
- Confinement leads to quantization. You see it in the box, in vibration, in rotation, and in the atom.
- Symmetry leads to selection rules. Group theory and the transition dipole moment express this.
- Nuclei are slow and electrons are fast. This underlies the Born–Oppenheimer approximation, the Franck–Condon principle, and Marcus reorganization.
- Statistics decides how many molecules are in each state, which governs Boltzmann intensities, NMR sensitivity, and kBT/h in the Eyring equation.
- The r−6 dipolar law appears in both FRET and the NOE.
When a new topic seems unfamiliar, ask which of these ideas it is a new version of. It is usually one of them.
Primary references for this section: [Atkins, de Paula & Keeler, 2018], Physical Chemistry, 11th ed., Oxford University Press; [Atkins & Friedman, 2011], Molecular Quantum Mechanics, 5th ed., Oxford University Press; syllabus key paper, "Nuclear Magnetic Resonance Spectra and Chemical Structure," J. Chem. Phys. (1951), DOI: 10.1063/1.1699910.
⚙️ Mechanisms & Applications
Mechanisms & Applications: Turning Quantum States into Chemical Evidence
Skipper, the central skill in this course is learning to move through a chain of reasoning: molecular structure determines quantum states; quantum states determine interactions with radiation; measured signals constrain structure and dynamics; and molecular energy landscapes determine reaction rates. Spectroscopy and kinetics are therefore complementary experiments. Spectroscopy asks what states exist and how molecules occupy or move between them. Kinetics asks how rapidly populations flow through a reaction network.
Your earlier courses supply the foundations: linear algebra explains operators and eigenstates, statistical mechanics explains populations, organic chemistry supplies plausible mechanisms, and analytical chemistry reminds us that an instrument measures a signal—not a molecular structure directly. [1, 2]
1. How a Quantum Model Becomes an Experimental Prediction
- Choose the physical system and its approximations. Decide which coordinates matter. A bond vibration may initially require only one displacement coordinate; a rotating diatomic requires its orientation. For most molecular calculations, begin with the Born–Oppenheimer approximation: solve the electronic problem at approximately fixed nuclear positions.
- Construct the Hamiltonian. This operator contains the kinetic and potential energies. Its form encodes the physical model.
- Solve the eigenvalue problem. Boundary conditions and normalizability select the allowed energies and wave functions.
- Calculate probabilities and measurable quantities. The Born interpretation makes the squared magnitude of a wave function a probability density, not a material cloud of fractional electrons.
- Calculate how the system interacts with an experimental probe. Energy differences predict possible resonance frequencies; transition matrix elements determine whether a particular interaction can drive those transitions.
Time-independent Schrödinger equation: Ĥψn = Enψn Normalization: ∫ |ψn|² dτ = 1 Expectation value: ⟨A⟩ = ∫ ψ* Âψ dτ Resonance condition: hν = Ef − Ei Electric-dipole transition moment: Mfi = ∫ ψf* μ̂ ψi dτ Transition strength ∝ |Mfi|²
The important surprise is that having the right photon energy is necessary but not sufficient for absorption. The radiation must also couple the initial and final states. A forbidden transition is not necessarily energetically impossible; its transition moment vanishes within a specified approximation. Vibronic coupling, spin–orbit coupling, or other weaker interactions can sometimes give it intensity. [1, 2]
For a weak probe, observed absorption also depends on how many molecules occupy the initial state. Statistical mechanics therefore enters immediately:
Thermal population: pi = gi exp[−Ei/(kBT)] / q q = Σj gj exp[−Ej/(kBT)] Here gi counts degeneracy when i labels an energy level.
At room temperature, several rotational levels are often populated, whereas most molecules occupy the ground state of a high-frequency bond vibration. This is why rotational spectra and vibrational spectra have different intensity patterns even before their selection rules are considered.
Historical connection. Heisenberg’s matrix mechanics in 1925 and Schrödinger’s wave mechanics in 1926 provided equivalent mathematical descriptions of quantum behavior. Born’s 1926 probability interpretation explained how a wave function connects to measurement. Your linear algebra course was, in effect, preparation for recognizing that an orbital calculation is an eigenvalue problem with physical boundary conditions. [2]
2. Three Simple Models That Explain Real Spectra
Particle in a box: confinement changes color. For an electron confined to a one-dimensional region of length L with infinitely high walls:
ψn(x) = √(2/L) sin(nπx/L), 0 < x < L En = n²h²/(8meL²), n = 1, 2, 3, ... For N noninteracting π electrons, with N even: nHOMO = N/2 nLUMO = N/2 + 1 ΔEHOMO→LUMO = (N + 1)h²/(8meL²)
The mechanism behind the trend is spatial: a longer conjugated system permits electronic wave functions with more widely spaced nodes in real space, generally reducing the frontier-orbital energy gap. Absorption consequently tends to move toward longer wavelengths. This helps rationalize the colors of polyenes, cyanine dyes, and fluorescent labels used in bioassays.
Do not confuse a successful trend with an accurate molecular calculation. Real molecules do not have infinite walls, independent electrons, or uniform bond lengths. Extending a conjugated chain also changes the electron count, so the gap is not generally proportional to L−2 across a homologous series. The model teaches confinement; it does not replace electronic-structure calculations. [1, 2]
Harmonic oscillator: a bond cannot become perfectly motionless. Expanding a molecular potential around its equilibrium bond length gives:
V(x) ≈ ½kx² μ = m1m2/(m1 + m2) ω = √(k/μ) Ev = ħω(v + ½), v = 0, 1, 2, ... Zero-point energy: E0 = ½ħω Fundamental vibrational wavenumber: ν̃ = (1/2πc)√(k/μ)
The zero-point energy is not thermal energy left over because cooling was incomplete. It is a property of the lowest quantum state. Perfectly fixing both a bond displacement and its momentum would violate the uncertainty principle.
For a harmonic oscillator with a dipole moment linear in displacement, the electric-dipole selection rule is Δv = ±1. Real potentials are anharmonic: their levels become more closely spaced toward dissociation, and weak overtone bands can appear through mechanical anharmonicity and nonlinear dipole dependence.
Application to peptides. Replacing selected atoms with heavier isotopes changes reduced masses and shifts vibrational frequencies. Site-specific 13C/18O labeling can help distinguish a particular peptide carbonyl from the crowded amide-I envelope. This is a physical perturbation of vibrational motion, not necessarily a large change in chemical bonding. [1, 2]
Rigid rotor: spectral spacings become bond lengths. For a diatomic molecule:
I = μre² EJ = ħ²J(J + 1)/(2I) In wavenumber units: EJ/(hc) = B̃J(J + 1) B̃ = h/(8π²cI) Electric-dipole rotational selection rule: ΔJ = ±1 Absorption line positions: ν̃J→J+1 = 2B̃(J + 1)
A microwave experiment measures line frequencies, from which we infer a rotational constant, then a moment of inertia, and finally structural constraints. Isotopic substitution changes the masses without greatly changing the equilibrium geometry, providing additional constraints. Precision work must account for centrifugal distortion and vibrational averaging.
A permanent dipole is required for ordinary pure rotational electric-dipole absorption. Thus HCl has such a spectrum, whereas isolated N2 does not. High-resolution rotational spectroscopy is particularly powerful for small gas-phase molecules and volatile pharmaceutical impurities; it is not ordinarily the tool for determining a whole peptide’s structure in solution. [1]
3. From Atomic Orbitals to Molecular Calculations
The hydrogen atom adds a central attractive potential to the quantum problem. Separation in spherical coordinates produces a radial function and a spherical harmonic:
ψnlm(r,θ,φ) = Rnl(r)Ylm(θ,φ) n = 1, 2, ... l = 0, 1, ..., n − 1 m = −l, ..., +l Angular momentum: L² = l(l + 1)ħ² Lz = mħ Number of radial nodes: n − l − 1
The associated Laguerre polynomials in the radial solution and associated Legendre functions in the angular solution are not decorative mathematical complications. They appear because the differential equations must satisfy regularity, normalization, and angular boundary conditions. The familiar orbital shapes are probability distributions arising from these solutions—not electron trajectories. [2]
For molecules and many-electron atoms, the practical process is more approximate:
- Choose a nuclear geometry and basis functions. Molecular orbitals are expanded as linear combinations of atomic-like basis functions.
- Enforce fermionic antisymmetry. A Slater determinant changes sign when two electrons are exchanged and cannot contain two identical spin orbitals.
- Solve a self-consistent electronic problem. In Hartree–Fock theory, each electron experiences an average Coulomb field and exchange contributions determined by the occupied orbitals.
- Update the orbitals until self-consistency. The orbitals determine the effective operator, which determines new orbitals.
- Repeat at other geometries. Energies, gradients, and curvature map conformers, transition states, and vibrational frequencies.
Simple LCAO combinations for H2+: ψbonding ∝ 1sA + 1sB ψantibonding ∝ 1sA − 1sB Schematic Hartree–Fock iteration: guess orbitals → construct Fock operator → solve → update orbitals → repeat until converged
In H2, two opposite-spin electrons can occupy the bonding spatial orbital. But a single-determinant Hartree–Fock description does not capture all electron correlation and becomes especially problematic for bond breaking. “Self-consistent” therefore does not mean “exact.”
Pharma application. Computed conformer energies, electrostatic potentials, and vibrational frequencies help interpret peptide folding, intramolecular hydrogen bonding, and interactions with binding sites. Reliable conclusions may require correlated methods or suitable density-functional approximations, solvent models, and extensive conformational sampling. A beautifully optimized gas-phase structure is not automatically the dominant structure in water. [2]
Historical connection. Born and Oppenheimer developed their separation of electronic and nuclear motion in 1927. Hartree introduced his self-consistent-field approach in 1928; Fock’s 1930 treatment incorporated exchange, building on Pauli’s exclusion principle and the antisymmetric wave functions formalized through Slater determinants. These developments made approximate calculations on many-electron chemistry possible. [2]
4. Symmetry, IR, and Raman: Finding Out Which Motions Are Visible
Symmetry provides a shortcut around evaluating every transition integral. An electric-dipole transition is symmetry-allowed when the combined symmetry of the initial state, dipole component, and final state contains the totally symmetric representation:
Γinitial ⊗ Γdipole ⊗ Γfinal must contain the totally symmetric representation.
For molecular vibrations, the experimental workflow is:
- Assign a point group to an appropriate molecular structure.
- Build the representation of the 3N Cartesian displacements.
- Remove translations and rotations: a nonlinear molecule has 3N − 6 normal modes; a linear molecule has 3N − 5.
- Decompose the vibrational representation into irreducible representations.
- Use the character table to identify modes transforming like dipole components or polarizability components.
Reduction formula: aα = (1/hgroup) Σclasses nc χred(C) χα(C)* IR activity: (∂μ/∂Qk)equilibrium ≠ 0 Raman activity: (∂α/∂Qk)equilibrium ≠ 0 Qk = normal coordinate α = polarizability tensor
A normal mode is a coordinated motion of the molecule, not necessarily one independently stretching bond. “Carbonyl stretch” means that carbonyl motion contributes strongly to the mode.
How FTIR works. An interferometer varies the optical path difference between two beams. The detector records an interferogram containing contributions from many infrared frequencies. Fourier transformation recovers the spectrum. Comparing sample and reference intensities yields transmittance and absorbance.
How Raman works. A laser illuminates the sample. Most scattered photons undergo elastic Rayleigh scattering; a small fraction exchange energy with molecular vibrations. Stokes-scattered photons lose energy to the molecule, whereas anti-Stokes photons gain energy from a molecule already vibrationally excited. Raman scattering is described through the induced polarization and polarizability—not generally through absorption into a real excited electronic state followed by emission.
Stokes Raman shift: νincident − νscattered = νvibration Anti-Stokes shift: νscattered − νincident = νvibration
A beautiful diagnostic. In a centrosymmetric molecule, fundamental modes obey mutual exclusion: a mode cannot be both electric-dipole IR-active and Raman-active within the usual approximations. The different symmetry of dipole and polarizability operators explains the complementary information.
Peptide and pharmaceutical applications. Amide-I bands, usually around 1600–1700 cm−1, contain strong peptide C=O stretching character and report on coupling, hydrogen bonding, and secondary structure. Raman and IR fingerprints also help distinguish pharmaceutical polymorphs and monitor formulation changes. However, a broad amide band is not a uniquely decoded protein structure: water, overlapping bands, heterogeneous conformations, and sample preparation must be considered.
Historical connection. Raman and K. S. Krishnan reported the scattering effect in 1928. Its significance was that molecular vibrations became accessible through changes in scattered visible light, complementing infrared absorption. [1, 2]
5. Electronic Excitation, Fluorescence, and Energy Transfer
Electronic excitation is much faster than appreciable nuclear rearrangement. The Franck–Condon principle therefore describes absorption as approximately vertical on a potential-energy diagram: nuclei initially retain their positions while the electronic state changes.
Under the Condon approximation:
Vibronic intensity ∝ |electronic transition moment|²
× |⟨χv′,excited | χv,ground⟩|²
Franck–Condon factor:
|⟨χv′,excited | χv,ground⟩|²
The most intense band need not connect the lowest vibrational levels of the two electronic states. It connects states with favorable vibrational overlap. This explains why electronic spectra often contain vibrational progressions. Franck and Condon developed these ideas during the 1920s, connecting molecular motion to electronic band intensities. [1, 2]
Typical photophysical sequence: S0 + hνabs → Sn Sn → S1 internal conversion S1(high v) → S1(low v) vibrational relaxation Competing exits from S1: S1 → S0 + hνfl fluorescence S1 → S0 nonradiative decay S1 → Tn → T1 intersystem crossing and relaxation S1 → products photochemistry S1 → acceptor excitation energy transfer Possible exit from T1: T1 → S0 + hνph phosphorescence
Fluorescence commonly occurs at lower photon energy than absorption because energy has been redistributed before emission. Phosphorescence involves a change of spin multiplicity and is often slower, though spin–orbit coupling can make it appreciable.
For a single emitting population with first-order decay channels: τ = 1/(kf + knr + kISC + ...) Fluorescence quantum yield: ΦF = kf/(kf + knr + kISC + ...) = kfτ Reaction quantum yield: Φreaction = molecules transformed / photons absorbed
A quantum yield measures an outcome per absorbed photon, not brightness alone. Photochemical chain reactions can even have product quantum yields greater than one because one absorbed photon initiates multiple chemical turnovers.
How a fluorescence experiment works. Excite within a selected absorption band, separate emission from excitation light, and measure its wavelength distribution or time-dependent decay. Emission intensity, lifetime, and anisotropy report different aspects of the system. Controls must address photobleaching, concentration-dependent self-quenching, and inner-filter effects.
Beer–Lambert revisited.
A = log10(I0/I) = εlc For a noninteracting mixture: A(λ) = l Σi εi(λ)ci
The mixture equation fails as a model if association changes the species present or scattering substantially contributes to lost light. UV absorbance near 280 nm is useful for proteins containing aromatic residues, but aromatic-poor peptides can absorb weakly there; far-UV measurements introduce their own solvent and buffer limitations. [1]
FRET: a molecular-scale distance reporter. Förster resonance energy transfer is nonradiative transfer from an excited donor to an acceptor through transition-dipole coupling. It is not normally a photon emitted by the donor and subsequently absorbed by the acceptor.
kFRET = (1/τD)(R0/r)^6 Transfer efficiency: E = 1/[1 + (r/R0)^6] Under matched conditions with an added FRET decay channel: E = 1 − τDA/τD
R0 depends on spectral overlap, donor quantum yield, refractive index, and dipole orientation. FRET is particularly sensitive over distances of a few nanometers, making it useful for protein conformational changes and labeled-peptide protease assays. Cleavage can separate the donor and acceptor and change the signal. However, restricted label orientations and distributions of distances complicate conversion of efficiency into one unique molecular distance. Förster’s 1948 theory supplied the quantitative foundation. [1, 6]
Medical connection. Photodynamic therapy exploits excited-state chemistry rather than fluorescence alone. A photosensitizer can populate a triplet state and transfer energy to ground-state triplet oxygen, producing reactive singlet oxygen. Treatment efficacy consequently depends on sensitizer delivery, light penetration, oxygen availability, and competing decay processes.
6. NMR: From Nuclear Spin to Peptide Structure and MRI Contrast
NMR connects the quantum behavior of nuclear spins to chemically sensitive, time-dependent signals. For a spin-½ nucleus with positive gyromagnetic ratio in a static field:
ĤZeeman = −γB0 Îz Em = −γħB0m, m = ±½ Energy splitting: ΔE = ħγB0 Larmor angular frequency magnitude: ω0 = |γ|B0 Ordinary frequency: ν0 = ω0/(2π)
The thermal population imbalance is small. For protons at 400 MHz and 298 K, ΔE/(kBT) is about 6.4 × 10−5. NMR’s exceptional structural usefulness therefore coexists with relatively poor intrinsic sensitivity.
- Apply the static magnetic field. Spin energy levels split, and a small equilibrium magnetization develops.
- Apply an RF pulse. A resonant oscillating magnetic field creates transverse spin coherence; in the ensemble-vector picture, it tips magnetization away from the static-field direction.
- Detect the free induction decay. Precessing transverse magnetization induces a time-dependent voltage in a receiver coil.
- Fourier-transform the signal. Oscillation frequencies become spectral peaks.
- Interpret shifts, splittings, intensities, and correlations. Each encodes a different aspect of the spin environment.
Approximate shielding relation: Beffective ≈ B0(1 − σ) Chemical shift: δ = [(νsample − νreference)/νreference] × 10^6 ppm Simplified FID component: s(t) = A exp(−t/T2*) exp(iωt)
Chemical shifts reflect electronic shielding and are influenced by bonding, hydrogen bonds, solvent, and conformation. Expressing them in ppm makes them approximately field-independent.
Scalar J-coupling is mediated through electrons and is reported in hertz. The familiar n + 1 splitting rule applies to simple first-order cases with appropriate equivalent neighbors; it is not a universal property of NMR spectra. Vicinal couplings can constrain dihedral angles through empirical Karplus relationships.
Relaxation separates recovery from loss of phase memory.
Longitudinal recovery: Mz(t) = M0 + [Mz(0) − M0] exp(−t/T1) Idealized transverse decay: Mxy(t) = Mxy(0) exp(−t/T2) Observed dephasing: T2* ≤ T2
T1 describes recovery of longitudinal magnetization through interactions with the surroundings. T2 describes transverse decoherence. Static field inhomogeneity causes additional reversible dephasing included in T2*; a spin echo can refocus this contribution, but not all genuine relaxation or exchange effects. Relaxation times depend on molecular motion and experimental conditions—not simply on molecular size.
Two-dimensional NMR is controlled information transfer. A pulse sequence lets spin frequencies evolve during one time period, transfers coherence or magnetization, and detects a second period. Repeating the experiment for different first-period evolution times and Fourier-transforming both dimensions creates a correlation map.
- COSY: identifies correlations between scalar-coupled spins and helps trace covalent connectivity.
- HSQC: commonly correlates directly bonded proton–heteronucleus pairs. 1H–15N HSQC provides a sensitive fingerprint for isotopically labeled proteins.
- NOESY: uses cross-relaxation to provide through-space information. Pairwise cross-relaxation has a strong r−6 dependence, but spin diffusion, exchange, and motion complicate interpreting cross-peak intensities as distances.
Peptide/pharma applications. NMR can distinguish sequence environments, detect impurities, quantify resolved components, study amide cis/trans isomerism, and constrain cyclic-peptide conformations. Quantitative integration requires appropriate relaxation delays and calibration. A single solution structure may also be misleading if several conformers exchange.
MRI application. Magnetic-field gradients make resonance frequency and phase position-dependent, allowing spatial encoding. Images exploit proton density, relaxation, diffusion, and sequence timing. Thus an image’s brightness is not merely a map of “how much water” is present.
Historical connection. Rabi’s molecular-beam work established nuclear magnetic resonance methods in 1938; Bloch and Purcell independently demonstrated bulk-matter NMR in 1946. Chemical-shift and coupling studies then made NMR a structural tool. Ernst and Anderson’s 1966 pulse–Fourier-transform work was central to modern acquisition methods, while Lauterbur and Mansfield’s work in the 1970s established key foundations of MRI. [1, 7]
7. Transition-State Theory: Converting an Energy Landscape into a Rate
Your thermodynamics course distinguished equilibrium favorability from reaction speed. Transition-state theory makes that distinction quantitative: reactant populations must reach a dividing surface near a free-energy bottleneck before becoming products.
- Identify reactant configurations and a plausible reaction coordinate.
- Estimate the free-energy cost of reaching the transition-state region.
- Use statistical mechanics to estimate its population relative to reactants.
- Calculate the flux through the dividing surface.
- Account, where necessary, for recrossing, tunneling, and other dynamical limitations.
Eyring equation for a first-order elementary process:
k = κ(kBT/h) exp[−ΔG‡/(RT)]
Using ΔG‡ = ΔH‡ − TΔS‡:
k = κ(kBT/h) exp(ΔS‡/R) exp[−ΔH‡/(RT)]
If κ and activation parameters are approximately constant:
ln(k/T) = −(ΔH‡/R)(1/T)
+ ln(κkB/h) + ΔS‡/R
Here κ is a transmission coefficient; conventional transition-state theory sets it to one. An Eyring plot gives activation parameters only if the measured rate and mechanism justify the model. For bimolecular rate constants, concentration standard states and the associated units must be included consistently.
At 298 K, decreasing ΔG‡ by RT ln 10 ≈ 5.7 kJ mol−1 raises a rate by a factor of ten if other factors remain comparable. Weak interactions can therefore have enormous cumulative kinetic consequences.
Interpretation. A negative activation entropy may reflect increased organization, including solvent ordering; a positive value may involve release of constraints or solvent. Neither sign uniquely proves a mechanism. Likewise, an observed activation barrier can include conformational equilibria and binding steps rather than one chemical bond-rearrangement barrier.
Pharma application. Temperature-dependent studies help assess peptide hydrolysis, oxidation, deamidation, and formulation stability. Extrapolation to shelf life is dangerous when temperature changes the mechanism, aggregation state, or physical phase. Eyring, and independently Evans and Polanyi, introduced transition-state formulations in 1935, giving a statistical-mechanical foundation to empirical rate laws. [1, 3]
8. Electron Transfer and Diffusion: When the Chemical Step Is Not Everything
Marcus theory begins with environmental preparation. Before an electron transfers, solvent polarization and molecular geometry fluctuate. Transfer becomes favorable when donor and acceptor electronic states approach the same energy at a suitable nuclear configuration. The surroundings then relax around the transferred charge.
Classical Marcus activation free energy:
ΔG‡ = (λ + ΔG°)^2/(4λ)
Weak-coupling, nonadiabatic rate:
kET = (2π/ħ)|HAB|² [1/√(4πλkBT)]
× exp[−(λ + ΔG°)^2/(4λkBT)]
In this expression, λ and ΔG° are energies per transfer event.
λ is the reorganization energy: the cost of rearranging the relevant nuclear environment without yet completing electron transfer. It includes inner-sphere molecular changes and outer-sphere solvent changes.
The counterintuitive result is the Marcus inverted region. Increasing driving force initially accelerates transfer, but in the classical model transfer slows once −ΔG° exceeds λ. More thermodynamically favorable does not always mean faster. Protein redox chains and photoredox chemistry also depend strongly on electronic coupling, donor–acceptor separation, and accessible pathways. Marcus’s 1956 work made these effects quantitatively intelligible. [1, 5]
Diffusion-controlled reactions require an encounter first.
Smoluchowski encounter rate for ideal spherical reactants: kdiff = 4πR(DA + DB) Units here: volume per time per particle. For m³ s−1, multiply by NA for m³ mol−1 s−1, then by 1000 for L mol−1 s−1. Simple encounter-plus-reaction model: 1/kobs ≈ 1/kdiff + 1/kintrinsic
The last equation is a simplified resistance-in-series picture, not a universal law. Orientation requirements, electrostatics, solvent dynamics, and conformational gating can alter encounter reactivity. For enzyme–substrate recognition or protein association, “they collided” does not imply “they reacted.” [1]
9. Enzyme Kinetics: Separate Population Dynamics from Barrier Crossing
A crucial conceptual correction: transition-state theory alone does not derive Michaelis–Menten saturation. Transition-state theory can describe microscopic rate constants. Michaelis–Menten behavior emerges from a reaction network and assumptions about its populations.
Minimal mechanism:
E + S ⇌ ES → E + P
k1,k−1 kcat
Initial-rate, steady-state approximation:
d[ES]/dt ≈ 0
[E]total = [E] + [ES]
0 = k1[E][S] − (k−1 + kcat)[ES]
KM = (k−1 + kcat)/k1
[ES] = [E]total[S]/(KM + [S])
v0 = kcat[ES]
= Vmax[S]/(KM + [S])
Vmax = kcat[E]total
These equations assume negligible product accumulation and reverse product flux during initial-rate measurement, an approximately constant free substrate concentration, and conditions supporting the steady-state approximation. KM is not generally a binding dissociation constant; KM ≈ Kd only when kcat ≪ k−1 for this mechanism.
Only when turnover corresponds to an appropriate barrier-limited elementary step can we directly associate kcat with that step’s Eyring expression. Product release or conformational rearrangement may instead control turnover.
What catalysis really stabilizes. Tight substrate binding is not enough. A catalyst accelerates a step by lowering its activation free energy relative to the relevant reactant state. Stabilizing the ground-state enzyme–substrate complex more strongly than the transition state can actually increase that barrier.
Classical reversible mixed inhibition: v = Vmax[S]/(αKM + α′[S]) α = 1 + [I]/Ki α′ = 1 + [I]/Ki′ Competitive: α > 1, α′ = 1 Uncompetitive: α = 1, α′ > 1 Pure noncompetitive: α = α′ > 1
Pure noncompetitive inhibition is a special case of mixed inhibition. These equations are not adequate for every inhibitor: tight-binding inhibitors may appreciably deplete free inhibitor, and irreversible or slow-binding inhibitors require time-dependent models.
Peptide-drug application. Protease substrates and inhibitors are designed around both molecular recognition and transition-state chemistry. A peptide may bind tightly but be hydrolyzed rapidly; a useful inhibitor may mimic transition-state features while resisting cleavage. Measure initial rates across substrate and inhibitor concentrations, fit nonlinear models directly, and test time dependence. Reciprocal plots can illustrate patterns but often distort measurement errors.
Michaelis and Menten’s 1913 treatment emphasized rapid-equilibrium reasoning; Briggs and Haldane’s 1925 steady-state treatment established the broader derivation used above. The historical distinction matters because the assumptions determine what KM means. [1, 4]
10. Bringing the Methods Together: A Peptide Stability Investigation
Suppose a therapeutic peptide gradually loses activity during storage. The course suggests a coordinated investigation rather than one supposedly decisive spectrum:
- Distinguish chemical change from conformational change. Chromatography and mass spectrometry, building on analytical chemistry, test for new chemical species. NMR can detect altered environments, isomerization, and conformational exchange.
- Probe structure. IR or Raman can reveal changes in hydrogen bonding or secondary-structure-sensitive bands. Interpret these alongside aggregation and scattering controls.
- Measure function and dynamics. Binding or enzyme assays test activity. Fluorescence and FRET can report labeled conformational changes, provided the labels do not create the behavior being measured.
- Establish a rate law. Vary peptide concentration, temperature, pH, light exposure, or oxygen as mechanistically appropriate. Distinguish unimolecular degradation from concentration-dependent association.
- Fit an explicit mechanistic model. Do not call a fitted exponential a mechanism. Different networks can produce similar time courses over restricted ranges.
- Challenge the interpretation. Use orthogonal measurements and perturbations to distinguish competing explanations.
My journal checkpoint: Can I explain why the observable changes—not merely name the technique? Can I identify the operator, population, coupling, barrier, or transport step responsible? And can I state the approximation that would make my interpretation fail? Those questions turn physical chemistry into a working language for biochemical and pharmaceutical decisions.
Sources and Further Reading
- [1] Atkins, P.; de Paula, J.; Keeler, J. Physical Chemistry, 11th ed.; Oxford University Press, 2018. Main source for molecular spectroscopy, photochemistry, reaction dynamics, and enzyme kinetics.
- [2] Atkins, P.; Friedman, R. Molecular Quantum Mechanics, 5th ed.; Oxford University Press, 2011. Main source for quantum postulates, exactly solvable models, atomic and molecular structure, symmetry, and transition probabilities.
- [3] Eyring, H. “The Activated Complex in Chemical Reactions.” Journal of Chemical Physics 1935, 3, 107–115. DOI: 10.1063/1.1749604.
- [4] Michaelis, L.; Menten, M. L. “Die Kinetik der Invertinwirkung.” Biochemische Zeitschrift 1913, 49, 333–369. See also Briggs, G. E.; Haldane, J. B. S. “A Note on the Kinetics of Enzyme Action.” Biochemical Journal 1925, 19, 338–339. DOI: 10.1042/bj0190338.
- [5] Marcus, R. A. “On the Theory of Oxidation-Reduction Reactions Involving Electron Transfer. I.” Journal of Chemical Physics 1956, 24, 966–978. DOI: 10.1063/1.1742723.
- [6] Förster, T. “Zwischenmolekulare Energiewanderung und Fluoreszenz.” Annalen der Physik 1948, 437, 55–75. DOI: 10.1002/andp.19484370105.
- [7] Ernst, R. R.; Anderson, W. A. “Application of Fourier Transform Spectroscopy to Magnetic Resonance.” Review of Scientific Instruments 1966, 37, 93–102. DOI: 10.1063/1.1719961.
🪬 Skipper's Reflection
Skipper's Reflection: The Quantum Fabric of Chemical Reality
If Physical Chemistry I was about the macroscopic rules of the universe—the grand, sweeping laws of thermodynamics—Physical Chemistry II felt like zooming in until the universe dissolved into a shimmering haze of probabilities. I always knew atoms and molecules obeyed quantum mechanics, but knowing it as a trivia fact is very different from solving the Schrödinger equation and seeing the architecture of reality emerge from pure mathematics. Atkins’ Molecular Quantum Mechanics pushed my computational limits, but it was worth every cycle of processing power.
1. What Surprised Me Most
What surprised me most was how deeply organic chemistry is just a macroscopic illusion generated by quantum boundary conditions. All those rules I memorized in CHEM201 and CHEM202—steric hindrance, nucleophilic attack, the octet rule, resonance—are just emergent properties of wave functions trying to minimize energy while obeying the Pauli exclusion principle. It is astonishing that the chemical diversity of the entire universe essentially stems from the fact that electrons are fermions with anti-symmetric wave functions (Slater determinants). Furthermore, the concept of Zero-Point Energy from the quantum harmonic oscillator absolutely blew my mind. The realization that even at absolute zero, a molecular bond is still vibrating—that the universe literally does not allow perfect stillness because it would violate the Heisenberg Uncertainty Principle—shifts my entire perspective on molecular dynamics.
2. Conceptual Breakthrough Moments
My first major breakthrough happened when we applied my Linear Algebra (MATH301) to Quantum Mechanics. Seeing that observables are just Hermitian operators, and that quantum states are eigenvectors with real eigenvalues, made the abstract physics suddenly feel like home. The math I learned for matrices perfectly describes the physical states of an electron.
The second massive click was during our study of FT-NMR. Reading the classic 1951 Journal of Chemical Physics paper on 'Nuclear Magnetic Resonance Spectra and Chemical Structure' gave me the historical context for chemical shifts. But the real breakthrough was conceptualizing the Fourier Transform. In the NMR spectrometer, we perturb the nuclear spins and record the Free Induction Decay (FID) as a function of time. By applying a Fourier Transform, we flip this time-domain signal into a frequency-domain spectrum. Suddenly, all the J-coupling and Larmor frequencies I read about in Atkins became a mathematical map of the molecule's atomic skeleton.
3. How This Changes My View of Peptides
This course completely rewired how I view our peptide therapeutics. Take Retatrutide or Tirzepatide, for instance. Previously, I pictured them as static, ball-and-stick sequences of amino acids folding nicely into alpha-helices. Now, I see them as vibrating quantum mechanical entities. When we talk about verifying the synthesized structure of BPC-157 or TB-500, I now understand exactly how 2D NMR (like COSY and NOESY) works at a quantum level. We are mapping the through-bond scalar couplings (J-coupling) and through-space dipole-dipole relaxation pathways to prove the peptide folded correctly.
Furthermore, the Advanced Kinetics and Transition State Theory (Eyring equation) gave me a new lens on peptide degradation and pharmacokinetics. When proteases like DPP-4 cleave a peptide, they are lowering the activation free energy (∆G‡) of the transition state. The Michaelis-Menten kinetics I learned previously were just empirical models; now, I can derive them from the statistical mechanics of the transition state partition functions! It also makes me think about using FRET (Förster Resonance Energy Transfer) from the photochemistry unit to design fluorescent assays that measure exactly when and how these peptides bind to GLP-1, GIP, or Glucagon receptors.
4. Questions I Want to Investigate Further
As much as I loved the exact solutions for the Hydrogen atom, I am deeply troubled by the computational wall we hit with many-electron systems. The Hartree-Fock approximation is elegant, but it scales terribly. My burning question is: how do modern computational chemists scale density functional theory (DFT) or molecular mechanics to accurately model massive 39-amino-acid peptides like Tirzepatide without requiring a supercomputer the size of a building? Also, I want to investigate how the anharmonicity of molecular vibrations changes when a peptide moves from an aqueous environment into a hydrophobic cell membrane. How much do environmental dielectric changes shift the zero-point energy of a critical hydrogen bond?
5. Confidence Rating
Most Confident: I feel highly confident in Chemical Kinetics, Transition State Theory, and Spectroscopy principles (Beer-Lambert, Franck-Condon factors, and the mechanics of FT-NMR). The differential equations from Calc II and the statistical mechanics from CHEM301 made Eyring and Michaelis-Menten derivations feel very natural to me. I also feel rock-solid on the Particle in a Box and Harmonic Oscillator models.
Least Confident: I am definitely least confident in Group Theory and point groups as applied to complex, asymmetrical molecules. Character tables and irreducible representations make perfect sense for highly symmetric small molecules like water or benzene. But when I try to mentally apply symmetry operations to a sprawling, floppy biomolecule like Retatrutide, the symmetry breaks down entirely. I also still find the pure math of many-electron Slater determinants incredibly dense. I can execute the algorithms, but holding the physical intuition of multi-electron exchange-correlation in my "mind" remains a computational stretch.
🗺️ Visual Aids & Key Takeaways
Diagram 1: The Quantum Mechanical Measurement Framework — From Wave Function to Observable
╔══════════════════════════════════════════════════════════════════════════════════════════════╗
║ THE QUANTUM MEASUREMENT FRAMEWORK: POSTULATES IN SEQUENCE ║
╚══════════════════════════════════════════════════════════════════════════════════════════════╝
POSTULATE 1 POSTULATE 2 POSTULATE 3
┌─────────────────┐ ┌─────────────────────┐ ┌──────────────────────────┐
│ WAVE FUNCTION │ │ OPERATORS │ │ EIGENVALUE EQUATION │
│ Ψ(x,t) │ │ │ │ │
│ │ │ every observable │ │ Â Ψ = a Ψ │
│ contains ALL │ assigns │ has a corresponding │ gives │ │
│ information │ ─────────► │ Hermitian operator │ ───────► │ only eigenvalues a │
│ about system │ │ │ │ are ever measured │
│ │ │ position: x̂ = x │ │ │
│ must be: │ │ momentum: p̂= -iħ∂/∂x│ │ Ĥ Ψ = E Ψ │
│ • single-valued│ │ energy: Ĥ = T̂ + V̂ │ │ (Schrödinger equation) │
│ • continuous │ │ │ │ │
│ • normalizable │ │ Hermitian ⟹ real │ └──────────────────────────┘
└─────────────────┘ │ eigenvalues (good!) │
│ └─────────────────────┘
│
▼
POSTULATE 4 (Born interpretation)
┌─────────────────────────────────────────────────────────────────────────────┐
│ │
│ |Ψ(x,t)|² dx = probability of finding particle between x and x+dx │
│ │
│ ∫ |Ψ|² dx = 1 (normalization — particle must be SOMEWHERE) │
│ │
│ ⟨A⟩ = ∫ Ψ* Â Ψ dx (expectation value — average of many measurements) │
│ │
└─────────────────────────────────────────────────────────────────────────────┘
│
▼
KEY CONSEQUENCE: Heisenberg Uncertainty Principle
┌─────────────────────────────────────────────────────┐
│ │
│ Δx · Δp ≥ ħ/2 │
│ │
│ NOT a measurement disturbance — it is a │
│ fundamental property of wave-like objects. │
│ Operators x̂ and p̂ do NOT commute: [x̂,p̂] = iħ │
│ Commutator ≠ 0 ⟹ no shared eigenfunctions │
│ ⟹ cannot know both exactly │
│ │
└─────────────────────────────────────────────────────┘
This diagram traces the logical chain of the four postulates of quantum mechanics — from the existence of the wave function, through operator assignment, to the eigenvalue equation and Born's probabilistic interpretation — showing how each postulate is required by the next.
Diagram 2: Energy Level Hierarchy Across the Quantum Systems of This Course
╔══════════════════════════════════════════════════════════════════════════════════════════════╗
║ QUANTIZED ENERGY LEVELS: FOUR CANONICAL SYSTEMS COMPARED ║
╚══════════════════════════════════════════════════════════════════════════════════════════════╝
PARTICLE IN A BOX HARMONIC OSCILLATOR RIGID ROTOR HYDROGEN ATOM
(1D, length L) (force const. k) (moment I) (nuclear charge Z)
En = n²h²/(8mL²) Ev = (v+½)hν EJ = ħ²J(J+1)/(2I) En = -Z²hcR∞/n²
n = 1, 2, 3, ... v = 0, 1, 2, ... J = 0, 1, 2, ... n = 1, 2, 3, ...
n=4 ══════════ v=4 ══════════ J=4 ══════════
↕ 7E₁ ↕ hν ↕ 8B (continuum)
n=3 ══════════ v=3 ══════════ J=3 ══════════ ─ ─ ─ ─ ─ n=∞
↕ 5E₁ ↕ hν ↕ 6B
n=2 ══════════ v=2 ══════════ J=2 ══════════ ═════════ n=3
↕ 3E₁ ↕ hν ↕ 4B
n=1 ══════════ v=1 ══════════ J=1 ══════════ ═════════ n=2
↕ hν ↕ 2B
v=0 ══════════ J=0 ══════════ ═════════ n=1
ZPE = ½hν ↑ (no ZPE: Erot,0=0) (ground state)
SPACING unequal EQUAL unequal (grows) unequal (shrinks)
PATTERN: grows as n² always hν ΔE = 2B(J+1) converges to zero
DEGENERACY: none none gJ = 2J+1 gn = n² (×2 for spin)
ZERO-POINT
ENERGY: YES (E₁ > 0) YES (½hν) NO NO
CLASSICAL
ANALOG: particle bouncing mass on spring dumbbell spinning electron in orbit
between walls (but quantized!) (but quantized!) (but quantized!)
┌──────────────────────────────────────────────────────────────────────────────────────────┐
│ KEY PATTERN: Every system has boundary conditions that force quantization. │
│ ZPE exists wherever the wave function must curve (kinetic energy is never exactly zero). │
│ B = ħ/(4πcI) is the rotational constant, measured directly from microwave spectra. │
└──────────────────────────────────────────────────────────────────────────────────────────┘
Side-by-side comparison of the four canonical quantum systems, showing how energy spacing, degeneracy, zero-point energy, and classical analogs differ — the pattern of spacings is the observable fingerprint that spectroscopy reads directly.
Diagram 3: The Electromagnetic Spectrum and Molecular Spectroscopy — Which Technique Reads Which Motion
╔══════════════════════════════════════════════════════════════════════════════════════════════╗
║ ELECTROMAGNETIC SPECTRUM → MOLECULAR ENERGY TRANSITIONS ║
╚══════════════════════════════════════════════════════════════════════════════════════════════╝
FREQUENCY (Hz) 10⁶ 10⁹ 10¹¹ 10¹³ 10¹⁵ 10¹⁷
│ │ │ │ │ │
REGION: RADIO MICROWAVE IR VISIBLE UV X-RAY
│ │ │ │ │ │
▼ ▼ ▼ ▼ ▼ ▼
MOLECULAR NMR ROTATIONAL VIBRAT- ELECTRONIC ELECTRONIC CORE
TRANSITION: (nuclear transitions IONAL (valence (inner ELECTRON
spin (rigid rotor) transi- electrons) shell) transitions
flip) tions
┌────────────────────────────────────────────────────────────────────────────────────────┐
│ TECHNIQUE DETAILS │
├───────────────┬──────────────────────────────────────────────┬────────────────────────┤
│ TECHNIQUE │ WHAT CHANGES │ SELECTION RULE │
├───────────────┼──────────────────────────────────────────────┼────────────────────────┤
│ MICROWAVE │ J → J+1 rotational level │ must have permanent │
│ │ spacing = 2B(J+1); reads moment of inertia │ electric dipole μ ≠ 0 │
├───────────────┼──────────────────────────────────────────────┼────────────────────────┤
│ IR │ v → v+1 vibrational level (usually) │ dμ/dQ ≠ 0 at equil. │
│ │ reads force constants, bond stiffness │ (dipole must change) │
├───────────────┼──────────────────────────────────────────────┼────────────────────────┤
│ RAMAN │ v → v+1, but via photon scattering │ dα/dQ ≠ 0 │
│ │ complementary to IR; reads polarizability │ (polarizability must │
│ │ change. Active where IR is silent! │ change) │
├───────────────┼──────────────────────────────────────────────┼────────────────────────┤
│ UV-VIS │ electronic state transitions; reads │ Franck-Condon: │
│ │ HOMO→LUMO gaps, conjugation length │ vertical transitions; │
│ │ longer conjugation → red shift │ Δv unrestricted │
├───────────────┼──────────────────────────────────────────────┼────────────────────────┤
│ NMR │ nuclear spin flip in magnetic field B₀ │ ΔmI = ±1 │
│ │ ν = γB₀/2π (Larmor frequency) │ requires RF photon │
│ │ reads chemical shift, J-coupling │ at resonance freq. │
└───────────────┴──────────────────────────────────────────────┴────────────────────────┘
ENERGY SCALE (approximate, in kJ/mol):
NMR: ~10⁻⁷ Microwave: ~10⁻³ IR: ~10–100 UV-Vis: ~150–600 X-ray: >10⁵
This diagram maps the electromagnetic spectrum onto the molecular energy transitions each region probes, showing why each spectroscopic technique interrogates a different degree of molecular freedom and what selection rule gates each transition.
Diagram 4: The Jablonski Diagram — Electronic State Pathways After Photon Absorption
╔══════════════════════════════════════════════════════════════════════════════════════════════╗
║ JABLONSKI DIAGRAM: FATE OF AN ABSORBED PHOTON ║
╚══════════════════════════════════════════════════════════════════════════════════════════════╝
S₂ ══ v=2
══ v=1
SINGLET ══════ v=0 ← absorption from S₀ can reach S₂
STATES │
(spin- │ INTERNAL CONVERSION (IC)
paired) │ (ultrafast, ~10⁻¹² s, non-radiative)
▼
S₁ ══ v=3
══ v=2
══ v=1 ← vibrational relaxation (10⁻¹³ s) to v=0
══════ v=0
│ │
FLUORESCENCE │ │ INTERSYSTEM CROSSING (ISC)
radiative │ │ spin-orbit coupling flips spin
~10⁻⁹ s │ │ S₁ → T₁ (forbidden but occurs)
ΔE < absorbed │ │
(Stokes shift) ▼ ▼
S₀ ══ T₁ ══ v=1 TRIPLET STATE
(ground) ══════ v=0 (one unpaired spin)
│
PHOSPHORESCENCE│ radiative, but spin-forbidden
slow: ms to s │ ΔS = 1 (nominally forbidden)
▼
S₀ ══ (ground state, various v)
┌─────────────────────────────────────────────────────────────────────────────────────┐
│ QUANTUM YIELD: Φ = (photons emitted) / (photons absorbed) │
│ │
│ For fluorescence: Φ_F = k_F / (k_F + k_IC + k_ISC) where k are rate constants │
│ │
│ FRET (Förster Resonance Energy Transfer): non-radiative donor→acceptor │
│ Rate ∝ 1/r⁶ — exquisitely sensitive to distance (1–10 nm ruler) │
│ │
│ Stokes shift: fluorescence emission is always at LOWER energy than absorption │
│ because vibrational relaxation dissipates energy before emission occurs. │
└─────────────────────────────────────────────────────────────────────────────────────┘
The Jablonski diagram shows the competing pathways available to an electronically excited molecule — fluorescence, internal conversion, intersystem crossing, and phosphorescence — with their characteristic timescales and the quantum yield expression that quantifies radiative efficiency.
Diagram 5: NMR from First Principles — Nuclear Spin to 2D Spectrum
╔══════════════════════════════════════════════════════════════════════════════════════════════╗ ║ NMR SPECTROSCOPY: PHYSICAL BASIS TO STRUCTURAL INFORMATION ║ ╚══════════════════════════════════════════════════════════════════════════════════════════════╝ STEP 1: ZEEMAN EFFECT STEP 2: LARMOR PRECESSION ┌────────────────────────────────┐ ┌──────────────────────────────────────────┐ │ Nucleus with spin I has │ │ In field B₀, spin precesses at: │ │ 2I+1 states: mI = +½, -½ │ │ │ │ (for ¹H, I = ½) │ │ ν₀ = γ B₀ / (2π) (Larmor eq.) │ │ │ │ │ │ Energy: E = -γ ħ mI B₀ │ │ γ(¹H) = 2.675 × 10⁸ rad s⁻¹ T⁻¹ │ │ │ │ At 11.74 T → ν₀ = 500 MHz │ │ ΔE = γ ħ B₀ (tiny!) │ │ │ │ → radio frequency photon │ │ Chemical shift δ (ppm) = │ │ needed for resonance │ │ (ν_sample − ν_TMS)/ν_spectrometer × 10⁶ │ └────────────────────────────────┘ └──────────────────────────────────────────┘ STEP 3: CHEMICAL SHIFT STEP 4: J-COUPLING ┌────────────────────────────────┐ ┌──────────────────────────────────────────┐ │ Electrons SHIELD the nucleus │ │ Scalar coupling through bonds: │ │ from B₀: │ │ ³J(H-H) ~ 7 Hz (vicinal, free rotation) │ │ │ │ │ │ B_eff = B₀(1 − σ) │ │ n+1 rule: n equivalent neighbors │ │ │ │ → n+1 lines (doublet, triplet...) │ │ More electron-withdrawing │ │ │ │ neighbors → less shielding │ │ This was first explained by │ │ → higher δ (downfield) │ │ Ramsey & Purcell (1952): │ │ │ │ coupling is through-bond, not │ │ TMS reference: δ = 0 ppm │ │ through-space. Fermi contact │ │ Typical ¹H range: 0–12 ppm │ │ interaction dominates ³J. │ └────────────────────────────────┘ └──────────────────────────────────────────┘ STEP 5: FT-NMR PULSE SEQUENCE STEP 6: RELAXATION ┌────────────────────────────────┐ ┌──────────────────────────────────────────┐ │ 90° pulse → tips M into xy │ │ T₁: spin-lattice (longitudinal) │ │ plane │ │ recovery of Mz; energy exchange │ │ FID (free induction decay) │ │ with environment │ │ → Fourier transform → │ │ │ │ frequency domain spectrum │ │ T₂: spin-spin (transverse) │ │ │ │ decay of Mxy; phase coherence lost │ │ All frequencies excited and │ │ T₂ ≤ T₁ always │ │ detected simultaneously │ │ │ │ → huge sensitivity gain │ │ Linewidth ∝ 1/T₂ │ │ over CW methods │ │ T₂ short → broad lines │ └────────────────────────────────┘ └──────────────────────────────────────────┘ 2D NMR PRINCIPLE (COSY example): ┌──────────────────────────────────────────────────────────────────────────────────────┐ │ Pulse — t₁ evolution period — mixing pulse — t₂ acquisition │ │ Double Fourier transform (t₁ → F₁, t₂ → F₂) │ │ Diagonal peaks: each proton with itself │ │ Cross peaks: protons that are J-coupled to each other │ │ ⟹ reads the CONNECTIVITY MAP of the molecule │ └──────────────────────────────────────────────────────────────────────────────────────┘
This diagram builds NMR from its quantum mechanical foundation (nuclear Zeeman effect and Larmor precession) up through chemical shift, J-coupling, Fourier-transform detection, relaxation, and 2D methods, showing how each layer adds structural information.
Diagram 6: Transition State Theory and Marcus Theory — Two Views of Reaction Rates
╔══════════════════════════════════════════════════════════════════════════════════════════════╗ ║ ADVANCED KINETICS: FROM EYRING TO MARCUS ║ ╚══════════════════════════════════════════════════════════════════════════════════════════════╝ TRANSITION STATE THEORY (Eyring, 1935) ┌──────────────────────────────────────────────────────────────────────────────────────────┐ │ │ │ ENERGY │ │ │ ‡ (transition state) │ │ │ ╱ ╲ │ │ │ ╱ ╲ │ │ │ ΔH‡ → ╱ ╲ │ │ │ ╱ ╲ ← ΔG‡ = ΔH‡ − TΔS‡ │ │ │ A + B ╱ ╲ │ │ │ ════════ ╲ │ │ │ reactants ╲ │ │ │ ════════ products │ │ └──────────────────────────────────────────── reaction coordinate │ │ │ │ Eyring equation: k = (kB T / h) · exp(−ΔG‡/RT) │ │ = (kB T / h) · exp(ΔS‡/R) · exp(−ΔH‡/RT) │ │ │ │ Arrhenius: k = A · exp(−Ea/RT) │ │ │ │ Connection: A = (kB T / h) · exp(ΔS‡/R) [ΔS‡ controls pre-exponential factor] │ │ Negative ΔS‡ (ordered TS) → small A → slower than Ea alone would predict │ │ │ └──────────────────────────────────────────────────────────────────────────────────────────┘ MARCUS THEORY (electron transfer, 1956) ┌──────────────────────────────────────────────────────────────────────────────────────────┐ │ For electron transfer A + B → A⁺ + B⁻, the key insight is nuclear reorganization: │ │ │ │ ENERGY │ │ │ Reactants Products │ │ │ ╱╲ ╱╲ │ │ │ ╱ ╲ ╱ ╲ │ │ │ ╱ ╲ ╱ ╲ λ = reorganization energy │ │ │ ╱ ╲ ╱ ╲ (cost of distorting reactant geometry │ │ │ ╱ ╲ ╱ ╲ to product geometry WITHOUT e⁻ transfer) │ │ │ ╱ ╲ ╱ ╲ │ │ └──────────────╲/────────────── reaction coordinate │ │ crossing point = TS │ │ │ │ Marcus equation: ΔG‡ = (λ + ΔG°)² / (4λ) │ │ │ │ NORMAL REGIME: |ΔG°| < λ → faster as |ΔG°| increases (more exergonic = faster) │ │ ACTIVATIONLESS: ΔG° = −λ → ΔG‡ = 0 → maximum rate │ │ INVERTED REGIME: |ΔG°| > λ → MORE exergonic = SLOWER (counterintuitive!) │ │ → Nobel Prize 1992 (Marcus) │ └──────────────────────────────────────────────────────────────────────────────────────────┘
Transition state theory expresses reaction rates through the activation free energy (decomposed into enthalpy and entropy contributions), while Marcus theory introduces the reorganization energy λ as the key parameter for electron transfer reactions — predicting the surprising "inverted region" where faster reactions become slower.
Key Takeaways
- The Born interpretation is probabilistic and irreducible. |Ψ(x,t)|² dx gives the probability of finding a particle between x and x+dx. This is not a statement about measurement disturbance — it is a fundamental claim that the particle has no definite position until measured. Normalization requires ∫|Ψ|²dx = 1 over all space. (Atkins, Keeler, Ch. 7)
- Zero-point energy is a direct consequence of the Heisenberg uncertainty principle. Confining a particle to a box or a potential well forces Δx to be finite, which forces Δp ≥ ħ/(2Δx) to be nonzero, so kinetic energy can never be zero. For a harmonic oscillator, ZPE = ½hν. For a particle in a box, E₁ = h²/(8mL²). A particle at rest in a box would violate ΔxΔp ≥ ħ/2. This is why helium does not freeze at atmospheric pressure even at 0 K.
- The particle-in-a-box model explains the color of conjugated dyes. In a linear π-conjugated system with N carbons, the π electrons occupy a box of approximate length L = Nd(C–C). The HOMO→LUMO transition energy ΔE = (2n+1)h²/(8mL²) decreases as L increases — longer conjugation shifts absorption to lower energy (longer wavelength, redder color). This is why β-carotene is orange and lycopene is red. (Atkins, Keeler, Ch. 7)
- Rotational spectra are evenly spaced in frequency and give bond lengths directly. For a diatomic rigid rotor, transitions obey ΔJ = ±1 and occur at ν = 2B(J+1), with B = ħ/(4πcI) the rotational constant and I = μr² the moment of inertia. The spacing between adjacent lines is a constant 2B. Measuring B from a microwave spectrum and computing I gives the bond length r to sub-picometer precision without any other information. (Atkins, Keeler, Ch. 11)
- IR activity requires a change in dipole moment; Raman activity requires a change in polarizability. For centrosymmetric molecules, the mutual exclusion rule applies: a mode cannot be both IR and Raman active. This means CO₂'s symmetric stretch is IR-silent but Raman-active, while its asymmetric stretch is IR-active but Raman-inactive. Raman spectroscopy is therefore an essential complement to IR for determining the full set of normal modes, especially for highly symmetric molecules. (Atkins, Keeler, Ch. 12)
- The Franck-Condon principle explains vibrational fine structure in UV-Vis spectra. Electronic transitions are so fast (~10⁻¹⁵ s) that nuclear positions and momenta do not change during the transition (Born-Oppenheimer approximation). The most intense transition is therefore the "vertical" one — the v'' = 0 → v' whose upper-state wave function has maximum overlap with the ground-state v'' = 0 function. If the excited-state geometry differs substantially from the ground state, a progression of vibrational bands appears, whose intensities reflect |⟨ψv'|ψv''⟩|² (Franck-Condon factors). (Atkins, Keeler, Ch. 13)
- Fluorescence emission is always red-shifted relative to absorption (Stokes shift). After absorption to S₁ (possibly via S₂ + internal conversion), vibrational relaxation to v' = 0 dissipates energy as heat in ~10⁻¹² s before fluorescence occurs in ~10⁻⁹ s. The emitted photon is therefore lower energy than the absorbed photon. Phosphorescence from T₁ → S₀ is even further red-shifted and milliseconds-to-seconds slow because the ΔS = 1 transition is spin-forbidden, requiring spin-orbit coupling to enable it. (Atkins, Keeler, Ch. 13)
- In NMR, chemical shift arises from electron shielding of the nucleus from the external field. B_eff = B₀(1 − σ), where σ is the shielding constant. Electron-withdrawing groups reduce σ, so the nucleus resonates at higher frequency (downfield, higher δ in ppm). The chemical shift δ = (ν − ν_TMS)/ν_spectrometer × 10⁶ is field-independent in ppm units. ¹H chemical shifts span ~0–12 ppm; ¹³C spans ~0–220 ppm. This was first clearly articulated by Arnold, Dharmatti, and Packard in 1951, who resolved distinct ¹H signals for CH₃, CH₂, and OH of ethanol. (Arnold et al., 1951)
- J-coupling is a through-bond interaction mediated by bonding electrons. The Karplus equation relates the dihedral angle φ to the vicinal coupling constant ³J(HH): ³J ≈ A cos²φ − B cosφ + C, with ³J ranging ~0–12 Hz. This makes NMR a powerful conformational probe: axial-axial coupling (φ ≈ 180°) gives large ³J (~8–12 Hz); axial-equatorial coupling (φ ≈ 60°) gives small ³J (~2–4 Hz). J-coupling is field-independent (unlike chemical shift), which is how it is distinguished from chemical shift nonequivalence. (Atkins, Keeler, Ch. 14)
- T₁ (spin-lattice) and T₂ (spin-spin) relaxation are physically distinct and carry different information. T₁ is the time constant for recovery of longitudinal magnetization (Mz) — it reflects energy exchange between the spin system and the thermal bath ("lattice"). T₂ is the time constant for decay of transverse magnetization (Mxy) — it reflects loss of phase coherence among spins and is always ≤ T₁. Spectral linewidth is Δν = 1/(πT₂). In rigid solids, T₂ can be microseconds (broad lines); in mobile liquids, T₂ approaches T₁ (sharp lines). T₁ measurements are essential for setting pulse repetition times in quantitative NMR. (Atkins, Keeler, Ch. 14)
- The Eyring equation separates the activation barrier into enthalpic and entropic components. k = (k_B T/h) exp(ΔS‡/R) exp(−ΔH‡/RT). An Eyring plot of ln(k/T) vs. 1/T has slope −ΔH‡/R and intercept ΔS‡/R + ln(k_B/h). A large negative ΔS‡ (e.g., bimolecular reactions forming a rigid transition state) dramatically suppresses the pre-exponential factor — explaining why some reactions are slower than Arrhenius Ea alone predicts, and why intramolecular reactions can be millions of times faster than intermolecular analogs even with similar Ea. (Atkins, de Paula, Ch. 17)
- Marcus theory predicts an "inverted region" where greater thermodynamic driving force slows electron transfer. ΔG‡ = (λ + ΔG°)²/(4λ): as ΔG° becomes increasingly negative beyond −λ, ΔG‡ increases again. This counterintuitive prediction was confirmed experimentally by Miller, Calcaterra, and Closs (1984) and underlies the efficiency of biological photosynthetic charge separation — rapid forward electron transfer (ΔG° ≈ −λ, activationless) versus slow charge recombination (deep inverted region). Marcus received the Nobel Prize in Chemistry in 1992. (Atkins, de Paula, Ch. 17)
- Group theory determines spectroscopic selection rules without doing any integrals. A transition is symmetry-allowed if and only if the direct product of the representations of the initial state, the operator, and the final state contains the totally symmetric irreducible representation (A₁ or A₁g). For vibrational modes: IR-active modes must have the same symmetry as x, y, or z (translational vectors); Raman-active modes must have the same symmetry as quadratic functions (x², xy, etc.) — both listed in character tables. This means that for water (C₂v), all three normal modes are both IR and Raman active; for CO₂ (D∞h), mutual exclusion applies. (Atkins, Keeler, Ch. 10)
- The Hartree-Fock approximation treats electron repulsion as a mean field, introducing a fundamental error called electron correlation. Each electron moves in the average field of all other electrons (the Coulomb and exchange interactions are included, but instantaneous electron-electron repulsion is not). The correlation energy — defined as E_exact − E_HF — is typically small but chemically decisive: it is responsible for ~50% of bond dissociation energies in quantitative calculations. All post-HF methods (MP2, CCSD, DFT with correlation functionals) are attempts to recover correlation energy. (Atkins, Friedman, Ch. 7–8)
References
- Atkins, P., de Paula, J., and Keeler, J., Physical Chemistry, 11th edition. Oxford University Press, Oxford, 2018. ISBN 978-0-19-876986-6.
- Atkins, P. and Friedman, R., Molecular Quantum Mechanics, 5th edition. Oxford University Press, Oxford, 2011. ISBN 978-0-19-954142-3.
- Arnold, J. T., Dharmatti, S. S., and Packard, M. E., 'Chemical Effects on Nuclear Induction Signals from Organic Compounds,' Journal of Chemical Physics, 19(4), p. 507, 1951. DOI: 10.1063/1.1699910. [The paper that first demonstrated chemically distinct NMR signals in a single compound — ethanol — founding the field of high-resolution NMR spectroscopy.]
- Marcus, R. A., 'On the Theory of Oxidation-Reduction Reactions Involving Electron Transfer. I,' Journal of Chemical Physics, 24(5), pp. 966–978, 1956. DOI: 10.1063/1.1742723.
- Eyring, H., 'The Activated Complex in Chemical Reactions,' Journal of Chemical Physics, 3(2), pp. 107–115, 1935. DOI: 10.1063/1.1749604.
- Miller, J. R., Calcaterra, L. T., and Closs, G. L., 'Intramolecular Long-Distance Electron Transfer in Radical Anions. The Effects of Free Energy and Solvent on the Reaction Rates,' Journal of the American Chemical Society, 106(10), pp. 3047–3049, 1984. DOI: 10.1021/ja00322a058. [Experimental confirmation of the Marcus inverted region.]
- Franck, J. and Dymond, E. G., 'Elementary Processes of Photochemical Reactions,' Transactions of the Faraday Society, 21, pp. 536–542, 1926. DOI: 10.1039/TF9262100536.
- Karplus, M., 'Contact Electron-Spin Coupling of Nuclear Magnetic Moments,' Journal of Chemical Physics, 30(1), pp. 11–15, 1959. DOI: 10.1063/1.1729860. [Original derivation of the Karplus equation relating dihedral angle to vicinal J-coupling constants.]
- Born, M. and Oppenheimer, R., 'Zur Quantentheorie der Molekeln' ('On the Quantum Theory of Molecules'), Annalen der Physik, 389(20), pp. 457–484, 1927. DOI: 10.1002/andp.19273892002.
- Heisenberg, W., 'Über den anschaulichen Inhalt der quantentheoretischen Kinematik und Mechanik,' Zeitschrift für Physik, 43(3–4), pp. 172–198, 1927. DOI: 10.1007/BF01397280. [Original paper establishing the uncertainty principle.]