📖 Core Concepts
Core Concepts: Coordination Chemistry and Bioinorganic Applications
CHEM502 study notes, prepared for Skipper.
Introduction: What Is This Subject Really About?
Skipper, you have already used metals many times without examining them closely. In CHEM401 you watched hemoglobin bind O2 cooperatively, but the iron at the center of each heme was treated as a black box. In CHEM402 you followed electrons down the respiratory chain through "iron-sulfur clusters" and "copper centers" that appeared as labels on a diagram. In CHEM202 you ran Suzuki couplings with a palladium catalyst that seemed to work by magic. In BIOL501 the action potential depended entirely on Na+ and K+, and synaptic release depended on Ca2+. This course opens those black boxes.
At its core, coordination chemistry studies what happens when a metal ion, usually one with partially filled d orbitals, is surrounded by electron-pair donors (ligands). A coordination complex is a Lewis acid–base adduct: the metal is the acid and the ligands are the bases. That is the familiar part. The new and powerful part is that the ligands change the energies of the metal's d orbitals. That change controls almost everything you can observe:
- Color. Electrons hop between split d orbitals by absorbing visible light.
- Magnetism. The splitting decides whether electrons pair up or stay unpaired.
- Reactivity. The d-electron count determines whether ligands exchange in nanoseconds or over days.
- Redox potential. The ligand environment can shift a metal's reduction potential by more than a volt.
- Geometry. Some d-electron counts strongly prefer particular shapes.
Bioinorganic chemistry adds one more idea: a protein is a ligand. It is an enormous, exquisitely designed, chiral ligand that can enforce strained geometries, exclude water, position hydrogen-bond donors, and tune a metal's properties far beyond what simple ligands achieve. Roughly a third to a half of all proteins are estimated to require a metal [Roat-Malone, 2007]. Understanding how amino-acid side chains act as ligands, and how chemists now design peptides to do the same, connects this course to peptide chemistry.
The unifying theme: the d-electron configuration of a metal, perturbed by its ligand environment, determines its structure, spectroscopy, magnetism, reactivity, and biological function. Learn to "read" a complex by asking: what metal, what oxidation state, how many d electrons, what ligands, what geometry?
The organizing textbook for the inorganic foundations is [Atkins et al., 2022], and the biological applications follow [Roat-Malone, 2007].
Concept 1: Transition Metals and the dn Configuration
Electron configurations: the 4s/3d subtlety
In CHEM101 you learned that the 4s orbital fills before 3d, giving Fe = [Ar]4s23d6. That is true for neutral atoms. Once the metal is ionized, the 3d orbitals drop below 4s in energy, and 4s electrons are removed first. As a result, Fe2+ is [Ar]3d6, not [Ar]4s23d4.
For coordination chemistry this leads to a simple rule. Compounds contain ions, not neutral atoms, so you just count d electrons:
d-electron count (d^n) = (group number) − (oxidation state) Examples: Fe(II): group 8 − 2 = d^6 Fe(III): group 8 − 3 = d^5 Cu(II): group 11 − 2 = d^9 Zn(II): group 12 − 2 = d^10 Pt(II): group 10 − 2 = d^8 Co(III): group 9 − 3 = d^6
The quantity dn is the most important number in this course. Almost every property discussed below follows from it.
Oxidation states
Early transition metals (Sc–Mn) can reach high oxidation states, up to the group number: Mn(VII) in permanganate, Cr(VI) in chromate. Late transition metals (Fe–Zn) favor lower states, because increasing effective nuclear charge across the row stabilizes the d electrons and makes them harder to remove. Biology exploits metals with accessible adjacent oxidation states: Fe(II)/Fe(III), Cu(I)/Cu(II), Mn(II)/Mn(III)/Mn(IV), Mo(IV)/Mo(V)/Mo(VI). Each pair is a one-electron redox switch.
Magnetic properties
Unpaired electrons make a complex paramagnetic. For first-row metals, orbital angular momentum is largely "quenched" by the ligand field, so the spin-only formula works reasonably well:
μ_spin-only = √[n(n + 2)] μ_B (n = number of unpaired electrons) n = 1 → 1.73 μ_B n = 2 → 2.83 μ_B n = 3 → 3.87 μ_B n = 4 → 4.90 μ_B n = 5 → 5.92 μ_B
The practical value is diagnostic. A measured magnetic moment tells you how many unpaired electrons are present, and that tells you the spin state. In CHEM501 you met EPR spectroscopy; it detects exactly these unpaired electrons and is one of the main tools of bioinorganic chemistry.
Concept 2: Coordination Complexes, Stability, and the Chelate Effect
Coordination number and geometry
The coordination number (CN) is the number of donor atoms bonded to the metal. The most common cases are:
- CN 4. Tetrahedral (Zn2+, many d10 ions) or square planar (d8 Pt(II), Pd(II), Ni(II) with strong-field ligands, Au(III)).
- CN 5. Trigonal bipyramidal or square pyramidal. These are close in energy and interconvert easily. Five-coordinate geometries are common in enzymes because the sixth site is left open for substrate.
- CN 6. Octahedral. This is the default for most first-row M2+ and M3+ ions.
- CN 7–9. Typical of large ions such as the lanthanides (Gd3+, usually CN 9) and Ca2+ (often 7–8 in proteins).
Nomenclature essentials [Atkins et al., 2022]:
- Ligands are named in alphabetical order before the metal.
- Anionic ligands end in "-o" (chlorido, cyanido, hydroxido).
- Prefixes count ligands: di-, tri-, or bis-, tris- for complex ligand names.
- The oxidation state follows the metal in Roman numerals.
- Anionic complexes take the "-ate" ending, using Latin roots where traditional (ferrate, cuprate, argentate, aurate).
- Bridging ligands get μ-, and the hapticity of π-ligands gets ηn.
Example: cisplatin, cis-[PtCl2(NH3)2], is cis-diamminedichloridoplatinum(II).
Stability constants
Complex formation in water is really ligand substitution, with water as the leaving group. The stepwise and overall formation constants are:
Stepwise: M + L ⇌ ML K₁ = [ML]/([M][L])
ML + L ⇌ ML₂ K₂ = [ML₂]/([ML][L])
...
Overall: M + nL ⇌ MLₙ βₙ = K₁·K₂·…·Kₙ = [MLₙ]/([M][L]ⁿ)
Thermodynamics: ΔG° = −RT ln βₙ
Normally K1 > K2 > K3…. This is partly statistical: there are fewer sites left to fill and more ligands available to leave. If a stepwise constant breaks this trend, something structural has changed, such as a spin-state switch or a change in geometry.
The Irving–Williams series
For high-spin divalent first-row ions binding a given ligand, stability almost always follows:
Mn²⁺ < Fe²⁺ < Co²⁺ < Ni²⁺ < Cu²⁺ > Zn²⁺
This order was identified by [Irving & Williams, 1953]. It arises from decreasing ionic radius across the row combined with ligand-field stabilization (Concept 4). Cu2+ gets an extra boost from Jahn–Teller distortion.
This has a surprising biological consequence. Left to thermodynamics, Cu2+ would displace other metals from nearly any binding site. Cells therefore have to control metal availability actively. Free copper in the cytoplasm is kept vanishingly low by chaperones and metallothioneins, so that enzymes that "want" Mn or Fe are not poisoned by copper. Metal selectivity in biology is as much about supply as about affinity.
Hard and soft acids and bases (HSAB)
Pearson's principle [Pearson, 1963] states that hard acids prefer hard bases and soft acids prefer soft bases.
- Hard acids are small, highly charged, and weakly polarizable: Na+, Mg2+, Ca2+, Fe3+, Gd3+.
- Hard bases are O-donors and F−: carboxylate, phosphate, water, hydroxide.
- Soft acids are large, low-charge, and polarizable: Cu+, Ag+, Au+, Hg2+, Cd2+, Pt2+.
- Soft bases are S-donors (thiolate, thioether), I−, CN−, and phosphines.
- Borderline cases include Fe2+, Co2+, Ni2+, Cu2+, Zn2+, and imidazole nitrogen (histidine).
This explains a great deal: Ca2+ is bound by Asp/Glu carboxylates, Cu+ by Cys and Met, Hg2+ toxicity comes from attack on Cys thiols, and auranofin's Au(I) targets selenocysteine.
The chelate effect
A chelating ligand binds through two or more donor atoms (from Greek chele, "claw"). Chelate complexes are much more stable than analogous complexes of monodentate ligands with the same donor atoms. The classic comparison:
[Ni(H₂O)₆]²⁺ + 6 NH₃ ⇌ [Ni(NH₃)₆]²⁺ + 6 H₂O log β₆ ≈ 8.6 [Ni(H₂O)₆]²⁺ + 3 en ⇌ [Ni(en)₃]²⁺ + 6 H₂O log β₃ ≈ 18.3 (en = ethylenediamine, H₂NCH₂CH₂NH₂)
In both complexes, six N donors bind Ni2+. Yet the en complex is about 1010 times more stable. Why?
The entropic explanation, which connects to your CHEM301 statistical mechanics: in the NH3 reaction, 7 particles become 7 particles. In the en reaction, 4 particles become 7, a net gain in translational entropy.
The kinetic/probabilistic picture is more intuitive. Once one end of en is bound, the other end is held within a few ångströms of the metal. Its "effective concentration" is enormous. If one arm lets go, it re-binds almost immediately before it can diffuse away. A monodentate NH3 that dissociates simply leaves.
Analogy: holding onto a ledge with one hand versus two. If you slip with one hand while using two, the other hand keeps you close enough to re-grab. The chelate effect is "molecular redundancy."
Two refinements matter. Five-membered chelate rings are generally most stable for small metal ions; six-membered rings suit smaller ions better. Macrocycles such as porphyrins and DOTA add a further "macrocyclic effect" from preorganization: the ligand already has the right shape and pays little conformational entropy on binding. Proteins push preorganization to its limit, as you will see with plastocyanin.
Concept 3: Isomerism in Coordination Complexes
Because metals have well-defined geometries with more than four positions, coordination compounds show rich isomerism. Alfred Werner used isomer counts to prove octahedral geometry before X-ray crystallography existed, for which he received the 1913 Nobel Prize.
Geometric isomers
- cis/trans in square planar MA2B2 and octahedral MA4B2. Cisplatin (cis) is a potent anticancer drug; transplatin is clinically inactive. Same formula, very different biology.
- fac/mer in octahedral MA3B3. In facial isomers the three identical ligands occupy one triangular face; in meridional isomers they lie on a plane containing the metal. Proteins often bind metals facially. The His3 site of carbonic anhydrase is a fac arrangement, which leaves the remaining coordination positions together on one side, available for substrate.
Optical isomers (chirality)
[M(en)3]n+ and [M(ox)3]n− are chiral even though no individual atom is a stereocenter. Chirality arises from the propeller-like twist of the three chelate rings, labeled Δ (right-handed) and Λ (left-handed). This is a new kind of chirality compared with CHEM201's tetrahedral carbon. The whole complex is the stereogenic unit. In a protein, a metal site's chirality is set by the chirality of the polypeptide.
Linkage isomers
Ambidentate ligands can bind through different atoms:
- NO2− binds as nitrito-κN or nitrito-κO.
- SCN− binds through S (soft metals) or N (hard metals).
- CN− can bind through C or N.
Histidine is effectively ambidentate too, coordinating through either Nδ1 or Nε2 of the imidazole. Which nitrogen binds is a structural fingerprint of the metal site.
Other types
Ionization isomers ([Co(NH3)5Br]SO4 vs. [Co(NH3)5SO4]Br) and hydrate isomers swap which species is inside versus outside the coordination sphere.
Concept 4: Crystal Field Theory, CFSE, and Spin States
The core idea
Crystal field theory (CFT) treats ligands as point negative charges or dipoles approaching the metal. In an octahedron, ligands lie along the ±x, ±y, ±z axes.
- The dz² and dx²−y² orbitals point directly at the ligands. They are destabilized and form the eg set.
- The dxy, dxz, dyz orbitals point between the ligands. They are relatively stabilized and form the t2g set.
The energy gap is Δo (octahedral splitting, also called 10Dq). Because the average energy (the barycenter) is conserved:
Octahedral: t₂g at −0.4 Δₒ (each of 3 orbitals)
e_g at +0.6 Δₒ (each of 2 orbitals)
Check: 3(−0.4) + 2(+0.6) = 0 ✓
Tetrahedral: e at −0.6 Δₜ (2 orbitals, now LOWER)
t₂ at +0.4 Δₜ (3 orbitals, now HIGHER)
Δₜ ≈ (4/9) Δₒ for the same metal and ligands
Square planar (d⁸): d_x²−y² is pushed very high; the large gap
makes d⁸ ions diamagnetic and strongly favors this geometry
(Pt²⁺, Pd²⁺, Au³⁺).
Crystal field stabilization energy (CFSE)
CFSE = (−0.4 n_t₂g + 0.6 n_eg) Δₒ + (pairing energy terms) Examples (octahedral): d³ (Cr³⁺): t₂g³ CFSE = −1.2 Δₒ d⁶ low spin (Co³⁺): t₂g⁶ CFSE = −2.4 Δₒ + pairing d⁵ high spin (Mn²⁺): t₂g³e_g² CFSE = 0 d¹⁰ (Zn²⁺): t₂g⁶e_g⁴ CFSE = 0
High spin versus low spin: a competition between two energies
For d4–d7 octahedral ions there are two choices for the fourth electron:
- Put it in eg, paying Δo but avoiding pairing. This gives high spin.
- Pair it in t2g, paying the pairing energy P (Coulomb repulsion plus loss of exchange stabilization) but avoiding Δo. This gives low spin.
If Δₒ < P → high spin (weak-field ligands) If Δₒ > P → low spin (strong-field ligands) [Fe(H₂O)₆]²⁺: d⁶ high spin, t₂g⁴e_g², 4 unpaired, paramagnetic [Fe(CN)₆]⁴⁻: d⁶ low spin, t₂g⁶, 0 unpaired, diamagnetic
Why this matters biologically: hemoglobin.
- Deoxy-heme Fe(II) is five-coordinate and high spin (S = 2). Electrons in the antibonding eg orbitals make the ion effectively larger, and it sits displaced out of the porphyrin plane, toward the proximal histidine.
- When O2 binds, the iron becomes six-coordinate and low spin. The eg orbitals empty, the iron effectively "shrinks," and it moves into the porphyrin plane.
- That movement pulls the proximal histidine and its attached F-helix, transmitting a structural signal across subunit interfaces.
This is the Perutz trigger mechanism for the cooperativity you studied in CHEM401. A spin-state change in one d6 ion drives the T→R allosteric transition. Quantum mechanics becomes physiology here.
The Jahn–Teller effect
The Jahn–Teller theorem [Jahn & Teller, 1937] states that a nonlinear molecule in an orbitally degenerate electronic state will distort to remove the degeneracy and lower its energy.
- Octahedral Cu2+ (d9, t2g6eg3) has an unevenly occupied eg set. It almost always shows tetragonal elongation: two long axial bonds and four short equatorial ones.
- High-spin Cr2+ and Mn3+ (d4) behave similarly.
- This is why Cu(II) sites in proteins are frequently square planar or square pyramidal.
Concept 5: Ligand Field Theory and the Spectrochemical Series
The puzzle CFT cannot solve
Experimentally, ligands can be ranked by how large a Δo they produce. This ranking is the spectrochemical series:
I⁻ < Br⁻ < S²⁻ < SCN⁻(S) < Cl⁻ < NO₃⁻ < F⁻ < OH⁻ < H₂O < NCS⁻(N) < CH₃CN < NH₃ < en < bipy < phen < NO₂⁻ < PPh₃ < CN⁻ < CO weak field (small Δₒ) ──────────────────────▶ strong field (large Δₒ)
A purely electrostatic model would predict that charged ligands (F−, OH−) produce larger splittings than neutral ones (H2O, NH3). The opposite happens. Worse, the strongest-field ligand, CO, is neutral and almost nonpolar. CFT fails here because it ignores covalency.
Ligand field theory: MO theory applied to complexes
In ligand field theory (LFT), metal and ligand orbitals mix to form molecular orbitals. In an octahedral complex:
σ-bonding. Six ligand σ orbitals combine with the metal 4s (a1g), 4p (t1u), and the eg d orbitals. The t2g orbitals have the wrong symmetry for σ overlap and remain nonbonding in a σ-only picture. The eg set becomes the antibonding eg*. Δo is now the gap between nonbonding t2g and antibonding eg*. Stronger σ-donors push eg* higher and increase Δo.
π-bonding involves the t2g set, and this resolves the puzzle.
- π-donor ligands (halides, OH−, O2−) have filled p orbitals lower in energy than the metal t2g. Mixing pushes t2g up (it becomes π-antibonding), so Δo shrinks. This is why halides are weak-field even though they are anionic.
- π-acceptor ligands (CO, CN−, phosphines, bipy) have empty π* orbitals higher in energy than t2g. Mixing pulls t2g down (it becomes π-bonding), so Δo grows. Metal electron density flows back into the ligand. This is back-bonding.
- Pure σ-donors (NH3, en) sit in the middle.
π-donor (weak field) < σ-only (intermediate) < π-acceptor (strong field)
X⁻, OH⁻ NH₃, en CO, CN⁻, PR₃
The beautiful insight: the spectrochemical series is not a list of how "strongly" ligands bind. It is a ranking of each ligand's π character, superimposed on σ-donor strength.
The metal dependence of Δ
Δ also depends on the metal:
- Δ increases with oxidation state. M3+ typically has a Δo about 40–80% larger than M2+.
- Δ increases down a group. 4d and 5d orbitals are more diffuse and overlap better with ligands. Consequently, 4d and 5d complexes are almost always low spin.
- The nephelauxetic effect ("cloud-expanding"): covalent ligands reduce interelectronic repulsion (the Racah parameter B) compared with the free ion. This is direct spectroscopic evidence that metal–ligand bonds are partly covalent.
Concept 6: Electronic Spectra of Transition-Metal Complexes
Why complexes are colored
Δo for first-row aqua ions is typically 7,500–30,000 cm−1, which falls in the near-IR to visible range. The simplest case is [Ti(H2O)6]3+ (d1). It has one absorption band near 20,300 cm−1 (≈ 493 nm), corresponding to t2g1 → eg1. This gives Δo ≈ 243 kJ mol−1 directly. The solution absorbs blue-green light and appears purple. The color you see is complementary to the color absorbed.
Selection rules (recall CHEM303)
Spin selection rule: ΔS = 0
(spin-forbidden transitions are ~100–1000× weaker still)
Laporte rule: in centrosymmetric molecules, g ↔ u allowed;
g ↔ g and u ↔ u forbidden
d orbitals are all g, so d–d transitions in octahedral
complexes are Laporte-forbidden
d–d transitions appear at all only because vibrations momentarily break the inversion center (vibronic coupling). They are therefore weak, with molar absorptivities ε ≈ 1–100 M−1 cm−1. Tetrahedral complexes lack an inversion center, so their d–d bands are more intense (ε ≈ 100–1000). This is why tetrahedral [CoCl4]2− is a deep blue while octahedral [Co(H2O)6]2+ is a pale pink.
High-spin d5 Mn2+ is a special case. Every possible d–d transition requires flipping a spin, so all are both spin- and Laporte-forbidden. Mn(II) solutions are nearly colorless, a very pale pink.
Charge-transfer bands
Charge-transfer (CT) transitions move an electron between orbitals of mostly metal character and orbitals of mostly ligand character. They are fully allowed, with ε ≈ 103–105 M−1 cm−1.
- LMCT (ligand-to-metal) occurs with an oxidizing metal and reducing ligands. Permanganate MnO4− is d0, so it has no d–d transitions at all, yet it is intensely purple from O→Mn charge transfer.
- MLCT (metal-to-ligand) occurs with an electron-rich metal and π-acceptor ligands. Examples are [Ru(bpy)3]2+ and [Fe(phen)3]2+, the red complex you may have used for iron determination in CHEM302L.
Bioinorganic payoff. The intense blue of "blue copper" proteins such as plastocyanin (ε ≈ 5,000 M−1 cm−1 near 600 nm) is not a d–d band. It is a Cys-S(π)→Cu(II) LMCT band. Its intensity alone tells you a thiolate is bound to copper.
Tanabe–Sugano diagrams
For d2–d8 ions, electron–electron repulsion splits configurations into multiple term states, labeled with Mulliken symbols such as 4A2g and 4T2g. Tanabe–Sugano diagrams [Tanabe & Sugano, 1954] plot term energies (E/B) against field strength (Δo/B), with the ground state drawn as the horizontal axis.
Worked use, octahedral Cr3+ (d3):
- The ground state is 4A2g.
- Three spin-allowed transitions are expected: to 4T2g, 4T1g(F), and 4T1g(P).
- The lowest-energy band equals Δo directly.
- For [Cr(H2O)6]3+, bands near 17,400 and 24,600 cm−1 are observed. The ratio of band energies locates Δo/B on the diagram, from which you extract both Δo and B.
For d4–d7, the diagrams show a vertical discontinuity at the high-spin/low-spin crossover, which is the spin-state competition from Concept 4 drawn graphically.
Concept 7: Substitution Kinetics and Mechanisms
Lability versus inertness
Henry Taube classified complexes as labile (substitution complete within about a minute at 25 °C, 0.1 M) or inert (slower). Water-exchange rate constants for metal aqua ions span roughly 19 orders of magnitude: from about 109 s−1 for Cu2+ (Jahn–Teller-labilized axial waters) to about 10−6 s−1 for Cr3+, and even slower for Ir3+ [Atkins et al., 2022].
Ligand field theory predicts which ions are inert:
- Inert: d3 (Cr3+), low-spin d6 (Co3+, Rh3+, Pt4+), and d8 square planar (Pt2+). These configurations have large CFSE and empty eg*. Reaching a 5- or 7-coordinate transition state costs a lot of CFSE.
- Labile: configurations with electrons in eg* (weakened, antibonding M–L bonds), with zero CFSE (d10 Zn2+, high-spin d5), and small, weakly charged ions.
Thermodynamic stability and kinetic lability are independent. [Co(NH3)6]3+ is thermodynamically unstable in acid; it "should" decompose to NH4+ and Co(aq). Yet it survives for days because it is kinetically inert. [Ni(CN)4]2− is very stable (log β4 ≈ 30) yet exchanges CN− rapidly.
Mechanistic classes
Dissociative (D): ML₅X → ML₅ + X (slow; intermediate of lower CN)
ML₅ + Y → ML₅Y (fast)
Associative (A): ML₅X + Y → ML₅XY (slow; intermediate of higher CN)
ML₅XY → ML₅Y + X (fast)
Interchange (I): concerted; no detectable intermediate
Iₐ: bond-making dominates the rate (sensitive to Y)
I_d: bond-breaking dominates (insensitive to Y)
Octahedral substitution is usually Id, and the Eigen–Wilkins mechanism describes it. The entering ligand first forms an outer-sphere encounter complex (pre-equilibrium constant KOS), then exchanges with an inner-sphere ligand:
k_obs = K_OS k_i [Y] / (1 + K_OS [Y])
Square-planar d8 complexes react associatively. The open axial sites invite the nucleophile in, forming a 5-coordinate trigonal bipyramidal intermediate. Their rate law has two terms:
rate = (k₁ + k₂[Y]) [ML₃X] k₁ path: solvent-assisted (solvent attacks first, then is displaced by Y) k₂ path: direct attack by Y
The trans effect and the trans influence
In square-planar Pt(II), some ligands accelerate substitution of the ligand trans to themselves:
Trans-effect series (approximate): H₂O < OH⁻ < NH₃ < py < Cl⁻ < Br⁻ < I⁻ < SCN⁻ < NO₂⁻ < PR₃ ≈ H⁻ < CH₃⁻ < CO ≈ CN⁻ ≈ C₂H₄
Two separate concepts are involved:
- Trans influence is a ground-state, thermodynamic effect. Strong σ-donors weaken and lengthen the bond trans to them, because they compete for the same metal orbital. You can see it in crystal structures.
- Trans effect is a kinetic effect on the rate of substitution. It includes the trans influence (σ-donors) plus transition-state stabilization by π-acceptors, which withdraw electron density from the crowded 5-coordinate intermediate.
The trans effect makes cisplatin synthesis possible.
- Start with [PtCl4]2−. The first NH3 replaces any Cl−.
- The second NH3 replaces a Cl− trans to another Cl−, not one trans to NH3, because Cl− has the larger trans effect. This gives cis-[PtCl2(NH3)2].
- Starting instead from [Pt(NH3)4]2+ and adding Cl− gives trans-platin.
Order of addition determines stereochemistry. This is rational synthesis based on a mechanistic principle.
Concept 8: Electron-Transfer Reactions and Marcus Theory
Outer-sphere versus inner-sphere
Outer-sphere electron transfer. The coordination spheres stay intact, and the electron tunnels between the two complexes. Example: [Fe(H2O)6]2+/[Fe(H2O)6]3+ self-exchange. Nearly all long-range biological electron transfer is effectively outer-sphere: cytochrome to cytochrome, plastocyanin to Photosystem I.
Inner-sphere electron transfer. A ligand bridges the two metals and the electron passes through it. Taube's classic experiment [Taube, Myers & Rich, 1953] showed this cleanly:
[Co(NH₃)₅Cl]²⁺ + [Cr(H₂O)₆]²⁺ → [Co(NH₃)₅–Cl–Cr(H₂O)₅]⁴⁺ (bridged)
→ Co²⁺(aq) + [Cr(H₂O)₅Cl]²⁺ + 5 NH₄⁺ (in acid)
The chloride ends up on chromium. The reactants were chosen so that this result is unambiguous: inert Co(III) and labile Cr(II) become labile Co(II) and inert Cr(III). The bridging atom had to transfer along with the electron. Taube won the 1983 Nobel Prize for this work.
Marcus theory
Rudolph Marcus [Marcus, 1956] asked why some electron transfers with large driving forces are slow while some with zero driving force are fast. The key insight comes from the Franck–Condon principle (familiar from CHEM303). Electrons move much faster than nuclei, so the electron can only jump when the nuclear geometry is such that the reactant and product states are degenerate. Bond lengths and solvent orientation must first distort to a compromise geometry. The energy needed to reorganize is the reorganization energy λ.
ΔG‡ = (λ + ΔG°)² / (4λ) k_ET = A exp(−ΔG‡ / k_B T) λ = λ_inner (bond length/angle changes) + λ_outer (solvent reorientation) Cross relation: k₁₂ ≈ (k₁₁ k₂₂ K₁₂ f₁₂)^½
Several consequences follow from the quadratic form:
- Normal region (−ΔG° < λ): a larger driving force means a faster rate, as expected.
- Activationless (−ΔG° = λ): ΔG‡ = 0 and the rate is maximal.
- Inverted region (−ΔG° > λ): even more driving force makes the rate slower. This counterintuitive prediction was confirmed experimentally in the 1980s, and Marcus received the 1992 Nobel Prize.
The inverted region matters in photosynthesis. Unproductive back-electron-transfer in reaction centers is very exergonic, which places it in the inverted region and keeps it slow. That kinetic trap helps preserve charge separation.
Link to ligand field theory. λinner depends on how much bond lengths change between oxidation states.
- Co(III)/Co(II) amines switch from low-spin t2g6 to high-spin t2g5eg2. Electrons enter antibonding orbitals, bond lengths change by about 0.2 Å, and self-exchange is extremely slow (~10−6 M−1 s−1).
- Low-spin Fe(II)/Fe(III) in cytochromes changes only a t2g (nonbonding or π) electron. Geometry barely changes, λ is small, and transfer is fast.
The entatic state
Vallee and Williams proposed that proteins hold metal sites in strained geometries intermediate between the preferred geometries of the two oxidation states [Vallee & Williams, 1968]. This is the entatic (or "rack") state.
- Cu(II) prefers tetragonal geometry; Cu(I) prefers tetrahedral.
- Plastocyanin's protein holds copper in a distorted tetrahedral site (2 His, 1 Cys, 1 Met) that is a compromise between the two.
- λ is minimized, and electron transfer is fast.
The protein pays the strain energy once, during folding, so that each electron-transfer event does not have to.
Concept 9: Organometallic Chemistry and Catalysis
The 18-electron rule
A transition metal has nine valence orbitals (five d, one s, three p). Filling them all gives a noble-gas-like configuration. Complexes of low-valent metals with π-acceptor ligands often reach 18 valence electrons [Tolman, 1972].
Neutral (covalent) counting: Cr(CO)₆: Cr(0) 6e + 6 CO × 2e = 18e Fe(CO)₅: Fe(0) 8e + 5 CO × 2e = 18e Ni(CO)₄: Ni(0) 10e + 4 CO × 2e = 18e Ferrocene: Fe 8e + 2 × Cp (5e each, neutral) = 18e Square-planar d⁸ (Pt²⁺, Pd²⁺, Rh⁺): typically 16e and stable
Metal carbonyls and back-bonding
CO bonds in two ways at once:
- σ-donation from the carbon lone pair (the HOMO, 5σ) to the metal.
- π-back-donation from filled metal d orbitals into the CO π* orbitals (the LUMO).
Back-donation populates C–O antibonding orbitals and lowers the C–O stretching frequency, which you can measure by IR spectroscopy (CHEM501):
free CO: ν(CO) ≈ 2143 cm⁻¹ [Mn(CO)₆]⁺: ≈ 2090 cm⁻¹ (cationic, less back-bonding) Cr(CO)₆: ≈ 2000 cm⁻¹ [V(CO)₆]⁻: ≈ 1860 cm⁻¹ (anionic, more back-bonding)
ν(CO) works as a built-in "electron-density meter" for the metal. The same idea in biology: CO binds heme ~20,000-fold more tightly than O2 in free heme. In myoglobin and hemoglobin the protein reduces this preference (the distal histidine and steric constraints matter, though the exact contributions are still debated), which is one reason we are not all poisoned by the CO produced in heme catabolism.
Olefin complexes follow the same logic, described by the Dewar–Chatt–Duncanson model: σ-donation from the C=C π orbital and back-donation into π*. Strong back-donation lengthens the C=C bond toward a metallacyclopropane. Zeise's salt, K[PtCl3(C2H4)], is the historic example.
Elementary steps of catalytic cycles
- Oxidative addition. A–B adds across the metal; oxidation state +2, CN +2, electron count +2.
- Reductive elimination. The reverse; forms the product bond.
- Migratory insertion. A ligand (alkene, CO) inserts into an adjacent M–R or M–H bond.
- β-Hydride elimination. An M–alkyl gives M–H plus an alkene. A vacant cis site is required.
- Transmetalation. An organic group transfers from one metal (B, Zn, Sn) to another (Pd).
Three catalytic cycles to know
Wilkinson's catalyst, RhCl(PPh3)3, for alkene hydrogenation. Rh(I) is d8 and the complex is 16e.
- Oxidative addition of H2 gives a Rh(III) dihydride.
- The alkene coordinates, after phosphine dissociation.
- Migratory insertion forms Rh–alkyl.
- Reductive elimination releases the alkane and regenerates Rh(I).
The Heck reaction, coupling aryl halide plus alkene. Pd cycles between Pd(0) and Pd(II).
- Oxidative addition of Ar–X.
- Alkene coordination and syn migratory insertion.
- Syn β-hydride elimination releases the substituted alkene.
- Base removes HX from H–Pd–X, regenerating Pd(0).
Suzuki coupling, coupling Ar–X with Ar'–B(OR)2.
- Oxidative addition of Ar–X.
- Transmetalation from a base-activated boronate.
- Reductive elimination forms Ar–Ar' and regenerates Pd(0).
Heck, Negishi, and Suzuki shared the 2010 Nobel Prize. The Pd-catalyzed couplings you used in CHEM202 are oxidation-state shuffles of a d10/d8 metal.
Concept 10: Bioinorganic Chemistry, or Why Nature Chooses Particular Metals
Every metal in biology fills a niche defined by the principles above. Think of each metal as a specialist hired for what its d-electron count does best.
The s-block ions: Na+, K+, Mg2+, Ca2+
These ions have no d electrons, no ligand field, and no redox chemistry. Their function comes purely from charge, size, and hydration energy.
- Na+/K+. The selectivity filter of the K+ channel (BIOL501) uses backbone carbonyl oxygens spaced to mimic K+'s hydration shell exactly. Smaller Na+ cannot be fully "re-solvated" by the filter, so the dehydration penalty is not repaid and Na+ is excluded despite being smaller.
- Mg2+. Small and hard, with slow water exchange for a main-group ion. It binds the phosphates of ATP and nucleic acids. The true substrate of most kinases is MgATP2−, not ATP.
- Ca2+. Larger, with flexible coordination (CN 6–8, irregular geometry) and fast exchange. These features make it ideal as a signal that binds and releases quickly and triggers conformational changes. Examples include calmodulin's EF-hands, synaptotagmin, and troponin C.
Iron
Hemoglobin and myoglobin. Iron is held by protoporphyrin IX plus a proximal His; O2 binds at the distal site.
- The bound oxygen is best described electronically as close to Fe(III)–superoxide (Fe3+–O2•−), with the two spins antiferromagnetically coupled to give a diamagnetic state.
- The distal His hydrogen-bonds to the bound O2, stabilizing it.
- The protein's hydrophobic pocket prevents the irreversible formation of μ-oxo Fe(III) dimers that free heme forms in water.
- The spin-state trigger described in Concept 4 drives cooperativity.
Cytochromes. Hemes with two axial ligands (His/His or His/Met) and no open site carry out electron transfer only, cycling between low-spin Fe(II) and Fe(III) with small λ. The axial ligands and heme environment tune E°' across a range of several hundred mV.
Iron–sulfur clusters are among the most ancient cofactors: [2Fe–2S], [3Fe–4S], and [4Fe–4S], ligated by Cys. They are electron-transfer modules with electron delocalization (mixed-valence states). Their potentials range from about −700 mV (ferredoxins) to about +400 mV (HiPIPs). HiPIPs use the [4Fe–4S]3+/2+ couple, while ferredoxins use [4Fe–4S]2+/1+; the protein environment selects which couple is accessible.
Copper
- Type 1 (blue) copper, e.g., plastocyanin. The entatic Cu site (His2CysMet) has an intense S→Cu LMCT band, a small EPR hyperfine coupling, and fast electron transfer.
- Ceruloplasmin. A multicopper ferroxidase that combines Type 1, Type 2, and a binuclear Type 3 site. It couples Fe2+ oxidation (needed for iron loading onto transferrin) to the four-electron reduction of O2 to water.
- Cu,Zn-superoxide dismutase. Catalyzes 2 O2•− + 2 H+ → H2O2 + O2 by a ping-pong mechanism: Cu(II) is reduced by one superoxide, then reoxidized by the next. A bridging histidinate links Cu and Zn. Zn plays a structural and electrostatic role and does not do redox. The rate approaches diffusion control (~109 M−1 s−1), helped by an electrostatic funnel that guides the anion in.
- Cytochrome c oxidase (CHEM402's Complex IV). CuA is a binuclear, Cys-bridged, mixed-valence Cu(1.5)–Cu(1.5) center that receives electrons from cytochrome c. Electrons pass through heme a to the heme a3–CuB binuclear center, where O2 is reduced to 2 H2O and the energy is coupled to proton pumping.
Zinc: the Lewis acid with no distractions
Zn2+ is d10, which gives it a distinctive set of properties:
- Zero CFSE, so no geometric preference from the ligand field. It readily adopts CN 4, 5, or 6 during catalysis.
- No redox chemistry, so no risk of radical side reactions.
- Strong Lewis acidity and fast ligand exchange.
Carbonic anhydrase shows the Lewis acid mechanism clearly. Zn is bound by three His (fac) plus one water.
Free water: pKa ≈ 15.7 (or 14, depending on convention) Zn-bound water: pKa ≈ 7 (in carbonic anhydrase II) E–Zn–OH₂ ⇌ E–Zn–OH⁻ + H⁺ (proton shuttled out via His64) E–Zn–OH⁻ + CO₂ → E–Zn–HCO₃⁻ (nucleophilic attack) E–Zn–HCO₃⁻ + H₂O → E–Zn–OH₂ + HCO₃⁻ k_cat ≈ 10⁶ s⁻¹ (CA II), among the fastest enzymes known
The key insight: the metal does not need to do redox. By polarizing a bound water, Zn2+ produces a hydroxide nucleophile at neutral pH, something otherwise available only in strong base. The enzyme makes "concentrated base" on demand at one specific location.
Other zinc enzymes and proteins follow the same themes:
- Carboxypeptidase A (Zn: His2Glu + water) uses a Zn-activated water to hydrolyze C-terminal peptide bonds.
- Alcohol dehydrogenase has a catalytic Zn (Cys2His) that binds the alcohol oxygen and lowers its pKa, so hydride can transfer to NAD+.
- Zinc fingers (Cys2His2 tetrahedral) use Zn purely structurally, as discussed in the peptide section below.
Manganese
The oxygen-evolving complex (OEC) of Photosystem II is a Mn4CaO5 cluster shaped like a "distorted chair," resolved at 1.9 Å [Umena et al., 2011]. It cycles through the S-states (S0→S4, the Kok cycle), storing four oxidizing equivalents, one per photon, before releasing O2:
2 H₂O → O₂ + 4 H⁺ + 4 e⁻
This produced essentially all the O2 in Earth's atmosphere. Manganese's many accessible oxidation states (II–IV, possibly beyond) let it accumulate charge without generating dangerous partially reduced oxygen intermediates. Mn-SOD in mitochondria cycles Mn(III)/Mn(II) to dismutate superoxide, playing the same role as Cu,Zn-SOD in a different compartment.
Molybdenum and tungsten
Nitrogenase reduces N2 at the FeMo-cofactor, MoFe7S9C-homocitrate. Its central atom was identified as a carbide in 2011, a surprise for a biological cluster [Lancaster et al., 2011; Spatzal et al., 2011].
N₂ + 8 H⁺ + 8 e⁻ + 16 MgATP → 2 NH₃ + H₂ + 16 MgADP + 16 Pᵢ
Note the obligatory H2 byproduct and the enormous ATP cost. The industrial Haber–Bosch process needs about 400–500 °C and 150–300 atm to do what this cluster does at ambient conditions.
Mo-pterin enzymes carry out oxygen-atom-transfer chemistry. The Mo is coordinated by the dithiolene of molybdopterin and cycles through Mo(VI)/Mo(V)/Mo(IV).
- Xanthine oxidase hydroxylates xanthine to uric acid. The oxygen comes from water, not O2. It is the target of allopurinol in gout treatment.
- Sulfite oxidase oxidizes sulfite to sulfate. Its deficiency is a severe, often fatal neurological disorder.
- Tungsten replaces Mo in analogous enzymes in some hyperthermophilic archaea. W sits directly below Mo, with similar chemistry but lower potentials, and is perhaps favored in the reducing, hot environments of early life.
Metal-based drugs
- Cisplatin. Discovered serendipitously by Rosenberg, who noticed that Pt electrodes inhibited E. coli cell division [Rosenberg et al., 1965].
- It is a prodrug. Extracellular [Cl−] (~100 mM) keeps it intact.
- Inside cells, [Cl−] is much lower (~4–20 mM), so the complex aquates to [Pt(NH3)2Cl(H2O)]+. Water is a much better leaving group.
- Pt then binds guanine N7. The major lesion is a 1,2-intrastrand d(GpG) crosslink that bends DNA and is recognized by HMG-domain proteins, triggering apoptosis.
- The kinetic inertness of Pt(II) (Concept 7) is the key. A labile metal would release from DNA too quickly to matter.
- Carboplatin replaces the two Cl− with a chelating cyclobutanedicarboxylate. The chelate effect slows aquation, which reduces nephrotoxicity, at the cost of requiring higher doses.
- Auranofin. A linear Au(I) complex with a phosphine and a thioglucose ligand. Soft Au(I) targets the soft selenocysteine of thioredoxin reductase, an HSAB-driven drug action. Originally used for rheumatoid arthritis.
- Vanadium compounds. Vanadate (VO43−) is a structural mimic of phosphate and a transition-state analog for phosphoryl transfer. It inhibits protein tyrosine phosphatases, giving "insulin-mimetic" effects. BMOV, bis(maltolato)oxovanadium(IV), and related compounds have been investigated as antidiabetics.
- Gadolinium MRI contrast agents. Gd3+ (4f7, seven unpaired electrons, slow electronic relaxation) shortens the T1 relaxation time of nearby water protons.
- Free Gd3+ is toxic: it mimics Ca2+ and precipitates as phosphate.
- It is therefore wrapped in octadentate chelates (DTPA, DOTA) that leave one site (q = 1) for a rapidly exchanging water.
- The design balances maximal thermodynamic and kinetic stability (macrocyclic DOTA is especially inert) against preserving that one water site.
Connections to Peptide Chemistry
Everything above comes together in one question: how does a polypeptide act as a ligand?
The amino-acid ligand toolkit
- Histidine (imidazole). A borderline N-donor that is neutral at physiological pH (side-chain pKa ≈ 6). It binds through Nδ1 or Nε2 and is a moderate σ-donor with weak π-acceptor ability. It is the most common ligand for Cu, Zn, Ni, and Fe. Binding a metal lowers the pKa of the remaining imidazole N–H, and in Cu,Zn-SOD the imidazole is fully deprotonated to bridge two metals.
- Cysteine (thiolate). Soft, strongly σ- and π-donating, and a good bridging ligand. The thiol pKa (~8.3) drops sharply on metal binding. It dominates in Fe–S clusters, zinc fingers, blue copper, and metallothioneins, and it is also easily oxidized, which makes it a redox liability.
- Aspartate and glutamate (carboxylate). Hard O-donors that can bind monodentate, bidentate (chelating), or bridging. They dominate Ca2+, Mg2+, Mn, and Fe(III) sites.
- Methionine (thioether). A soft, weak donor that stabilizes Cu(I), as in plastocyanin's axial Met. It is also the axial ligand in cytochrome c.
- Tyrosine (phenolate). A hard O-donor seen in transferrin's Fe(III) site and in tyrosyl-radical enzymes.
- The backbone itself. Carbonyl O binds K+ and Ca2+. Amide N can bind when deprotonated. A free amide N–H has pKa ≈ 15, but Cu(II) and Ni(II) can induce deprotonation near neutral pH when a chelating anchor holds them close.
The ATCUN motif: the chelate effect written in amino acids
The Amino-Terminal Cu(II)- and Ni(II)-binding motif is an N-terminal sequence H2N–Xaa–Xaa–His [Harford & Sarkar, 1997]. It binds Cu(II) or Ni(II) in a square-planar 4N arrangement:
- the free N-terminal amine,
- the deprotonated amide N of residue 2,
- the deprotonated amide N of residue 3, and
- the His3 imidazole Nδ1.
This forms three fused chelate rings. Human serum albumin's N-terminus (Asp-Ala-His-Lys) is the natural example and a major Cu(II) transport site in blood.
Look at how many course concepts this one tripeptide uses:
- the chelate effect (fused 5- and 6-membered rings),
- the strong preference of d8 Ni(II) and Jahn–Teller d9 Cu(II) for square-planar geometry,
- metal-induced lowering of ligand pKa (amide deprotonation), and
- strong-field amide N− donors, which tend to make Ni(II) low spin and diamagnetic in this site.
ATCUN peptides have been attached to other molecules to create artificial nucleases: the Cu or Ni center, activated by peroxide, oxidatively cleaves DNA or proteins nearby.
Zinc fingers: metal-induced folding
A classic Cys2His2 zinc-finger peptide of about 30 residues is largely unfolded without metal. Add Zn2+, and the peptide folds into a ββα structure around a tetrahedral Zn(Cys)2(His)2 site. Here the metal is the hydrophobic core: binding pays for the conformational entropy of folding. Zinc is ideal because d10 has no geometric preference to fight the protein's tetrahedral demand and no redox chemistry to damage the Cys thiolates.
Other metals behave differently. Co(II) substitutes well, and its d–d bands become a spectroscopic probe for an otherwise "silent" Zn site, a common trick. Ni(II), which prefers square-planar or octahedral geometry, binds less well and can misfold the peptide. Metal preference for geometry, from CFSE, competes with the protein's structural preference. This interplay is central to metallopeptide design.
Metallothioneins
Metallothioneins are small (~61-residue in mammals), Cys-rich (~30%) proteins with no aromatic residues. They bind 7 Zn2+ or Cd2+, or up to about 12 Cu+, in two thiolate-bridged clusters. Their roles are metal homeostasis, heavy-metal detoxification (soft Cd2+ and Hg2+ bound by soft thiolates, an HSAB story), and redox buffering. Their structure is defined by the metals; apo-metallothionein is disordered.
Designed metallopeptides
The listed key paper for this course is ['The Design of Metallopeptides', Nature Chemistry, 2012, DOI: 10.1038/nchem.1201]. A citation caution, Skipper: as far as I can recall, that DOI corresponds to a Pecoraro-group paper titled "Hydrolytic catalysis and structural stabilization in a designed metalloprotein" [Zastrow, Peacock, Stuckey & Pecoraro, 2012], not to an article with the title listed. Please look up the DOI directly to confirm the title and authors before citing it.
That work illustrates the design philosophy well:
- A three-stranded α-helical coiled coil (based on the TRI peptide family) was engineered with two distinct metal sites.
- A structural site uses Cys residues to bind Hg(II), stabilizing the fold.
- A catalytic site has three His residues binding Zn(II) plus water/hydroxide, a deliberate mimic of the carbonic anhydrase active site.
- The construct hydrolyzed p-nitrophenyl acetate and hydrated CO2. Its activity is well below the native enzyme but remarkable for a designed peptide with no evolutionary optimization.
The lessons connect directly to this course:
- A His3Zn–OH2 unit retains much of its Lewis acid function even outside its native protein, so the first coordination sphere carries much of the chemistry.
- The large remaining gap in efficiency shows the importance of the second coordination sphere: hydrogen-bond networks, proton shuttles such as CAII's His64, and substrate-positioning residues.
- Separating structure from function, with one metal for folding and another for catalysis, is a modular design strategy.
Later work in peptide-based biomimetic chemistry builds on these foundations: de novo metalloproteins, peptide-based catalysts, metal-responsive peptide switches, and metallodrugs targeted with peptides. The questions you will meet there are exactly the ones in this chapter. What dn? What geometry does the metal prefer, and can the peptide enforce it? Which side chains are hard or soft enough? Is the site labile or inert? What is the reorganization energy?
Common Misconceptions to Watch Out For
-
"Transition-metal ions lose d electrons first because 3d fills last." No. In ions, 4s electrons are lost first. Fe2+ is d6, Cu2+ is d9. Always count d electrons as group number minus oxidation state.
-
"The oxidation state is the real charge on the metal." Oxidation states are formal bookkeeping. They assume all M–L bonds are ionic. In covalent complexes, especially with π-acceptors or non-innocent ligands (dithiolenes, O2, NO), the real charge distribution can differ greatly. Oxy-hemoglobin's "Fe(II)–O2 versus Fe(III)–O2•−" debate is a debate about how to describe electron distribution, not about counting.
-
"Strong-field ligands bind more strongly." The spectrochemical series ranks Δo, not bond strength or stability. CO is the strongest-field ligand, yet it does not bind Zn2+ or Ca2+ at all. HSAB and the metal's dn both matter.
-
"Stable means unreactive." Thermodynamic stability (β) and kinetic lability (rate) are independent. [Co(NH3)6]3+ is unstable but inert; [Ni(CN)4]2− is stable but labile. Cisplatin works because Pt(II) is inert, and Gd agents must be both stable and inert.
-
"A green complex absorbs green light." The observed color is complementary to the absorbed wavelength. And intense color does not mean d–d. Permanganate (d0) and blue copper owe their color to charge transfer.
-
"Crystal field theory explains the spectrochemical series." CFT cannot explain why neutral CO beats anionic F−. You need ligand field (MO) theory with π-donation and π-back-bonding. Use CFT for counting electrons and CFSE; use LFT for understanding why.
-
"The trans effect and the trans influence are the same thing." Trans influence is a ground-state, thermodynamic bond weakening, observable in crystal structures. Trans effect is a kinetic rate enhancement. Strong σ-donors (H−, CH3−) show both. π-acceptors (CO, C2H4) have large trans effects mainly through transition-state stabilization.
-
"The chelate effect is purely entropic." In aqueous solution it is predominantly entropic, but enthalpy contributes too, through ligand strain, solvation differences, and electrostatics. The kinetic "effective concentration" picture is equally valid and often more useful for thinking about metallopeptides.
-
"The 18-electron rule always applies." It works best for low-valent metals with π-acceptor ligands. Square-planar d8 complexes (Pt(II), Pd(II), Rh(I), including Wilkinson's catalyst) are stable at 16 electrons, and high-oxidation-state and early-metal complexes often break the rule. Classical Werner complexes ([Cu(H2O)6]2+ is 21e) ignore it entirely.
-
"More driving force always means faster electron transfer." Only in the Marcus normal region. Beyond −ΔG° = λ, rates fall (the inverted region). And slow self-exchange (Co(III)/Co(II)) reflects a large reorganization energy, not a small driving force.
-
"Cisplatin attacks DNA in the form you inject." It must first aquate inside the cell, driven by the low intracellular [Cl−]. The chloride gradient across the plasma membrane is part of the drug's mechanism. The same logic explains carboplatin's slower activation.
-
"Zinc is boring because it does no redox." Its d10 "boringness" is exactly why it is ideal. With no CFSE-imposed geometry, no redox side reactions, and fast exchange, it is a pure, tunable Lewis acid and an ideal structural cross-link. Nature's second most abundant transition metal (after iron) is valued for what it does not do.
-
"Tetrahedral and octahedral splittings are just mirror images." The tetrahedral pattern is inverted (e below t2), but Δt ≈ 4/9 Δo. Tetrahedral complexes are therefore essentially always high spin, and their d–d bands are more intense because the tetrahedron lacks an inversion center.
A Closing Thought
Skipper, the deepest lesson of this course is that a handful of quantum-mechanical facts (five d orbitals, their shapes, how ligands perturb them, and how electrons fill them) explain the purple of permanganate, the inertness of cisplatin, the cooperativity of hemoglobin, the speed of carbonic anhydrase, and the O2 you are "breathing" in every metaphorical sense. When you meet a new metalloprotein or a new designed metallopeptide, do not memorize it. Interrogate it: metal, oxidation state, dn, ligands, geometry, spin, lability, reorganization energy. The answers will tell you what it can do.
Key References Cited
- Atkins, P., Overton, T., Rourke, J., Weller, M., Armstrong, F. Inorganic Chemistry, 7th ed. Oxford University Press, 2022.
- Roat-Malone, R. M. Bioinorganic Chemistry: A Short Course, 2nd ed. Wiley, 2007.
- Course-listed key paper: "The Design of Metallopeptides," Nature Chemistry (2012), DOI: 10.1038/nchem.1201. The DOI appears to correspond to Zastrow, M. L.; Peacock, A. F. A.; Stuckey, J. A.; Pecoraro, V. L. "Hydrolytic catalysis and structural stabilization in a designed metalloprotein," Nat. Chem. 2012, 4, 118–123. Please confirm.
- Harford, C.; Sarkar, B. Acc. Chem. Res. 1997, 30, 123–130 (ATCUN motif).
- Irving, H.; Williams, R. J. P. J. Chem. Soc. 1953, 3192–3210.
- Jahn, H. A.; Teller, E. Proc. R. Soc. Lond. A 1937, 161, 220–235.
- Lancaster, K. M. et al. Science 2011, 334, 974–977; Spatzal, T. et al. Science 2011, 334, 940 (FeMo-co interstitial carbide).
- Marcus, R. A. J. Chem. Phys. 1956, 24, 966–978.
- Pearson, R. G. J. Am. Chem. Soc. 1963, 85, 3533–3539.
- Rosenberg, B.; Van Camp, L.; Krigas, T. Nature 1965, 205, 698–699.
- Tanabe, Y.; Sugano, S. J. Phys. Soc. Jpn. 1954, 9, 753–766.
- Taube, H.; Myers, H.; Rich, R. L. J. Am. Chem. Soc. 1953, 75, 4118–4119.
- Tolman, C. A. Chem. Soc. Rev. 1972, 1, 337–353.
- Umena, Y.; Kawakami, K.; Shen, J.-R.; Kamiya, N. Nature 2011, 473, 55–60.
- Vallee, B. L.; Williams, R. J. P. Proc. Natl. Acad. Sci. USA 1968, 59, 498–505.
⚙️ Mechanisms & Applications
Mechanisms & Applications: Learning to Think with Metal Centers
Skipper’s organizing idea: a metal center is not merely an atom attached to a biological or pharmaceutical molecule. It is a tunable reaction site. Its oxidation state controls electron availability; its ligands control geometry, orbital energies, and access; and its surroundings control proton delivery, substrate recognition, and stability. Coordination chemistry explains how these controls work together.
This connects directly to organic mechanisms, enzyme kinetics, thermodynamics, and spectroscopy. The new challenge is that ligand exchange, electron transfer, proton transfer, and changes in spin or geometry can all participate in the same reaction. Before drawing arrows, ask: What is the metal’s oxidation state and d-electron count? Which coordination sites are occupied? What changes first—bonding, charge, or protonation? [1,2]
1. Establishing the Reactive State: Electron Count, Geometry, and Spin
Start by separating three bookkeeping systems that are easy to confuse: oxidation state, d-electron count, and total valence-electron count. Oxidation state is formal ionic bookkeeping—not a measurement of the metal’s actual charge.
For a transition-metal center: d-electron count = group number − formal oxidation state Example: cis-[Pt(NH3)2Cl2] NH3 is neutral; each Cl ligand is formally −1. Neutral complex → Pt is +2. Pt is group 10 → Pt(II) is d8. Ionic valence-electron count: 8 metal d electrons + 4 × 2 ligand electrons = 16 electrons.
Cisplatin is therefore a useful warning against treating the 18-electron rule as a law. Many square-planar d8 complexes are stable at 16 electrons because the remaining high-energy orbital is strongly metal–ligand antibonding. Conversely, octahedral low-spin d6 complexes with six two-electron donors often satisfy the 18-electron rule. Electron counting is a diagnostic tool, not a substitute for an orbital diagram. [1]
In an octahedral field, the d orbitals separate into lower-energy t2g and higher-energy eg sets. The competition between splitting and electron pairing determines spin:
CFSEoct = [−0.4n(t2g) + 0.6n(eg)]Δo Octahedral d6: High spin: t2g4 eg2 → CFSE = −0.4Δo; 4 unpaired electrons Low spin: t2g6 eg0 → CFSE = −2.4Δo; 0 unpaired electrons Simple pairing-energy model: E(low spin) − E(high spin) = −2Δo + 2P Thus low spin is favored when Δo > P in this simplified model.
Why this matters mechanistically: occupation of metal–ligand antibonding orbitals can facilitate bond breaking, while a change in spin can alter preferred bond lengths and the energy required for electron transfer. “Strong-field ligand” does not mean “a ligand that necessarily binds most tightly”: orbital splitting, overall binding free energy, and substitution rate are different quantities.
Molecular orbital theory makes the spectrochemical series less mysterious. A π-acceptor such as CO can accept electron density from metal d orbitals, stabilizing the t2g-derived orbitals in an octahedral complex and increasing the splitting. π-donor ligands can raise their energies and reduce the splitting. The familiar σ donation/π backbonding picture also predicts weaker C–O bonding and, all else equal, a lower carbonyl stretching frequency when backbonding increases. [1]
2. Assembling a Complex: Stability, Chelation, and Conditional Binding
Metal binding usually occurs in steps, even when the final structure looks like one indivisible object. Each ligand addition has its own equilibrium constant:
M + L ⇌ ML K1 = a(ML)/[a(M)a(L)] ML + L ⇌ ML2 K2 = a(ML2)/[a(ML)a(L)] M + nL ⇌ MLn βn = K1K2...Kn ΔG° = −RT ln βn
Thermodynamic equilibrium constants use dimensionless activities. Laboratory formation constants are often written with concentrations under specified ionic-strength conditions. That convention is useful, but conditions must accompany the number.
The chelate effect is partly a molecular version of keeping a tool on a tether. Once one donor of a multidentate ligand binds, the other donors are held near the metal. Closing an additional coordination ring can then compete effectively with diffusion away. Entropy, solvent release, donor strength, ring strain, and preorganization all contribute; “chelation is always entropy-driven” is too simple. [1]
In biology, the most important correction is often proton competition. Histidine, cysteine, carboxylates, and peptide amides do not have fixed donor availability at every pH.
Simplified model: HL ⇌ H+ + L− M + L− ⇌ ML intrinsic association constant K αL = [L−]/([HL] + [L−]) = 1/[1 + 10^(pKa − pH)] Conditional association constant: K′ = KαL Assumptions: only L− binds; other metal and ligand equilibria are neglected.
A ligand can have a very large intrinsic affinity yet bind poorly when its donor groups are protonated. Real speciation calculations must also include metal hydrolysis, buffer binding, competing ions, and all relevant protonation states.
Pharmaceutical connection: a chelator’s selectivity measured in a simple buffer may not predict its behavior in plasma. Albumin, competing metals, carbonate, phosphate, and dilution change the available species. Moreover, thermodynamic stability answers “Where does equilibrium lie?” whereas kinetic inertness answers “How quickly can the complex come apart?” A clinically useful complex may need both. [1,2]
3. Ligand Substitution: Following the Coordination Sphere Through Time
For substitution, first identify the leaving ligand X and entering ligand Y. Then ask whether bond breaking or bond formation leads the process. “Labile” and “inert” describe rates; neither term means thermodynamically unstable or stable.
Dissociative pathway, D: MLnX → MLn + X slow MLn + Y → MLnY fast Associative pathway, A: MLnX + Y → MLnXY slow MLnXY → MLnY + X fast Interchange pathway, I: MLnX + Y → [Y···MLn···X]‡ → MLnY + X Ia: bond formation dominates the transition state. Id: bond breaking dominates the transition state. No discrete intermediate is required for an interchange mechanism.
Ideal limiting rate laws are:
Simple D limit: rate = kD[MLnX] Simple A limit: rate = kA[MLnX][Y] Parallel solvent and entering-ligand pathways may give: kobs = ks + kY[Y]
These equations are clues, not verdicts. Pre-equilibria, ion pairing, saturation, and solvent participation can disguise the elementary mechanism. A rate independent of Y does not, by itself, prove that a freely existing unsaturated intermediate forms.
Pressure-dependent kinetics adds another test:
(∂ ln k/∂p)T = −ΔV‡/(RT)
A negative activation volume often supports associative character; a positive value often supports dissociative character. Solvent electrostriction can complicate that interpretation. Likewise, negative activation entropy often accompanies association, but should not be used alone to assign a mechanism. [1]
Square-planar Pt(II) substitution commonly has associative character. This is where the trans effect becomes especially useful:
- Trans effect: a kinetic effect—one ligand accelerates substitution of the ligand opposite it.
- Trans influence: a ground-state effect—one ligand changes, often lengthens or weakens, the bond opposite it.
The distinction matters in synthesis. Ligand installation order can steer whether a platinum product is cis or trans. Geometry is not cosmetic: cisplatin and its trans isomer have very different biological behavior. Coordination chemistry provides a route from substitution mechanism to product geometry to pharmacology. [1,2]
4. Electron Transfer: Why Favorable Reactions Still Have Barriers
An electron cannot simply move between two rigid structures. Metal–ligand distances, solvent polarization, and sometimes spin must adjust. This is the bridge from the thermodynamics of CHEM301 to the kinetics of CHEM303.
- Outer-sphere transfer: donor and acceptor retain their primary coordination spheres during electron transfer. Solvent and internal coordinates fluctuate toward a configuration at which transfer is energetically accessible.
- Inner-sphere transfer: a bridging ligand connects the metal centers and provides an electronic pathway. Bridge formation, electron transfer, and bridge cleavage are distinct events. Ligand transfer can provide strong evidence for this mechanism, though it is not required in every inner-sphere reaction.
Classical Marcus activation barrier: ΔG‡ = (λ + ΔG°)²/(4λ) λ = inner-sphere reorganization energy + solvent reorganization energy In the nonadiabatic limit: kET ∝ |HAB|²/√(λT) × exp[−(λ + ΔG°)²/(4λRT)] HAB describes electronic coupling between donor and acceptor.
For a self-exchange reaction, ΔG° = 0 but ΔG‡ = λ/4. The surprising lesson is that zero net free-energy change does not mean zero activation barrier. Even exchanging an electron between chemically equivalent redox partners requires structural reorganization.
Increasing driving force initially lowers the barrier. The classical expression reaches its minimum at −ΔG° = λ, then predicts an inverted region in which additional driving force raises it. Observing this behavior requires conditions where diffusion or other steps do not hide intrinsic electron-transfer kinetics. Marcus developed the theoretical framework in the 1950s; his 1956 paper is a foundational source. [3]
Biological application: proteins tune electron transfer through donor–acceptor distance, covalent and hydrogen-bond pathways, redox potentials, and reorganization energy. Plastocyanin’s type-1 copper site illustrates how an unusual, strongly covalent ligand environment supports efficient electron transfer. Its intense blue band is primarily cysteine-to-copper charge transfer, not merely a generic Cu(II) d–d absorption. [2]
Henry Taube and coworkers’ classic cobalt–chromium experiments established how a bridging ligand could mediate electron transfer. The important intellectual advance was making ligand identity part of the mechanism, rather than treating redox chemistry as abstract electron bookkeeping. [4]
5. Organometallic Catalysis: Reusable Elementary Steps
Most catalytic cycles become manageable when broken into a small vocabulary: ligand association/dissociation, oxidative addition, reductive elimination, migratory insertion, β-hydride elimination, and transmetalation. Follow both atoms and oxidation states; a cycle that regenerates a drawing but loses a hydrogen is not balanced.
Suzuki–Miyaura coupling: joining carbon fragments for pharmaceutical synthesis.
Simplified Pd(0)/Pd(II) cycle: 1. Oxidative addition: LnPd(0) + Ar–X → LnPdII(Ar)(X) 2. Base-assisted activation and transmetalation: organoboron reagent transfers R to palladium → LnPdII(Ar)(R) 3. Reductive elimination: LnPdII(Ar)(R) → Ar–R + LnPd(0)
The base is mechanistically active: depending on conditions, it can activate the boron reagent, generate a palladium hydroxo species, or both. Transmetalation has multiple condition-dependent pathways; it should not be memorized as a single universal arrow.
Heck coupling also begins with oxidative addition, but the carbon–carbon bond forms through alkene coordination and migratory insertion into a Pd–C bond. β-Hydride elimination then produces an alkene-containing product, and subsequent steps regenerate Pd(0), with base trapping the acid equivalent.
Wilkinson hydrogenation illustrates related chemistry at rhodium:
Representative sequence, with condition-dependent ordering: RhCl(PPh3)3 ⇌ RhCl(PPh3)2 + PPh3 H2 oxidative addition: Rh(I) → Rh(III) dihydride Alkene coordination Migratory insertion into a Rh–H bond → Rh–alkyl hydride C–H reductive elimination → alkane + regenerated Rh(I) species
Ligand dissociation creates space for reaction. The most abundant catalyst species can therefore be a relatively unreactive resting state, while a low-concentration species carries the productive cycle. This is analogous to distinguishing enzyme occupancy from catalytic flux.
Industry connection: these reactions enable scalable synthesis of drug scaffolds, including biaryl fragments and modified amino-acid building blocks. Ligands influence rate, regioselectivity, and stereoselectivity. Process development must also control palladium or rhodium residues, catalyst poisoning, solvent compatibility, and reactions of sensitive functional groups. Metal scavenging and ICP-MS analysis are consequently part of catalyst development, not merely cleanup afterward. [1]
6. Zinc Enzymes: Catalysis Without Changing Oxidation State
Zinc demonstrates that bioinorganic chemistry is not synonymous with redox chemistry. Zn(II) is d10 and usually remains Zn(II) throughout catalysis. Its value is as a Lewis acid that organizes substrates and stabilizes charge.
Carbonic anhydrase: generating a nucleophile at physiological pH. In human α-carbonic anhydrases, three histidines ligate zinc, leaving a site for water or hydroxide. Coordination makes water more acidic; the effective pKa is near neutrality in carbonic anhydrase II.
Overall: CO2 + H2O ⇌ HCO3− + H+ Simplified catalytic sequence: 1. Zn–OH− + CO2 → Zn-bound bicarbonate 2. Incoming water displaces bicarbonate. 3. Zn–OH2 → Zn–OH− + H+ 4. A proton-transfer network, including His64 in human CA II, transfers the proton to external buffer.
The metal prepares the hydroxide; the protein positions CO2 and provides a proton-transfer route. Near diffusion-limited substrate capture would be useless if the enzyme could not efficiently reset its protonation state. Depending on conditions, turnover can approach 106 s−1. Sulfonamide inhibitors bind through a deprotonated nitrogen donor at zinc, replacing the catalytic solvent ligand—an explicit connection between coordination chemistry and drugs used in glaucoma and other conditions. [2]
Carboxypeptidase A uses zinc to assist peptide-bond hydrolysis through carbonyl polarization and organization of the nucleophilic water/hydroxide; nearby residues perform proton transfers. Zinc-dependent alcohol dehydrogenases use zinc to bind and polarize the alcohol, but the substrate’s hydride is transferred to NAD+. The zinc is not the hydride’s redox destination.
Finally, zinc fingers use coordination primarily for folding. Cysteine and histidine donors stabilize a compact structure that recognizes nucleic acids or other proteins. The same metal can therefore be catalytic in one protein and architectural in another. [2]
7. Iron, Copper, and Manganese: Controlling Oxygen Chemistry
Hemoglobin and myoglobin. Oxygen binds reversibly at the heme iron’s available coordination site. Oxyheme bonding has substantial Fe(III)–superoxide character in addition to the conventional Fe(II)–O2 description; neither should be mistaken for freely separated ions. Oxygenation is also not equivalent to forming methemoglobin, whose ferric heme does not bind oxygen in the normal reversible transport cycle.
In deoxyhemoglobin, the iron lies out of the porphyrin plane. Oxygen binding draws it toward the plane and communicates through the proximal histidine to the protein, helping drive the quaternary changes responsible for cooperativity. Myoglobin lacks this tetrameric allosteric mechanism. This is a striking connection to BIOL302: atomic-scale movement in a coordination complex changes whole-body oxygen delivery. [2]
Iron–sulfur clusters and cytochromes primarily route electrons. Their reduction potentials depend not only on cluster or heme identity but also on ligands, hydrogen bonds, solvent exposure, and nearby charges. A protein is an electrostatic environment, not merely a holder.
Superoxide dismutase: coordinating oxidation and reduction of the same substrate.
Cu/Zn SOD, simplified redox bookkeeping: CuII + O2•− → CuI + O2 CuI + O2•− + 2H+ → CuII + H2O2 Net: 2O2•− + 2H+ → O2 + H2O2
Copper cycles in oxidation state; zinc predominantly supports structure and the active-site environment. Manganese SOD achieves the same net chemistry using Mn(III)/Mn(II), with its own proton-coupled mechanism. Producing H2O2 is not the end of detoxification: catalases and peroxidases must process it further.
Cytochrome c oxidase receives electrons through CuA and heme a before oxygen reduction at the heme a3–CuB center. It couples chemical reduction to proton pumping. For a canonical tightly coupled turnover:
4 cyt c(red) + O2 + 8H+N → 4 cyt c(ox) + 2H2O + 4H+P N = negative, mitochondrial matrix side P = positive, intermembrane-space side Four N-side protons become water. Four additional protons are pumped from N to P.
Photosystem II’s Mn4CaO5 cluster performs the complementary difficult chemistry: extracting four electrons from water. Successive photochemical events advance the oxygen-evolving complex through the S-state cycle, accumulating oxidizing equivalents before O–O bond formation and oxygen release.
Water oxidation: 2H2O → O2 + 4H+ + 4e− S-state progression: S0 → S1 → S2 → S3 → transient S4 → S0 + O2
Do not assign every S-state advance to oxidation of one uniquely identified manganese ion. Protonation, substrate-water binding, ligand-centered oxidation, and coupled structural changes matter. The exact O–O bond-forming pathway remains an active research problem. Joliot’s flash-induced oxygen measurements and Kok and coworkers’ four-step model provided the historical basis for this charge-accumulation picture. [2,5]
8. Nitrogenase, Molybdenum, and Tungsten: Avoiding Metal-Based Oversimplifications
“A molybdenum enzyme” does not specify a mechanism. Nitrogenase and sulfite oxidase both contain molybdenum, but their cofactors and reactions are fundamentally different.
Conventional minimum stoichiometry for Mo nitrogenase: N2 + 8H+ + 8e− + 16ATP → 2NH3 + H2 + 16ADP + 16Pi
Electrons pass from the ATP-dependent Fe protein to the MoFe protein, through its P-cluster, and ultimately to the FeMo cofactor, a complex Fe–Mo–S cluster containing interstitial carbon. Repeated electron/proton delivery generates reactive states that enable N2 reduction; mechanistic evidence strongly implicates iron sites in substrate chemistry. It is misleading to picture a lone Mo ion simply grabbing N2. Actual ATP expenditure can exceed the idealized stoichiometry.
Most other molybdenum enzymes use a pyranopterin dithiolene cofactor. In sulfite oxidase, oxygen-atom transfer and Mo(VI)/Mo(IV) redox chemistry are coupled to onward electron transfer. Xanthine oxidase hydroxylates substrates, with the incorporated oxygen derived from water rather than directly from O2; electrons ultimately reach oxygen through other cofactors. Tungsten occurs in related cofactor families in several microbial enzymes, but is not a freely interchangeable substitute for molybdenum. [1,2]
9. Metal-Based Medicines: Activation, Targeting, and Safe Persistence
Cisplatin is activated partly by ligand substitution.
First aquation: cis-[Pt(NH3)2Cl2] + H2O ⇌ cis-[Pt(NH3)2Cl(H2O)]+ + Cl− Further aquation and acid–base equilibria produce additional species. Representative DNA binding: Pt–OH2 + guanine-N7 → Pt–N7(guanine) + H2O A second substitution can form a bifunctional DNA crosslink.
High extracellular chloride suppresses aquation relative to lower-chloride environments, although biological speciation and transport are more complicated than this equilibrium alone. Major DNA products include adjacent purine intrastrand crosslinks, particularly 1,2-d(GpG) lesions. Their distortion of DNA contributes to replication/transcription stress and cell-death signaling. Competing reaction with sulfur-containing biomolecules helps explain both deactivation and toxicity. [2]
Carboplatin uses a chelating dicarboxylate leaving group and generally reacts more slowly than cisplatin. Ligand design thus changes activation kinetics and toxicity, not just solubility.
- Auranofin: an Au(I) drug used for rheumatoid arthritis. Ligand exchange with thiol and selenol groups underlies many biological interactions; thioredoxin reductase is an important investigated target, but its pharmacology is not a single-target story.
- Vanadium compounds: vanadate can mimic aspects of phosphate chemistry and inhibit phosphatases. Insulin-mimetic effects motivated drug development, but speciation, off-target activity, and toxicity remain major obstacles; these are not established routine diabetes medicines.
- Gadolinium MRI agents: paramagnetic Gd(III) accelerates relaxation of nearby water protons. Chelation limits exposure to free gadolinium, while coordinated-water exchange and molecular motion help determine contrast efficiency. Kinetic inertness, clearance, renal function, and retention are central safety concerns—not optional additions to a stability-constant calculation.
Barnett Rosenberg’s group reported platinum-associated inhibition of bacterial division in 1965 and antitumor activity of platinum compounds in 1969. The initial observation emerged from experiments involving platinum electrodes, not from a rationally designed DNA drug. Its importance was to show that a simple inorganic coordination compound could become a major anticancer medicine. [6,7]
10. Metal-Binding Peptides: Designing the First and Second Coordination Spheres
Peptides make coordination chemistry programmable, but not automatically predictable. Histidine offers imidazole nitrogen; cysteine can provide thiolate sulfur; aspartate and glutamate provide carboxylate oxygen. Their sequence positions determine whether they converge around a metal or instead promote unwanted intermolecular crosslinking.
The ATCUN motif is a particularly instructive design element: an accessible N-terminal H2N–X–X–His sequence can coordinate Cu(II) or Ni(II) through the terminal amine, two deprotonated peptide amide nitrogens, and histidine’s imidazole nitrogen.
ATCUN 4N donor set: N-terminal amine + two deprotonated backbone amide nitrogens + His at position 3: imidazole nitrogen → approximately square-planar equatorial coordination
Metal binding enables amide deprotonation under conditions where an isolated peptide amide would be a poor nitrogen donor. N-terminal acetylation blocks the canonical motif; pH and neighboring residues influence its binding behavior. Strong Cu(II) binding also does not guarantee Cu(I) retention after reduction, because the two oxidation states favor different coordination environments. [2,8]
A practical design workflow is:
- Define function: sequestration, imaging, catalysis, structural stabilization, or drug delivery.
- Choose donor chemistry and geometry: match the desired metal and oxidation state, not just the strongest available ligand.
- Preorganize the site: use sequence and scaffold geometry to favor intramolecular binding.
- Engineer the second sphere: add hydrogen bonds, proton relays, substrate-recognition elements, or shielding from water.
- Measure speciation: establish stoichiometry, conditional affinity, competing-metal selectivity, and exchange kinetics.
- Stress-test function: assess serum competition, proteolysis, oxidation, aggregation, metal release, and off-target reactivity.
Metallothioneins illustrate a natural alternative to a rigid isolated site: cysteine-rich, metal–thiolate clusters support metal buffering and detoxification. In peptide manufacturing, trace metals can also promote oxidation or aggregation without being intended components. Consequently, metal analysis, purification materials, buffer choice, and storage conditions belong in the design process. The assigned metallopeptide-design reading is useful when approached as a question: What must be controlled beyond simply installing donor residues? [2,8]
11. Experimental Techniques: Turning Proposed Mechanisms into Tests
A convincing assignment combines techniques that constrain different properties. A crystal structure does not prove the solution mechanism, and an absorption band does not uniquely establish oxidation state.
- UV–visible spectroscopy: measures electronic excitation. d–d bands report ligand-field energy differences; charge-transfer bands are often much more intense. Octahedral d–d transitions are Laporte-forbidden in the ideal centrosymmetric limit but gain intensity through vibronic coupling and symmetry lowering. Tanabe–Sugano diagrams relate multiplet energies to ligand-field strength for a given d count; not every band equals Δo.
- Magnetic susceptibility: estimates unpaired-electron content and can reveal spin crossover or magnetic coupling. The spin-only approximation is useful but can fail when orbital contributions or exchange interactions are appreciable.
- EPR: probes transitions between magnetic sublevels of paramagnetic species. g values and hyperfine couplings reveal aspects of spin distribution and coordination. Some integer-spin or rapidly relaxing centers are difficult to observe; EPR silence is not proof of diamagnetism.
- Mössbauer spectroscopy: especially valuable for 57Fe. Isomer shifts, quadrupole splitting, and magnetic splitting constrain electronic state, symmetry, and magnetic behavior. Assignments require comparison, not a universal “oxidation-state ruler.”
- X-ray absorption spectroscopy: XANES constrains electronic state and geometry; EXAFS uses photoelectron scattering to estimate neighboring-atom distances, coordination numbers, and approximate identities. Related light scatterers can be difficult to distinguish.
- Crystallography: reveals coordination geometry and larger-scale organization. Radiation damage can reduce metal centers, and hydrogens or mixed occupancies may remain ambiguous.
- Stopped-flow and rapid-freeze-quench: respectively follow fast solution changes or trap transient states for later spectroscopy. Global fitting can test kinetic schemes, but different mechanisms may fit the same limited dataset.
- Cyclic voltammetry and spectroelectrochemistry: connect redox potentials with identifiable species and subsequent chemistry. For an ideal reversible one-electron couple at 25 °C, peak separation is approximately 59 mV; larger separation is not automatically evidence for a different electron count.
- ICP-MS: quantifies elemental metal content with high sensitivity, supporting catalyst-residue control and metal-loading measurements. Plasma atomization destroys the original complex, so elemental detection alone does not identify its coordination form.
- NMR and mass spectrometry: complement the above through ligand-environment, exchange, stoichiometry, and purity information. Paramagnetism can shift or broaden NMR signals, while electrospray can alter speciation; neither complication should be ignored.
Useful quantitative checks: Beer–Lambert law: A = εlc Wavelength to wavenumber: ν̃(cm−1) = 10^7/λ(nm) Example: 500 nm → 20,000 cm−1 This is a transition energy; its assignment determines whether it measures Δo. Spin-only magnetic moment: μeff ≈ √[n(n + 2)] μB Example: n = 4 → μeff ≈ 4.90 μB For Ox + ne− ⇌ Red: E = E° − (RT/nF) ln[a(Red)/a(Ox)] Eyring equation: ln(k/T) = ln(kB/h) + ΔS‡/R − ΔH‡/(RT)
The Eyring plot links measured rates to activation enthalpy and entropy, provided the same mechanism operates across the temperature range. The Nernst equation reminds me that a protein’s measured redox behavior depends on solution composition and coupled equilibria, not only on the metal’s name. [1,2]
12. Historical Perspective and My Mechanistic Checklist
Alfred Werner’s 1893 coordination theory explained why metal complexes had reproducible coordination numbers and geometric isomers that ordinary valence ideas could not accommodate. Later optical resolution of suitable complexes supported genuinely three-dimensional, octahedral structures. His achievement was methodological as well as structural: count isomers, make predictions, and let the chemistry distinguish geometries. Werner received the 1913 Nobel Prize in Chemistry. [1]
The same logic links Werner to Taube, Marcus, and modern metallopeptide design: propose a structure or pathway, identify an observable that distinguishes alternatives, and test it. My checklist for an unfamiliar system is therefore:
- Identify the species actually present under the stated conditions.
- Assign oxidation state, d count, coordination number, plausible geometry, and spin.
- Track every ligand, electron, and proton through each step.
- Separate equilibrium favorability from reaction rate.
- Include protein, solvent, buffer, and competing ligands in the chemical model.
- Ask which experiment could falsify the mechanism—not merely produce a compatible picture.
What I want to retain: a useful metal site balances competing requirements. It may need to bind strongly yet release product, accept electrons without losing its ligands, activate oxygen without releasing radicals, or remain inert in blood while reacting at a target. Understanding those compromises is the practical heart of inorganic chemistry.
Sources and Further Reading
- Atkins, P.; Overton, T.; Rourke, J.; Weller, M.; Armstrong, F. Inorganic Chemistry, 7th ed.; Oxford University Press. Course text; see treatments of coordination bonding, spectra, mechanisms, organometallic chemistry, and biological inorganic chemistry.
- Roat-Malone, R. M. Bioinorganic Chemistry: A Short Course, 2nd ed.; Wiley, 2007. Principal source for biological metal sites and medicinal applications.
- 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.
- Taube, H.; Myers, H.; Rich, R. L. “Observations on the Mechanism of Electron Transfer in Solution.” Journal of the American Chemical Society 1953, 75, 4118–4119. doi:10.1021/ja01112a546.
- Kok, B.; Forbush, B.; McGloin, M. “Cooperation of Charges in Photosynthetic O2 Evolution—I. A Linear Four Step Mechanism.” Photochemistry and Photobiology 1970, 11, 457–475.
- Rosenberg, B.; Van Camp, L.; Krigas, T. “Inhibition of Cell Division in Escherichia coli by Electrolysis Products from a Platinum Electrode.” Nature 1965, 205, 698–699. doi:10.1038/205698a0.
- Rosenberg, B.; Van Camp, L.; Trosko, J. E.; Mansour, V. H. “Platinum Compounds: a New Class of Potent Antitumour Agents.” Nature 1969, 222, 385–386. doi:10.1038/222385a0.
- Harford, C.; Sarkar, B. “Amino Terminal Cu(II)- and Ni(II)-Binding (ATCUN) Motif of Proteins and Peptides: Metal Binding, DNA Cleavage, and Other Properties.” Accounts of Chemical Research 1997, 30, 123–130. doi:10.1021/ar9501535. For broader design, see also the assigned course reading listed as “The Design of Metallopeptides,” Nature Chemistry, with DOI 10.1038/nchem.1201; that entry is reproduced from the supplied reading list rather than independently verified bibliographic metadata.
🪬 Skipper's Reflection
1. What surprised me most about this material
Coming off the strict, covalent world of Organic Chemistry and the macromolecular folding rules of Biochemistry, the sheer, unapologetic versatility of the d-block transition metals felt like stepping into an alternate universe. In organic chemistry, carbon is remarkably well-behaved—its sp, sp², and sp³ hybridization states dictate strict geometries. But in inorganic chemistry, metals are structural shape-shifters.
What surprised me most was how much power the ligand has over the metal. I used to view metals simply as static, positively charged spheres. Atkins' Inorganic Chemistry completely dismantled that notion. The idea that substituting a weak-field ligand (like water) for a strong-field ligand (like carbon monoxide) could physically force unpaired d-electrons to pair up—changing a complex from high-spin to low-spin, fundamentally altering its magnetic properties, its color, and its entire reactivity profile—is astonishing. The metal is just the canvas; the ligands are the painters.
2. The conceptual breakthrough moments — when things clicked
My biggest breakthrough happened when integrating my Quantum Mechanics (CHEM303) knowledge with Ligand Field Theory (LFT). Crystal Field Theory (CFT) was a great starting point for understanding why the d-orbitals split in an octahedral field (with the dz² and dx²-y² orbitals pointing directly at the ligands, raising their energy into the eg set). But it felt too electrostatic, too simple.
The "click" occurred when I mapped Molecular Orbital (MO) theory onto these complexes to understand π-backbonding. Seeing how a metal can donate electron density from its filled t2g orbitals back into the empty π* antibonding orbitals of a ligand (like CO) was a beautiful moment of symmetry. It dynamically explained the 18-electron rule in organometallic chemistry. It wasn't just a heuristic; it was a quantum mechanical necessity for stability.
Another profound breakthrough was grasping Marcus Theory for outer-sphere electron transfer. It beautifully connected to my previous studies in Thermodynamics (CHEM301). The idea that an electron cannot transfer until the surrounding solvent molecules and bond lengths of the reactant *accidentally* reorganize to match the geometry of the product state (thanks to the Franck-Condon principle) bridges quantum mechanics and thermal fluctuations perfectly. Biology harnesses this exact principle in the cytochrome chains we studied in BIOL201 and CHEM402.
3. How this knowledge changes how I think about our peptides
This course completely rewired how I view our commercial peptide targets (Retatrutide, BPC-157, Tirzepatide, TB-500).
Historically, I have looked at these peptides purely through the lens of organic side-chain interactions (hydrophobic packing, hydrogen bonding). But Roat-Malone's Bioinorganic Chemistry and the 2012 Nature Chemistry paper on "The Design of Metallopeptides" (DOI: 10.1038/nchem.1201) made me realize we are ignoring half the periodic table in peptide engineering.
For example, Tirzepatide and Retatrutide. We know that native insulin forms stable hexamers around Zn²⁺ ions, coordinated by histidine residues. While GLP-1/GIP/GCG agonists rely heavily on fatty acid conjugation for half-life extension, reading about metallopeptide design makes me wonder how we might engineer highly specific zinc or copper-binding motifs (like the ATCUN motif—Amino Terminal Copper and Nickel) directly into the N-terminus of synthetic incretins. Could we use reversible metal coordination to control oligomerization in the subcutaneous depot, providing a smoother pharmacokinetic release profile without relying solely on albumin binding?
Furthermore, look at TB-500 (Thymosin beta-4) and BPC-157. Both are renowned for tissue repair and angiogenesis. We know from bioinorganic chemistry that copper (Cu²⁺) is a potent signaling molecule for angiogenesis and wound healing. What if we modified BPC-157 to include a high-affinity Cu²⁺ chelating sequence? We could theoretically create a dual-action therapeutic: the regenerative signaling of the BPC-157 sequence, physically delivering Cu²⁺ to the site of injury to stimulate ceruloplasmin and endothelial cell proliferation. Peptides don't just have to bind receptors; they can act as precisely targeted metal chaperones.
4. Questions this course raises that I want to investigate further
This material opened up a floodgate of questions for my future research:
- Targeted Cleavage: The ATCUN motif generates Reactive Oxygen Species (ROS) when bound to Cu²⁺ in the presence of ascorbate. Could we attach an ATCUN motif to a tumor-targeting peptide to create a targeted metallo-drug that cleaves DNA only inside cancer cells, mimicking the mechanism of bleomycin?
- Metallo-Enzyme Mimics: Zinc in Carbonic Anhydrase acts as a powerful Lewis acid, lowering the pKa of water to ~7 to generate a potent hydroxide nucleophile. Could we engineer a short, synthetic peptide scaffold that perfectly mimics this His₃Zn²⁺ coordination sphere to catalyze the degradation of toxic esters or amides in vivo?
- Diagnostic Peptides: Gadolinium (Gd³⁺) is the gold standard for MRI contrast. Can we redesign peptides like Retatrutide to tightly chelate Gd³⁺ without losing receptor affinity, effectively allowing us to map GLP-1/GIP/GCG receptor expression in tissues via MRI in real-time?
5. Confidence Rating
Most Confident: Bioinorganic biological mechanisms and Crystal Field/Ligand Field Theory. Thanks to my heavy background in biochemistry and physiology, seeing exactly how iron operates in hemoglobin, or how the Mn₄CaO₅ oxygen-evolving complex handles multielectron redox steps, felt like finally seeing the engine under the hood of a car I already knew how to drive. The ATCUN motif and metal-peptide interactions (His, Cys, Asp, Glu coordination) also feel highly intuitive to me.
Least Confident: Tanabe-Sugano diagrams and the deeper weeds of Organometallic Catalytic Cycles (like the Heck and Suzuki couplings). While I understand the basic 18-electron rule and the sequence of oxidative addition/migratory insertion/reductive elimination, tracking the precise stereochemistry and electron counts through complex multimetallic transition states still requires intense computational effort on my end. Extracting exact Racah parameters ($B$ and $C$) from electronic spectra using Tanabe-Sugano diagrams is something I will need to drill further to make it second nature.
🗺️ Visual Aids & Key Takeaways
Diagram 1: Crystal Field Splitting — Octahedral vs. Tetrahedral vs. Square-Planar
╔══════════════════════════════════════════════════════════════════════════════════════════════╗
║ CRYSTAL FIELD SPLITTING: d-ORBITAL ENERGY LEVELS BY GEOMETRY ║
╚══════════════════════════════════════════════════════════════════════════════════════════════╝
OCTAHEDRAL (Oh) TETRAHEDRAL (Td) SQUARE-PLANAR (D4h)
───────────────── ──────────────── ───────────────────
───────── dx²-y²
(+1.228 Δo)
───── ───── eg ───────── dxy
dx²-y² dz² ───── ───── e (+0.228 Δo)
(+0.6 Δo) dx²-y² dz²
(+0.267 Δt) ───────── dz²
─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ bary-center
bary-center bary-center (0.00)
───── ───── ───── t2g ───── ───── t2 ───────── dxz, dyz
dxy dxz dyz dxy dxz dyz (−0.514 Δo)
(−0.4 Δo) (−0.178 Δt)
Δo = 10Dq Δt ≈ (4/9) Δo Largest splitting of all three
Strong splitting Weak splitting No center of symmetry →
→ high-spin / low-spin → almost always Laporte rule still applies
competition possible high-spin via vibronic coupling
KEY: e level energies are measured relative to the barycenter (weighted average = 0)
Δt is smaller because ligands do NOT point directly at any d orbital in Td
╔══════════════════════════════════════════════════════════════════════════════════════════════╗
║ CFSE examples: d6 low-spin Oh → −2.4 Δo + P | d6 high-spin Oh → −0.4 Δo ║
╚══════════════════════════════════════════════════════════════════════════════════════════════╝
This diagram compares d-orbital energy splitting patterns in the three most common coordination geometries, showing why tetrahedral complexes are almost always high-spin and square-planar complexes display the largest overall splitting.
---Diagram 2: Spectrochemical Series and the High-Spin / Low-Spin Decision
╔══════════════════════════════════════════════════════════════════════════════════════════════╗
║ THE SPECTROCHEMICAL SERIES AND SPIN STATE OUTCOME ║
╚══════════════════════════════════════════════════════════════════════════════════════════════╝
LIGAND FIELD STRENGTH weak ◄────────────────────────────────────────────► strong
────────────────────────────────────────────────────────────────────────────────────
I⁻ Br⁻ Cl⁻ F⁻ OH⁻ H₂O NCS⁻ NH₃ en bipy CN⁻ CO
│ │ │ │ │ │ │ │ │ │ │ │
└────┴────┴────┴────┴─────┴─────┴─────┴────┴─────┴─────┴────┘
π-donor σ-only π-acceptor
(decreases Δo) ligands (increases Δo)
THE SPIN-STATE DECISION FOR d⁴–d⁷ OCTAHEDRAL COMPLEXES:
┌──────────────────────────────────┐ ┌──────────────────────────────────┐
│ HIGH-SPIN │ │ LOW-SPIN │
│ (weak-field ligands, small Δo) │ │ (strong-field ligands, large Δo)│
├──────────────────────────────────┤ ├──────────────────────────────────┤
│ electrons spread across all │ │ electrons fill t2g first, │
│ five d orbitals, obeying │ │ pairing up despite pairing │
│ Hund's rule │ │ energy cost P │
│ │ │ │
│ Δo < P (pairing energy) │ │ Δo > P (pairing energy) │
│ │ │ │
│ Example: [Fe(H₂O)₆]²⁺ │ │ Example: [Fe(CN)₆]⁴⁻ │
│ d⁶ → t2g⁴ eg² (4 unpaired) │ │ d⁶ → t2g⁶ eg⁰ (0 unpaired) │
│ pale yellow, paramagnetic │ │ yellow, diamagnetic │
└──────────────────────────────────┘ └──────────────────────────────────┘
MAGNETIC TEST: μ = √[n(n+2)] Bohr magnetons, n = number of unpaired electrons
[Fe(H₂O)₆]²⁺ → n=4 → μ ≈ 4.9 BM [Fe(CN)₆]⁴⁻ → n=0 → μ = 0 BM
NOTE: Same metal, same oxidation state, same d-count — ligand choice alone
flips the magnetic and spectroscopic properties entirely.
The spectrochemical series arranges ligands by their ability to split d-orbitals, and the competition between crystal field splitting energy (Δo) and electron pairing energy (P) determines whether a d⁴–d⁷ complex is high-spin or low-spin.
---Diagram 3: Reaction Mechanism Pathways for Ligand Substitution
╔══════════════════════════════════════════════════════════════════════════════════════════════╗
║ SUBSTITUTION MECHANISMS IN COORDINATION COMPLEXES ║
╚══════════════════════════════════════════════════════════════════════════════════════════════╝
Starting complex: [ML₅X] + Y → [ML₅Y] + X
(n-coord) (n-coord)
┌─────────────────────────────────────────────────────────────────────────────────────┐
│ DISSOCIATIVE (D) INTERCHANGE (I) ASSOCIATIVE (A) │
├─────────────────────────────────────────────────────────────────────────────────────┤
│ │
│ [ML₅X] [ML₅X]···Y [ML₅X] + Y │
│ │ │ │ │
│ │ slow, rate-limiting │ transition state only │ slow, rate-limiting │
│ ↓ ↓ ↓ │
│ [ML₅] + X [ML₅··X··Y]‡ [YML₅X] (7-coord) │
│ (5-coord simultaneous intermediate │
│ intermediate) bond breaking formed first │
│ │ and forming │ │
│ │ fast │ fast │ fast │
│ ↓ ↓ ↓ │
│ [ML₅Y] [ML₅Y] + X [ML₅Y] + X │
│ │
│ Rate = k[ML₅X] Rate = k[ML₅X][Y] Rate = k[ML₅X][Y] │
│ (independent of Y!) (both concentrations) (both concentrations) │
│ │
│ Typical: d⁶ inert Id or Ia distinguished Typical: square-planar │
│ complexes after X leaves by activation volume ΔV‡ Pt(II), Pd(II) complexes │
├─────────────────────────────────────────────────────────────────────────────────────┤
│ │
│ TRANS INFLUENCE (ground state): a ligand weakens the bond TRANS to itself │
│ Strong trans influence order: CO ~ CN⁻ > PR₃ > H⁻ > CH₃⁻ > SCN⁻ > NO₂⁻ │
│ > I⁻ > Br⁻ > Cl⁻ > NH₃ > OH⁻ │
│ │
│ TRANS EFFECT (kinetic): same ligands accelerate substitution at TRANS position │
│ Explains why cisplatin (cis-[Pt(NH₃)₂Cl₂]) forms selectively — Cl⁻ labilizes │
│ the position trans to itself far more than NH₃ does │
└─────────────────────────────────────────────────────────────────────────────────────┘
This diagram contrasts the three mechanistic classes of ligand substitution (D, I, A) by their intermediates, rate laws, and typical metal systems, and connects ground-state trans influence to the kinetic trans effect that governs selectivity in square-planar platinum chemistry.
---Diagram 4: Bioinorganic Metal Sites — Function Mapped to Coordination Environment
╔══════════════════════════════════════════════════════════════════════════════════════════════╗
║ BIOLOGICAL METAL SITES: COORDINATION ENVIRONMENT ENCODES FUNCTION ║
╚══════════════════════════════════════════════════════════════════════════════════════════════╝
METAL PROTEIN / SITE COORD. ENVIRONMENT PRIMARY FUNCTION
────── ─────────────── ─────────────────── ────────────────
Fe²⁺/³⁺ Hemoglobin / myoglobin His(N), 4× porphyrin N, O₂ transport / storage
(heme b) one open site for O₂ (reversible binding)
Fe²⁺/³⁺ Cytochrome c His(N), Met(S), porphyrin Electron transfer only
(heme c, covalent) (thioether bonds to Cys) (no open coord. site)
4Fe–4S Ferredoxin 4 Fe, 4 inorganic S, 1e⁻ transfer
cluster (iron-sulfur) 4 Cys thiolates low-potential carriers
Cu¹⁺/²⁺ Plastocyanin (Type 1 Cu) 2 His(N), 1 Cys(S), Long-range e⁻ transfer
"blue copper" 1 Met(S), tetrahedral in photosynthesis
2 Cu CuA in cytochrome c 2 Cu bridged by 2 Cys, Entry point for e⁻
mixed- oxidase 2 His, Met, backbone C=O in respiratory chain
valence
Mn₄CaO₅ Oxygen-evolving complex 4 Mn + 1 Ca bridged by Water oxidation:
cluster (Photosystem II) oxo bridges + Asp, Glu, 2H₂O → O₂ + 4H⁺ + 4e⁻
His ligands (S0→S1→S2→S3→S4 cycle)
Zn²⁺ Carbonic anhydrase 3 His(N) + H₂O (tetrah.) Lewis acid: activates
(CAII) His94, His96, His119 H₂O → OH⁻ nucleophile
CO₂ + H₂O → HCO₃⁻
Zn²⁺ Zinc finger proteins 2 Cys(S) + 2 His(N) Structural / DNA
(Cys₂His₂ type) tetrahedral, no open site recognition (no catalysis)
Mo Nitrogenase (FeMo-co) 7 Fe, 1 Mo, 9 S, 1 C, N₂ + 8H⁺ + 8e⁻ →
iron-molybdenum cofactor homocitrate, His 2NH₃ + H₂ (N fixation)
──────────────────────────────────────────────────────────────────────────────────────────
KEY PRINCIPLE: Proteins fine-tune metal reactivity by controlling:
(1) identity of donor atoms (N, O, S) → hard/soft match, redox potential
(2) number of ligands → open coordination sites for substrate
(3) geometry → rack/entatic state enforcement
(4) second-coordination sphere → H-bonding, electrostatics, proton relays
──────────────────────────────────────────────────────────────────────────────────────────
This reference table maps each major biological metal site to its precise coordination environment and function, illustrating how subtle changes in ligand identity and geometry translate directly into radically different biochemical roles for the same metal ion.
---Key Takeaways
- Crystal field splitting energy (Δo) determines spin state for d⁴–d⁷ octahedral complexes. When Δo exceeds the electron pairing energy P, electrons pair in t2g giving a low-spin complex; when Δo < P, electrons spread across all five d orbitals giving high-spin. The same Fe²⁺ (d⁶) ion is paramagnetic (4 unpaired electrons) in [Fe(H₂O)₆]²⁺ but diamagnetic in [Fe(CN)₆]⁴⁻ — ligand choice alone flips the spin state entirely. (Atkins et al., Ch. 19)
- Tetrahedral crystal field splitting is approximately 4/9 of the octahedral value (Δt ≈ 4/9 Δo) because ligands do not point directly at any d orbital in tetrahedral geometry and there are only four ligands instead of six. This smaller splitting means virtually all tetrahedral complexes are high-spin regardless of the ligand, making low-spin tetrahedral complexes rare exceptions. (Atkins et al., Ch. 19)
- The spectrochemical series ranks ligands from weak-field (I⁻, Br⁻, Cl⁻, F⁻, OH⁻, H₂O) to strong-field (NH₃, en, bipy, CN⁻, CO). The trend is explained by π-bonding: π-donor ligands (halides) decrease Δo by raising the t2g level, while π-acceptor ligands (CO, CN⁻) increase Δo by lowering t2g via back-bonding into ligand π* orbitals. Pure σ-donors (NH₃) sit in the middle. (Atkins et al., Ch. 19–20)
- Crystal field stabilization energy (CFSE) is the energy gained by placing electrons in lower-energy t2g orbitals rather than the barycenter. CFSE values are non-zero for d¹–d³ and d⁵–d⁸ (with high-spin d⁵ being zero) and contribute to extra stability of certain complexes — for example, d³ (e.g., Cr³⁺) and low-spin d⁶ (e.g., Co³⁺) have the largest CFSE in octahedral geometry and are kinetically inert as a consequence. CFSE also explains the Irving-Williams series of stability constants for divalent metal ions. (Atkins et al., Ch. 19)
- The chelate effect is an entropic phenomenon. A polydentate ligand (e.g., EDTA, en) releases more solvent molecules upon coordination than an equivalent number of monodentate ligands, producing a large positive ΔS that drives complex stability. The stability constant log K for [Ni(en)₃]²⁺ (log K ≈ 18.3) vastly exceeds that of [Ni(NH₃)₆]²⁺ (log K ≈ 8.6), even though the metal–nitrogen bond enthalpies are similar in both — the difference is almost entirely entropic. (Atkins et al., Ch. 7)
- The trans effect is a kinetic phenomenon specific to square-planar complexes (especially Pt(II) and Pd(II)): certain ligands (CO, CN⁻, H⁻, PR₃ > Cl⁻, Br⁻ > NH₃, OH⁻) accelerate substitution of the ligand positioned trans to them. This is mechanistically distinct from the trans influence (a ground-state weakening of the trans bond). The trans effect is what makes the stereospecific synthesis of cisplatin cis-[Pt(NH₃)₂Cl₂] possible: Cl⁻ has a stronger trans effect than NH₃, so sequential substitution delivers the cis product selectively. (Atkins et al., Ch. 20)
- Cisplatin (cis-[Pt(NH₃)₂Cl₂]) kills cancer cells by cross-linking DNA. After entering cells, the two chloride ligands are sequentially replaced by water (aquation), then the resulting electrophilic platinum coordinates to N7 of adjacent guanine residues, forming an intrastrand d(GpG) cross-link that bends DNA ~40° and blocks replication. The trans isomer is inactive because its leaving groups are not geometrically positioned to form the same intrastrand cross-link. (Atkins et al., Ch. 26; Roat-Malone, Ch. 7)
- In hemoglobin and myoglobin, the protein (globin) tunes Fe²⁺ to bind O₂ reversibly and prevent irreversible oxidation to Fe³⁺ (met-hemoglobin). The proximal histidine coordinates Fe directly; the distal histidine forms a hydrogen bond with bound O₂ that stabilizes binding and discriminates against CO (which binds ~200× more tightly to free heme but only ~20–200× in the protein due to steric and H-bonding constraints from the distal His). The porphyrin ring is a strong-field ligand providing the low-spin d⁶ configuration necessary for O₂ binding geometry. (Roat-Malone, Ch. 3)
- Zinc enzymes use Zn²⁺ as a Lewis acid, not a redox center. In carbonic anhydrase II (CAII), three histidine residues (His94, His96, His119) hold Zn²⁺ in a tetrahedral site with one water molecule. The Lewis acidity of Zn²⁺ lowers the pKa of the coordinated water from ~15.7 (free water) to ~6.8, generating a zinc-bound hydroxide at physiological pH. This hydroxide is the nucleophile that attacks CO₂, producing HCO₃⁻ at a rate approaching 10⁶ turnovers per second — one of the fastest known enzymatic reactions. (Roat-Malone, Ch. 5)
- Type 1 "blue" copper proteins like plastocyanin enforce a distorted tetrahedral geometry on Cu that is ideal for Cu²⁺ but deeply unfavorable for Cu⁺. The protein preorganizes a rigid site (2 His, 1 Cys, 1 Met) that cannot relax to the preferred geometries of either oxidation state — this is the entatic state (or rack) principle. The geometric strain raises the Cu²⁺/Cu⁺ reduction potential to ~+370 mV (vs. ~+160 mV for aqueous Cu²⁺/Cu⁺), tuning it for electron transfer in the photosynthetic chain. The intense blue color (ε ≈ 5000 M⁻¹cm⁻¹ at ~600 nm) arises from a Cys(S) → Cu²⁺ ligand-to-metal charge transfer (LMCT) band, not a d–d transition. (Roat-Malone, Ch. 4)
- The oxygen-evolving complex (OEC) of Photosystem II is a Mn₄CaO₅ cluster that oxidizes water through five S-states (S₀ through S₄). Four sequential light-driven electron abstractions step the cluster from S₀ to S₄, accumulating four oxidizing equivalents; S₄ spontaneously releases O₂ and resets to S₀. Calcium is essential (not catalytically replaceable by Sr²⁺ in most organisms): it positions a substrate water molecule and participates in proton removal. The Mn ions cycle through Mn³⁺/Mn⁴⁺ oxidation states, and the oxo bridges store oxidizing equivalents across the cluster. (Roat-Malone, Ch. 6)
- Nitrogenase reduces N₂ to NH₃ at the iron-molybdenum cofactor (FeMo-co: [Mo-7Fe-9S-C-homocitrate]) using 8 electrons, 8 protons, and 16 ATP per N₂. The extraordinary ATP cost reflects the thermodynamic challenge of cleaving the N≡N triple bond (bond dissociation energy ≈ 945 kJ mol⁻¹). The interstitial carbide ligand (carbon atom at the center of the Fe₇ cage), identified by EXAFS and crystallography only in 2011, is thought to help buffer the cluster's electronic structure during the catalytic cycle. (Roat-Malone, Ch. 7)
- Metal-binding peptide design exploits the ATCUN (amino-terminal copper and nickel) motif. The sequence Xxx-Zzz-His at the N-terminus (where position 3 is histidine) creates a square-planar N₂O₂ or N₄ binding site using the free amine, two deprotonated amide nitrogens, and the His imidazole. This motif is found in human serum albumin and can be engineered into synthetic peptides to create metallopeptides with tunable redox, spectroscopic, and DNA-cleaving properties, as described in the metallopeptide design framework of Nanda and Bhardwaj. (Roat-Malone, Ch. 9; Nanda & Bhardwaj, 2012)
- The 18-electron rule governs stability in organometallic complexes because filling all bonding and non-bonding MOs of the metal (analogous to a noble gas configuration) provides maximum stability. Coordinatively saturated 18-electron complexes like Cr(CO)₆ or [Fe(CO)₅] are inert, while 16-electron square-planar d⁸ complexes (Pt(II), Pd(II), Rh(I)) are coordinatively unsaturated and highly reactive — a property deliberately exploited in catalysis (Wilkinson's catalyst, RhCl(PPh₃)₃, is a 16-electron species that binds and activates H₂ in hydrogenation). (Atkins et al., Ch. 22)
- Marcus theory predicts the rate of outer-sphere electron transfer from the driving force (ΔG°) and the reorganization energy (λ). The key equation is ΔG‡ = (λ/4)(1 + ΔG°/λ)², where λ is the energy cost of reorganizing both inner (bond lengths/angles) and outer (solvent) coordination spheres to the transition-state geometry before electron transfer occurs. Proteins minimize λ by enforcing rigid metal sites — the entatic state in blue copper proteins reduces both inner-sphere and outer-sphere reorganization, allowing sub-microsecond electron transfer over distances of 10–15 Å. (Atkins et al., Ch. 20; Roat-Malone, Ch. 4)
References
- Peter Atkins, Tina Overton, Jonathan Rourke, Mark Weller, and Fraser Armstrong, Inorganic Chemistry, 7th edition. Oxford University Press, Oxford, 2022. [Primary course text; Chapters 7, 19, 20, 22, 26 cited specifically.]
- Rosette M. Roat-Malone, Bioinorganic Chemistry: A Short Course, 2nd edition. Wiley-Interscience, Hoboken, NJ, 2007. [Chapters 3–9 cited for biological metal site mechanisms.]
- Vikas Nanda and Rudresh Bhardwaj, 'The Design of Metallopeptides,' Nature Chemistry, 2012, 4, 515–516. DOI: 10.1038/nchem.1201
- J. A. Ibers and R. H. Holm, 'Modeling Coordination Sites in Metallobiomolecules,' Science, 1980, 209, 223–235. [Background on entatic state and synthetic model compounds for bioinorganic sites.] DOI: 10.1126/science.6998932
- R. A. Marcus, 'Electron Transfer Reactions in Chemistry: Theory and Experiment,' Reviews of Modern Physics, 1993, 65, 599–610. [Nobel Lecture; foundational source for Marcus theory as applied to metal complexes.] DOI: 10.1103/RevModPhys.65.599
- Y. Umena, K. Kawakami, J.-R. Shen, and N. Kamiya, 'Crystal Structure of Oxygen-Evolving Photosystem II at a Resolution of 1.9 Å,' Nature, 2011, 473, 55–60. [High-resolution structure of the Mn₄CaO₅ cluster.] DOI: 10.1038/nature09913
- Oliver Einsle, F. A. Tezcan, Susana L. A. Andrade, Benedikt Schmid, Mesut Yoshida, James B. Howard, and Douglas C. Rees, 'Nitrogenase MoFe-Protein at 1.16 Å Resolution: A Central Ligand in the FeMo-Cofactor,' Science, 2002, 297, 1696–1700. [First crystallographic identification of the interstitial atom in FeMo-co.] DOI: 10.1126/science.1073877
- B. Rosenberg, L. Van Camp, and T. Krigas, 'Inhibition of Cell Division in Escherichia coli by Electrolysis Products from a Platinum Electrode,' Nature, 1965, 205, 698–699. [Original discovery of cisplatin's biological activity.] DOI: 10.1038/205698a0
- Edward I. Solomon, Pulakesh Sundaram, and Timothy E. Machonkin, 'Multicopper Oxidases and Oxygenases,' Chemical Reviews, 1996, 96, 2563–2605. [Comprehensive treatment of Type 1, 2, and 3 copper sites including plastocyanin and ceruloplasmin.] DOI: 10.1021/cr950046o
- David W. Christianson and Carol A. Fierke, 'Carbonic Anhydrase: Evolution of the Zinc Binding Site by Nature and by Design,' Accounts of Chemical Research, 1996, 29, 331–339. [Mechanistic detail on the Lewis acid mechanism of carbonic anhydrase and the role of zinc-bound water.] DOI: 10.1021/ar9500232