The electron transport chain and oxidative phosphorylation

Updated November 2026

Most of the ATP from glucose is made at the end, in the inner mitochondrial membrane. Electrons from NADH and FADH₂ pass down a chain of protein complexes to oxygen; the energy released pumps protons across the membrane, and the protons flowing back through ATP synthase drive ATP production.

The short version

The chain, complex by complex

Everything here sits in or on the inner mitochondrial membrane, which is folded into cristae to give more surface. Electrons move from carriers that hold them loosely (high energy) to carriers that hold them tightly, ending with oxygen, which pulls hardest of all.

Electron transport chain components
ComponentTakes electrons fromPasses them toProtons pumped (per 2 e⁻)
Complex I (NADH dehydrogenase)NADHCoenzyme Q4
Complex II (succinate dehydrogenase)FADH₂ (from succinate)Coenzyme Q0
Coenzyme Q (ubiquinone)Complexes I and IIComplex IIImobile carrier in the membrane
Complex III (cytochrome bc₁ complex)Coenzyme QCytochrome c4
Cytochrome cComplex IIIComplex IVsmall protein, carries one electron at a time
Complex IV (cytochrome c oxidase)Cytochrome cO₂ → H₂O2

Adding up the pumping column: one NADH moves about 4 + 4 + 2 = 10 protons. One FADH₂ skips Complex I, so it moves about 4 + 2 = 6 protons. That difference is the whole reason FADH₂ is worth less ATP than NADH.

Chemiosmosis and ATP synthase

In 1961 Peter Mitchell proposed that the link between electron transport and ATP synthesis is not a chemical intermediate but a proton gradient across the membrane. He received the 1978 Nobel Prize in Chemistry for this chemiosmotic theory. The gradient has two parts: a concentration difference (the intermembrane space is more acidic) and an electrical difference (it is more positive). Together they are called the proton-motive force.

The inner membrane is nearly impermeable to protons, so the only easy way back into the matrix is through ATP synthase. Its membrane part (Fo, the “o” for oligomycin) is a ring that rotates as protons pass through it; the rotation turns a central stalk inside the head (F₁), which projects into the matrix and changes shape to join ADP and phosphate into ATP. It is a genuine rotary motor.

Where 2.5 and 1.5 come from

Measurements suggest that making one ATP and exporting it to the cytoplasm costs about 4 protons (about 3 through the synthase itself, plus about 1 to bring in phosphate and swap ADP for ATP across the membrane). So:

These ratios of ATP made per oxygen atom reduced are called P/O ratios. They are experimental averages, not whole-number rules, and they vary somewhat between organisms.

ATP per glucose, and why the numbers differ

Approximate ATP per glucose, modern versus older counting
SourceCarriersModern (2.5 / 1.5)Older (3 / 2)
Glycolysis, direct2 ATP22
Glycolysis NADH (cytosol)2 NADH3 or 54 or 6
Link reaction2 NADH56
Krebs cycle, direct2 GTP/ATP22
Krebs cycle NADH6 NADH1518
Krebs cycle FADH₂2 FADH₂34
Total30 or 3236 or 38

The "or" in the glycolysis row is real biology. NADH made in the cytoplasm cannot cross the inner membrane, so its electrons are ferried in by a shuttle. The malate–aspartate shuttle (used in liver and heart) delivers them to the chain as NADH, worth about 2.5 each; the glycerol-3-phosphate shuttle (prominent in skeletal muscle and brain) delivers them to coenzyme Q, worth about 1.5 each. Real cells also leak some protons and use the gradient for other transport, so the true yield is usually a little below these maximums. When you see 36–38 in an older book and 30–32 in a newer one, the biology has not changed; the measured P/O ratios have.

Uncouplers and poisons

An uncoupler lets protons back into the matrix without going through ATP synthase. Electron transport and oxygen use continue, often faster, but the energy is released as heat instead of ATP. Brown fat does this on purpose with an uncoupling protein (UCP1, thermogenin) to keep newborns and hibernating animals warm. The chemical 2,4-dinitrophenol (DNP) does it uncontrollably: it was sold as a weight-loss drug in the 1930s and can cause fatal overheating.

Inhibitors block the chain itself. Cyanide and carbon monoxide block Complex IV, so electrons cannot reach oxygen, proton pumping stops and ATP production collapses. Rotenone blocks Complex I and oligomycin blocks ATP synthase.

Memory aid

Think of a hydroelectric dam. The electron transport chain is the set of pumps that push water (protons) uphill into a reservoir (the intermembrane space). ATP synthase is the turbine in the dam wall: water rushing back down spins it and generates power (ATP). Oxygen is the drain at the bottom of the electron line; block it (cyanide) and the pumps stop. An uncoupler is a hole in the dam: water flows, but the turbine does not turn.

For the order, "New Owls Quietly Think, Cats Fear Oxygen": NADH, One, Q, Three, c, Four, O₂.

A verse to remember it

Tune: Camptown RacesFrom NADH to Complex One, Down the chain, down the chain; Q to Three, then c will run To Four: O₂ ends the train. FADH₂ comes in at Two, Skips a pump, skips a pump, Hands to Q and rides the same way through, Less ATP, it's true. Pumping protons out, letting protons in, The synthase rotor spins around and ATP's made again.

Check yourself

Why is FADH₂ worth about 1.5 ATP but NADH about 2.5?

FADH₂ hands its electrons to Complex II, which pumps no protons, so they skip Complex I's pumping. About 6 protons are moved instead of 10, and at roughly 4 protons per ATP that gives 1.5 instead of 2.5.

What is the final electron acceptor, and what does it become?

Oxygen. Each O₂ takes 4 electrons and 4 protons and becomes 2 water molecules.

Cyanide blocks Complex IV. Why does ATP synthesis stop?

Electrons back up along the chain, so Complexes I and III stop pumping too. Without a proton gradient, ATP synthase has nothing to drive it.

A textbook says 38 ATP per glucose, another says 30–32. Which is wrong?

Neither is a typo. 36–38 uses the older whole-number ratios of 3 ATP per NADH and 2 per FADH₂; 30–32 uses the measured averages of about 2.5 and 1.5. The range within each depends on which shuttle brings glycolytic NADH into the mitochondrion.

Sources