How many ATP does beta oxidation produce?

Beta oxidation does not make ATP directly. Each round of the spiral trims two carbons from a fatty acid and produces one FADH₂, one NADH and one acetyl-CoA; the ATP comes later, when mitochondria oxidize those products. The two electron carriers are worth about 4 ATP per round, and counting the acetyl-CoA as well brings a round to 14 ATP.

For palmitate, the 16-carbon saturated fatty acid used in the standard calculation, seven rounds followed by complete oxidation of the products give 106 net ATP, after subtracting two ATP equivalents spent activating the fatty acid. Older textbooks give 129 for the same molecule because they assume more ATP per NADH and FADH₂. Figures such as 17 ATP per round or 131 per palmitate are the same bookkeeping under different assumptions.

Does beta oxidation produce ATP?

None of the four reactions in a round of beta oxidation forms ATP. An acyl-CoA dehydrogenase first removes two hydrogens to create a double bond, passing the electrons to FAD. Water is then added across the double bond, a second dehydrogenation passes electrons to NAD⁺, and a thiolase cuts off acetyl-CoA, leaving a fatty acyl-CoA two carbons shorter.^[1][2]

The ATP is made downstream, by oxidative phosphorylation. NADH hands its electrons to complex I of the respiratory chain. The electrons held by the dehydrogenase's FAD take a shorter route: electron transfer flavoprotein (ETF) and ETF dehydrogenase deliver them straight to ubiquinone, bypassing complex I.^[1]

That difference in entry point sets their value. Hinkle's review of measurements since 1937 put the ATP made per pair of electrons, the P/O ratio, at about 2.5 for NADH-linked substrates and 1.5 for succinate, whose electrons also enter at ubiquinone.^[3] The pathway even costs ATP before it starts: activating a fatty acid to its acyl-CoA runs through an acyl-adenylate intermediate and consumes the equivalent of two ATP.^[2]

How many ATP does one round of beta oxidation produce?

One round yields about 4 ATP through oxidative phosphorylation: 2.5 from its NADH and 1.5 from its FADH₂.^[2][3] The acetyl-CoA it releases is worth another 10 ATP once the Krebs cycle oxidizes it, because each acetyl-CoA yields 3 NADH, 1 FADH₂ and 1 GTP.^[2] A round and its acetyl-CoA therefore support 14 ATP.

The older figures of 5 and 17 ATP per round count the same products at 3 ATP per NADH and 2 per FADH₂, which also raises each acetyl-CoA to 12 ATP.^[4] Table 1 lines up the numbers most often quoted for beta oxidation and shows what each one counts.

What is counted	Modern ratios (ATP)	Older ratios (ATP)
One round, NADH and FADH₂ only	4	5
One acetyl-CoA through the Krebs cycle	10	12
One round plus its acetyl-CoA	14	17
Palmitate, NADH and FADH₂ from seven rounds	28	35
Palmitate, seven rounds minus activation	26	33
Palmitate, complete oxidation before activation	108	131
Palmitate, complete oxidation, net	106	129

How is the 106 ATP yield from palmitate calculated?

Palmitate needs seven rounds, not eight. Each of the first six removes one acetyl-CoA, and the seventh splits the remaining four-carbon chain into two, so the spiral produces 7 NADH, 7 FADH₂ and 8 acetyl-CoA.^[2][1] The eight acetyl-CoA then turn the Krebs cycle eight times, bringing the totals to 31 NADH, 15 FADH₂ and 8 GTP.^[2]

Product	Made in	Molecules	ATP each	ATP
NADH	Beta oxidation	7	2.5	17.5
FADH₂	Beta oxidation	7	1.5	10.5
NADH	Krebs cycle	24	2.5	60
FADH₂	Krebs cycle	8	1.5	12
GTP	Krebs cycle	8	1	8
Activation	Acyl-CoA synthetase	1	−2	−2
Net				106

Only 28 of the 108 gross ATP, about a quarter, trace back to the spiral itself; the rest comes from the acetyl-CoA. That split is why the question has two honest answers, depending on whether the acetyl-CoA is counted as part of beta oxidation's output.

How does fatty-acid length change ATP yield?

A saturated fatty acid with an even number of carbons, n, runs through n/2 − 1 rounds and releases n/2 acetyl-CoA. Counting 4 ATP per round, 10 per acetyl-CoA and 2 for activation reduces the whole calculation to one expression:

$$\begin{aligned} \text{net ATP} &= 4\left(\tfrac{n}{2} - 1\right) + 10 \cdot \tfrac{n}{2} - 2 \\ &= 7n - 6 \end{aligned}$$

Palmitate, with 16 carbons, returns 106. Each extra pair of carbons adds one round and one acetyl-CoA, or 14 ATP, so stearate (C18) yields 120 and arachidate (C20) 134. We calculated the values in Figure 1 with this rule.

The rule does not hold for odd chains or double bonds. An odd-chain fatty acid finishes with one acetyl-CoA and one three-carbon propionyl-CoA, which is converted to succinyl-CoA and joins the Krebs cycle partway round instead of being oxidized as acetyl-CoA.^[5]

Unsaturated fatty acids need extra enzymes: an isomerase moves a double bond into the position the spiral expects, and polyunsaturated chains also need a reductase.^[1] Where a double bond is already in place, that round skips the FAD-linked first step and makes no FADH₂, so an unsaturated fatty acid yields slightly less than a saturated one of the same length.

Why do some sources report 129 ATP from palmitate?

The 129 figure is the same calculation with the historical ratios of 3 ATP per NADH and 2 per FADH₂.^[4] Seven rounds then supply 35 ATP instead of 28, eight acetyl-CoA supply 96 instead of 80, and subtracting activation from 131 leaves 129. Beignon and colleagues trace the fall from 129 to 106 entirely to the revised ratios, the same revision that lowered glucose from 38 to about 30 to 32 ATP.^[4]

The modern ratios are themselves a convention rather than a constant. Hinkle noted that the structure of ATP synthase could imply values as low as 2.3 and 1.4, while calculations from the proton cost of mammalian ATP synthase give 2.7 and 1.6.^[3][4] Either set moves the palmitate total by several ATP, but 2.5 and 1.5 match most measurements and remain the standard teaching values.

Living cells usually recover less than any of these maxima. Protons leak back across the inner membrane, uncoupling proteins in brown fat turn the gradient into heat, and the effective P/O ratio of resting liver cells oxidizing glucose has been estimated at about 1.3.^[4] During fasting, the liver also diverts acetyl-CoA from beta oxidation into ketone bodies rather than burning it in the Krebs cycle.^[2]

Read as one ledger, beta oxidation releases the electrons and carbon, the Krebs cycle and respiratory chain turn them into ATP, and the chosen P/O ratios set the final count. On the modern ledger one palmitate is worth 106 ATP, more than three times the 30 to 32 ATP from one glucose.