Free Energy, Spontaneity, and Coupled Reactions

A yeast cell manufactures glucose-6-phosphate by spending ATP. This happens even though the phosphorylation reaction alone is thermodynamically uphill. Left to itself, glucose plus inorganic phosphate will never assemble into glucose-6-phosphate in appreciable yield. Yet inside the cell this reaction runs essentially to completion. This happens every second, in every dividing cell on Earth. The trick is not a violation of thermodynamics. It is a masterclass in how thermodynamics actually works when reactions share a common intermediate. This article is about the arithmetic of life's energy economy. It covers how Gibbs free energy determines whether a reaction proceeds. It explains how the standard free energy change relates to the actual free energy change under cellular conditions. It also shows how cells stitch an energetically unfavorable reaction to a favorable one so that the sum runs forward. This is the conceptual machine underneath The Chemical Logic of the Cell. It explains why ATP hydrolysis can "pay for" biosynthesis, ion pumping, and mechanical work. It is also why a pathway's overall thermodynamics can never be judged from a single step in isolation. The big picture is deceptively simple once the algebra is unpacked. $\Delta G$ is not a fixed property of a reaction. It is a function of concentration. $\Delta G^{\circ\prime}$ (the biochemist's standard state) is merely the value of $\Delta G$ at one particular, often unphysiological, concentration point. Because $\Delta G$ depends logarithmically on the reaction quotient $Q$ , a reaction with an unfavorable $\Delta G^{\circ\prime}$ can be pulled forward. This happens simply by keeping product concentration low or substrate concentration high. Coupling two reactions through a shared intermediate is the biological embodiment of this fact. It is thermodynamic bookkeeping, not new chemistry. By the end you will be able to compute $\Delta G$ from $\Delta G^{\circ\prime}$ and real concentrations. You will also convert between $\Delta G^{\circ\prime}$ and $K'_{eq}$ . You will correctly add coupled-reaction free energies. And you will diagnose why a pathway step that looks impossible in isolation is entirely routine in vivo. This is Thermodynamics of Living Systems made quantitative and operational.

Key takeaways

$\Delta G^{\circ\prime}$ is the free energy change at defined standard conditions (1 M, pH 7, 298 K); it is a constant for a given reaction, but it is not what actually happens in the cell.
$\Delta G$ is the free energy change at actual cellular concentrations; it, not $\Delta G^{\circ\prime}$ , determines the direction and spontaneity of a reaction as it runs.
$\Delta G^{\circ\prime}$ and $K'_{eq}$ are two faces of one coin: $\Delta G^{\circ\prime} = -RT\ln K'_{eq}$ .
Free energies of consecutive reactions sharing an intermediate are additive; this lets an unfavorable step be driven by a favorable one.
ATP hydrolysis ( $\Delta G^{\circ\prime} \approx -30.5\ \text{kJ/mol}$ ) is the universal energy currency because it couples to otherwise endergonic biosynthetic and transport steps.
Cells maintain far-from-equilibrium concentrations (high [ATP]/[ADP][Pi]) precisely so that $\Delta G$ for ATP hydrolysis is much more negative than $\Delta G^{\circ\prime}$ .
Key idea: spontaneity in the cell is governed by the actual $\Delta G$ , and coupling through shared intermediates lets a favorable reaction's negative $\Delta G$ absorb an unfavorable reaction's positive $\Delta G$ so the net process is spontaneous.

Where this lives today

Every metabolic pathway diagram in modern biochemistry is, at bottom, a free-energy coupling diagram: glycolysis, the citric acid cycle, oxidative phosphorylation, and fatty acid synthesis are all sequences engineered so unfavorable steps are paired with favorable ones through shared ATP/ADP, NAD⁺/NADH, or acyl-CoA intermediates. Pharmaceutical enzyme inhibitor design routinely uses $\Delta G^{\circ\prime}$ and $K'_{eq}$ data to predict whether a target reaction is likely to be reversible in vivo (a good drug target is often one sitting at a large, physiologically irreversible $\Delta G$ , like hexokinase or phosphofructokinase, since blocking it commits the cell rather than merely nudging an equilibrium). Metabolic engineering and synthetic biology use flux-balance and thermodynamic feasibility analysis (software such as OptMDFpathway or the "max-min driving force" method) built directly on the $\Delta G = \Delta G^{\circ\prime} + RT\ln Q$ relationship to redesign microbial pathways for biofuel or pharmaceutical precursor production, checking that every engineered step has $\Delta G < 0$ under achievable intracellular metabolite concentrations. In exercise physiology and clinical medicine, the actual cellular $\Delta G$ of ATP hydrolysis (not the textbook −30.5 kJ/mol) is measured noninvasively by ³¹P NMR spectroscopy in muscle, since it governs whether ATP-dependent processes like the sodium-potassium pump or myosin cross-bridge cycling can be sustained under ischemia or intense exertion. Even the chemiosmotic proton-motive force that drives ATP synthase is analyzed with the same $\Delta G = \Delta G^{\circ\prime} + RT\ln Q$ machinery, generalized to include an electrical term for charged species crossing a membrane — the electrochemical potential — making this single equation the quantitative backbone connecting Gibbs's 1876 thermodynamics to the operation of a living mitochondrion.

:::quiz[If [ATP], [ADP], and [Pi] are all set to exactly 1 M, what is ΔG for ATP hydrolysis?]

Always zero, since the reaction is at equilibrium
Equal to ΔG°′, since Q = 1 and RT ln Q = 0 ✓
Undefined, since 1 M is never physiological
More negative than ΔG°′, since concentrations are high

Ask the tutor: How does this free-energy coupling framework extend to reactions that move charged species across a membrane, where an electrical potential term must be added to $\Delta G$ ?

The physical picture — free energy is a moving target, not a label

The single most important habit to build is refusing to ask "is this reaction spontaneous?" without immediately asking "at what concentrations?" Gibbs free energy change, $\Delta G$ , is the criterion for spontaneity at constant temperature and pressure — the conditions of a living cell. A process with $\Delta G < 0$ is exergonic and proceeds spontaneously as written; $\Delta G > 0$ is endergonic and requires energy input; $\Delta G = 0$ means the system sits at equilibrium. But $\Delta G$ is a function of the instantaneous concentrations of reactants and products, and those concentrations change as the reaction proceeds — indeed $\Delta G \to 0$ is precisely what happens because the reaction proceeds, since it is running downhill towards equilibrium. $\Delta G^{\circ\prime}$ , the biochemical standard free energy change, freezes this into a single reference number: it is $\Delta G$ evaluated when every reactant and product is at 1 M (or, for water, at its pure-liquid activity of 1; for H⁺, at pH 7 rather than 1 M — the biochemist's modification of the chemist's standard state, hence the prime). This is a bookkeeping convenience, not a claim about what is actually in a cell. No cell has 1 M glucose-6-phosphate. $\Delta G^{\circ\prime}$ is useful precisely because it is a fixed, tabulable, comparable number — but the moment you want to know whether a reaction actually runs forward in vivo, you must convert back to real concentrations. The figure below traces the logic: a reaction's free energy surface as a function of extent of reaction, showing why $\Delta G^{\circ\prime}$ (the standard-state point) and $\Delta G$ (the actual operating point) can even have opposite sign.

Figure

Free energy versus extent of reaction: standard-state point vs. actual cellular point

Notice the essential point encoded in the picture: the slope of the $G$ -vs- $\xi$ curve at any point is $\Delta G$ at that composition. At the standard-state point the slope can be positive (reaction unfavorable at 1 M) while at the actual cellular composition — far from 1 M, with substrate kept high and product kept low by downstream consumption — the local slope is negative. Same reaction, same molecules, opposite sign of $\Delta G$ , because concentration is doing the work. The second organizing idea is coupling through a shared intermediate. If reaction 1 produces species X and reaction 2 consumes the same X, the two reactions can be summed: X cancels, and the overall $\Delta G$ is the arithmetic sum $\Delta G_1 + \Delta G_2$ . This is legal because Gibbs free energy is a state function — path-independent — so free energies of sequential steps genuinely add, exactly like $\Delta H$ in Hess's law. If reaction 2 is enough exergonic, the sum can be negative even though reaction 1 alone is endergonic. This is not "borrowing energy" in any mystical sense; it is the same algebra as adding two signed numbers.

Figure

Coupling: an endergonic step summed with an exergonic step through a shared intermediate

The core ideas — the governing equations

The central relationship connects the actual free energy change to the standard free energy change and the instantaneous composition through the reaction quotient $Q$ :

\Delta G = \Delta G^{\circ\prime} + RT\ln Q

where:

$\Delta G$ = actual free energy change at the current concentrations (kJ/mol) — the true arbiter of spontaneity,
$\Delta G^{\circ\prime}$ = biochemical standard free energy change (1 M solutes, pH 7, 298 K, 1 atm),
$R$ = 8.314 J mol⁻¹K⁻¹, $T$ = absolute temperature (K),
$Q$ = the reaction quotient, $[\text{products}]^{\text{coeffs}}/[\text{reactants}]^{\text{coeffs}}$ , evaluated at the actual instantaneous concentrations (water and H⁺ excluded/folded into the prime convention, solids/pure liquids at activity 1).

Reading: the first term is a fixed offset (the reaction's intrinsic thermodynamic preference at the reference point); the second term is how far actual conditions are displaced from that reference. Because it enters as $RT\ln Q$ , doubling or halving concentrations produces only a modest, logarithmic shift in $\Delta G$ — but the shift is unbounded as $Q\to 0$ or $Q\to\infty$ , meaning sufficiently extreme concentration ratios can force $\Delta G$ to any sign, however unfavorable $\Delta G^{\circ\prime}$ is. At equilibrium $\Delta G = 0$ by definition, and $Q$ takes its equilibrium value $K'_{eq}$ , giving the second core relationship:

0 = \Delta G^{\circ\prime} + RT\ln K'_{eq} \quad\Longrightarrow\quad \Delta G^{\circ\prime} = -RT\ln K'_{eq}

where:

$K'_{eq}$ = the equilibrium constant under biochemical standard-state conventions (concentrations at equilibrium, water/H⁺ folded in as above).

Reading: $\Delta G^{\circ\prime}$ and $K'_{eq}$ carry exactly the same information — one is the log of the other, scaled by $RT$ . A large positive $K'_{eq}$ (products favored at equilibrium) corresponds to a large negative $\Delta G^{\circ\prime}$ , and vice versa. This equation is how tables of $\Delta G^{\circ\prime}$ are actually constructed — nobody measures free energy directly; they measure equilibrium concentrations and back out $\Delta G^{\circ\prime}$ . The third relationship is additivity for coupled/sequential reactions:

\Delta G_{\text{overall}} = \sum_i \Delta G_i \qquad \Delta G^{\circ\prime}_{\text{overall}} = \sum_i \Delta G^{\circ\prime}_i

where:

the sum runs over each elementary or overall step in a sequence whose intermediates cancel when the equations are added.

Reading: because $G$ is a state function, this additivity is exact and requires no correction term — unlike, say, naively adding equilibrium constants (which must be multiplied, not added; $K'_{eq,\text{overall}} = \prod_i K'_{eq,i}$ , consistent with the log relationship above).

Derivation AFrom

Q

-dependence to the equilibrium-constant relation

Goal. Derive $\Delta G = \Delta G^{\circ\prime} + RT\ln Q$ from the chemical potentials of reactants and products, then obtain $\Delta G^{\circ\prime} = -RT\ln K'_{eq}$ as the special case $\Delta G = 0$ .

Step 1 — chemical potential of each species. For a reaction $aA + bB \rightleftharpoons cC + dD$ , each species $i$ present in dilute solution has a chemical potential

\mu_i = \mu_i^{\circ} + RT\ln a_i

where $\mu_i^{\circ}$ is the chemical potential at unit activity ( $a_i = 1$ , i.e., the standard state) and $a_i$ is the activity (approximated here by concentration in M, relative to the 1 M reference, so $a_i$ is dimensionless).

Step 2 — free energy of reaction as a sum of chemical potentials. The free energy change for the reaction as written is the weighted difference between product and reactant potentials:

\Delta G = \left[c\,\mu_C + d\,\mu_D\right] - \left[a\,\mu_A + b\,\mu_B\right]

Step 3 — substitute Step 1 into Step 2 and collect standard-state terms. Expanding each $\mu_i$ :

\Delta G = \left[c\,\mu_C^{\circ} + d\,\mu_D^{\circ} - a\,\mu_A^{\circ} - b\,\mu_B^{\circ}\right] + RT\left[c\ln a_C + d\ln a_D - a\ln a_A - b\ln a_B\right]

The bracketed standard-state term is by definition $\Delta G^{\circ\prime}$ (the free energy change when every species sits at unit activity). The logarithmic bracket collapses using $\ln x^n = n\ln x$ and $\ln(xy)=\ln x+\ln y$ into a single logarithm of a ratio:

\Delta G = \Delta G^{\circ\prime} + RT\ln\left(\frac{a_C^{\,c}\,a_D^{\,d}}{a_A^{\,a}\,a_B^{\,b}}\right) = \Delta G^{\circ\prime} + RT\ln Q

which is the first core equation, with $Q \equiv a_C^{\,c}a_D^{\,d}/(a_A^{\,a}a_B^{\,b})$ evaluated at the actual instantaneous concentrations.

Step 4 — impose the equilibrium condition. At equilibrium the system has stopped moving downhill in $G$ ; formally this is the condition $\Delta G = 0$ (the tangent to the $G$ -vs- $\xi$ curve is flat — recall the minimum in the earlier figure). Substituting $\Delta G = 0$ and renaming $Q$ at equilibrium as $K'_{eq}$ :

0 = \Delta G^{\circ\prime} + RT\ln K'_{eq}

Rearranging isolates the standard free energy change:

\Delta G^{\circ\prime} = -RT\ln K'_{eq}

This is exact — no approximation was introduced beyond activity ≈ concentration, which is the standard biochemical convention. The sign is essential: a favorable equilibrium ( $K'_{eq} > 1$ , products dominate) gives $\ln K'_{eq} > 0$ and hence $\Delta G^{\circ\prime} < 0$ , matching intuition.

Go deeper · Derivation B · 1 minAdditivity of coupled reactions and the free-energy "budget" for ATP-driven biosynthesis

Derivation BAdditivity of coupled reactions and the free-energy "budget" for ATP-driven biosynthesis

Goal. Show why free energies of two reactions sharing an intermediate add algebraically, and use this to derive the condition under which ATP hydrolysis can drive an endergonic biosynthetic step.

Step 1 — write the two half-reactions and their $\Delta G^{\circ\prime}$ . Consider glucose phosphorylation (unfavorable alone) and ATP hydrolysis (favorable):

\text{Reaction 1: glucose} + P_i \rightarrow \text{glucose-6-phosphate} + H_2O \qquad \Delta G^{\circ\prime}_1 = +13.8\ \text{kJ/mol}

\text{Reaction 2: ATP} + H_2O \rightarrow \text{ADP} + P_i \qquad \Delta G^{\circ\prime}_2 = -30.5\ \text{kJ/mol}

Step 2 — add the reactions as chemical equations, canceling species common to both sides. $P_i$ appears as a product of reaction 2 and a reactant of reaction 1; one $H_2O$ appears as reactant of reaction 1 and product of reaction 2. Adding and cancelling:

\text{glucose} + \text{ATP} \rightarrow \text{glucose-6-phosphate} + \text{ADP}

Step 3 — justify additivity of $\Delta G^{\circ\prime}$ from Hess's-law logic. Because $G$ (and hence $\Delta G^{\circ\prime}$ , a difference of $G$ values at fixed standard states) is a state function, the free energy change of the sum of two equations equals the sum of the free energy changes of the individual equations — this holds regardless of mechanism or of whether the intermediate ( $P_i$ ) is ever released into free solution, exactly as $\Delta H$ sums in Hess's law of heat summation. Formally, writing $\Delta G^{\circ\prime}_i = G^{\circ\prime}_{\text{products},i} - G^{\circ\prime}_{\text{reactants},i}$ for each step, summing over $i$ telescopes: intermediate species' $G^{\circ\prime}$ contributions appear once as a product (positive) and once as a reactant (negative) and cancel, leaving only the net reactants and products of the combined equation.

Step 4 — compute the coupled $\Delta G^{\circ\prime}$ .

\Delta G^{\circ\prime}_{\text{overall}} = \Delta G^{\circ\prime}_1 + \Delta G^{\circ\prime}_2 = (+13.8) + (-30.5)

\Delta G^{\circ\prime}_{\text{overall}} = -16.7\ \text{kJ/mol}

Step 5 — state the general coupling condition. An endergonic step with standard free energy $+\Delta G^{\circ\prime}_{\text{endergonic}}$ can in principle be driven by coupling to ATP hydrolysis provided

\Delta G^{\circ\prime}_{\text{ATP hydrolysis}} + \Delta G^{\circ\prime}_{\text{endergonic}} < 0 \quad\Longleftrightarrow\quad \Delta G^{\circ\prime}_{\text{endergonic}} < 30.5\ \text{kJ/mol}

so ATP hydrolysis (−30.5 kJ/mol standard) can in principle offset any single endergonic step whose own standard cost is below about 30.5 kJ/mol — and, crucially, because the actual cellular $\Delta G$ of ATP hydrolysis is substantially more negative than −30.5 kJ/mol (worked below), the true energetic budget available is even larger. This is why almost all "uphill" biosynthetic steps in metabolism are written as ATP-consuming: the coupling budget is generous, not fine-tuned.

Worked examples

Example 1 — Foundational: computing actual ΔG for ATP hydrolysis inside a real cell. ATP hydrolysis, $\text{ATP} + H_2O \rightarrow \text{ADP} + P_i$ , has $\Delta G^{\circ\prime} = -30.5\ \text{kJ/mol}$ at 298 K. Typical intracellular concentrations (asymmetric, realistic, non-round numbers, as measured in hepatocytes) are

[\text{ATP}] = 3.4\times10^{-3}\ \text{M}, \quad [\text{ADP}] = 1.3\times10^{-4}\ \text{M}, \quad [P_i] = 4.8\times10^{-3}\ \text{M}

Water is the solvent and is folded into the standard-state convention (activity 1), so it does not appear in $Q$ . The reaction quotient is

Q = \frac{[\text{ADP}][P_i]}{[\text{ATP}]} = \frac{(1.3\times10^{-4})(4.8\times10^{-3})}{3.4\times10^{-3}}

Numerator: $(1.3\times10^{-4})(4.8\times10^{-3}) = 6.24\times10^{-7}$ .

Q = \frac{6.24\times10^{-7}}{3.4\times10^{-3}} = 1.835\times10^{-4}

Now apply the master equation with $R = 8.314\ \text{J mol}^{-1}\text{K}^{-1}$ , $T = 310\ \text{K}$ (physiological body temperature, not 298 K — another asymmetric, deliberately non-round choice):

RT\ln Q = (8.314)(310)\ln(1.835\times10^{-4})

$RT = 2577\ \text{J/mol} = 2.577\ \text{kJ/mol}$ . $\ln(1.835\times10^{-4}) = \ln(1.835) + \ln(10^{-4}) = 0.6070 - 9.2103 = -8.603$ .

RT\ln Q = 2.577\times(-8.603) = -22.17\ \text{kJ/mol}

\Delta G = \Delta G^{\circ\prime} + RT\ln Q = -30.5 + (-22.17) = -52.7\ \text{kJ/mol}

The actual free energy released by ATP hydrolysis under these cellular conditions, $\approx -52.7$ kJ/mol, is nearly double the standard value. Limiting-case check: if we artificially set all three concentrations to exactly 1 M, then $Q = (1)(1)/1 = 1$ , $\ln Q = 0$ , and $\Delta G = \Delta G^{\circ\prime} = -30.5$ kJ/mol exactly — confirming the formula correctly reduces to the definition of the standard state when concentrations are standard. The cell deliberately keeps [ATP] high and [ADP][Pi] comparatively low (via continuous respiration and coupled consumption), which is exactly the displacement from equilibrium that makes $Q \ll 1$ and drives $\Delta G$ strongly negative — the cell is not at equilibrium, and staying away from equilibrium is the whole point.

Example 2 — Applied: can ATP hydrolysis drive the first step of glycolysis under real hepatocyte concentrations? The hexokinase reaction, glucose + ATP → glucose-6-phosphate + ADP, has $\Delta G^{\circ\prime} = -16.7$ kJ/mol (computed above by coupling). But we should check the actual $\Delta G$ , not just the standard value, using realistic concentrations:

[\text{glucose}] = 5.0\times10^{-3}\ \text{M},\quad [\text{G6P}] = 8.3\times10^{-5}\ \text{M},\quad [\text{ATP}] = 3.4\times10^{-3}\ \text{M},\quad [\text{ADP}] = 1.3\times10^{-4}\ \text{M}

Q = \frac{[\text{G6P}][\text{ADP}]}{[\text{glucose}][\text{ATP}]} = \frac{(8.3\times10^{-5})(1.3\times10^{-4})}{(5.0\times10^{-3})(3.4\times10^{-3})}

Numerator: $(8.3\times10^{-5})(1.3\times10^{-4}) = 1.079\times10^{-8}$ . Denominator: $(5.0\times10^{-3})(3.4\times10^{-3}) = 1.70\times10^{-5}$ .

Q = \frac{1.079\times10^{-8}}{1.70\times10^{-5}} = 6.35\times10^{-4}

At $T = 310$ K, $\ln(6.35\times10^{-4}) = \ln(6.35) + \ln(10^{-4}) = 1.8485 - 9.2103 = -7.362$ .

RT\ln Q = 2.577\times(-7.362) = -18.97\ \text{kJ/mol}

\Delta G = -16.7 + (-18.97) = -35.7\ \text{kJ/mol}

The coupled reaction is strongly exergonic in vivo — even more favorable than its already-favorable standard value — confirming that hexokinase runs essentially irreversibly under cellular conditions. This is precisely why hexokinase is a committed, regulated step in glycolysis: its large negative $\Delta G$ (not merely $\Delta G^{\circ\prime}$ ) makes the reaction physiologically one-directional, a hallmark of the pathway's rate-controlling steps.

Example 3 — Exam/competition level: back-calculating an unknown $K'_{eq}$ from a multi-step coupled pathway and checking self-consistency. Suppose a hypothetical two-step pathway is:

\text{Step A: } X \rightarrow Y, \qquad \Delta G^{\circ\prime}_A = +21.4\ \text{kJ/mol}

\text{Step B: } Y \rightarrow Z, \qquad \Delta G^{\circ\prime}_B = -9.6\ \text{kJ/mol}

(a) Find $\Delta G^{\circ\prime}$ for the overall process $X \rightarrow Z$ , and the corresponding $K'_{eq}$ at $T = 298$ K. By additivity (justified in the second derivation above, since Y is the shared intermediate and cancels):

\Delta G^{\circ\prime}_{\text{overall}} = \Delta G^{\circ\prime}_A + \Delta G^{\circ\prime}_B = 21.4 + (-9.6) = +11.8\ \text{kJ/mol}

Solve $\Delta G^{\circ\prime} = -RT\ln K'_{eq}$ for $K'_{eq}$ :

\ln K'_{eq} = -\frac{\Delta G^{\circ\prime}}{RT} = -\frac{11.8\times10^3}{(8.314)(298)}

$RT = 2478.6$ J/mol $= 2.479$ kJ/mol.

\ln K'_{eq} = -\frac{11.8}{2.479} = -4.761

K'_{eq} = e^{-4.761} = 8.55\times10^{-3}

So at equilibrium, [Z]/[X] $\approx 0.0086$ — the overall reaction, taken in isolation at standard concentrations, disfavors Z. This alone would predict the pathway does not run forward.

(b) Now suppose Z is continuously siphoned off by a third, highly exergonic reaction that keeps $[Z]/[X] = 2.0\times10^{-4}$ in the steady-state cell (well below the equilibrium ratio of $8.55\times10^{-3}$ ). Compute the actual $\Delta G$ for $X \rightarrow Z$ under these conditions and interpret.

Q = \frac{[Z]}{[X]} = 2.0\times10^{-4}

\ln Q = \ln(2.0\times10^{-4}) = \ln(2.0) - 4\ln(10) = 0.6931 - 9.2103 = -8.517

RT\ln Q = 2.479\times(-8.517) = -21.12\ \text{kJ/mol}

\Delta G = \Delta G^{\circ\prime} + RT\ln Q = 11.8 + (-21.12) = -9.32\ \text{kJ/mol}

Even though $\Delta G^{\circ\prime}$ for X → Z is positive (+11.8 kJ/mol, thermodynamically unfavorable at standard concentrations), the actual $\Delta G$ under the cell's maintained steady-state concentrations is negative (−9.3 kJ/mol): the pathway runs forward. Limiting-case check: if the downstream consumption of Z were switched off and $[Z]/[X]$ were allowed to rise back to the equilibrium ratio $K'_{eq} = 8.55\times10^{-3}$ , then $Q = K'_{eq}$ and $\Delta G = \Delta G^{\circ\prime} + RT\ln K'_{eq} = \Delta G^{\circ\prime} - \Delta G^{\circ\prime} = 0$ exactly, recovering equilibrium as required. This example is the general mechanism by which "thermodynamically unfavorable" pathway steps are made to run in cells: not by changing $\Delta G^{\circ\prime}$ , but by never letting $Q$ approach $K'_{eq}$ .

Limiting cases, scope & common traps

Situation	What happens to the equations	Interpretation
$Q = 1$ (all species at unit activity)	$\Delta G = \Delta G^{\circ\prime}$	Recovers the definition of standard state exactly
$Q = K'_{eq}$	$\Delta G = 0$	System at equilibrium; no net flux either direction
$Q \to 0$ (products depleted)	$RT\ln Q \to -\infty$ , $\Delta G \to -\infty$	Reaction becomes arbitrarily favorable — cells exploit this by continuous product removal
$Q \to \infty$ (products flooded)	$RT\ln Q \to +\infty$ , $\Delta G \to +\infty$	Reaction is forced backward regardless of $\Delta G^{\circ\prime}$
$T \to 0$	$RT\ln Q \to 0$	$\Delta G \to \Delta G^{\circ\prime}$ regardless of $Q$ (entropic driving vanishes)
Coupling two reactions, shared intermediate	$\Delta G^{\circ\prime}_{\text{tot}} = \sum \Delta G^{\circ\prime}_i$	Valid always for $\Delta G^{\circ\prime}$ and for actual $\Delta G$ of the summed reaction, provided the intermediate concentration is the same value in both half-reactions as written
Multi-step $K'_{eq}$	$K'_{eq,\text{tot}} = \prod_i K'_{eq,i}$	Constants multiply (from exponentiating the additive $\Delta G^{\circ\prime}$ sum) — never add equilibrium constants

Common trapCommon misconception 2: "Coupling means ATP literally transfers energy as heat or 'charges up' the substrate."

Coupling is a statement about free energies adding when a real chemical intermediate is shared — most commonly, ATP transfers its terminal phosphoryl group covalently onto the substrate (e.g., forming glucose-6-phosphate directly, with ADP as the other product), not via some diffuse transfer of "high-energy" heat. The enzyme (hexokinase) physically catalyzes phosphoryl transfer from ATP to glucose in a single active site; the two half-reactions (ATP hydrolysis, glucose phosphorylation) are a bookkeeping decomposition of one real mechanistic step, and their $\Delta G^{\circ\prime}$ values simply happen to add correctly because $G$ is a state function. There is no reaction in which ATP is "hydrolyzed to release energy" as a separate physical event before that energy is independently absorbed elsewhere.

The chemist behind it

Josiah Willard Gibbs (1839–1903) spent nearly his entire career at Yale, publishing the foundational papers of chemical thermodynamics — "On the Equilibrium of Heterogeneous Substances" (1876, 1878) — in the relatively obscure Transactions of the Connecticut Academy of Arts and Sciences, so quietly that it took a decade for European scientists like Maxwell and Ostwald to recognize the work's importance and popularize it. Gibbs was trained as an engineer (the first American PhD in engineering, from Yale in 1863) and brought an engineer's insistence on operationally defined, measurable quantities to what had been a physicist's and chemist's somewhat ad hoc subject. His free energy function — what we write today as $G = H - TS$ — provided exactly the single scalar quantity needed to predict, from tabulated data alone, whether any process at constant temperature and pressure would proceed spontaneously, without needing to separately track heat and disorder. Gibbs never applied his function to a living cell; that synthesis, connecting $\Delta G^{\circ\prime}$ , $K'_{eq}$ , and the biochemical standard state, was worked out over the following seven decades by biochemists including Fritz Lipmann, who in 1941 crystallized the idea of "high-energy phosphate bonds" and energy coupling that makes ATP the currency it is.

Origins · 1 minJosiah Willard Gibbs and how this was discovered

Josiah Willard Gibbs

1839–1903 · Yale University; founder of chemical thermodynamics and statistical mechanics

Gibbs formalized the free energy function that bears his name and derived the phase rule, giving chemistry its first fully general, quantitative criterion for spontaneity and equilibrium — the mathematical foundation on which every subsequent biochemical energy calculation, including ATP coupling, ultimately rests.

History

- 1873 — Gibbs introduces graphical thermodynamic surfaces relating energy, entropy, and volume.

- 1876/1878 — Gibbs publishes "On the Equilibrium of Heterogeneous Substances," defining the free energy function and chemical potential.

- 1923 — Gilbert N. Lewis and Merle Randall popularize activities and the modern chemical-thermodynamics formalism (

\Delta G = \Delta G^{\circ} + RT\ln Q

) in their landmark textbook.

- 1941 — Fritz Lipmann proposes the concept of the "high-energy phosphate bond" and articulates ATP as the common intermediate coupling catabolism to biosynthesis.

- 1957 — Systematic tables of biochemical standard free energies of hydrolysis (ATP, other phosphorylated compounds) are compiled, enabling routine

\Delta G^{\circ\prime}

-based pathway analysis.

Why does biochemistry use ΔG°′ instead of the chemist's ΔG°?

Chemists define the standard state with all species (including H⁺) at 1 M activity. But cellular pH is about 7, meaning [H⁺] ≈ 10⁻⁷ M, not 1 M — using 1 M H⁺ would make every biological standard state absurdly acidic and physiologically meaningless. Biochemists therefore define a modified standard state, denoted with a prime, in which pH is fixed at 7 (and water's concentration, ~55.5 M, is folded into the constant rather than treated as a variable) while all other solutes remain at 1 M. $\Delta G^{\circ\prime}$ values in biochemistry tables always use this convention; converting between $\Delta G^{\circ}$ and $\Delta G^{\circ\prime}$ for a reaction involving H⁺ requires knowing the stoichiometric number of protons and applying the corresponding $RT\ln[\text{H}^+]$ correction.

Is it true that ATP has a special "high-energy bond" that stores energy?

Not in the sense of a uniquely strong or strained covalent bond storing potential energy like a coiled spring. The large negative $\Delta G^{\circ\prime}$ of ATP hydrolysis arises from several cooperating factors: electrostatic repulsion among the closely spaced negative charges on the triphosphate chain is relieved upon hydrolysis, the products (ADP²⁻ and Pi²⁻) are stabilized by greater resonance delocalization than the triphosphate, and both products are more effectively solvated by water than ATP is. "High-energy bond" is a historically entrenched shorthand (coined by Lipmann) for "a bond whose hydrolysis has a large negative $\Delta G^{\circ\prime}$ ," not a statement about unusual bond strength — in fact the P–O bond being broken is not unusually weak in a bond-dissociation-energy sense.

Can a reaction with a favorable ΔG (negative) still fail to occur at any appreciable rate?

Yes, and this is a crucial distinction from kinetics. $\Delta G$ tells you the thermodynamic feasibility and direction of a reaction — whether it can happen — but says nothing about the rate at which it happens, which is governed by the activation energy barrier and is the domain of enzyme kinetics, not thermodynamics. The hydrolysis of ATP to ADP and Pi is enormously exergonic under cellular conditions, yet ATP is kinetically stable in water for hours in the absence of a catalyst, because the uncatalyzed reaction has a substantial activation barrier. This kinetic stability is itself biologically essential — ATP would be useless as an energy currency if it decomposed spontaneously the instant it was made; enzymes selectively lower the barrier only when and where the energy is needed.

How is ΔG°′ actually measured or looked up for a given reaction?

In practice, $\Delta G^{\circ\prime}$ is rarely measured calorimetrically. Instead, the reaction is allowed to reach equilibrium in vitro (often catalyzed by the relevant enzyme to reach equilibrium quickly, since the enzyme changes only rate, not the equilibrium position), the equilibrium concentrations of all reactants and products are measured (by spectrophotometry, HPLC, or coupled enzymatic assays), $K'_{eq}$ is computed from those concentrations, and $\Delta G^{\circ\prime} = -RT\ln K'_{eq}$ is then calculated. Comprehensive tables (originally compiled by researchers such as Robert Alberty) list $\Delta G^{\circ\prime}_f$ (standard free energy of formation) for individual metabolites, from which the $\Delta G^{\circ\prime}$ of any reaction connecting them can be computed as a difference of formation energies, analogous to using $\Delta H^{\circ}_f$ tables in general chemistry.

Does energetic coupling require the two reactions to occur in the same active site?

No — thermodynamic coupling via a shared intermediate works regardless of whether the two half-reactions occur at the same active site, in different active sites of the same enzyme, or in entirely different enzymes and even different cellular compartments, as long as the intermediate species (ATP, NADH, a proton gradient, an acyl-CoA thioester) is genuinely shared between them. However, kinetic efficiency is often far higher when coupling happens within one active site (as in hexokinase, where phosphoryl transfer occurs directly from ATP to glucose without ever releasing free phosphate into solution), since this avoids diffusional loss and unwanted side reactions of a reactive intermediate. Both mechanisms — direct group transfer in one active site, and diffusible shared intermediates connecting separate reactions — are thermodynamically identical in their free-energy bookkeeping.

Why is it wrong to add K'eq values for sequential reactions the way ΔG°′ values are added?

Because $\Delta G^{\circ\prime} = -RT\ln K'_{eq}$ is a logarithmic relationship: summing $\Delta G^{\circ\prime}$ values corresponds to summing logarithms of $K'_{eq}$ values, which by the laws of logarithms means the $K'_{eq}$ values themselves must be multiplied, not added, to get the overall equilibrium constant. Concretely, $K'_{eq,\text{overall}} = \prod_i K'_{eq,i}$ follows directly from exponentiating $\Delta G^{\circ\prime}_{\text{overall}} = \sum_i \Delta G^{\circ\prime}_i = -RT\sum_i \ln K'_{eq,i} = -RT\ln\left(\prod_i K'_{eq,i}\right)$ . Adding equilibrium constants directly has no thermodynamic justification and will generally give a badly wrong answer, especially when the constants span many orders of magnitude, as biochemical ones often do.

Free Energy, Spontaneity, and Coupled Reactions

Where this lives today

The physical picture — free energy is a moving target, not a label

The core ideas — the governing equations

Worked examples

Limiting cases, scope & common traps

The chemist behind it

Before this

Where this leads