A thermodynamic potential (or more accurately, a thermodynamic potential energy)^{[1]}^{[2]} is a scalar quantity used to represent the thermodynamic state of a system. The concept of thermodynamic potentials was introduced by Pierre Duhem in 1886. Josiah Willard Gibbs in his papers used the term fundamental functions. One main thermodynamic potential that has a physical interpretation is the internal energy U. It is the energy of configuration of a given system of conservative forces (that is why it is called potential) and only has meaning with respect to a defined set of references (or data). Expressions for all other thermodynamic energy potentials are derivable via Legendre transforms from an expression for U. In thermodynamics, external forces, such as gravity, are typically disregarded when formulating expressions for potentials. For example, while all the working fluid in a steam engine may have higher energy due to gravity while sitting on top of Mount Everest than it would at the bottom of the Mariana Trench, the gravitational potential energy term in the formula for the internal energy would usually be ignored because changes in gravitational potential within the engine during operation would be negligible. In a large system under even homogeneous external force, like the earth atmosphere under gravity, the intensive parameters () should be studied locally having even in equilibrium different values in different places far from each other (see thermodynamic models of troposphere].
Thermodynamics  

The classical Carnot heat engine  


Five common thermodynamic potentials are:^{[3]}
Name  Symbol  Formula  Natural variables 

Internal energy  
Helmholtz free energy  
Enthalpy  
Gibbs free energy  
Landau Potential (Grand potential)  , 
where T = temperature, S = entropy, p = pressure, V = volume. The Helmholtz free energy is in ISO/IEC standard called Helmholtz energy^{[1]} or Helmholtz function. It is often denoted by the symbol F, but the use of A is preferred by IUPAC,^{[4]} ISO and IEC.^{[5]} N_{i} is the number of particles of type i in the system and μ_{i} is the chemical potential for an itype particle. The set of all N_{i} are also included as natural variables but may be ignored when no chemical reactions are occurring which cause them to change.
These five common potentials are all potential energies, but there are also entropy potentials. The thermodynamic square can be used as a tool to recall and derive some of the potentials.
Just as in mechanics, where potential energy is defined as capacity to do work, similarly different potentials have different meanings:
From these meanings (which actually apply in specific conditions, e.g. constant pressure, temperature, etc), we can say that ΔU is the energy added to the system, ΔF is the total work done on it, ΔG is the nonmechanical work done on it, and ΔH is the sum of nonmechanical work done on the system and the heat given to it. Thermodynamic potentials are very useful when calculating the equilibrium results of a chemical reaction, or when measuring the properties of materials in a chemical reaction. The chemical reactions usually take place under some constraints such as constant pressure and temperature, or constant entropy and volume, and when this is true, there is a corresponding thermodynamic potential that comes into play. Just as in mechanics, the system will tend towards lower values of potential and at equilibrium, under these constraints, the potential will take on an unchanging minimum value. The thermodynamic potentials can also be used to estimate the total amount of energy available from a thermodynamic system under the appropriate constraint.
In particular: (see principle of minimum energy for a derivation)^{[6]}
The variables that are held constant in this process are termed the natural variables of that potential.^{[7]} The natural variables are important not only for the abovementioned reason, but also because if a thermodynamic potential can be determined as a function of its natural variables, all of the thermodynamic properties of the system can be found by taking partial derivatives of that potential with respect to its natural variables and this is true for no other combination of variables. On the converse, if a thermodynamic potential is not given as a function of its natural variables, it will not, in general, yield all of the thermodynamic properties of the system.
The set of natural variables for the above four potentials are formed from every combination of the TS and PV variables, excluding any pairs of conjugate variables. There is no reason to ignore the N_{i} − μ_{i} conjugate pairs, and in fact we may define four additional potentials for each species.^{[8]} Using IUPAC notation in which the brackets contain the natural variables (other than the main four), we have:
Formula  Natural variables 
If there is only one species, then we are done. But, if there are, say, two species, then there will be additional potentials such as and so on. If there are D dimensions to the thermodynamic space, then there are 2^{D} unique thermodynamic potentials. For the most simple case, a single phase ideal gas, there will be three dimensions, yielding eight thermodynamic potentials.
The definitions of the thermodynamic potentials may be differentiated and, along with the first and second laws of thermodynamics, a set of differential equations known as the fundamental equations follow.^{[9]} (Actually they are all expressions of the same fundamental thermodynamic relation, but are expressed in different variables.) By the first law of thermodynamics, any differential change in the internal energy U of a system can be written as the sum of heat flowing into the system and work done by the system on the environment, along with any change due to the addition of new particles to the system:
where δQ is the infinitesimal heat flow into the system, and δW is the infinitesimal work done by the system, μ_{i} is the chemical potential of particle type i and N_{i} is the number of type i particles. (Neither δQ nor δW are exact differentials. Small changes in these variables are, therefore, represented with δ rather than d.)
By the second law of thermodynamics, we can express the internal energy change in terms of state functions and their differentials. In case of reversible changes we have:
where
and V is volume, and the equality holds for reversible processes.
This leads to the standard differential form of the internal energy in case of a quasistatic reversible change:
Since U, S and V are thermodynamic functions of state, the above relation holds also for arbitrary nonreversible changes. If the system has more external variables than just the volume that can change, the fundamental thermodynamic relation generalizes to:
Here the X_{i} are the generalized forces corresponding to the external variables x_{i}.^{[10]}
Applying Legendre transforms repeatedly, the following differential relations hold for the four potentials:
The infinitesimals on the righthand side of each of the above equations are of the natural variables of the potential on the lefthand side. Similar equations can be developed for all of the other thermodynamic potentials of the system. There will be one fundamental equation for each thermodynamic potential, resulting in a total of 2^{D} fundamental equations.
The differences between the four thermodynamic potentials can be summarized as follows:
We can use the above equations to derive some differential definitions of some thermodynamic parameters. If we define Φ to stand for any of the thermodynamic potentials, then the above equations are of the form:
where x_{i} and y_{i} are conjugate pairs, and the y_{i} are the natural variables of the potential Φ. From the chain rule it follows that:
Where y_{i ≠ j} is the set of all natural variables of Φ except y_{i} . This yields expressions for various thermodynamic parameters in terms of the derivatives of the potentials with respect to their natural variables. These equations are known as equations of state since they specify parameters of the thermodynamic state.^{[11]} If we restrict ourselves to the potentials U, F, H and G, then we have:
where, in the last equation, ϕ is any of the thermodynamic potentials U, F, H, G and are the set of natural variables for that potential, excluding N_{i} . If we use all potentials, then we will have more equations of state such as
and so on. In all, there will be D equations for each potential, resulting in a total of D 2^{D} equations of state. If the D equations of state for a particular potential are known, then the fundamental equation for that potential can be determined. This means that all thermodynamic information about the system will be known, and that the fundamental equations for any other potential can be found, along with the corresponding equations of state.
The above equations of state suggest methods to experimentally measure changes in the thermodynamic potentials using physically measureable parameters. For example the free energy expressions
and
can be integrated at constant temperature and quantities to obtain:
which can be measured by monitoring the measureable variables of pressure, temperature and volume. Changes in the enthalpy and internal energy can be measured by calorimetry (which measures the amount of heat ΔQ released or absorbed by a system). The expressions
can be integrated:
Note that these measurements are made at constant {N_{j} } and are therefore not applicable to situations in which chemical reactions take place.
Again, define x_{i} and y_{i} to be conjugate pairs, and the y_{i} to be the natural variables of some potential Φ. We may take the "cross differentials" of the state equations, which obey the following relationship:
From these we get the Maxwell relations.^{[3]}^{[12]} There will be (D − 1)/2 of them for each potential giving a total of D(D − 1)/2 equations in all. If we restrict ourselves the U, F, H, G
Using the equations of state involving the chemical potential we get equations such as:
and using the other potentials we can get equations such as:
Again, define x_{i} and y_{i} to be conjugate pairs, and the y_{i} to be the natural variables of the internal energy. Since all of the natural variables of the internal energy U are extensive quantities
it follows from Euler's homogeneous function theorem that the internal energy can be written as:
From the equations of state, we then have:
Substituting into the expressions for the other main potentials we have:
As in the above sections, this process can be carried out on all of the other thermodynamic potentials. The Euler integrals are sometimes also referred to as fundamental equations.
Deriving the Gibbs–Duhem equation from basic thermodynamic state equations is straightforward.^{[9]}^{[13]}^{[14]} Equating any thermodynamic potential definition with its Euler integral expression yields:
Differentiating, and using the second law:
yields:
Which is the Gibbs–Duhem relation. The Gibbs–Duhem is a relationship among the intensive parameters of the system. It follows that for a simple system with I components, there will be I + 1 independent parameters, or degrees of freedom. For example, a simple system with a single component will have two degrees of freedom, and may be specified by only two parameters, such as pressure and volume for example. The law is named after Josiah Willard Gibbs and Pierre Duhem.
Changes in these quantities are useful for assessing the degree to which a chemical reaction will proceed. The relevant quantity depends on the reaction conditions, as shown in the following table. Δ denotes the change in the potential and at equilibrium the change will be zero.
Constant V  Constant p  

Constant S  ΔU  ΔH 
Constant T  ΔF  ΔG 
Most commonly one considers reactions at constant p and T, so the Gibbs free energy is the most useful potential in studies of chemical reactions.