Intuition The big picture
A thermodynamic potential is an energy-like quantity whose natural variables you can control in a lab, and whose minimum tells you the equilibrium state. Just as a ball rolls to the bottom of a valley (minimum of potential energy), a thermodynamic system "rolls" to the minimum of the right potential — but which potential depends on what you hold fixed (entropy? temperature? pressure? volume?).
WHY four of them? Because in different experiments different things are held constant. You don't control entropy easily, but you can control temperature (thermostat) and pressure (open beaker). Each potential is engineered to make a particular pair of variables the "natural" ones.
Definition Internal energy
U U U
U U U is the total energy stored in a system. The combined 1st + 2nd law for a reversible process (closed system, no other work) is
d U = T d S − p d V dU = T\,dS - p\,dV d U = T d S − p d V
WHY this form? First law: d U = δ Q − δ W dU = \delta Q - \delta W d U = δ Q − δ W . For a reversible path δ Q = T d S \delta Q = T\,dS δ Q = T d S (definition of entropy) and δ W = p d V \delta W = p\,dV δ W = p d V . Substitute:
d U = T d S − p d V . dU = T\,dS - p\,dV. d U = T d S − p d V .
This equation is the mother of all potentials. Reading it as U = U ( S , V ) U=U(S,V) U = U ( S , V ) , the natural variables of U U U are = = S and V = = ==S \text{ and } V== == S and V == , because d U dU d U is naturally expressed in terms of d S dS d S and d V dV d V .
Intuition What is a Legendre transform (Feynman style)?
Suppose your energy is written in a variable you can't control (like S S S ). You'd rather use its partner T T T (which you can control with a thermostat). A Legendre transform swaps a variable for its conjugate slope by subtracting off the product .
HOW it works. We have T = ( ∂ U / ∂ S ) V T = (\partial U/\partial S)_V T = ( ∂ U / ∂ S ) V . To switch from S S S to T T T , define a new function by subtracting T S TS T S :
F = U − T S . F = U - TS. F = U − T S .
Then
d F = d U − T d S − S d T = ( T d S − p d V ) − T d S − S d T = − S d T − p d V . dF = dU - T\,dS - S\,dT = (T\,dS - p\,dV) - T\,dS - S\,dT = -S\,dT - p\,dV. d F = d U − T d S − S d T = ( T d S − p d V ) − T d S − S d T = − S d T − p d V .
The T d S T\,dS T d S terms cancel , leaving d F dF d F in terms of d T dT d T and d V dV d V . That's the whole magic.
The same trick swapping − p d V → + V d p -p\,dV \to +V\,dp − p d V → + V d p uses + p V +pV + p V .
U (internal energy) U \quad\text{(internal energy)} U (internal energy)
H = U + p V (enthalpy) H = U + pV \quad\text{(enthalpy)} H = U + p V (enthalpy)
F = U − T S (Helmholtz free energy) F = U - TS \quad\text{(Helmholtz free energy)} F = U − T S (Helmholtz free energy)
G = H − T S = U + p V − T S (Gibbs free energy) G = H - TS = U + pV - TS \quad\text{(Gibbs free energy)} G = H − T S = U + p V − T S (Gibbs free energy)
Derivation of d H dH d H : H = U + p V ⇒ d H = d U + p d V + V d p = ( T d S − p d V ) + p d V + V d p = T d S + V d p . H=U+pV \Rightarrow dH = dU + p\,dV + V\,dp = (T\,dS - p\,dV)+p\,dV+V\,dp = T\,dS+V\,dp. H = U + p V ⇒ d H = d U + p d V + V d p = ( T d S − p d V ) + p d V + V d p = T d S + V d p . Why this step? The − p d V -p\,dV − p d V from d U dU d U cancels the + p d V +p\,dV + p d V from d ( p V ) d(pV) d ( p V ) .
Derivation of d F dF d F : done above. Why? Subtracting T S TS T S kills the T d S T\,dS T d S term, promoting T T T to a natural variable.
Derivation of d G dG d G : G = U + p V − T S G=U+pV-TS G = U + p V − T S , combine both tricks:
d G = d U + p d V + V d p − T d S − S d T = − S d T + V d p . dG = dU + p\,dV+V\,dp - T\,dS - S\,dT = -S\,dT + V\,dp. d G = d U + p d V + V d p − T d S − S d T = − S d T + V d p .
Why? Both T d S T\,dS T d S and p d V p\,dV p d V cancel, leaving the lab-friendly pair ( T , p ) (T,p) ( T , p ) .
Intuition Which potential is "the boss"?
The 2nd law says the total entropy of the universe never decreases . Translating that constraint to the system alone under different lab conditions turns "entropy maximized" into "the right potential minimized."
Held constant
Minimized potential
Lab situation
S , V S,V S , V
U U U
isolated, rigid (rare)
S , p S,p S , p
H H H
adiabatic, constant pressure
T , V T,V T , V
F F F
thermostat, rigid box
T , p T,p T , p
G G G
thermostat, open to atmosphere (chemistry! )
Intuition Why "free energy"?
F = U − T S F=U-TS F = U − T S : the part of internal energy free to do work at constant T T T — you must "pay" T S TS T S to the entropy bath. Δ G < 0 \Delta G<0 Δ G < 0 tells a chemist a reaction at room conditions runs spontaneously.
Worked example Example 1 — Enthalpy = heat at constant pressure
Show δ Q p = d H \delta Q_p = dH δ Q p = d H .
Step 1: d H = T d S + V d p dH = T\,dS + V\,dp d H = T d S + V d p . Why? Use the differential we derived.
Step 2: At constant pressure d p = 0 dp=0 d p = 0 , so d H = T d S = δ Q rev dH = T\,dS = \delta Q_{\text{rev}} d H = T d S = δ Q rev . Why? δ Q = T d S \delta Q=T\,dS δ Q = T d S .
Result: heat absorbed at constant p p p equals Δ H \Delta H Δ H — that's why enthalpy is the natural "heat content" for chemistry done in open flasks. C p = ( ∂ H / ∂ T ) p C_p=(\partial H/\partial T)_p C p = ( ∂ H / ∂ T ) p .
Worked example Example 2 — Get an entropy from
G G G
Given G ( T , p ) G(T,p) G ( T , p ) , find S S S .
Step 1: d G = − S d T + V d p dG=-S\,dT+V\,dp d G = − S d T + V d p . Why? Natural-variable differential.
Step 2: At constant p p p , ( ∂ G ∂ T ) p = − S \left(\frac{\partial G}{\partial T}\right)_p=-S ( ∂ T ∂ G ) p = − S . Why? Match the d T dT d T coefficient.
Result: S = − ( ∂ G ∂ T ) p S=-\left(\dfrac{\partial G}{\partial T}\right)_p S = − ( ∂ T ∂ G ) p , and likewise V = ( ∂ G ∂ p ) T V=\left(\dfrac{\partial G}{\partial p}\right)_T V = ( ∂ p ∂ G ) T .
Worked example Example 3 — A Maxwell relation from
F F F
Derive ( ∂ S ∂ V ) T = ( ∂ p ∂ T ) V \left(\frac{\partial S}{\partial V}\right)_T=\left(\frac{\partial p}{\partial T}\right)_V ( ∂ V ∂ S ) T = ( ∂ T ∂ p ) V .
Step 1: d F = − S d T − p d V dF=-S\,dT-p\,dV d F = − S d T − p d V , so S = − ( ∂ F / ∂ T ) V S=-(\partial F/\partial T)_V S = − ( ∂ F / ∂ T ) V , p = − ( ∂ F / ∂ V ) T p=-(\partial F/\partial V)_T p = − ( ∂ F / ∂ V ) T .
Step 2: Mixed second derivatives of F F F are equal: ∂ 2 F ∂ V ∂ T = ∂ 2 F ∂ T ∂ V \frac{\partial^2 F}{\partial V\,\partial T}=\frac{\partial^2 F}{\partial T\,\partial V} ∂ V ∂ T ∂ 2 F = ∂ T ∂ V ∂ 2 F . Why? F F F is a state function (exact differential).
Step 3: That gives − ( ∂ S ∂ V ) T = − ( ∂ p ∂ T ) V -\left(\frac{\partial S}{\partial V}\right)_T = -\left(\frac{\partial p}{\partial T}\right)_V − ( ∂ V ∂ S ) T = − ( ∂ T ∂ p ) V .
Result: ( ∂ S ∂ V ) T = ( ∂ p ∂ T ) V \boxed{\left(\frac{\partial S}{\partial V}\right)_T=\left(\frac{\partial p}{\partial T}\right)_V} ( ∂ V ∂ S ) T = ( ∂ T ∂ p ) V — a hard-to-measure entropy slope replaced by an easy gas-law quantity.
Common mistake "Free energy means energy that's free / costs nothing."
Why it feels right: the word "free." Truth: F F F /G G G is the maximum useful work extractable at fixed T T T (and p p p for G G G ). The entropy term T S TS T S is energy locked away by the second law. Fix: read "free" as "available to do work," not "free of charge."
Common mistake "All potentials are minimized at equilibrium, always."
Why it feels right: symmetry of the four definitions. Truth: minimization holds only for the matching constraints (e.g. G G G at constant T , p T,p T , p ). Under the wrong constraints a potential need not be extremal. Fix: match potential ⇄ its natural variables held fixed.
Common mistake Wrong sign in
d H dH d H .
Writing d H = T d S − V d p dH=T\,dS-V\,dp d H = T d S − V d p . Why it feels right: d U dU d U has a minus sign too. Fix: the p V pV p V added flips the work sign: + p d V +p\,dV + p d V in d ( p V ) d(pV) d ( p V ) cancels − p d V -p\,dV − p d V and leaves + V d p +V\,dp + V d p .
Recall Feynman: explain to a 12-year-old
Energy is like money in your wallet. But how much you can actually spend depends on the rules of where you are. In a hot room (fixed temperature) a "tax" called T S TS T S gets taken away — what's left to spend is the free energy F F F . If the room can also push on you with air pressure, you must also reserve some money for that (p V pV p V ), and what's left is G G G . Each "wallet" (U , H , F , G U,H,F,G U , H , F , G ) is just your energy after accounting for which taxes apply in your situation , and the system always settles where its relevant wallet is smallest.
Mnemonic Remembering the four & their variables
"Good Physicists Have Studied Under Very Fine Teachers" isn't it — use the square :
Corners clockwise: U U U –H H H –G G G –F F F with variables on the sides. Or just remember the cheat:
Add p V pV p V → enthalpy; subtract T S TS T S → free; both → Gibbs.
Sign rule: "S follows T, V follows p." When T T T is natural you get − S d T -S\,dT − S d T ; when p p p is natural you get + V d p +V\,dp + V d p .
What is the fundamental differential of U U U ? d U = T d S − p d V dU = T\,dS - p\,dV d U = T d S − p d V (natural variables
S , V S,V S , V )
Define enthalpy H H H and give d H dH d H . H = U + p V H=U+pV H = U + p V ;
d H = T d S + V d p dH=T\,dS+V\,dp d H = T d S + V d p (natural variables
S , p S,p S , p )
Define Helmholtz free energy and give d F dF d F . F = U − T S F=U-TS F = U − T S ;
d F = − S d T − p d V dF=-S\,dT-p\,dV d F = − S d T − p d V (natural variables
T , V T,V T , V )
Define Gibbs free energy and give d G dG d G . G = U + p V − T S G=U+pV-TS G = U + p V − T S ;
d G = − S d T + V d p dG=-S\,dT+V\,dp d G = − S d T + V d p (natural variables
T , p T,p T , p )
Which potential is minimized at constant T , p T,p T , p ? Gibbs free energy
G G G (the chemistry case)
Which potential is minimized at constant T , V T,V T , V ? Helmholtz free energy
F F F Why does enthalpy matter for constant-pressure heat? At
d p = 0 dp=0 d p = 0 ,
d H = T d S = δ Q dH=T\,dS=\delta Q d H = T d S = δ Q , so
Δ H \Delta H Δ H is heat absorbed
Express S S S and V V V from G ( T , p ) G(T,p) G ( T , p ) . S = − ( ∂ G / ∂ T ) p S=-(\partial G/\partial T)_p S = − ( ∂ G / ∂ T ) p ,
V = ( ∂ G / ∂ p ) T V=(\partial G/\partial p)_T V = ( ∂ G / ∂ p ) T What is a Legendre transform doing here? Swapping a variable for its conjugate by subtracting their product, to change natural variables
State the Maxwell relation from F F F . ( ∂ S / ∂ V ) T = ( ∂ p / ∂ T ) V (\partial S/\partial V)_T=(\partial p/\partial T)_V ( ∂ S / ∂ V ) T = ( ∂ p / ∂ T ) V What does "free" in free energy mean? Maximum work extractable at fixed
T T T (and
p p p for
G G G ); not "costless"
First Law of Thermodynamics — source of d U = T d S − p d V dU=TdS-pdV d U = T d S − p d V
Entropy and the Second Law — why minimization replaces maximization
Maxwell Relations — direct consequence of exact differentials of U , H , F , G U,H,F,G U , H , F , G
Legendre Transform — the mathematical engine
Heat Capacities Cp and Cv — C p = ( ∂ H / ∂ T ) p C_p=(\partial H/\partial T)_p C p = ( ∂ H / ∂ T ) p , C v = ( ∂ U / ∂ T ) V C_v=(\partial U/\partial T)_V C v = ( ∂ U / ∂ T ) V
Gibbs Free Energy and Chemical Equilibrium — Δ G < 0 \Delta G<0 Δ G < 0 spontaneity
swaps variable for conjugate
Intuition Hinglish mein samjho
Dekho, thermodynamic potential ka matlab hai ek energy-jaisa quantity (U , H , F , G U, H, F, G U , H , F , G ) jiska minimum batata hai ki system equilibrium me kahan settle hoga — bilkul jaise ball valley ke neeche tak roll karke ruk jaati hai. Lekin twist ye hai: kaunsa potential minimize hoga ye depend karta hai ki tum lab me kya constant rakh rahe ho. Agar temperature aur pressure fix hain (jaise open beaker, room temperature — yani chemistry), to Gibbs free energy G G G boss hai.
Sab kuch nikalta hai ek hi master equation se: d U = T d S − p d V dU = T\,dS - p\,dV d U = T d S − p d V . Ab problem ye hai ki entropy S S S ko lab me directly control karna mushkil hai, par temperature T T T ko thermostat se easily control kar sakte ho. To hum Legendre transform ka trick use karte hain — variable ko uske partner se swap karne ke liye unka product subtract/add kar dete hain. T S TS T S subtract karo to F = U − T S F = U - TS F = U − T S banta hai aur T d S T\,dS T d S term cancel ho jaata hai, leaving d F = − S d T − p d V dF = -S\,dT - p\,dV d F = − S d T − p d V . Isi tarah p V pV p V add karke H H H aur dono milake G G G banta hai.
"Free energy" me "free" ka matlab muft nahi hai bhai — iska matlab hai kitna useful kaam nikaal sakte ho fixed temperature pe. T S TS T S wala part second law ke kaaran "lock" ho jaata hai, use nahi kar sakte. Chemistry me Δ G < 0 \Delta G < 0 Δ G < 0 ho to reaction apne aap (spontaneously) chalti hai — isliye G G G itna important hai. Aur ek bonus: in potentials ke exact differential se Maxwell relations nikalti hain, jisse hard-to-measure cheezein (jaise ∂ S / ∂ V \partial S/\partial V ∂ S / ∂ V ) easy cheezon (jaise ∂ p / ∂ T \partial p/\partial T ∂ p / ∂ T ) se replace ho jaati hain. Yahi 80/20 hai — chaar definitions yaad rakho, baaki sab derive ho jayega.