Intuition The one-line pictureChemical potential is the energy cost (or payoff) of adding one more particle to a system while keeping temperature and pressure fixed. If you slip an extra molecule into a glass of water, the Gibbs free energy changes by exactly μ \mu μ high μ \mu μ μ \mu μ , just as heat flows from high T T T T T T
The first law gives d U = T d S − P d V dU = T\,dS - P\,dV d U = T d S − P d V number of particles is fixed . The moment particles can move (diffusion, evaporation, chemical reaction, phase change), we need a new "knob" that tells us how the energy of the system responds to changing N N N
T T T S S S P P P V V V = = μ = = ==\mu== == μ == N N N particle flow ).
Definition Chemical potentialThe chemical potential is the rate of change of a thermodynamic potential with particle number, holding the natural variables of that potential fixed :
μ ≡ ( ∂ G ∂ N ) T , P = ( ∂ U ∂ N ) S , V = ( ∂ F ∂ N ) T , V = − T ( ∂ S ∂ N ) U , V \mu \equiv \left(\frac{\partial G}{\partial N}\right)_{T,P} = \left(\frac{\partial U}{\partial N}\right)_{S,V} = \left(\frac{\partial F}{\partial N}\right)_{T,V} = -T\left(\frac{\partial S}{\partial N}\right)_{U,V} μ ≡ ( ∂ N ∂ G ) T , P = ( ∂ N ∂ U ) S , V = ( ∂ N ∂ F ) T , V = − T ( ∂ N ∂ S ) U , V same number μ \mu μ T T T P P P
Formula Derivation from the first lawStep 1 — Generalize the first law. Allow N N N S , V , N S,V,N S , V , N d U = T d S − P d V + μ d N dU = T\,dS - P\,dV + \mu\,dN d U = T d S − P d V + μ d N Why this step? We define μ \mu μ d N dN d N S , V S,V S , V
Step 2 — Build Gibbs free energy. By definition
G ≡ U − T S + P V . G \equiv U - TS + PV. G ≡ U − T S + P V . Why this step? G G G T , P , N T,P,N T , P , N μ \mu μ G G G
Step 3 — Differentiate G G G
d G = d U − T d S − S d T + P d V + V d P . dG = dU - T\,dS - S\,dT + P\,dV + V\,dP. d G = d U − T d S − S d T + P d V + V d P . Why this step? Just the product rule on − T S -TS − T S + P V +PV + P V
Step 4 — Substitute the first law for d U dU d U
d G = ( T d S − P d V + μ d N ) − T d S − S d T + P d V + V d P . dG = (T\,dS - P\,dV + \mu\,dN) - T\,dS - S\,dT + P\,dV + V\,dP. d G = ( T d S − P d V + μ d N ) − T d S − S d T + P d V + V d P . T d S T\,dS T d S P d V P\,dV P d V cancel :
d G = − S d T + V d P + μ d N \boxed{dG = -S\,dT + V\,dP + \mu\,dN} d G = − S d T + V d P + μ d N Why this step? This is the master differential of G G G G G G d T , d P , d N dT,dP,dN d T , d P , d N
Step 5 — Read off the partial derivative. Holding T T T P P P μ = ( ∂ G ∂ N ) T , P . ■ \mu = \left(\frac{\partial G}{\partial N}\right)_{T,P}. \qquad \blacksquare μ = ( ∂ N ∂ G ) T , P . ■
Intuition Why μ is just Gibbs energy per particleG G G extensive : double the system at fixed T , P T,P T , P G G G G G G N N N G ( T , P , λ N ) = λ G ( T , P , N ) . G(T,P,\lambda N) = \lambda\,G(T,P,N). G ( T , P , λ N ) = λ G ( T , P , N ) . G = N ( ∂ G ∂ N ) T , P = N μ ⇒ μ = G N . G = N\left(\frac{\partial G}{\partial N}\right)_{T,P} = N\mu \quad\Rightarrow\quad \mu = \frac{G}{N}. G = N ( ∂ N ∂ G ) T , P = N μ ⇒ μ = N G . single pure substance , μ \mu μ per particle (or per mole if N N N partial molar Gibbs energies μ i = ( ∂ G / ∂ N i ) T , P , N j ≠ i \mu_i = (\partial G/\partial N_i)_{T,P,N_{j\ne i}} μ i = ( ∂ G / ∂ N i ) T , P , N j = i
Worked example Example 1 — Diffusive equilibrium between two boxesTwo gas chambers A and B share a wall that lets particles through, at common T , P T,P T , P N = N A + N B N = N_A + N_B N = N A + N B d N A = − d N B dN_A = -dN_B d N A = − d N B G G G d G = 0 dG=0 d G = 0 d G = μ A d N A + μ B d N B = ( μ A − μ B ) d N A = 0. dG = \mu_A\,dN_A + \mu_B\,dN_B = (\mu_A - \mu_B)\,dN_A = 0. d G = μ A d N A + μ B d N B = ( μ A − μ B ) d N A = 0. d N A dN_A d N A μ A = μ B \mu_A = \mu_B μ A = μ B Why this step? Equilibrium = no net drive to move particles, which means equal chemical potentials. Particles flow until μ levels out.
Worked example Example 2 — Ideal gas chemical potentialFor an ideal gas one can show (from μ = G / N \mu = G/N μ = G / N G G G μ ( T , P ) = μ ∘ ( T ) + k B T ln P P ∘ . \mu(T,P) = \mu^\circ(T) + k_BT\ln\!\frac{P}{P^\circ}. μ ( T , P ) = μ ∘ ( T ) + k B T ln P ∘ P . Why this step? At fixed T T T d G = V d P dG = V\,dP d G = V d P V = N k B T / P V=Nk_BT/P V = N k B T / P d μ = ( k B T / P ) d P d\mu = (k_BT/P)\,dP d μ = ( k B T / P ) d P P ∘ P^\circ P ∘ P P P ln P \ln P ln P P P P raises its chemical potential — it gets "more eager" to escape to lower-pressure regions.
Worked example Example 3 — Which way does matter flow?Box A has μ A = − 0.30 \mu_A = -0.30 μ A = − 0.30 μ B = − 0.50 \mu_B = -0.50 μ B = − 0.50 T , P T,P T , P d N dN d N G G G d G = ( μ B − μ A ) d N = ( − 0.50 − ( − 0.30 ) ) d N = − 0.20 d N dG = (\mu_B - \mu_A)\,dN = (-0.50-(-0.30))\,dN = -0.20\,dN d G = ( μ B − μ A ) d N = ( − 0.50 − ( − 0.30 )) d N = − 0.20 d N d G < 0 dG<0 d G < 0 d N > 0 dN>0 d N > 0 particles flow A→B (high μ → low μ) until μ A = μ B \mu_A=\mu_B μ A = μ B Why this step? Spontaneous = lowers G G G μ B − μ A \mu_B-\mu_A μ B − μ A
Common mistake "μ is the same as potential energy of a particle."Why it feels right: it's called a potential and has energy units. The fix: μ \mu μ free energy per particle — it bundles in entropy (μ = ( ∂ U / ∂ N ) − T ( ∂ S / ∂ N ) … \mu = (\partial U/\partial N) - T(\partial S/\partial N)\dots μ = ( ∂ U / ∂ N ) − T ( ∂ S / ∂ N ) … different μ \mu μ k B T ln ( P ) k_BT\ln(P) k B T ln ( P )
Common mistake "μ = (∂G/∂N) means just take any derivative of G."Why it feels right: G G G The fix: the subscript T , P T,P T , P . ( ∂ G / ∂ N ) T , P = μ (\partial G/\partial N)_{T,P}=\mu ( ∂ G / ∂ N ) T , P = μ ( ∂ G / ∂ N ) S , V (\partial G/\partial N)_{S,V} ( ∂ G / ∂ N ) S , V
Common mistake "More particles always means higher μ."Why it feels right: adding stuff costs energy. The fix: μ \mu μ concentration/pressure and T T T , not raw count. Diluting (lower P P P lowers μ \mu μ lower μ \mu μ equal ).
Recall Feynman: explain it to a 12-year-oldImagine each room in a house can hold kids, and every kid has a "comfort score." A crowded room has a low comfort score, so kids wander toward emptier rooms. Chemical potential is that comfort score for tiny particles — but for the whole crowd, not one kid. Particles always shuffle from rooms where it's "expensive" to be (high μ \mu μ μ \mu μ
"μ is the Membership fee to join the club G, paid per new member, with the Thermostat (T) and Pressure (P) held."
And the flow rule: "μ flows like a river — Mountains (high μ) to mUd (low μ)."
Recall Test yourself before looking
State the master differential of G G G
Why does μ = G / N \mu = G/N μ = G / N
In which direction do particles flow between two reservoirs?
What is the definition of chemical potential μ in terms of G? μ = ( ∂ G ∂ N ) T , P \mu = \left(\frac{\partial G}{\partial N}\right)_{T,P} μ = ( ∂ N ∂ G ) T , P — the change in Gibbs free energy per added particle at fixed T and P.
What is the master differential of Gibbs free energy with variable N? d G = − S d T + V d P + μ d N dG = -S\,dT + V\,dP + \mu\,dN d G = − S d T + V d P + μ d N .
Which thermodynamic variable is μ conjugate to? Particle number
N N N (just as
T T T ↔
S S S and
P P P ↔
V V V ).
Give all four equivalent definitions of μ. ( ∂ U / ∂ N ) S , V = ( ∂ F / ∂ N ) T , V = ( ∂ G / ∂ N ) T , P = − T ( ∂ S / ∂ N ) U , V (\partial U/\partial N)_{S,V} = (\partial F/\partial N)_{T,V} = (\partial G/\partial N)_{T,P} = -T(\partial S/\partial N)_{U,V} ( ∂ U / ∂ N ) S , V = ( ∂ F / ∂ N ) T , V = ( ∂ G / ∂ N ) T , P = − T ( ∂ S / ∂ N ) U , V .
Why does μ = G/N for a pure substance? Because
G G G is extensive (homogeneous degree 1 in N); Euler's theorem gives
G = N μ G = N\mu G = N μ .
Condition for diffusive equilibrium between connected systems? Equal chemical potentials:
μ A = μ B \mu_A = \mu_B μ A = μ B (at common T, P).
Which way do particles spontaneously flow? From high μ to low μ (lowering total G).
Chemical potential of an ideal gas? μ ( T , P ) = μ ∘ ( T ) + k B T ln ( P / P ∘ ) \mu(T,P) = \mu^\circ(T) + k_BT\ln(P/P^\circ) μ ( T , P ) = μ ∘ ( T ) + k B T ln ( P / P ∘ ) .
Steel-man: is μ the same as a particle's potential energy? No — μ is a free energy per particle including entropy (
− T ∂ S / ∂ N -T\,\partial S/\partial N − T ∂ S / ∂ N ), so flow can be entropy-driven.
Why is the subscript T,P essential in (∂G/∂N)_{T,P}? Different held variables give different physical quantities; only fixing T,P yields μ from G.
defines coefficient of dN
combined via G eq U minus TS plus PV
differentiate and substitute dU
First law dU eq TdS minus PdV
Particle number N can vary
Generalized first law adds mu dN
mu eq partial G over partial N at T,P
G is extensive degree 1 in N
mu eq G over N pure substance
Particles flow high mu to low mu
Intuition Hinglish mein samjhoDekho, chemical potential μ \mu μ T T T P P P μ = ( ∂ G / ∂ N ) T , P \mu = (\partial G/\partial N)_{T,P} μ = ( ∂ G / ∂ N ) T , P G G G N N N high μ \mu μ μ \mu μ ki taraf bhaagte hain. Jab dono jagah μ \mu μ
Derivation yaad rakhna easy hai. Pehle first law ko generalize karo: d U = T d S − P d V + μ d N dU = TdS - PdV + \mu\,dN d U = T d S − P d V + μ d N μ \mu μ d N dN d N G = U − T S + P V G = U - TS + PV G = U − T S + P V d U dU d U T d S TdS T d S P d V PdV P d V d G = − S d T + V d P + μ d N dG = -S\,dT + V\,dP + \mu\,dN d G = − S d T + V d P + μ d N T T T P P P μ = ( ∂ G / ∂ N ) T , P \mu = (\partial G/\partial N)_{T,P} μ = ( ∂ G / ∂ N ) T , P μ = G / N \mu = G/N μ = G / N G G G
Ek important galti se bachna: μ \mu μ − T ∂ S / ∂ N -T\,\partial S/\partial N − T ∂ S / ∂ N μ = μ ∘ + k B T ln ( P / P ∘ ) \mu = \mu^\circ + k_BT\ln(P/P^\circ) μ = μ ∘ + k B T ln ( P / P ∘ ) μ \mu μ
Yeh concept kyun important hai? Kyunki phase change (paani se bhaap), chemical reactions, diffusion, aur quantum statistics (Fermi level toh μ \mu μ μ \mu μ μ \mu μ
