Visual walkthrough — Fermi-Dirac statistics — fermions, Fermi energy

2.4.17 · D2 · Physics › Thermodynamics & Statistical Mechanics (Advanced) › Fermi-Dirac statistics — fermions, Fermi energy

Hum assume karte hain ki tumhe parent se sirf do cheezein pata hain: fermions woh particles hain jo ek state share karne se mana karte hain (the Pauli Exclusion Principle), aur har allowed configuration ek statistical "weight" carry karta hai. Baaki sab — har symbol — hum yahan build karte hain.

---

Step 1 — Ek shelf slot, do possibilities

KYA. Bilkul andar tak zoom karo. Poora metal bhool jao. Sirf ek single quantum state dekho — socho iske baare mein jaise ek bookshelf ka ek slot ho. Ek slot ka ek energy label $\epsilon$ hota hai ("epsilon", bas ek naam jo batata hai ki yahan baithne wala particle kitni energy carry karega).

KYUN. Kyunki fermions antisocial hote hain, slots ek doosre ki counting mein interfere nahi karte. Har slot apni ek tiny duniya hai jo ya to empty ($n=0$) ya full ($n=1$) ho sakti hai. Kabhi $n=2$ nahi — woh do identical fermions ek state mein hoga, jo Pauli Exclusion Principle forbid karta hai. Yahi independence ki wajah se ek mushkil problem asaan one-slot problems mein toot jaati hai.

PICTURE. Do slots side by side: ek dark (empty), ek glowing (ek single electron se occupied). Har ek ke neeche uska occupation $n$ likha hai.

---

Step 2 — Har possibility ko weigh karo (the two-term partition function)

KYA. Hum har possibility ko ek statistical weight dete hain. Grand Canonical Ensemble mein — jahan hamara slot ek giant reservoir ke saath energy aur particles dono trade kar sakta hai — $n$ particles wala configuration yeh weight leta hai: $w(n)=e^{-n(\epsilon -\mu )/k_{B}T}.$

Main har symbol ko wahan pe hi name karta hoon jahan woh exist karta hai:

• $n$ — occupation (0 ya 1), woh cheez jo humne abhi define ki.
• $\epsilon$ — slot ki energy: yahan particle rakhne ka "cost".
• $\mu$ — chemical potential: woh energy "budget" jo reservoir ek particle hand over karne ke liye spend karne ko taiyar hai. Bada $\mu$ = reservoir slots fill karne ke liye eager hai.
• $k_B$ — Boltzmann's constant: temperature aur energy ke beech exchange rate ($1.38\times 10^{-23}\text{J/K}$). Yeh bas $T$ ko energy units mein convert karta hai.
• $T$ — temperature: reservoir kitna thermal "jiggle" supply karta hai.
• $e^{-x}$ — the exponential. Yeh tool kyun, plain fraction nahi? Kyunki independent random choices apne weights multiply karte hain, aur sirf exponential hi energies add karne ko weights multiply karne mein badalta hai ($e^{-a}{e}^{-b}=e^{-(a+b)}$). "Budget se $E$ energy upar" ke liye Nature ka weight hamesha $e^{-E/k_{B}T}$ hota hai — the Boltzmann factor.

KYUN. Probability pane ke liye humein divide karne ke liye ek total chahiye. Sabhi allowed configurations ke weights add karo. Exactly do hain: $\mathcal{Z}=\underset{=1}{\underbrace{w(0)}}+\underset{=e^{-(\epsilon -\mu )/k_{B}T}}{\underbrace{w(1)}}=1+e^{-(\epsilon -\mu )/k_{B}T}.$ $\mathcal{Z}$ (the grand partition function) woh normalizing total hai — yeh slot jitne bhi tarike se ho sakta hai unka sum.

PICTURE. Ek balance scale: left pan mein empty state hai weight $1$ ke saath; right pan mein full state hai weight $e^{-(\epsilon -\mu )/k_{B}T}$ ke saath. Kaun sa pan bhaari hai yeh decide karta hai ki slot full rehna prefer karta hai ya empty.

---

Step 3 — Occupation ka average nikalo → the Fermi-Dirac distribution

KYA. Average occupation $\bar{n}$ har possible $n$ aur uski probability ka product hai. Kisi configuration ki probability = uska weight ÷ $\mathcal{Z}$: $\bar{n}=\frac{0\cdot 1+1\cdot e^{-(\epsilon -\mu )/k_{B}T}}{1+e^{-(\epsilon -\mu )/k_{B}T}}.$ $n=0$ wala term vanish ho jaata hai (zero se multiply hota hai), sirf upar "full" weight bachti hai.

Top aur bottom ko $e^{-(\epsilon -\mu )/k_{B}T}$ se kyun divide karein? Algebra ko ek yaadgar shape mein clean karne ke liye. Numerator aur denominator dono ko $e^{+(\epsilon -\mu )/k_{B}T}$ se multiply karo: $\bar{n}=\frac{1}{e^{+(\epsilon -\mu )/k_{B}T}+1}\equiv f(\epsilon ).$

PICTURE. $f$ ka famous S-curve $\epsilon$ ke against, ek warm temperature ke liye drawn. Yeh $1$ (full, low energy) se slide karke $\frac{1}{2}$ ($\epsilon =\mu$ par) se hote hue $0$ (empty, high energy) tak jaata hai.

---

Step 4 — Curve se teeno regimes padho

KYA. Ek formula, teen stories, $(\epsilon -\mu )$ ke sign ke hisaab se:

Regime	Exponent $(\epsilon -\mu )/k_{B}T$	$e^{(\epsilon -\mu )/k_{B}T}$	$f$
$\epsilon \ll \mu$	large negative	$\to 0$	$\to \frac{1}{0+1}=1$ (full)
$\epsilon =\mu$	exactly $0$	$=1$	$\frac{1}{1+1}=\frac{1}{2}$ (half)
$\epsilon \gg \mu$	large positive	$\to \infty$	$\to \frac{1}{\infty }=0$ (empty)

KYUN yeh matter karta hai. Middle row ek gift hai: $\epsilon =\mu$ par temperature bilkul cancel ho jaata hai, isliye $f=\frac{1}{2}$ har $T>0$ ke liye. Yahi wajah hai ki experimentalists physically $\mu$ pin down karte hain — woh energy dhundho jo exactly half-occupied hai. (Maxwell-Boltzmann Statistics se compare karo, jiska $f=e^{-(\epsilon -\mu )/k_{B}T}$ ka koi aisa fixed crossing nahi hai aur woh 1 se exceed ho sakta hai — fermion probability ke liye yeh nonsense hai.)

PICTURE. Wahi S-curve, ab teeno regions shaded hain aur half-filling point $\epsilon =\mu$ vertical line par crosshair se mark kiya gaya hai.

---

Step 5 — Absolute zero par freeze karo: curve ek cliff ban jaata hai

KYA. $T\to 0$ bhejo. Exponent $(\epsilon -\mu )/k_{B}T$ dekho: jaise $T\to 0$, denominator $k_{B}T\to 0$, isliye koi bhi nonzero numerator $\pm \infty$ tak blow up ho jaata hai.

• $\epsilon <\mu$: numerator negative → exponent $\to -\infty$ → $e^{-\infty }=0$ → $f=1$.
• $\epsilon >\mu$: numerator positive → exponent $\to +\infty$ → $e^{+\infty }=\infty$ → $f=0$.

KYUN. Ab hum $E_F\equiv \mu (T=0)$ define karte hain, yahi Fermi energy hai — absolute zero par chemical potential ki value. Smooth S-curve ek perfect step mein collapse ho jaata hai: $E_F$ se neeche har slot full hai, upar har slot empty hai.

PICTURE. Shrinking $T$ ke liye warm S-curves, $E_F$ par sharp vertical cliff par converge karte hue. Baaye fill block hi Fermi sea hai.

---

Step 6 — k-space fill karo: the Fermi sphere aur slots count karo

KYA. Ab hum poochte hain ki $E_F$ kitni oonchi hai. Volume $V$ ke ek box mein, ek free electron ki state ek wavevector $\mathbf{k}$ se label hoti hai (uska momentum $\hslash \mathbf{k}$ hai). Allowed $\mathbf{k}$-points ek uniform grid par baithe hain; $\mathbf{k}$-space mein volume per state $(2\pi {)}^3/V$ hai. Energy origin se door jaane par badhti hai: $\epsilon ={\hslash }^2{k}^2/2m$, isliye neeche se upar fill karna matlab origin par centered radius $k_F$ ki ek solid sphere fill karna hai.

Sphere kyun? Kyunki $\epsilon$ sirf $|\mathbf{k}|=k$ par depend karta hai, same energy wale saare states same radius par hain. "Pehle lowest energies fill karo" = "pehle smallest radii fill karo" = ek ball grow karo. Us ball ki surface, radius $k_F$ par, Fermi surface hai.

Electrons count karo = (sphere ke andar states) × (spin factor 2, kyunki har $\mathbf{k}$ ek spin-up aur ek spin-down electron hold karta hai): $N=\underset{\text{spin}}{\underbrace{2}}\times \underset{\mathbf{k}\text{-points ki density}}{\underbrace{\frac{V}{(2\pi {)}^3}}}\times \underset{\text{Fermi sphere ka volume}}{\underbrace{\frac{4}{3}\pi k_F^3}}=\frac{V}{3{\pi }^2}{k}_F^3.$

Fermi radius ke liye solve karo: $k_F={\left(3{\pi }^2\frac{N}{V}\right)}^{1/3}=(3{\pi }^{2}n{)}^{1/3},n\equiv \frac{N}{V}\text{(number density)}.$

PICTURE. $\mathbf{k}$-space ka cutaway: dots ka ek grid, radius $k_F$ ki ek filled violet sphere, bahar dots empty. Label padhta hai "fill lowest energy = fill smallest radius".

---

Step 7 — Radius ko energy mein convert karo → Fermi energy formula

KYA. $k_F$ ko surface par evaluate kiye gaye $\epsilon ={\hslash }^2{k}^2/2m$ mein daalo: $E_F=\frac{{\hslash }^2{k}_F^2}{2m}=\frac{{\hslash }^2}{2m}[(3{\pi }^{2}n{)}^{1/3}{]}^2.$

Term by term:

• $\hslash$ — reduced Planck constant, quantum scale set karta hai ($1.055\times 10^{-34}\text{J⋅s}$).
• $m$ — electron mass ($9.11\times 10^{-31}\text{kg}$); halke particles → zyada $E_F$.
• $(3{\pi }^{2}n{)}^{1/3}$ ko power $2$ tak raise karne par neeche wala $2/3$ milta hai.

PICTURE. Ek "ladder to energy" panel: $k_F$ par flat sea surface parabola $\epsilon ={\hslash }^2{k}^2/2m$ ke through $E_F$ height tak map hoti hai.

---

Step 8 — Edge case: ek warm sea sirf surface par ripple karta hai

KYA. Thoda sa $T$ wapas laao jahan $k_{B}T\ll E_F$ ho (room temperature par har metal ke liye sach). Step 5 ki cliff ek narrow ramp mein soften ho jaati hai. Parent ke Example 3 se, $f$ width ke upar $0.9$ se $0.1$ tak girta hai: $\Delta \epsilon \approx 4.4k_{B}T,$ $\mu$ ke centered. $f=0.9$ solve karo: $e^x+1=1/0.9$ deta hai $x=\ln (1/9)\approx -2.20$; $f=0.1$ deta hai $x=\ln 9\approx +2.20$; gap hai $2\times 2.20=4.4$, jahan $x=(\epsilon -\mu )/k_{B}T$.

KYUN yeh matter karta hai. Sirf is thin surface shell ke electrons ek thermal energy nudge absorb karke move ho sakte hain — deep sea exclusion se frozen solid hai (paas mein jump karne ke liye koi empty slot nahi). Isliye metals ki heat capacity classical guess se itni tiny hoti hai, aur yeh Sommerfeld Expansion ka launch pad hai, jo yeh bhi reveal karta hai ki $\mu (T)$, $T$ badhne par $E_F$ se thoda neeche dip karta hai — isliye $\mu =E_F$ sirf $T=0$ par hai.

PICTURE. Fermi surface ka zoom: sharp cliff ke saath ek shaded "fuzzy shell" of width $\sim k_{B}T$ jahan $f$ ramp karta hai, arrows dikhate hain electrons just-below se just-above jump karte hue.

---

Ek-picture summary

Ek canvas par poori derivation: ek slot (0 ya 1) → do states weigh karo → average karke S-curve $f$ banao → ise step mein freeze karo → step ko $\mathbf{k}$-space mein filled sphere mein stack karo → uski surface height $E_F$ ke taur par padho → warm karo aur sirf surface shell ripple karta hai.

Recall Feynman retelling — ise plain words mein wapas bolo

Socho energy levels ki ek giant staircase hai, aur electrons ki ek bheed jo ek doosre ke saath ek step par khade hone se bilkul mana karti hai. Sab kuch absolute zero tak cool karo aur bheed bottom par collapse nahi karti — woh staircase ko ground se upar tak pack karti hai jab tak log khatam na ho jayein. Aakhri occupied step hi Fermi energy hai. Usse neeche: kaandhe se kaandha, har step full. Usse upar: empty. Ab bheed ko thoda warm karo: neeche wale pin ho jaate hain — paas mein koi free step nahi move karne ke liye — lekin bilkul top par khade kuch log $k_{B}T$ ke ek band ke upar-neeche shuffle kar sakte hain. Woh thin, jittery top layer hi metal mein saara action hai. Aur yeh poora staircase picture kahan se aaya? Bas ek Boltzmann factor ke saath ek step ke liye "empty" aur "full" weigh karo, probability ke liye divide karo, aur un probabilities ko har step par stack karo.

Recall

Ek fermion slot ke do allowed occupations ::: $n=0$ (empty) aur $n=1$ (full) — 1 par cap hi Pauli exclusion hai. Weight mein exponential kyun aata hai ::: independent energies add hoti hain, isliye unke weights multiply hone chahiye, aur $e^{-a}{e}^{-b}=e^{-(a+b)}$ hi aisa karne wala akela function hai. $f$ ki value exactly $\epsilon =\mu$ par, aur yeh $T$-independent kyun hai ::: $f=\frac{1}{2}$; exponent $0$ hai isliye $e^0=1$ regardless of $T$. Fermi sphere kya hai ::: $T=0$ par radius $k_F=(3{\pi }^{2}n{)}^{1/3}$ ki filled $\mathbf{k}$-states ki ball. Sea surface energy mein kitni oonchi hai ::: $E_F=\frac{{\hslash }^2}{2m}(3{\pi }^{2}n{)}^{2/3}$, sirf density $n$ se set hoti hai. Thermal "fuzzy shell" ki width ::: $\mu$ ke around lagbhag $4.4k_{B}T$ ($f=0.9$ aur $f=0.1$ ke beech).