Boltzmann's entropy S = k_B ln(Ω): Visual walkthrough

2.4.9 · D2 · Physics › Thermodynamics & Statistical Mechanics (Advanced) › Boltzmann's entropy S = k_B ln(Ω)

Step 1 — "Ek tarika" kya hota hai? (the microstate)

KYA. Socho char chhote boxes ek line mein hain. Har box mein ek coin hai, aur har coin ya to Heads (H) dikhata hai ya Tails (T). Charon boxes ka ek poora filling — jaise H T T H — ek exact snapshot hai. Hum aise har complete snapshot ko microstate kehte hain.

KYUN. Tarike count karne se pehle, humein agree karna hoga ki "ek tarika" kya hai. Ek microstate counting ki sabse chhoti, fully-detailed unit hai: har coin decided, kuch bhi vague nahi.

PICTURE. Figure s01 dekho. Amber row ek microstate hai — har coin decided. Neeche faint cyan rows doosre microstates hain: ek bhi coin badlo aur yeh ek alag snapshot ban jata hai.

Figure — Boltzmann's entropy S = k_B ln(Ω)

Step 2 — Aise snapshots ko group karna jo "same lagte hain" (the macrostate)

KYA. Bahar se tum individual coins nahi dekh sakte — tum sirf ek bulk number padh sakte ho, jaise "total kitne Heads hain." To hum har microstate ko us bulk number ke label wale bins mein sort karte hain: 0 heads, 1 head, 2 heads, ... Ek bin ek macrostate hai.

KYUN. Real measurements coarse hoti hain. Tum energy $E$ , volume $V$ , particle number $N$ measure karte ho — exact microstate kabhi nahi. To physically meaningful object bin hai, snapshot nahi. "Yeh bin kitni likely hai?" ka sawaal ban jaata hai "kitne snapshots isme gire?"

PICTURE. Figure s02 mein har snapshot (cyan dot) apne head-count ke hisaab se ek bin mein girti hai. "2 heads" wali bin clearly sabse moti hai — zyada dots wahan gire. Yahi poori kahaani hai.

Term-by-term: $Ω$ = count; ek microstate = ek dot; ek macrostate = ek bin. Badi bin ⇒ bada $Ω$ ⇒ (hum dekhenge) zyada entropy.

Step 3 — Do systems ko jodhna: counts multiply hote hain

KYA. System $A$ (jiske $Ω_{A}$ snapshots hain) ko ek independent system $B$ (jiske $Ω_{B}$ snapshots hain) ke saath rakho. $A$ ke har snapshot ke liye tum $B$ ke koi bhi snapshot pair kar sakte ho. To joint snapshots ki total count product hai.

$Ω_{A B} = Ω_{A} \times Ω_{B}$

Term-by-term padhna:

$Ω_{A}$ = sirf left system ke liye tarike,
$Ω_{B}$ = sirf right system ke liye tarike,
$Ω_{A B}$ = dono ko ek combined system maante hue tarike,
$\times$ isliye aata hai kyunki choices independent hain — $A$ ke baare mein kuch bhi $B$ ko restrict nahi karta.

KYUN YEH TOOL (multiplication, addition nahi). Jab bhi do choices independently ki jaati hain, totals multiply hote hain — exactly yahi wajah hai ki do dice $6 \times 6 = 36$ outcomes dete hain, $6 + 6$ nahi. Hum yahan physics assume nahi kar rahe, sirf independent choices ki arithmetic.

PICTURE. Figure s03 ek grid hai: $A$ ke options rows mein hain, $B$ ke columns mein. Har cell ek joint microstate hai. Cells count karo — yeh rows $\times$ columns hai. Picture hi multiplication hai.

Step 4 — Entropy se hamari demand: yeh add karni chahiye

KYA. Hum chahte hain ek quantity $S$ — "entropy" — jo hum $Ω$ se somehow define karenge, $S = f (Ω)$ . Thermodynamics already entropy ko extensive treat karta hai: do identical blocks ko glue karo aur entropy double ho jaati hai, bilkul mass ya energy ki tarah. To hum insist karte hain:

$S_{A B} = S_{A} + S_{B}$

Term-by-term: $S_{A}$ = left block ki entropy, $S_{B}$ = right ki, $S_{A B}$ = combined block ki, aur $+$ woh property hai jo hum haath se impose kar rahe hain real thermodynamics se match karne ke liye.

KYUN. Yeh pivotal design choice hai. Step 3 mein nature ne humein products diye; thermodynamics yahan sums chahti hai. Hamara kaam woh ek function $f$ dhundhna hai jo dono ke beech bridge kare.

PICTURE. Figure s04 tension ko side by side dikhata hai: left mein, counts $\times$ se combine hote hain; right mein, humari required entropy $+$ se combine hoti hai. Double arrow us gap ko mark karta hai jise ek function ko close karna hai.

Step 5 — Woh ek function jo $\times$ ko $+$ mein badalta hai: the logarithm

KYA. Steps 3 aur 4 combine karo. Kyunki $S = f (Ω)$ aur $Ω_{A B} = Ω_{A} Ω_{B}$ hai, additivity demand karti hai ki

$f (Ω_{A} Ω_{B}) = f (Ω_{A}) + f (Ω_{B}) .$

Hum ise solve karte hain. Dono sides ko $Ω_{A}$ ke respect mein differentiate karo (ise $x$ kaho, aur $Ω_{B} = y$ ): $y f^{'} (x y) = f^{'} (x) .$ $x = 1$ set karo: $y f^{'} (y) = f^{'} (1) \equiv k$ , ek constant. Hence $f^{'} (y) = \frac{k}{y} ⟹ f (y) = k ln y + C .$ Aur $C = 0$ kyunki ek tarika matlab zero entropy: $f (1) = 0$ (agar sirf ek single snapshot hai, to tumhare paas koi missing information hi nahi).

$f (Ω) = k ln Ω$

KYUN LOGARITHM (kyun yeh tool aur koi nahi). Logarithm define hota hai identity $ln (x y) = ln x + ln y$ se. Yeh woh unique continuous function hai jo product nigal kar sum ugalta hai. Hamara poora tension — nature multiply karta hai, thermodynamics add karna chahti hai — ka exactly ek continuous ilaaj hai, aur uska naam $ln$ hai.

PICTURE. Figure s05 dikhata hai yeh hote hue: multiplying number-line par do counts, $ln$ curve se guzar kar, adding number-line par map hote hain. Dekho $Ω_{A} Ω_{B}$ bottom axis par land karta hai $ln Ω_{A} + ln Ω_{B}$ pe side axis par.

Step 6 — Constant fix karna: $k = k_{B}$

KYA. Functional equation ne $k$ ko free chhod diya. Hum ise ek known case — ideal gas — ko measured thermodynamic entropy se match karke fix karte hain. Yeh match $k$ ko Boltzmann's constant banne par majboor karta hai:

$S = k_{B} ln Ω, k_{B} = 1.380649 \times 1 0^{- 23} J/K .$

KYUN. $ln Ω$ ek pure number hai, lekin real entropy joules-per-kelvin mein measure hoti hai. $k_{B}$ woh exchange rate hai jo "number of ways" ko thermodynamic units mein convert karta hai — aur yahi exactly woh constant hai jo $1/ T = \partial S / \partial E$ ko kelvin mein sahi banata hai.

PICTURE. Figure s06: bare curve $ln Ω$ (cyan) ko tiny amber factor $k_{B}$ se rescale karke physical entropy $S$ (amber) milti hai. Same shape, physical units.

Step 7 — Degenerate case: $Ω = 1$ (Third Law)

KYA. Kya hoga agar kisi system mein sirf ek possible microstate ho? Absolute zero par ek perfect crystal apni single lowest-energy arrangement mein frozen baith jaata hai, to $Ω = 1$ . Tab

$S = k_{B} ln 1 = k_{B} \times 0 = 0.$

KYUN. $ln 1 = 0$ isliye kyunki $1$ paane ke liye tum kuch bhi multiply nahi karte — koi choices baaki nahi, koi hidden information nahi. Boltzmann ka counting formula predict karta hai Third Law of Thermodynamics: entropy tab gayab ho jaati hai jab sirf ek tarika bacha ho. Yeh add-on nahi hai; yeh seedha log se nikal aata hai.

PICTURE. Figure s07: jaise-jaise available snapshots ki sankhya ek single dot par collapse hoti hai, entropy bar smoothly floor $S = 0$ tak shrink ho jaata hai.

Step 8 — Doosra edge: energy badhne par entropy gir sakti hai

KYA. Hum usually sochte hain "zyada energy ⇒ zyada entropy." Lekin $S$ , $ln Ω$ track karta hai, $E$ nahi. Ek capped-energy system mein — spins jo magnetic field mein sirf upar ya neeche point kar sakte hain — energy pump karne se eventually arrangements ki sankhya kam ho jaati hai (saare spins align hone par majboor), to $Ω$ girta hai aur $S$ girta hai. Wahan $\partial S / \partial E < 0$ hai, jo negative temperature ki pehchaan hai.

KYUN. Yeh edge case us galat shortcut ke khilaf raksha karta hai ki "entropy = energy." Boltzmann ka formula sirf tarike count karta hai; agar tarike high energy par khatam ho jaate hain, to entropy girti hai. Ise cover karna hamari picture ko sabhi regimes mein honest rakhta hai, na ki sirf roz ke heating wale mein.

PICTURE. Figure s08 mein $Ω$ ko energy ke against plot kiya gaya hai two-state spin system ke liye: yeh utha hai, middle energy par peak kiya, phir wapas gira — ek hill, ramp nahi. Entropy bhi usi hill ko follow karti hai.

Ek-picture summary

Figure s09 poore walkthrough ko ek single flow mein compress karta hai: snapshots → unhe bin karo → systems join karo (counts multiply hote hain) → demand karo ki entropy add ho → ek hi bridge hai $ln$ → $k_{B}$ se scale karo → edge cases padho ( $Ω = 1$ se $S = 0$ milta hai; capped systems entropy kho sakte hain).

Recall Feynman: poora walk simple shabdon mein

Coins se shuru karo. Saare coins ka ek exact filling ek snapshot hai. Room ke doosre side se tum coins nahi dekh sakte — tum sirf "kitne heads" count kar sakte ho, to tum snapshots ko bins mein sort karte ho. Kuch bins bahut zyada snapshots rakhti hain; ek bin ka woh headcount $Ω$ hai.

Ab do coin-tables ko saath push karo. Left ka har snapshot right ke kisi bhi snapshot se pair ho sakta hai, to joint count multiply hota hai — bilkul us grid ki tarah jiske cells rows times columns hote hain. Lekin hum chahte hain ki hamara "entropy" number weight ki tarah behave kare: do tables ko double entropy deni chahiye, add karke, multiply karke nahi.

Exactly ek math gadget hai jo multiply kha kar add wapas deta hai: logarithm, kyunki $ln (ab) = ln a + ln b$ . To entropy zaroori hai ki count ka $ln$ ho. Chhoti si equation solve karna ise confirm karta hai aur leftover constant ko vanish hone par majboor karta hai jab sirf ek tarika ho ( $ln 1 = 0$ ). Aakhir mein hum $k_{B}$ sprinkle karte hain, ek tiny number jo "count of ways" ko real joules-per-kelvin mein badhata hai.

Do sanity checks loop close karte hain: apni single arrangement mein frozen ek perfect crystal ka $Ω = 1$ hai, to $S = 0$ — Third Law, free mein. Aur ek spin system mein jahan high energy par tarike khatam ho jaate hain, energy add karne par entropy actually gir sakti hai — proof ki entropy tarike count karta hai, energy kabhi nahi.

Recall Quick self-test

Kyun counts multiply hote hain lekin entropies add hoti hain? ::: Independent systems join karne par ek ka har microstate doosre ke har microstate se pair hota hai (cells ka grid = rows × columns), to counts multiply hote hain; hum demand karte hain ki entropy extensive ho isliye yeh add hoti hai — $ln$ dono ko reconcile karta hai. Kya cheez $f (Ω) = k ln Ω$ aur kisi aur function ko force nahi karti? ::: $ln$ $f (x y) = f (x) + f (y)$ ka unique continuous solution hai. $Ω = 1$ kya deta hai aur woh kaun sa law hai? ::: $S = k_{B} ln 1 = 0$ , Third Law. Energy add karne par entropy kab gir sakti hai? ::: Capped-energy systems mein (jaise spins) jahan high $E$ par $Ω$ decrease hota hai — negative-temperature regime.

Connections

Parent: Boltzmann's entropy
Microcanonical Ensemble
Gibbs Entropy $S=-k_B\sum p_i\ln p_i$
Temperature as $\partial S/\partial E$
Third Law of Thermodynamics
Negative Temperature & Spin Systems
Gibbs Paradox & Indistinguishability
Shannon Information Entropy
Entropy and the Second Law of Thermodynamics