Entropy and disorder — Boltzmann S = k·ln(W): Visual walkthrough

1.7.24 · D2 · Physics › Thermodynamics › Entropy and disorder — Boltzmann S = k·ln(W)

Agar tumne microstate aur macrostate words abhi tak nahi dekhe, toh ek baar Microstates and Macrostates mein padh lo — lekin hum unhe neeche dobara picture karenge.

Step 1 — Hum kya count kar rahe hain? (Microstates vs macrostates)

KYA. Socho ek row mein chaar coins hain. Jo cheez tum bahar se dekhte ho woh hai "kitne heads hain." Jo cheez tum nahi dekhte woh hai kaunse khaas coins heads hain. Entropy $S$ jo humne abhi naam diya woh aakhir mein is hidden count se bani hogi.

KYUN. Jo kuch bhi aage aata hai woh ek single number ke baare mein hai: kitne hidden arrangements ek hi visible result produce karte hain. Is picture ko rock-solid kar lo aur baaki sab bookkeeping hai.

PICTURE. Figure mein, top box ek macrostate hai: "4 mein se 2 heads." Neeche saare alag hidden arrangements hain — microstates — jo sab bahar se "2 heads" jaisi dikhti hain.

Figure — Entropy and disorder — Boltzmann S = k·ln(W)

Yahan $W$ sirf ek plain counting number hai. Abhi koi logarithm nahi — humne abhi zaroorat nahi padi.

Step 2 — Fundamental rule: har microstate equally likely hoti hai

KYA. Nature hidden arrangements ko weight nahi karti. Coin-pattern HTHT bilkul utni hi likely hai jitni HHTT. Har individual microstate ki same tiny probability hoti hai.

KYUN. Yeh poori theory ka ek physical assumption hai (equal a-priori probability postulate). Iske bina, "counting" kuch bhi predict nahi karti.

PICTURE. Neeche har choti tile ka same size hai — same probability. Macrostate ki probability phir sirf kitni tiles uske paas hain yeh hoti hai.

Abhi tak humaara "kitna likely / kitne tarike" ka measure $W$ khud hai. Ab dekho kyun $W$ akela entropy kehne ke liye galat quantity hai.

Step 3 — Do systems: counting multiply karta hai

KYA. System $A$ (multiplicity $W_{A}$ ) ko independent system $B$ (multiplicity $W_{B}$ ) ke paas rakho. Combined system kitne tarike se arrange ho sakta hai?

KYUN. Real thermodynamics chhote systems se bane bade systems ke baare mein hai. Hume jaanna chahiye ki count kaisa behave karta hai jab hum systems ko jodte hain.

PICTURE. Ek grid: har row $A$ ka ek arrangement hai, har column $B$ ka ek arrangement hai. Har cell ek valid combined microstate hai. Rows $\times$ columns $=$ total cells.

Aage ke liye key observation: combined count product se badhta hai, sum se nahi.

Step 4 — Lekin jo quantity hum chahte hain woh add karni chahiye

KYA. Hum chahte hain ek "entropy ki matra" $S$ . Energy, mass, volume extensive hain: do identical boxes side by side hoon toh matra double hoti hai. Hum demand karte hain ki entropy bhi aise hi behave kare.

KYUN. Entropy ko bulk "matra" honi chahiye jaise heat content — yeh scale honi chahiye kitna stuff hai usse. Do identical gas boxes ⇒ double entropy.

PICTURE. Do identical boxes. Unke volumes add hote hain ( $V + V = 2 V$ ), unki entropies add honi chahiye ( $S + S = 2 S$ ). Lekin Step 3 se unke counts multiply hote hain ( $W \times W = W^{2}$ ). Ek clash hai jise hume resolve karna hai.

Step 5 — Kaun sa function $\times$ ko $+$ mein badalta hai? Logarithm

KYA. Likho $S = f (W)$ kisi unknown function $f$ ke liye. Steps 3 aur 4 milke demand karte hain: $f (W_{A} W_{B}) = f (W_{A}) + f (W_{B})$

KYUN. Yeh single equation $f$ ko pin down karti hai — lekin sirf tabhi jab hum do mild physical demands add karein jo entropy ko obviously maanni chahiye:

Smoothness (continuity). Agar count $W$ thoda sa bhi change ho, toh entropy $S$ bhi thoda sa change honi chahiye — koi jumps nahi. Physically, arrangements ki kitni matra available hai ismein thodi si change entropy ko discontinuously leap nahi kara sakti. Yahi humein the logarithm pick karne deta hai na ki koi pathological product-to-sum function.
Increasing ( $d S / d W > 0$ ). Zyada available microstates ka matlab hona chahiye zyada entropy, kabhi kam nahi. Toh $f$ ko $W$ badhne par badhna chahiye. Yeh woh demand hai jo force karti hai ki agle step mein scaling constant positive ho.

Un dono demands ke saath, $f (x y) = f (x) + f (y)$ ka continuous increasing solution unique hai — woh logarithm hai.

PICTURE. Graph dikhata hai log kya karta hai: horizontal axis multiplicatively stretch hoti hai ( $1, 2, 4, 8, \dots$ equally spaced), aur log unhe ek evenly rising ladder mein badal deta hai ( $0, ln 2, 2 ln 2, 3 ln 2$ ). Notice karo curve smooth hai (koi jumps nahi) aur hamesha climbing hai (increasing) — dono demands ko match karta hai. Input ko multiply karna output mein ek fixed step add karta hai — exactly woh rule jo humein chahiye.

Recall Kyun

the logarithm aur koi bhi function kyun nahi? Requirements ::: $f (x y) = f (x) + f (y)$ sabhi positive $x, y$ ke liye, $f$ ke saath continuous (smooth, no jumps) aur increasing ( $d S / d W > 0$ ). Unique solution ::: $f (x) = c ln x$ jahan $c > 0$ — koi aur continuous, increasing function yeh nahi kar sakta.

Step 6 — Units fix karna: constant $k$

KYA. $ln W$ ek pure number hai (koi units nahi). Real entropy joules per kelvin (J/K) mein measure hoti hai. Hum inhe bridge karne ke liye constant $k$ se multiply karte hain. Kyunki $f$ ko increasing hona tha (Step 5), yeh constant positive hona chahiye: $k > 0$ .

KYUN. Abstract "count" logic joules ya kelvins ke baare mein nahi jaanta — woh Clausius ki thermometer-aur-heat wali duniya se aate hain. Constant $k$ "nats of counting" aur "J/K of physics" ke beech exchange rate hai, aur iska positivity guarantee karta hai ki zyada microstates ⇒ zyada entropy.

PICTURE. Do dials — ek pure-number dial ( $ln W$ ) aur ek physical J/K dial — ek gear se connected hain jis par $k = 1.38 \times 1 0^{- 23}$ J/K likha hai. Ek ko ghoomao toh dusra fixed proportion mein ghoomta hai.

Step 7 — Kya yeh real thermodynamics reproduce karta hai? (Free expansion check)

KYA. $N$ gas molecules, har ek volume $V$ mein, suddenly $2 V$ access karte hain (ek partition hata di jaati hai). Har molecule independently double jagahen paata hai rehne ke liye.

KYUN. Ek formula tabhi trustworthy hota hai jab woh kisi known result se match kare. Free expansion ka ek classical answer hai, isliye yeh perfect test hai. Dekho Free Expansion of an Ideal Gas.

PICTURE. Left: molecules volume $V$ mein crowded hain. Right: same molecules $2 V$ mein spread hain. Har molecule ke "position-choices" double ho jaate hain, aur choices molecules ke across multiply hoti hain.

Step 8 — Edge cases (kabhi koi gap mat chhodna)

KYA. Extreme inputs check karo taaki koi reader kisi untested scenario mein na pade.

KYUN. Ek derivation jis par tum trust kar sako use apni khud ki boundaries survive karni chahiye: sabse chhota possible count, aur "doubling $W$ doubles $S$ " wala behavior nahi hota — iska.

PICTURE. $ln W$ ki ek number line: $W = 1$ par yeh exactly $0$ par baithta hai; yeh kabhi neeche nahi ja sakta; aur $W$ ko double karna ise sirf $k ln 2$ se nudge karta hai, factor of two se nahi.

Ek-picture summary

Upar sab kuch ek single flow mein: count → log → scale, reality ke against tested.

Recall Feynman one-liner — woh single sentence jo yaad rakho

Hidden arrangements count karo ( $W$ ), log lo taaki do systems ke scores add hon multiply ki jagah, $k$ se scale karo joules-per-kelvin reach karne ke liye — aur result free expansion jaisi real experiments predict karta hai. Yahi hai $S = k ln W$ .

Connections

Parent: Entropy and disorder — woh topic jisme yeh page deep dive karta hai.
Microstates and Macrostates — Step 1 ki bookkeeping.
Second Law of Thermodynamics — Step 2 ka "biggest pile wins."
Free Expansion of an Ideal Gas — Step 7 ka test case.
Clausius Entropy dS = dQ_rev / T — thermodynamic twin jo $S = k ln W$ reproduce karta hai.
Boltzmann Distribution — energy un microstates mein kaise spread hoti hai jo humne count kiye.
Information Entropy (Shannon) — wahi $ln (count)$ idea bits mein.
Third Law of Thermodynamics — Step 8 ka $W \to 1 \Rightarrow S \to 0$ .