KAISE. Maano vx ko dvx mein paane ki probability g(vx)dvx hai, similarly y,z ke liye.
Independence ki wajah se joint distribution factorize hoti hai:
F(vx,vy,vz)=g(vx)g(vy)g(vz).
Isotropy ki wajah se, F sirf magnitudev=vx2+vy2+vz2 par depend kar sakta hai, yaani
vx2+vy2+vz2 par. To:
g(vx)g(vy)g(vz)=Φ(vx2+vy2+vz2).
To lng(vx)=a−bvx2, jisse milta hai:
g(vx)=Ae−bvx2.
Hamein b>0 chahiye (warna normalize nahi ho sakta — large speeds infinitely likely ho jaenge).
Normalization∫−∞∞g(vx)dvx=1, Gaussian integral
∫−∞∞e−bx2dx=π/b use karke:
Abπ=1⇒A=πb.
b ko physics se fix karo.b ko temperature se relate karna hoga. Equipartition use karo: har
translational degree of freedom average energy 21kBT carry karta hai:
⟨21mvx2⟩=21kBT⇒⟨vx2⟩=mkBT.
Gaussian g(vx)=Ae−bvx2 ke liye, variance hai ⟨vx2⟩=2b1.
T badhne par f(v) ka kya hoga? Pehle forecast karo, phir check karo.
Forecast: peak right shift karega aur curve flatten/broaden hogi.
Verify:vp∝T → peak right move karta hai. Area 1 rehta hai, isliye higher peak position
matlab lower, wider curve. Hotter gas = faster aur zyada spread-out speeds. ✓
Ek bumper cars ki bheed imagine karo jo randomly crash kar rahi hain. Kuch time baad, kuch cars zoom kar rahi hain, kuch crawl kar rahi hain, aur zyaatar ek medium speed par hain. Agar inhe speed ke hisaab se sort karo, ek hill-shaped graph milta hai: almost koi frozen nahi hain (kuch to speed chahiye), almost koi super-fast nahi hain (usme huge energy lagti hai), aur beech ka bump "usual" speed hai. Floor ko heat karo taaki cars zyada push karein — poori hill faster speeds ki taraf slide karti hai aur spread out hoti hai. Woh hill Maxwell-Boltzmann distribution hai.
f(v) ko shape karne wale do competing factors kaunse hain?
Geometric factor 4πv2 (high v par velocity-space mein zyada room) vs Boltzmann factor e−mv2/2kBT (energy cost).
f(0)=0 kyun hai?
v2 shell-volume factor v=0 par vanish karta hai — radius 0 ki shell mein koi velocity vector zero magnitude nahi rakhta.
Single-component distribution g(vx) ki form?
(2πkBTm)1/2e−mvx2/2kBT, ek Gaussian 0 par centered.
Full MB speed distribution?
f(v)=4π(2πkBTm)3/2v2e−mv2/2kBT.
Most probable speed vp?
2kBT/m, dvd(v2e−αv2)=0 se.
Mean speed ⟨v⟩?
8kBT/πm.
RMS speed vrms?
3kBT/m.
Teen speeds ka order?
vp<⟨v⟩<vrms (ratios 2:8/π:3).
g(vx) Gaussian (exponential of −vx2) kyun hona chahiye?
Independence joint density ko product mein factorize karta hai; isotropy use sirf vx2+vy2+vz2 par dependent banata hai; sirf exponential products ko sums mein convert karta hai.