Question bank — Bose-Einstein condensation — concept

False — ideal-gas BEC mein zero interaction hoti hai; crowding Bose-Einstein statistics se aati hai jo unlimited shared occupation allow karta hai, kisi potential se nahin.

False — $\mu$ $0^{-}$ par pin rahta hai; agar $\mu$ positive ho jaaye toh $\bar{n}({\epsilon }_0)=\frac{1}{e^{-\mu /k_{B}T}-1}$ negative ho jaayega, jo forbidden hai.

False — Density of states $g(\epsilon )\propto \sqrt{\epsilon }$, $\epsilon =0$ par zero hoti hai, toh integral ground state ko silently discard kar deta hai; below $T_c$ tumhe $N_0$ haath se wapas add karna padta hai.

False — 2D mein excited-state integral chhoti energy par diverge kar jaata hai, toh excited states kabhi saturate nahin hoti aur uniform 2D ideal gas ke liye koi BEC nahin hota.

False — $T_c$ par condensate fraction zero reach kar rahi hoti hai; $N_0/N=1-(T/T_c{)}^{3/2}=0$ wahaan hota hai aur sirf $T<T_c$ ke liye badhta hai.

True — $N_0/N=1-(T/T_c{)}^{3/2}\to 1$ jab $T\to 0$, toh saare atoms ek lowest state occupy karte hain.

Misleading — yeh atoms lowest momentum/quantum state mein hote hain, ek single delocalised wavefunction jo poore trap mein failti hai, koi spatial pile nahin.

True — yeh purely statistical phase transition hai; phase-space density $n{\lambda }_T^3$ ka $g\zeta (3/2)\approx 2.612$ (for $g=1$) cross karna hi poora trigger hai.

False — fermions Fermi-Dirac statistics follow karte hain jisme $\bar{n}=\frac{1}{e^{(\epsilon -\mu )/k_{B}T}+1}\le 1$ per state (Pauli exclusion), toh ek state mein macroscopic pile-up possible nahin.

True — lekin "overlap" precisely define hai: condensation exactly tab shuru hoti hai jab $n{\lambda }_T^3=g\zeta (3/2)\approx 2.612$ (degeneracy $g=1$ ke liye), yaani interparticle spacing ${\lambda }_T$ jaisi hoti hai; "$\sim 1$" slogan ka matlab bas yahi order-unity number hai.

Error: $\mu$ bounded hai by $\mu \le 0$; jab excited states $\mu \to 0^{-}$ par saturate ho jaati hain, $\mu$ aur upar nahin ja sakta aur surplus ka koi option nahin bachta siwaaye ground state ke.

Error: $\epsilon =0$ ke paas integrand $\sqrt{\epsilon }/\epsilon =1/\sqrt{\epsilon }$ ki tarah behave karta hai, jo 3D mein integrable hai — excited integral finite rehta hai, toh yeh saturate ho jaata hai. Sirf discrete $N_0$ term diverge karta hai.

Error: Boltzmann form excited occupation par koi upper bound nahin lagaata ($\mu$ unrestricted), toh koi overflow nahin hota aur koi condensation nahin hoti.

Error: $T_c\propto n^{2/3}$, toh kam density se kam $T_c$ milta hai — isliye real alkali BECs (bahut chhota $n$) sub-microkelvin cooling maangti hain.

Error: constant density of states se excited integral ${\int }_0^{\infty }\frac{dx}{e^x-1}$ chhote $x$ par logarithmically diverge karta hai, toh excited states koi bhi number absorb kar leti hain — koi saturation nahin, koi condensate nahin.

Error: Riemann zeta function $\zeta (3/2)$ energy integral ${\int }_0^{\infty }\frac{\sqrt{x}dx}{e^x-1}=\Gamma (3/2)\zeta (3/2)$ se aata hai, substitution $x=\epsilon /k_{B}T$ ke baad — spatial volume se nahin.

Error: dono linked hain lekin alag hain — Superfluidity mein interactions aur bina dissipation ke flow hota hai; ideal-gas BEC non-interacting hota hai aur superfluid hona zaroori nahin.

Kyunki $g(\epsilon )\to 0$ at $\epsilon =0$ ground state ko continuum integral mein zero weight deta hai, phir bhi below $T_c$ usme macroscopic $N_0$ hota hai — toh use discrete term ki tarah add karna padta hai.

Kyunki saturation condition $n=g\zeta (3/2)/{\lambda }_T^3$ density aur temperature ko ${\lambda }_T$ ke through baandhti hai; sirf unka combination measure karta hai jab matter waves overlap karti hain.

Kyunki jab $\mu \to 0^{-}$ hota hai toh denominator $\to 0^{+}$ hota hai, toh $N_0\to \infty$ — ek single state unbounded capacity ka "escape valve" deta hai.

Saari temperature dependence ko $(k_{B}T{)}^{3/2}$ ki tarah front mein nikalne ke liye aur baaki integral ko ek pure dimensionless number ($\Gamma (3/2)\zeta (3/2)$) mein reduce karne ke liye; upar line-by-line box dekhein.

Kyunki $N_0/N=1-(T/T_c{)}^{3/2}$; $T_c$ ke paas deviation $1-(T/T_c{)}^{3/2}\approx \frac{3}{2}(1-T/T_c)$ underlying integral mein fractional power ki tarah zero se badhta hai, jo ek non-analytic onset hai jo genuine phase transition indicate karta hai, smooth crossover nahin.

Kyunki yeh sirf identical-boson wavefunctions ki symmetrization se emerge hota hai (jo shared states allow karta hai), koi potential energy ya classical mechanism involve nahin hota.

Zyada $T$ se ${\lambda }_T$ chhota ho jaata hai (kyunki ${\lambda }_T\propto T^{-1/2}$), $n{\lambda }_T^3$ ko $g\zeta (3/2)$ se neeche le jaata hai; excited states ko sabko accommodate karne ke liye phir se kaafi jagah mil jaati hai.

Upar se approach karte hue, $\mu$ $0$ ki taraf chadh raha hota hai; $T_c$ par aur neeche, yeh $0^{-}$ par lock ho jaata hai aur wahin rehta hai — yahi condensed phase ki pahchaan hai.

Zero — $T_c$ ke upar koi condensate nahin hota; excited states mein abhi jagah baaki hai, toh $N_0$ microscopic hota hai (order 1, macroscopic nahin).

Iska matlab hai ki $N_0/N$ finite (nonzero) rehti hai jab $N\to \infty$ — ek single quantum state ek unbounded total ka fixed fraction rakhta hai.

Excited-state integral 2D se bhi zyada strongly diverge karta hai, toh ek uniform 1D ideal gas mein koi saturation nahin aur koi condensation nahin hoti.

Haan — trapping effective density of states ko alter kar deta hai, 2D harmonic trap mein BEC allow karta hai jahaan uniform 2D gas fail ho jaata hai; geometry, sirf spatial dimension nahin, $\epsilon$ ki power set karta hai.

Yeh Boltzmann form $e^{-(\epsilon -\mu )/k_{B}T}$ mein reduce ho jaata hai, kyunki $-1$ negligible ho jaata hai — high-energy tails hamesha classical lagte hain.

Distinguishability statistical enhancement hata deti hai; har species apni boson statistics se independently condense karti hai, lekin dono ek shared BEC state mein merge nahin hote.