Ion implantation and diffusion — Hinglish notes

4.3.11 · Hardware › Semiconductor Fabrication

Big picture: Ek transistor banane ke liye, hume dopant atoms (jaise Boron ya Phosphorus) ko silicon wafer ke precise regions mein, precise concentrations aur depths par dalna hota hai. Do tools yeh kaam karte hain: diffusion (thermal, atoms wander karke andar jaate hain) aur ion implantation (ions ko darts ki tarah shoot karte hain). Yeh note dono ko first principles se derive karta hai.

1. WHY do we dope at all?

Do knobs matter karte hain:

Dose $Q$ = dopant atoms ki sankhya per unit area (atoms/cm²).
Profile $N (x)$ = concentration vs depth (atoms/cm³).

2. Diffusion — atoms ek concentration gradient ke neeche wander karte hain

WHY the flux law (Fick's 1st law)

Socho depth $x$ par ek plane hai. Zyada atoms high-concentration side par hain, isliye unme se zyada plane ke us taraf hop karte hain rather than wapas. Net flux itni steep concentration ke saath proportional hoti hai:

$J = - D \frac{\partial N}{\partial x}$

Minus sign kyun? Flux gradient ke neeche jaati hai (high se low). Agar $N$ , $x$ ke saath decrease hota hai, toh $\partial N / \partial x < 0$ , isliye $J > 0$ (wafer mein flow). ✔
$D$ = diffusion coefficient (cm²/s), strongly temperature dependent: $D = D_{0} e^{- E_{a} / k T}$ Arrhenius kyun? Ek atom ko hop karne ke liye energy barrier $E_{a}$ clear karni hoti hai; jinke paas enough thermal energy hoti hai unka fraction Boltzmann factor $e^{- E_{a} / k T}$ hota hai.

WHY the diffusion equation (Fick's 2nd law)

$x$ aur $x + d x$ ke beech ek thin slab lo. Jo bhi andar aaye minus jo bahar jaaye, woh accumulate hona chahiye:

$\frac{\partial N}{\partial t} = - \frac{\partial J}{\partial x}$

Kyun? Yeh sirf atoms ka conservation hai (continuity). Fick's 1st law substitute karo (constant $D$ ke saath):

$\frac{\partial N}{\partial t} = D \frac{\partial ^{2} N}{\partial x ^{2}}$

HOW we solve it — do classic profiles

Case A: Constant surface concentration (predeposition). Surface fixed $N_{s}$ par rakha jaata hai (unlimited dopant supply). Solution:

$N (x, t) = N_{s} erfc (\frac{x}{2 D t})$

Case B: Fixed dose, drive-in (limited source). Ek dose $Q$ pehle se surface par hai, phir hum source hata ke heat karte hain. Solution ek Gaussian hai:

$N (x, t) = \frac{Q}{π D t} e^{- x^{2} / (4 D t)}$

Gaussian kyun? Yeh diffusion equation ka fundamental solution hai ek point/sheet source ke liye, aur yeh automatically total dose conserve karta hai $\int_{0}^{\infty} N d x = Q$ .

3. Ion Implantation — dopants ko shoot karo andar

HOW ions stop karte hain

Ions do tarike se energy lose karte hain:

Nuclear stopping — lattice nuclei se collisions (low energy par dominant, damage karta hai).
Electronic stopping — electrons se drag (high energy par dominant).

Stopping depths ka spread (bahut saari random collisions ki statistics) profile ko approximately Gaussian banata hai:

$N (x) = N_{p} exp [- \frac{( x - R _{p} ) ^{2}}{2 Δ R _{p}^{2}}]$

Peak surface par nahi, andar kyun hota hai: ions ko apni saari energy lose karne se pehle thoda travel karna padta hai, isliye max concentration $N_{p}$ , $x = R_{p}$ par, wafer ke andar hoti hai.

Peak ko dose se relate karna — total dose = curve ke neeche ka area:

\;\Rightarrow\; \boxed{N_p=\frac{Q}{\sqrt{2\pi}\,\Delta R_p}}$$ *Yeh step kyun?* Ek Gaussian $N_p e^{-(x-R_p)^2/2\sigma^2}$ integrate hokar $N_p\sigma\sqrt{2\pi}$ deta hai; isse known dose ke barabar set karne par $N_p$ pin down hota hai. ### Implantation ke baad: **annealing** Implantation crystal ko taud deta hai (damage) aur dopants ko interstitial (non-electrical) sites par chhod deta hai. Hum garam karte hain (~900–1000 °C) taaki: - **Lattice repair** ho, aur - **Dopants activate** hon (unhe lattice sites par move karo). Yeh anneal profile ko bhi *diffuse* karta hai: Gaussian ka straggle badhkar $\sqrt{\Delta R_p^2 + 2Dt}$ ho jaata hai. --- ## 4. Diffusion vs Implantation — the 80/20 table | | Diffusion | Ion implantation | |---|---|---| | Depth control | poor (√t, thermal) | ==excellent (energy se set hota hai)== | | Dose control | moderate | ==precise (charge count karo)== | | Temperature | high (~1000 °C) | ==low (room temp)== | | Lateral spread | large (isotropic) | small (anisotropic) | | Crystal damage | none | yes → annealing chahiye | | Profile shape | erfc / Gaussian at surface | ==Gaussian peaked at $R_p$== | --- ## 5. Worked Examples > [!example] E1 — Diffusion depth scaling > Ek drive-in at $T$, 30 min mein junction depth $x_j$ deta hai. **Double** depth ke liye kitna time chahiye (same $T$)? > *Kyun:* junction depth $\propto\sqrt{Dt}$ (from $L=2\sqrt{Dt}$). > Depth double karna ⇒ $\sqrt{t_2}=2\sqrt{t_1}\Rightarrow t_2=4t_1=120$ min. > **Lesson:** depth time ka ek *slow* function hai — small gains ke liye bahut bada cost. > [!example] E2 — Implant peak concentration > Boron implant karo, dose $Q=1\times10^{14}$ cm⁻², straggle $\Delta R_p=0.05\,\mu\text{m}=5\times10^{-6}$ cm. > *Kyun:* $N_p=Q/(\sqrt{2\pi}\,\Delta R_p)$ use karo. > $N_p=\dfrac{10^{14}}{2.5066\times5\times10^{-6}}\approx 7.98\times10^{18}\ \text{cm}^{-3}.$ > *Yeh step kyun?* Peak concentration ko puri dose ek Gaussian mein pack karni hoti hai jiska width $\Delta R_p$ hai. > [!example] E3 — Implant dose from beam > Beam current $I=10\ \mu$A, area $A=200\ \text{cm}^2$ par, $t=100$ s ke liye. $q=1.6\times10^{-19}$ C. > *Kyun:* $Q=It/(qA)$. > $Q=\dfrac{(10^{-5})(100)}{(1.6\times10^{-19})(200)}=\dfrac{10^{-3}}{3.2\times10^{-17}}\approx 3.1\times10^{13}\ \text{cm}^{-2}.$ > *Yeh step kyun?* Har incoming ion charge $q$ carry karta hai; total charge / area / $q$ = atoms/area. --- ## 6. Common Mistakes (Steel-manned) > [!mistake] "Depth diffusion time ke proportional hoti hai." > **Kyun sahi lagta hai:** zyada time → gehraai, linear lagta hai. > **Fix:** solution $x/\sqrt{Dt}$ par depend karta hai, isliye depth $\propto\sqrt{t}$. Time char guna karne se depth sirf double hoti hai. > [!mistake] "Implant peak surface par hota hai." > **Kyun sahi lagta hai:** ions *surface* par fire karte ho, toh surface = sabse zyada ions? > **Fix:** ions rukne se pehle andar penetrate karte hain. Gaussian $x=R_p$ par peak karta hai *wafer ke andar*; surface concentration kam hoti hai. > [!mistake] "Implantation akele ek working junction banata hai." > **Kyun sahi lagta hai:** dopants physically wahan hain. > **Fix:** zyada tar interstitial/damaged sites mein hain aur electrically **inactive** hain + lattice kharab hai. Activate karne aur repair karne ke liye *anneal karna zaroori* hai. > [!mistake] "$D$ temperature ke saath khas nahi badlta." > **Fix:** $D=D_0e^{-E_a/kT}$ — exponential. 100 °C ka change $D$ ko ek order of magnitude tak shift kar sakta hai. Temperature diffusion ke liye master knob hai. --- ## 7. Flashcards #flashcards/hardware Fick's 1st law aur uske minus sign ka matlab ::: $J=-D\,\partial N/\partial x$; flux gradient ke *neeche* flow karta hai (high → low), isliye minus. Fick's 2nd law (diffusion equation) ::: $\partial N/\partial t = D\,\partial^2 N/\partial x^2$, continuity + Fick's 1st law se. D ki temperature dependence ::: $D=D_0 e^{-E_a/kT}$ (Arrhenius); atoms ko hop karne ke liye barrier $E_a$ clear karni padti hai. Constant-source diffusion profile ::: $N(x,t)=N_s\,\mathrm{erfc}\big(x/2\sqrt{Dt}\big)$. Limited-source (drive-in) profile ::: $N(x,t)=\dfrac{Q}{\sqrt{\pi Dt}}e^{-x^2/4Dt}$ (Gaussian, dose conserve karta hai). Diffusion length ::: $L=2\sqrt{Dt}$; depth $\sqrt{t}$ ki tarah scale hoti hai. Implant profile shape ::: Gaussian, range $R_p$ par peak, width $\Delta R_p$. Peak implant concentration formula ::: $N_p=Q/(\sqrt{2\pi}\,\Delta R_p)$, Gaussian = dose integrate karne se. Dose from beam current ::: $Q=It/(qA)$. Implantation ke baad annealing kyun ::: lattice damage repair karo + dopants ko lattice sites par move karo taaki activate hon (profile bhi diffuse hoti hai). Do ion stopping mechanisms ::: nuclear stopping (nuclei collisions, low-E, damage) aur electronic stopping (electron drag, high-E). Depth control mein implant vs diffusion ::: implant bahut better — depth energy se set hoti hai, slow √t thermal spread nahi. --- > [!recall]- Feynman: explain to a 12-year-old > Socho silicon ek Lego wall hai aur hum usme khaas colored bricks (dopants) sprinkle karna chahte hain. > **Diffusion** = colored bricks upar pile karo aur wall ko garam karo taaki bricks slowly wiggle karke andar jaayein — lekin heating se woh *sideways bhi wiggle* karte hain, aur woh sirf slowly creep karte hain (double the time ≈ sirf 1.4× gehraai). > **Implantation** = colored bricks ko ek *toy gun* mein daalo aur shoot karo taaki woh ek chosen depth par lodge ho jaayein. Super precise hai, lekin wall thodi crack ho jaati hai jahan woh hit karte hain — isliye baad mein tum gently warm karte ho (annealing) cracks smooth karne aur bricks ko jagah par lock karne ke liye. > [!mnemonic] > **"DIFFUSE = √time, IMPLANT = √π".** > Diffusion depth **√(Dt)** ki tarah badhti hai; implant peak dose ko **√(2π)·ΔRp** se divide karta hai. Aur implant order ke liye **A.I.D.** yaad rakho: **A**ccelerate → **I**mplant → **D**amage-anneal. ## Connections - [[Semiconductor Fabrication]] - [[Doping and PN Junctions]] - [[Thermal Oxidation]] (high-T furnaces bhi use karta hai; oxide implants mask karta hai) - [[Photolithography]] (define karta hai *kahan* dopants jaate hain) - [[Fick's Laws]] / [[Diffusion Equation]] - [[MOSFET Structure]] (source/drain implant + anneal se banta hai) - [[Arrhenius Equation]] ## 🖼️ Concept Map ```mermaid flowchart TD DOPE[Need doping] -->|creates| NP[n-type and p-type regions] NP -->|form| JUNCTION[Junction = device] DOPE -->|controlled by| DOSE[Dose Q atoms/cm2] DOPE -->|controlled by| PROFILE[Profile N of x] DOPE -->|method 1| DIFF[Diffusion thermal] DOPE -->|method 2| IMPLANT[Ion implantation] DIFF -->|net flux| FICK1[Fick 1st law J = -D dN/dx] FICK1 -->|D from| ARR[Arrhenius D0 exp -Ea/kT] FICK1 -->|conservation gives| FICK2[Fick 2nd law diffusion eqn] FICK2 -->|const surface Ns| ERFC[erfc profile predeposition] FICK2 -->|fixed dose Q| GAUSS[Gaussian profile drive-in] ERFC -->|defines| PROFILE GAUSS -->|defines| PROFILE ```