FUNDAMENTALS OF MATERIALS SCIENCE · CHAPTER 7

Point Defects & Atomic Diffusion II

Fick's Laws, Analytical Solutions & Random-Walk Atomic Theory
TEXTBOOK § 7.4 – § 7.6 · POINT DEFECTS & DIFFUSION · PART 2
“In our derivation, individual atoms have no idea which way the concentration gradient points—every atom jumps blindly at random, and the net flux arises purely from the statistical difference in population between adjacent crystal planes.”Fundamentals of Materials Science · Ch. 7 Macroscopic PDE Meets Microscopic Random Walk
Liang Wumse3d.com · Smart Classroom
Top: macroscopic Gaussian profile spreads self-similarly as 2√(Dt) · Bottom: microscopic 1D random walk yields ⟨x²⟩ = 2Dt
ClickWheel→ NextP PenL LaserO OverviewF FullB Blank
OPENING · FROM STATICS TO KINETICS: THE FIRST PROCESS WITH A REAL TIME CLOCK

Thermodynamics Fixes the Endpoint; Diffusion Answers How Long and by What Scaling Law

QUESTION 01 · HYDROGEN LEAKAGE & PALLADIUM MEMBRANE PURIFIERS

Why does high-pressure \(\mathrm{H_2}\) easily permeate steel walls at room temperature while \(\mathrm{N_2}\) barely leaks at all? And why does doubling \(\mathrm{H_2}\) pressure on both sides of a \(\mathrm{Pd}\) membrane boost permeation flux by only \(\sqrt{2}\) rather than \(2\times\)?

QUESTION 02 · DOUBLING CASE DEPTH & HOMOGENIZING CAST DENDRITES

Why must furnace holding time be multiplied by \(4\) to double a gear's carburized case depth from \(0.5\text{ mm}\) to \(1.0\text{ mm}\)? And why does dendritic microsegregation in cast steel vanish in \(\sim 20\text{ h}\) at \(1100\,^\circ\mathrm{C}\), yet persist for trillions of years at room temperature?

QUESTION 03 · BLINDLY JUMPING ATOMS & BROKEN COPPER INTERCONNECTS

Near its melting point, a \(\mathrm{Cu}\) atom makes \(10^8\) jumps per second—traveling \(2.5\text{ cm}\) in total path—yet achieves a net displacement of only \(2.4\;\mu\text{m}\). Why? And why does a compositionally uniform \(\mathrm{Cu}\) chip interconnect spontaneously grow voids and break open under high current density?

Textbook § 7.4–7.6 · Macroscopic Fick's Laws & Atomic Theory of D
CHAPTER 7 PART 2 · FIVE-STOP ROADMAP

Macroscopic Continuum PDEs and Microscopic Lattice Jumps Converge on One Ruler: \(\sqrt{Dt}\)

  1. Fick's First LawPlanar & spherical steady states, \(P=DS\) & Sieverts' \(\sqrt{p}\) lawTextbook § 7.4
  2. Fick's Second LawContinuity equation, curvature smoothing & Boltzmann \(\eta=x/\sqrt{Dt}\)Textbook § 7.5.1–2
  3. Four Analytical ProfilesCarburizing \(\mathrm{erf}\), thin-film Gaussian, homogenization & \(2\sqrt{Dt}\)Textbook § 7.5.2–4
  4. Random-Walk TheoryAdjacent-plane jumps to \(D=\tfrac{1}{6}\lambda^2\Gamma\) & MSD \(\langle R^2\rangle=6Dt\)Textbook § 7.6.1
  5. Activation & DriftVacancy barrier \(Q=\Delta H_{\mathrm{v}}+Q_{\mathrm{m}}\), Einstein \(M=D/k_{\mathrm{B}}T\) & EMTextbook § 7.6.2–4
Chapter 7 · Part 2 (§ 7.4 – § 7.6)
01
PART ONE

Fick's First Law & Steady-State Permeation

When the concentration field no longer varies with time, matter entering every thin slab strictly equals matter leaving—making total diffusive transport constant across space.
  1. Fick's first law \(\boldsymbol{J}=-D\nabla C\), validity regime & Case 7.3 \(\mathrm{Pd}\) \(\mathrm{H_2}\) purifier05
  2. Permeability \(P=DS\): Henry's linear law vs. Sieverts' \(\sqrt{p}\) law06
§ 7.4.1 – § 7.4.2 · FICK'S FIRST LAW (1855) & STEADY-STATE MEMBRANES

Steady-State Planar Profiles Are Linear \(J=D\tfrac{C_1-C_2}{L}\); Spherical Radial Flux Scales as \(1/R\)

\(\begin{gathered}J=-D\,\frac{\partial C}{\partial x}\;\;(\text{3D: }\boldsymbol{J}=-D\nabla C,\;\mathrm{mol\,m^{-2}\,s^{-1}})\\ \text{Planar: }J=D\frac{C_1-C_2}{L},\quad\text{Spherical: }J(R)=D\frac{C_{\mathrm{s}}-C_\infty}{R}\end{gathered}\)
  • Minus Sign & Steady-State Lag: Flux points down the concentration gradient (positive entropy production requires \(-J\,\partial\mu/\partial x\ge 0\); \(D\lt 0\) uphill diffusion occurs only when \(\partial\mu/\partial C\lt 0\)). Establishing steady state takes \(t\gtrsim L^2/D\).
  • Case 7.3 · Palladium \(\mathrm{H_2}\) Purifier: A \(0.1\text{ mm}\) \(\mathrm{Pd}\) membrane at \(500\,^\circ\mathrm{C}\) (\(D_{\mathrm{H}}=10^{-8}\text{ m}^2/\text{s}\)) reaches steady state in \(\lt 1\text{ s}\), delivering \(J\approx 0.08\text{ mol/(m}^2\text{·s)}\) while blocking all non-dissociating impurity gases.
① Planar Steady-State Membrane: Constant Area ⇒ Constant J ⇒ Linear C(x) High-p Side C₁ Linear profile C(x) J = D(C₁ − C₂)/L → Low-p Side C₂ Lag time t ≳ L²/D ② Spherical Particle Steady Diffusion: 4πr²J = const ⇒ J(R) = D(Cs − C_∞)/R R Surface flux J(R) = D(Cs − C_∞) / R Smaller radius R drives steeper dissolution/growth flux → Drives Ch. 9 Ostwald ripening (small dissolve, large grow)
Textbook § 7.4.1–7.4.2 · Fick's First Law & Case 7.3 Pd Membrane
§ 7.4.3 · PERMEABILITY P = DS: HENRY'S LAW VS. SIEVERTS' SQUARE-ROOT LAW

Diatomic Gases Dissociate in Metals \((\mathrm{H_2\rightleftharpoons 2H})\), Fixing \(\sqrt{p}\)—Thinning Beats Pressurizing

\(\begin{gathered}\text{Molecular (polymers): }C=Sp\implies J=\frac{P}{L}\Delta p\\ \text{Dissociative (metals): }C=S\sqrt{p}\implies J=\frac{P}{L}(\sqrt{p_1}-\sqrt{p_2})\\ \text{Selectivity: }\alpha_{\mathrm{AB}}=\frac{P_{\mathrm{A}}}{P_{\mathrm{B}}}=\frac{D_{\mathrm{A}}}{D_{\mathrm{B}}}\cdot\frac{S_{\mathrm{A}}}{S_{\mathrm{B}}}\;\;(P\equiv DS)\end{gathered}\)
  • Why Thinning Beats Pressurizing: Doubling \(\mathrm{H_2}\) pressure on both sides of a \(\mathrm{Pd}\) membrane boosts flux by only \(\mathbf{\sqrt{2}\approx 1.41\times}\), whereas halving thickness \(L\) doubles \(J\) directly.
  • Answering Question 1 (\(\mathrm{H_2}\) vs. \(\mathrm{N_2}\) in Steel): In \(300\text{ K}\) \(\alpha\text{-Fe}\), \(D_{\mathrm{H}}\approx 7\times 10^{-9}\text{ m}^2/\text{s}\) vs. \(D_{\mathrm{N}}\sim 10^{-20}\text{ m}^2/\text{s}\) (\(11\) orders apart) and \(\mathrm{N_2}\) triple bonds barely dissociate at \(300\text{ K}\), so \(\mathrm{H_2}\) permeates steel while \(\mathrm{N_2}\) stays sealed.
Textbook § 7.4.3 · Permeability P = DS, Henry & Sieverts' Laws
02
PART TWO

Fick's Second Law & Diffusion Length \(2\sqrt{Dt}\)

Coupling Fick's first law with mass conservation yields a curvature-driven PDE that trims peaks and fills valleys—locking distance to the square root of time across carburizing, tracers, and annealing.
  1. Continuity equation, concentration-dependent \(D(C)\) & Boltzmann \(\eta=x/\sqrt{Dt}\)08
  2. Interactive model: diffusion couples, Case 7.4 gear carburizing & Gaussian tracers09
  3. Characteristic diffusion length \(L_{\mathrm{d}}=2\sqrt{Dt}\) & dendritic homogenization \(\tau\)10
§ 7.5.1 – § 7.5.2 · CONTINUITY EQUATION, FICK'S SECOND LAW & ERROR-FUNCTION DERIVATION

Curvature \(\partial^2 C/\partial x^2\) Drives Concentration Change; Scale Invariance Collapses PDEs to \(\eta=x/\sqrt{Dt}\)

\(\begin{gathered}\frac{\partial C}{\partial t}=-\frac{\partial J}{\partial x}=\frac{\partial}{\partial x}\!\left(D(C)\frac{\partial C}{\partial x}\right)\xrightarrow{D=\text{const}}D\frac{\partial^2 C}{\partial x^2}\\ \eta=\frac{x}{\sqrt{Dt}}\implies\frac{\mathrm{d}^2 C}{\mathrm{d}\eta^2}+\frac{\eta}{2}\frac{\mathrm{d}C}{\mathrm{d}\eta}=0\implies C(\eta)=A'\mathrm{erf}\!\left(\frac{\eta}{2}\right)+B\end{gathered}\)
  • Geometric Meaning of Curvature: Concave-up valleys (\(C''\gt 0\)) receive more influx than efflux (\(\partial C/\partial t\gt 0\)); concave-down peaks (\(C''\lt 0\)) drain (\(\partial C/\partial t\lt 0\)); straight lines (\(C''=0\)) remain at steady state.
  • Two Canonical \(\mathrm{erf}\) Profiles: ① Semi-infinite couple \(C=\frac{C_0}{2}\bigl[1-\mathrm{erf}(\frac{x}{2\sqrt{Dt}})\bigr]\) (pinned at \(C_0/2\) at \(x=0\)); ② Constant-surface carburizing \(\frac{C_{\mathrm{s}}-C}{C_{\mathrm{s}}-C_0}=\mathrm{erf}(\frac{x}{2\sqrt{Dt}})\) (semi-infinite valid while half-thickness \(\gt 4\sqrt{Dt}\)).
Variable Diffusivity \(D(C)\): Boltzmann–Matano AnalysisIn \(927\,^\circ\mathrm{C}\) austenite carburizing, high-carbon surface layers have nearly \(2\times\) higher \(D_{\mathrm{C}}\) than the low-carbon core, squaring off measured profiles relative to constant-\(D\) \(\mathrm{erfc}\). Locating the equal-area Matano plane on a diffusion-couple profile extracts the full interdiffusivity curve \(\tilde{D}(C)\).
Instantaneous Thin-Film Source (Gaussian Solution · Eq. 7.40)Depositing a fixed dose \(N_{\mathrm{s}}\) at \(x=0\) with zero further supply gives \(C(x,t)=\frac{N_{\mathrm{s}}}{\sqrt{\pi Dt}}\exp(-\frac{x^2}{4Dt})\). Total solute \(N_{\mathrm{s}}\) is conserved as peak height decays like \(t^{-1/2}\); radiotracer sectioning plots \(\ln C\) vs. \(x^2\) to extract \(D^*\) from slope \(-1/(4Dt)\).
Textbook § 7.5.1–7.5.3 · Continuity, Boltzmann η & erf/Gaussian
§ 7.5.2 · CASE 7.4 GAS CARBURIZING OF STEEL GEARS & TRANSIENT DIFFUSION LAB

Doubling Case Depth Quadruples Furnace Time: Drag \(C^*\) to Design Gear Carburizing

\(\begin{gathered}\frac{C_{\mathrm{s}}-C}{C_{\mathrm{s}}-C_0}=\frac{0.8-0.4}{0.8-0.2}=0.667=\mathrm{erf}(z)\\ z=\frac{x}{2\sqrt{Dt}}\approx 0.685\\ t=\frac{(5\times 10^{-4})^2}{4\times 1.28\times 10^{-11}\times 0.685^2}\approx 2.9\text{ h}\end{gathered}\)
  • Case 7.4 Gear Design: Base steel \(C_0=0.2\%\), gas carbon potential \(C_{\mathrm{s}}=0.8\%\), target \(0.4\%\text{ C}\) at \(x=0.5\text{ mm}\) at \(927\,^\circ\mathrm{C}\) requires \(t\approx 2.9\text{ h}\) (\(4\sqrt{Dt}=1.46\text{ mm}\), satisfying semi-infinite geometry for standard gear teeth).
  • Four Regimes in One Lab: Switch presets on the right to compare gas carburizing, diffusion couples, Gaussian tracers, and dendritic homogenization.
Textbook § 7.5.2–7.5.3 · Case 7.4 Gear Carburizing & Table 7.1
§ 7.5.3 – § 7.5.4 · DIFFUSION LENGTH Ld = 2√(Dt) & DENDRITIC HOMOGENIZATION

When \(T\) Hits Its Ceiling, Shortening Distance Beats Waiting: \(1/5\) Spacing Cuts Time to \(1/25\)

\(\begin{gathered}L_{\mathrm{d}}\equiv 2\sqrt{Dt}\;\;(1-\mathrm{erf}(1)\approx 15.7\%),\quad t\approx\frac{L^2}{4D}\\ C(x,t)=\bar{C}+\beta_0\sin\!\frac{\pi x}{l}\,\mathrm{e}^{-t/\tau},\quad\tau=\frac{l^2}{\pi^2 D}\end{gathered}\)
  • Why Fast-Charge Cathodes Use Fine Particles: In \(\mathrm{LiCoO_2}\) (\(D_{\mathrm{Li}}\sim 10^{-14}\text{ m}^2/\text{s}\)), a \(5\;\mu\text{m}\) particle needs \(\sim 10\text{ min}\) to lithiate, whereas refining to \(1\;\mu\text{m}\) cuts diffusion time to \(\sim 0.5\text{ min}\) (\(25\times\) faster).
  • Answering Question 2 (Cast-Steel Segregation): For half-spacing \(l=25\;\mu\text{m}\), substitutional \(\mathrm{Mn}\) at \(1100\,^\circ\mathrm{C}\) gives \(\tau\approx 10\text{ h}\) (\(90\%\) homogenized in \(22\text{ h}\)); at \(300\text{ K}\), \(D\lt 10^{-30}\text{ m}^2/\text{s}\) gives \(\tau\gt 10^{12}\text{ yr}\).
Textbook § 7.5.3–7.5.4 · Diffusion Length 2√(Dt) & Homogenization
03
PART THREE

Atomic Theory: Random Walk, Barriers & Drift

Derive Fick's law from blind jumps between adjacent crystal planes, connect Part 1 vacancies to diffusivity via Q = ΔHv + Qm, and tilt the periodic barrier with external forces to obtain Einstein's relation.
  1. Bidirectional plane jumps yield \(D=\frac{1}{6}\lambda^2\Gamma\) & \(\langle R^2\rangle=6Dt\)12
  2. Interactive model: \(2000\)-atom random walks converging to the \(\sqrt{t}\) law13
  3. Two-stage vacancy barrier \(Q=\Delta H_{\mathrm{v}}+Q_{\mathrm{m}}\) & Arrhenius plot14
  4. Einstein mobility \(M=D/(k_{\mathrm{B}}T)\), \(\mu\)-driven flux & Case 7.5 electromigration15
§ 7.6.1 · DERIVING FICK'S FIRST LAW & MEAN-SQUARE DISPLACEMENT FROM LATTICE JUMPS

Unbiased Single-Atom Jumps Across Adjacent Planes Automatically Yield \(D=\tfrac{1}{6}\lambda^2\Gamma\)

\(\begin{gathered}J=\frac{\Gamma}{6}(n_1-n_2)=\frac{\Gamma\lambda}{6}\bigl[C(x)-C(x+\lambda)\bigr]=-\underbrace{\frac{1}{6}\lambda^2\Gamma}_{D}\frac{\partial C}{\partial x}\\ \langle R^2\rangle=\sum_{i=1}^n\langle r_i^2\rangle+\!\!\sum_{i\neq j}\underbrace{\langle\boldsymbol{r}_i\cdot\boldsymbol{r}_j\rangle}_{0}=n\lambda^2=\Gamma\lambda^2 t=6Dt\end{gathered}\)
  • Universal Cubic Factor \(1/6\): In simple cubic (\(z=6\)), \(\pm x\) jumps each take \(1/6\); in fcc along \([100]\) (plane spacing \(a/2\), jump \(\lambda=a/\sqrt{2}\)), \(4\) of \(12\) neighbors cross to the next plane—giving the exact same \(D=\frac{1}{6}\Gamma\lambda^2\).
  • Macro–Micro Convergence: Projecting 3D \(\langle R^2\rangle=6Dt\) onto one axis gives \(\langle x^2\rangle=2Dt\)—matching the macroscopic thin-film Gaussian variance identically.
Fig. 7.13: Unbiased Plane Jumps Yield Fick's Law vs. Tilted Barrier Yields Drift (a) Plane Spacing λ · Blind Jumps (Γ/6) n₁ (Γ/6) n₂ Plane 1 n₁ = C(x)λ Plane 2 n₂ = C(x+λ)λ Net J = −(λ²Γ/6) ∂C/∂x (b) Force F Tilts Periodic Barrier Fwd lowered · Rev raised (Δg = Fλ) Γ₊ > Γ₋ ⇒ v = (D/kBT) F Einstein mobility: M = D / (kBT)
Textbook § 7.6.1 · Plane-Jump Derivation & Random-Walk MSD
§ 7.6.1 · RANDOM-WALK SCALING & ORDERS OF MAGNITUDE: 10⁸ JUMPS/S VS. ONE JUMP IN 300 BILLION YEARS

Back-and-Forth Cancellations Force \(\sqrt{n}\) Scaling: Room-Temperature \(\mathrm{Cu}\) Lattices Are Frozen Solid

\(\begin{gathered}R_{\mathrm{rms}}=\sqrt{\langle R^2\rangle}=\sqrt{n}\,\lambda=\sqrt{6Dt}\\ T_{\mathrm{m}}\text{ Cu: }\Gamma\sim 10^8\text{ s}^{-1}\;(2.5\text{ cm}\to 2.4\;\mu\text{m})\\ 300\text{ K Cu: }D\sim 10^{-39}\text{ m}^2/\text{s}\;(\Gamma^{-1}\approx 3\times 10^{11}\text{ yr})\end{gathered}\)
  • One Jump per \(10^5\) Vibrations: Even near \(T_{\mathrm{m}}\), against \(\nu_0\sim 10^{13}\text{ s}^{-1}\) phonons, an atom attempts the barrier \(\sim 10^5\) times per successful hop, and net RMS drift is only \(10^{-4}\) of total path length.
  • Sweep \(N\) from \(1\) to \(2000\): Watch noisy single-atom walks collapse onto the \(\langle R^2\rangle=n\lambda^2\) line as cross-correlation terms \(\sum_{i\neq j}\langle\boldsymbol{r}_i\cdot\boldsymbol{r}_j\rangle\) vanish.
Textbook § 7.6.1 · Random-Walk √n Scaling & Cu Diffusivity
§ 7.6.2 – § 7.6.3 · TWO-STAGE VACANCY DIFFUSION Q = ΔHv + Qm & ARRHENIUS PLOT

Adjacent Vacancy \((C_{\mathrm{v}})\) × Saddle-Point Squeeze \((\omega)\): Multiplying Boltzmann Factors Adds Their Barriers

\(\begin{gathered}\Gamma_{\mathrm{vac}}=z\,C_{\mathrm{v}}\,\omega\implies D=D_0\exp\!\left(-\frac{\Delta H_{\mathrm{v}}+Q_{\mathrm{m}}}{k_{\mathrm{B}}T}\right)\\ D_0=\frac{1}{6}fz\lambda^2\nu_0\exp\!\left(\frac{\Delta S_{\mathrm{v}}+\Delta S_{\mathrm{m}}}{k_{\mathrm{B}}}\right)\;\;(\text{fcc }f=0.78)\end{gathered}\)
  • Part 1 Meets Part 2: In \(\mathrm{Cu}\), \(\Delta H_{\mathrm{v}}\approx 1.28\text{ eV}\) plus migration \(Q_{\mathrm{m}}\approx 0.70\text{ eV}\) predicts \(1.98\text{ eV}\), matching measured \(2.04\text{ eV}\) (\(197\text{ kJ/mol}\)). Interstitial \(\mathrm{C}\) needs no vacancies (\(Q=Q_{\mathrm{m}}\)), diffusing \(\mathbf{4\times 10^5\times}\) faster than \(\mathrm{Fe}\) in \(927\,^\circ\mathrm{C}\) \(\gamma\text{-Fe}\).
  • Melting-Point Scaling Rule: Close-packed fcc/hcp metals obey \(Q/(RT_{\mathrm{m}})\approx 18\), clustering self-diffusivity near \(D(T_{\mathrm{m}})\approx \mathbf{10^{-12}\text{ m}^2/\text{s}}\).
Textbook § 7.6.2–7.6.3 · Vacancy Barrier Q = ΔHv + Qm & Arrhenius
§ 7.6.4 · EINSTEIN RELATION M = D/(kBT), GENERALIZED FLUX & CASE 7.5 ELECTROMIGRATION

Unifying Concentration Gradients and External Forces into \(-\partial\mu/\partial x\) Explains Electromigration & Uphill Diffusion

\(\begin{gathered}v=\lambda(\Gamma_+-\Gamma_-)=\frac{D}{k_{\mathrm{B}}T}F\implies M=\frac{D}{k_{\mathrm{B}}T}\;\;(\text{Einstein})\\ J=-D\frac{\partial C}{\partial x}+C\frac{D}{k_{\mathrm{B}}T}F=-\frac{DC}{k_{\mathrm{B}}T}\frac{\partial\mu}{\partial x}\\ \text{Non-ideal: }D=D^*\Phi,\quad\Phi=1+\frac{\partial\ln\gamma}{\partial\ln x_{\mathrm{B}}}\end{gathered}\)
  • Tracer \(D^*\) vs. Thermodynamic Factor \(\Phi\): Kinetic \(D^*\) is strictly positive, whereas \(\Phi\) tracks \(\partial\mu/\partial C\)—when \(\Phi\lt 0\), \(D=D^*\Phi\lt 0\) drives spontaneous uphill diffusion (spinodal decomposition).
  • Answering Question 3 · Case 7.5 \(\mathrm{Cu}\) Interconnects: At \(j\sim 10^6\text{ A/cm}^2\), electron-wind force \(F=Z^*e\rho j\) (\(Z^*\approx -5\)) sweeps \(\mathrm{Cu}\) atoms along grain boundaries; flux divergence depletes voids (open circuit) and piles up hillocks (short circuit)—mitigated by trace \(\mathrm{Mn/Al}\) doping or bamboo grains.
Equilibrium Consistency CheckUnder a conservative potential \(U=-Fx\), setting net flux \(J=0\) requires \(\frac{1}{C}\frac{\mathrm{d}C}{\mathrm{d}x}=\frac{F}{k_{\mathrm{B}}T}\), integrating directly to Boltzmann's law \(C(x)=C_0\exp(Fx/k_{\mathrm{B}}T)\)—proving that \(M=D/(k_{\mathrm{B}}T)\) is uniquely required for kinetic–thermodynamic consistency.
Chip Interconnect Electromigration LedgerAtomic drift flux obeys \(J_{\mathrm{atom}}=\frac{DC}{k_{\mathrm{B}}T}Z^*e\rho j\): single-hop work \(F\lambda\approx 2.5\times 10^{-7}\text{ eV}\ll k_{\mathrm{B}}T\) guarantees exact linear response, while moderate chip temperatures make \(D_{\mathrm{gb}}\gg D_{\mathrm{b}}\) so transport runs almost entirely along grain boundaries & cap interfaces.
Textbook § 7.6.4 · Einstein Mobility, Φ & Case 7.5 Electromigration
SUMMARY

Macroscopic Diffusion PDEs and Microscopic Random Walks Unite Under \(-\nabla\mu\)

Fick's 1st & √p Law→PDE ∂C/∂t = D∇²C→erf · Gauss · τ & 2√(Dt)→Random Walk D = λ²Γ/6→Q = ΔHv+Qm & M = D/kT
BACK TO QUESTIONS 01 & 02

Tiny interstitial \(\mathrm{H}\) diffuses \(11\) orders faster than \(\mathrm{N}\) in \(300\text{ K}\) \(\alpha\text{-Fe}\), while \(\mathrm{H_2\rightleftharpoons 2H}\) dissociation yields Sieverts' \(\sqrt{p}\) scaling; scale-free PDEs depend on \(x/\sqrt{Dt}\) so \(2\times\) case depth takes \(4\times\) time; cast-steel dendritic homogenization \(\tau=l^2/(\pi^2 D)\) takes \(\sim 20\text{ h}\) at \(1100\,^\circ\mathrm{C}\) vs. \(\gt 10^{12}\text{ yr}\) at \(300\text{ K}\).

BACK TO QUESTION 03

Uncorrelated jump directions cancel \(10^8\) steps down to \(\sqrt{n}\lambda\approx 2.4\;\mu\text{m}\) RMS drift; under \(10^6\text{ A/cm}^2\) electron wind, \(\mathrm{Cu}\) atoms drift along grain boundaries at \(J=(DC/k_{\mathrm{B}}T)Z^*e\rho j\), nucleating open-circuit voids at triple-junction flux divergences.

Next Lecture — Point Defects & Atomic Diffusion III: Solid-State Li-Battery Ionics, Kirkendall Marker Drift & Darken Equations, Parabolic Oxidation & Diffusional Creep.

Point Defects & Atomic Diffusion II
CONCEPT QUIZ

Three Quick Questions: Click to Check or Launch a Live Class Poll

1. At \(1000\,^\circ\mathrm{C}\), carbon in \(\gamma\text{-Fe}\) has \(D=3.5\times 10^{-11}\text{ m}^2/\text{s}\). After \(t=2\text{ h}\) (\(7200\text{ s}\)), what is the diffusion length \(L_{\mathrm{d}}=2\sqrt{Dt}\), and what total time doubles that depth?

\(\sqrt{Dt}=\sqrt{3.5\times 10^{-11}\times 7200}\approx 0.50\text{ mm}\), so \(L_{\mathrm{d}}=2\sqrt{Dt}\approx 1.00\text{ mm}\); since \(L_{\mathrm{d}}\propto\sqrt{t}\), doubling depth requires \(4\times 2\text{ h}=8\text{ h}\).

2. In a cubic crystal with jump distance \(\lambda=0.3\text{ nm}\) and jump rate \(\Gamma=10^8\text{ s}^{-1}\), what is \(D=\frac{1}{6}\lambda^2\Gamma\)? Why is \(Q\) for \(\mathrm{C}\) in \(\gamma\text{-Fe}\) far lower than \(\mathrm{Fe}\) self-diffusion?

\(D=\frac{1}{6}(3\times 10^{-10})^2\times 10^8=1.5\times 10^{-12}\text{ m}^2/\text{s}\); vacancy self-diffusion requires \(Q=\Delta H_{\mathrm{v}}+Q_{\mathrm{m}}\), whereas interstitial sites are mostly empty so \(Q=Q_{\mathrm{m}}\).

3. Combining Sieverts' permeation law, homogenization kinetics, and Einstein's relation, which statement is completely accurate?

\(\mathrm{H_2\rightleftharpoons 2H}\) gives \(C=S\sqrt{p}\), so doubling \(p\) raises \(J\) by \(\sqrt{2}\); negative thermodynamic factor \(\Phi\lt 0\) drives uphill diffusion; homogenization obeys \(\tau=l^2/(\pi^2 D)\propto l^2\).

Point Defects & Atomic Diffusion II
FUNDAMENTALS OF MATERIALS SCIENCE · CHAPTER 7

Thank You

2 Embedded Models: Diffusion Couple & Carburizing · Random Walk & √t Diffusion Law — mse3d.com