Lecture 6 - Knowledge Check

Assess your advanced understanding of Non-Equilibrium Transport & Excess Carriers.

Progress Question 1 of 25
1

High-Field Velocity Saturation

The total velocity of a charge carrier is the sum of its random thermal velocity and its drift velocity. As the applied electric field across a silicon sample is continuously increased into the high-field regime, the carrier drift velocity stops increasing linearly and plateaus at a saturation velocity ($v_{sat} \approx 10^7$ cm/s). What is the primary microscopic scattering mechanism responsible for this abrupt high-field velocity saturation?

Explanation

Correct Answer: C. At low electric fields, the drift velocity is linearly proportional to the field ($v_{drf} = \mu E$). However, at extremely high electric fields, the charge carriers gain immense kinetic energy. This excess energy is rapidly dissipated through frequent, highly energetic collisions with optical phonons in the crystal lattice. This mechanism acts as a speed limit, causing the drift velocity to completely saturate, regardless of further increases in the external field.
2

Diffusion Current Sign Convention

The macroscopic 1D diffusion current density for electrons is mathematically defined as $J_{nx|dif} = e D_n \frac{dn}{dx}$. However, the equivalent diffusion current density for holes is written as $J_{px|dif} = -e D_p \frac{dp}{dx}$. Why does the hole diffusion current equation explicitly carry a negative sign?

Step-by-Step Solution

Correct Answer: B.
  1. According to Fick's First Law, particles always physically diffuse from regions of high concentration to low concentration (they move down the gradient slope). Therefore, the physical particle flux vector is negative: $\Phi_p = -D_p \frac{dp}{dx}$.
  2. Current density is defined as the product of charge and particle flux: $J = \text{Charge} \times \Phi$.
  3. Because holes carry a positive elementary charge ($+e$), the current vector equation maintains the original negative sign from the flux: $J_{px} = (+e) \times \left( -D_p \frac{dp}{dx} \right) = -e D_p \frac{dp}{dx}$.
  4. Conversely, electrons carry a negative charge ($-e$). When computing their current, the two negatives perfectly cancel out: $J_{nx} = (-e) \times \left( -D_n \frac{dn}{dx} \right) = +e D_n \frac{dn}{dx}$.
3

Total Current Profile Analysis

The total 1D electron current density is mathematically expressed as the sum of drift and diffusion components: $J_n = e n \mu_n E_x + e D_n \frac{dn}{dx}$. Consider a perfectly shielded semiconductor region where the electric field is strictly zero ($E_x = 0$), but a steady-state, constant electron current $J_n$ still flows exclusively via diffusion. Assuming the diffusion coefficient $D_n$ is constant, what must be the physical shape of the spatial electron concentration profile $n(x)$?

Step-by-Step Solution

Correct Answer: B.
  1. Start with the full total current equation: $J_n = e n \mu_n E_x + e D_n \frac{dn}{dx}$.
  2. Apply the given zero-field condition ($E_x = 0$). The entire drift term evaluates to zero.
  3. The equation simplifies to pure diffusion: $J_n = e D_n \frac{dn}{dx}$.
  4. Rearrange the equation to isolate the concentration gradient: $\frac{dn}{dx} = \frac{J_n}{e D_n}$.
  5. Analyze the isolated fraction: The problem explicitly defines the total current $J_n$ and the diffusion coefficient $D_n$ as constant values. The elementary charge $e$ is naturally a constant. Therefore, the entire right side of the equation represents a single constant scalar (let's call it $C$).
  6. We now have the mathematical differential equation: $\frac{dn}{dx} = C$.
  7. Integrating both sides with respect to $x$ yields: $n(x) = Cx + n(0)$. This is the classic slope-intercept form ($y = mx + b$) representing a straight, linear profile.
4

Built-in Electric Field Direction

The built-in electric field induced by a non-uniform doping gradient is mathematically modeled as $E_x = -\frac{kT}{e} \frac{1}{N_d(x)} \frac{dN_d(x)}{dx}$. If the donor concentration $N_d(x)$ is highly concentrated on the left edge ($x=0$) and smoothly decreases as one moves to the right (increasing $x$), in what direction does the resulting built-in electric field vector point?

Step-by-Step Solution

Correct Answer: A.
  1. Analyze the physical setup: The donor concentration function $N_d(x)$ has a high value on the left ($x=0$) and structurally decreases toward the right (positive $x$).
  2. Evaluate the mathematical derivative: Because the concentration function decreases as the variable $x$ increases, the fundamental spatial derivative slope $\frac{dN_d(x)}{dx}$ is mathematically a negative value.
  3. Examine the built-in field equation: $E_x = -\frac{kT}{e} \frac{1}{N_d} \times (\text{Negative Value})$.
  4. Perform the sign calculation: The explicit negative sign placed at the front of the formula mathematically multiplies with the negative derivative value.
  5. Determine final vector direction: $(-) \times (-) = (+)$. The resulting electric field $E_x$ evaluates to a positive scalar, meaning the electrostatic vector points cleanly in the positive x-direction.
5

Doping Gradient Band Bending

In a non-uniformly doped n-type semiconductor at thermal equilibrium, the donor concentration $N_d(x)$ linearly decreases from left to right. When plotting the energy band diagram across this region, how do the conduction band $E_c$ and valence band $E_v$ geometrically align relative to the constant Fermi level $E_F$?

Step-by-Step Solution

Correct Answer: D.
  1. In thermal equilibrium, the Fermi level $E_F$ must be strictly constant (perfectly horizontal).
  2. The donor concentration $N_d(x)$ decreases from left to right. Because $n \approx N_d$, the electron concentration actively decreases.
  3. From the mass action equation $n = n_i e^{(E_F - E_{Fi})/kT}$, as $n$ decreases, the required gap between $E_F$ and $E_{Fi}$ must decrease.
  4. Since $E_F$ is locked flat, $E_{Fi}$ must physically move upward (closer to $E_F$) as we move right to reduce the gap.
  5. The conduction band $E_c$ and valence band $E_v$ always run perfectly parallel to $E_{Fi}$. Therefore, both bands must also slope upward (positive slope) from left to right to accurately reflect the internal built-in electric field.
6

Einstein Relation Derivation

When deriving the Einstein Relation ($\frac{D_n}{\mu_n} = \frac{kT}{e}$) from a non-uniformly doped equilibrium semiconductor, why is the assumption of macroscopic quasi-neutrality ($n \approx N_d(x)$) mathematically critical to completing the proof?

Step-by-Step Solution

Correct Answer: B.
  1. The derivation begins with the net zero current equation in equilibrium: $J_n = e \mu_n n E_x + e D_n \frac{dn}{dx} = 0$.
  2. To solve this, we must link the local electron concentration $n$ to the known built-in field $E_x$.
  3. The built-in field $E_x$ is derived based on the donor profile: $E_x = -\frac{kT}{e} \frac{1}{N_d(x)} \frac{dN_d(x)}{dx}$.
  4. By assuming quasi-neutrality ($n \approx N_d(x)$), we can substitute $N_d(x)$ directly into the current equation for $n$, and $\frac{dN_d}{dx}$ for $\frac{dn}{dx}$.
  5. This crucial substitution allows all concentration terms to algebraically cancel out, isolating the pure ratio $\frac{D_n}{\mu_n} = \frac{kT}{e}$.
7

Einstein Relation Failure Regimes

The elegant Einstein Relation $\frac{D}{\mu} = \frac{kT}{e}$ is a foundational link between carrier drift and diffusion. Under what specific physical operational regime does this classic relation begin to systematically fail in actual manufactured semiconductor devices?

Explanation

Correct Answer: C. The Einstein relation assumes that the carrier distribution follows classical Maxwell-Boltzmann statistics and that mobility $\mu$ is a strict constant. In very high electric fields (e.g., inside modern short-channel transistors), carriers gain so much kinetic energy they become "hot carriers." Their drift velocity saturates due to optical phonon scattering, causing $\mu$ to drop drastically as a function of the field $E$. This nonlinear behavior fundamentally breaks the simple $\frac{D}{\mu} = \frac{kT}{e}$ ratio.
8

Limits of the Exponential Decay Model

The elegant exponential decay equation for excess carriers, $\delta n(t) = \delta n(0) e^{-t/\tau_{n0}}$, is frequently used in device physics. However, under high-level injection ($\delta n \approx p_0$ or greater), this specific simplified mathematical model completely fails. Why?

Step-by-Step Solution

Correct Answer: B.
  1. The raw decay equation is: $\frac{d(\delta n)}{dt} = -\alpha_r \delta n [(n_0 + p_0) + \delta n]$.
  2. Under low-level injection in a p-type material, $p_0 \gg n_0$ and $p_0 \gg \delta n$. Thus, the bracketed term collapses to just $p_0$.
  3. This allows us to define a strict constant $\tau_{n0} = (\alpha_r p_0)^{-1}$, leading to a clean exponential differential equation.
  4. Under high-level injection, $\delta n$ is no longer negligible compared to $p_0$. The bracketed term $[p_0 + \delta n]$ changes as $\delta n$ decays over time.
  5. Because the "constant" $\tau$ is now actively changing with time, the simple exponential solution $e^{-t/\tau_{n0}}$ is no longer a mathematically valid solution to the differential equation.
9

Low-Level Injection Decay Scaling

A semiconductor is illuminated, reaching a steady-state excess minority carrier concentration of $\Delta n_0$. The light is abruptly turned off at $t=0$, initiating decay according to $\delta n(t) = \delta n_0 e^{-t/\tau_{n0}}$. At exactly $t = 3\tau_{n0}$ (three minority carrier lifetimes later), use the interactive chart to determine approximately what percentage of the initial excess carriers still remain.

Derivation

Correct Answer: C.
1. Evaluate the decay equation at the requested time: $\delta n(3\tau_{n0}) = \delta n_0 e^{-(3\tau_{n0})/\tau_{n0}}$.
2. The $\tau_{n0}$ terms cancel: $\delta n = \delta n_0 e^{-3}$.
3. Mathematically evaluate $e^{-3} \approx 0.04978$.
4. Convert to a percentage: $0.04978 \times 100 \approx 4.98\% \approx 5.0\%$.
This is a standard engineering rule of thumb: an exponential decay process is considered ~95% complete after 3 time constants.
10

Non-Equilibrium Mass Action Calculations

Given a physical semiconductor sample sitting in the dark where the thermal equilibrium concentrations are measured to be $n_0 = 10^{15} \text{ cm}^{-3}$, $p_0 = 10^5 \text{ cm}^{-3}$, and $n_i = 10^{10} \text{ cm}^{-3}$. A light source is turned on, injecting steady-state excess carriers $\delta n = \delta p = 10^{13} \text{ cm}^{-3}$. Calculate the exact non-equilibrium $n \cdot p$ product of the resulting system.

Derivation

Correct Answer: C.
1. Calculate the new total electron concentration: $n = n_0 + \delta n = 10^{15} + 10^{13} = 1.01 \times 10^{15} \text{ cm}^{-3}$.
2. Calculate the new total hole concentration: $p = p_0 + \delta p = 10^5 + 10^{13} \approx 10^{13} \text{ cm}^{-3}$. (Because $10^{13} \gg 10^5$, the baseline $10^5$ is statistically irrelevant).
3. Multiply the new totals together: $np = (1.01 \times 10^{15}) \times (10^{13})$.
4. The exponents add ($15 + 13 = 28$), resulting in a non-equilibrium product of $\approx 1.01 \times 10^{28} \text{ cm}^{-6}$.
This clearly demonstrates how drastically light injection shatters the equilibrium rule of $n_i^2 = 10^{20}$.
11

Continuity Equation Flux Derivative

In the differential derivation of the 1D Continuity Equation, the spatial term $-\frac{\partial F_p}{\partial x}dx$ emerges from expanding a Taylor series across a boundary. What physical phenomenon does a positive value for this specific partial derivative ($\frac{\partial F_p}{\partial x} > 0$) strictly imply for the differential volume?

Step-by-Step Solution

Correct Answer: B.
  1. The continuity equation tracks the change in concentration: $\frac{\partial p}{\partial t} = \text{Flux In} - \text{Flux Out} + \dots$
  2. The flux difference is $F_p(x) - F_p(x+dx)$. Using a Taylor series, this difference is explicitly equal to $-\frac{\partial F_p}{\partial x} dx$.
  3. If the derivative itself is positive ($\frac{\partial F_p}{\partial x} > 0$), it means the flux function $F_p(x)$ is increasing as $x$ increases.
  4. Physically, this means the flux leaving at $x+dx$ is mathematically larger than the flux entering at $x$.
  5. Because Flux Out > Flux In, the net contribution of the $-\frac{\partial F_p}{\partial x}$ term is negative, causing a net depletion (reduction) of carriers in that volume over time.
12

Time-Dependent Diffusion Solutions

The master time-dependent diffusion equation for holes is $\frac{\partial p}{\partial t} = D_p \frac{\partial^2 p}{\partial x^2} - \mu_p \frac{\partial (pE)}{\partial x} + g_p - \frac{p}{\tau_{pt}}$. If a steady-state condition is reached in a strictly field-free region ($E=0$) with no external optical generation ($g_p=0$), what mathematical form must the carrier concentration profile $p(x)$ fundamentally take?

Step-by-Step Solution

Correct Answer: B.
  1. Apply steady-state condition: $\frac{\partial p}{\partial t} = 0$.
  2. Apply field-free condition ($E=0$): The entire drift term $-\mu_p \frac{\partial (pE)}{\partial x}$ evaluates to zero.
  3. Apply no external generation: $g_p = 0$.
  4. The master equation simplifies drastically to: $0 = D_p \frac{\partial^2 p}{\partial x^2} - \frac{p}{\tau_{pt}}$.
  5. Rearrange into a standard 2nd-order ODE: $\frac{\partial^2 p}{\partial x^2} = \frac{p}{D_p \tau_{pt}}$.
  6. The mathematical solution to a differential equation where the second derivative of a function equals a positive constant times the function itself is a combination of exponentials: $p(x) = A e^{x/L_p} + B e^{-x/L_p}$ (where $L_p = \sqrt{D_p \tau_{pt}}$).
13

Continuity Equation Imbalances

Consider a specific 1D differential volume where the local optical generation rate $g_p$ is precisely equal to the local recombination rate $p/\tau_{pt}$. However, the local hole concentration $p(t)$ inside the volume is still observed to be rapidly dropping. What must be true about the spatial flux profile $F_p(x)$ across this volume?

Step-by-Step Solution

Correct Answer: B.
  1. The continuity equation is $\frac{\partial p}{\partial t} = -\frac{\partial F_p}{\partial x} + g_p - \frac{p}{\tau_{pt}}$.
  2. The problem states $g_p = p/\tau_{pt}$, so the generation and recombination terms perfectly cancel each other out.
  3. The equation reduces to: $\frac{\partial p}{\partial t} = -\frac{\partial F_p}{\partial x}$.
  4. The problem states that concentration is dropping, meaning $\frac{\partial p}{\partial t}$ is negative.
  5. For $-\frac{\partial F_p}{\partial x}$ to be a negative value, the derivative itself ($\frac{\partial F_p}{\partial x}$) must be a positive value.
  6. Physically, a positive flux derivative means more carriers are flowing out of the far boundary than are flowing into the near boundary, leading to the observed depletion.
14

Continuity Equation Lifetimes

When utilizing the simplified time-dependent diffusion equations, why are only the minority carrier lifetimes ($\tau_{n0}$ or $\tau_{p0}$) utilized in the recombination term $-\frac{p}{\tau}$, rather than the majority carrier lifetimes?

Explanation

Correct Answer: C. Recombination requires one electron and one hole to find each other. In a p-type material, there are trillions of holes everywhere, so an excess electron will find a hole almost instantly. The limiting factor determining the rate of the reaction is strictly the survival time of the scarce minority electrons ($\tau_{n0}$). The hole lifetime doesn't matter because the hole population doesn't realistically deplete.
15

The Physics of Ambipolar Transport

Use the interactive visualizer below to toggle internal ambipolar forces. When a localized pulse of excess electrons and holes is injected into a semiconductor under an applied electric field, why do they drift as a single unified packet rather than splitting entirely in opposite directions?

Explanation

Correct Answer: B. The external field naturally tries to pull electrons one way and holes the other. However, as soon as these huge clusters of charge begin to separate even slightly, an immense internal electrostatic field (Coulomb attraction) develops between them. This internal field ($E_{int}$) acts exactly opposite to the separation, acting like a tight rubber band binding the electron and hole clouds together into a single, unified traveling wave.
16

Simplifying Ambipolar Mobility

The full ambipolar mobility equation is $\mu' = \frac{\mu_n \mu_p (p - n)}{\mu_n n + \mu_p p}$. In a heavily extrinsic n-type semiconductor ($n_0 \gg p_0$) operating under low-level injection, this equation collapses to a simplified form. What is this form, and what does its sign signify physically about the unified packet?

Step-by-Step Solution

Correct Answer: C.
  1. Identify conditions: n-type means $n \gg p$. Low-level injection means $\delta n$ doesn't alter this dominance.
  2. Substitute the dominance into the numerator: $(p - n) \approx -n$.
  3. Substitute the dominance into the denominator: $(\mu_n n + \mu_p p) \approx \mu_n n$.
  4. Evaluate the simplified equation: $\mu' \approx \frac{\mu_n \mu_p (-n)}{\mu_n n}$.
  5. Cancel the $\mu_n n$ terms: $\mu' \approx -\mu_p$.
  6. Interpret the physics: The negative sign in the electron-based transport equation indicates that the unified electron-hole packet is being dragged along precisely at the drift velocity of the minority carriers (holes), which move in the same direction as the applied field.
17

Ambipolar Coefficient Intrinsic Limits

Evaluate the full ambipolar diffusion coefficient equation $D' = \frac{\mu_n n D_p + \mu_p p D_n}{\mu_n n + \mu_p p}$ for a perfectly intrinsic semiconductor where $n = p = n_i$. By substituting the Einstein relation ($\mu \propto D$), to what specific mathematical form does $D'$ ultimately simplify?

Derivation

Correct Answer: C.
1. Substitute $n = p = n_i$ into the equation: $D' = \frac{\mu_n n_i D_p + \mu_p n_i D_n}{\mu_n n_i + \mu_p n_i}$.
2. Factor out and cancel $n_i$: $D' = \frac{\mu_n D_p + \mu_p D_n}{\mu_n + \mu_p}$.
3. Use the Einstein relation $\mu = \frac{e}{kT}D$ to replace all mobilities: $D' = \frac{(\frac{e}{kT}D_n) D_p + (\frac{e}{kT}D_p) D_n}{(\frac{e}{kT}D_n) + (\frac{e}{kT}D_p)}$.
4. Factor out and cancel the $\frac{e}{kT}$ terms from numerator and denominator.
5. The numerator becomes $D_n D_p + D_p D_n = 2 D_n D_p$.
6. The denominator becomes $D_n + D_p$.
7. Final form: $D' = \frac{2 D_n D_p}{D_n + D_p}$.
18

Intrinsic Ambipolar Mobility

Consider an absolutely pure, intrinsic semiconductor operating in thermal equilibrium where $n = p = n_i$. If a tiny pulse of excess carriers is injected, what happens mathematically and physically to the ambipolar mobility $\mu' = \frac{\mu_n \mu_p (p - n)}{\mu_n n + \mu_p p}$?

Step-by-Step Solution

Correct Answer: C.
  1. Identify the specific condition: For an intrinsic semiconductor, $n = p$.
  2. Examine the numerator of the ambipolar mobility equation: $\mu_n \mu_p (p - n)$.
  3. Substitute the condition $p = n$ into the binomial: $(p - p) = 0$.
  4. The entire numerator collapses to zero, making $\mu' = 0$.
  5. Physical Interpretation: In an intrinsic material, there is a perfect balance of electron and hole concentrations. The applied electric field pulls the equal populations with equal and opposite force. The resulting internal ambipolar binding field perfectly cancels the external field, resulting in a unified packet that cannot drift at all (it can only diffuse).
19

Ambipolar Electric Field Sourcing

The internal ambipolar electric field $\mathcal{E}_{int}$ that holds the drifting electron-hole packet together is derived by setting the electron and hole diffusion currents mathematically equal. What is the fundamental physical source of the energy that establishes this powerful internal electrostatic field?

Explanation

Correct Answer: C. When the laser injects a neutral pulse of electrons and holes, the applied field immediately attempts to rip them in opposite directions. The electrons drift slightly upstream, and the holes drift slightly downstream. This tiny fractional separation of opposite charges creates an internal capacitor-like structure (a microscopic dipole). The energy comes from the work the external field did to slightly separate them. This separation generates an intense internal Coulomb field ($\mathcal{E}_{int}$) that perfectly balances the external force, preventing them from splitting further.
20

Microscopic Diffusion Derivation

In the derivation of the macroscopic diffusion current density $J_{nx|dif} = e D_n \frac{dn}{dx}$, the net electron flow crossing the $x=0$ plane is evaluated by comparing the carrier concentration at $x = -l$ and $x = +l$. What physical parameter does the length $l$ specifically represent in this microscopic model?

Explanation

Correct Answer: C. The derivation assumes that carriers crossing the $x=0$ plane last scattered at a distance $l$ away. This distance is the mean free path, which is the average distance a carrier travels between collisions in the lattice, calculated dynamically by $v_{th}\tau_{cn}$.
21

Continuity Equation & Uniform Injection

According to the 1D continuity equation $\frac{\partial p}{\partial t} = -\frac{\partial F_p}{\partial x} + g_p - \frac{p}{\tau_{pt}}$, what happens to the spatial derivative flux term $-\frac{\partial F_p}{\partial x}$ if a massive laser flash creates a strictly uniform, steady-state excess carrier profile across the entire infinite semiconductor?

Step-by-Step Solution

Correct Answer: A.
  1. If the excess carrier profile is perfectly uniform, there is no spatial concentration gradient ($\frac{dp}{dx} = 0$).
  2. With no gradient, there is no net diffusion flux anywhere.
  3. Consequently, the difference in flux between any $x$ and $x+dx$ is precisely zero.
  4. The entire spatial derivative term drops out, and the continuity equation simplifies strictly to balancing generation and recombination over time.
22

Uniform Doping and the Built-in Field

In a non-uniformly doped sample at thermal equilibrium, the built-in electric field is mathematically linked to the spatial derivative of the intrinsic Fermi level: $E_x = \frac{1}{e}\frac{dE_{Fi}}{dx}$. If the semiconductor instead has a perfectly uniform (constant) donor doping profile $N_d(x) = N_d$, what must be the geometric slope of the intrinsic Fermi level $E_{Fi}$ on the band diagram?

Step-by-Step Solution

Correct Answer: B.
  1. A uniform doping profile $N_d(x)$ means there is no concentration gradient.
  2. Without a gradient, there is no initial diffusion of carriers.
  3. With no diffusion, no built-in electric field ($E_x = 0$) is required or generated to oppose it.
  4. If $E_x = 0$, the spatial derivative $\frac{dE_{Fi}}{dx}$ must identically be zero, meaning the energy bands, including $E_{Fi}$, are perfectly horizontal.
23

Ambipolar Diffusion Physics

Under low-level injection in a heavily doped p-type material, the complex ambipolar diffusion coefficient $D'$ simplifies completely to $D_n$, the minority carrier diffusion coefficient. What physical reality does this mathematical simplification represent?

Step-by-Step Solution

Correct Answer: C. In a heavily doped material, the majority carriers are so abundant that they require almost zero effort to slightly rearrange and electrostatically screen the minority carriers. Because the majority holes can respond almost instantly, the overall movement of the bound ambipolar packet is restricted entirely by the slower, scarce minority electrons diffusing down their concentration gradient.
24

Conductivity and Majority Dominance

The macroscopic conductivity of a semiconductor is given by $\sigma = e(\mu_n n + \mu_p p)$. In a highly doped n-type semiconductor ($N_d \gg n_i$), why is the hole contribution to the total macroscopic drift conductivity generally ignored in practical engineering calculations?

Step-by-Step Solution

Correct Answer: B.
  1. In an n-type semiconductor, $n \approx N_d$.
  2. Because thermal equilibrium demands $n \cdot p = n_i^2$, a large $n$ forces $p$ to become extremely small (e.g., $10^{15} \times 10^5 = 10^{20}$).
  3. Because the hole concentration is billions of times smaller than the electron concentration, its term in the conductivity equation ($\mu_p p$) contributes virtually nothing to the total sum and is safely ignored.
25

Thermodynamic Scaling of Einstein Relation

The Einstein relation mandates that $\frac{D_n}{\mu_n} = \frac{kT}{e}$. If the operating temperature of an active semiconductor is doubled from 300 K to 600 K, and assuming the mobility $\mu_n$ decreases slightly due to increased phonon scattering at the higher temperature, what must happen to the diffusion coefficient $D_n$ to satisfy this rigid thermodynamic requirement?

Step-by-Step Solution

Correct Answer: C.
  1. The ratio of $D_n$ to $\mu_n$ must equal $kT/e$. If $T$ doubles, the ratio $D_n/\mu_n$ must strictly double.
  2. If the denominator $\mu_n$ shrinks, the numerator $D_n$ must grow even more substantially to ensure the overall ratio still doubles.
  3. Physically, at much higher temperatures, the massive increase in random thermal energy dramatically boosts the carriers' ability to spread out (diffuse).