Drift Velocity & the Microscopic Picture

Semiconductor engineers live and die by the two knobs in $I = nAv_d e$. In a silicon chip, doping sets $n$ atom by atom, and mobility engineering — straining the crystal, choosing materials like gallium arsenide where electrons scatter less — sets how much drift each volt buys. When a datasheet boasts about "high-mobility channels," it is talking about equation (3) with better $\tau$. The drift picture also explains a real hazard in chip design called electromigration: in interconnects thinner than a bacterium, the drifting electron wind literally blows metal atoms out of position over years of operation, and reliability teams model it with carrier-counting arguments one step removed from this article's derivation. At the other extreme of scale, the electron beams in particle accelerators and the ion drifts in battery electrolytes obey the same census: current = density × area × speed × charge. Whenever charge moves — in copper, in silicon, in a lithium cell, in a lightning channel — someone is applying $I = nAv_d e$ with the appropriate cast of carriers.

Picture a shopping mall on a crowded evening. Everyone is walking briskly — fast, random, colliding, changing direction. Photograph the crowd twice, a second apart, and the crowd as a whole has gone nowhere: for every person walking left there is one walking right. That is a metal with no battery attached. The free electrons dart about at around $10^6\ \mathrm{m/s}$ — a thousand times faster than a rifle bullet — but the velocities point every which way, the average is zero, and the ammeter reads zero. Now let a gentle breeze blow through the mall, nudging every walker slightly toward the exit. Nobody stops zig-zagging; the collisions continue exactly as before. But each person's path now tilts, imperceptibly, exit-ward, and the crowd acquires a slow collective creep. That breeze is the electric field the battery sets up inside the wire; the creep is the drift velocity $v_d$. It is not the speed of any individual electron — it is the average velocity of the whole sea, the tiny ordered whisper on top of a thermal hurricane. In a 1 mm copper wire carrying 2 A the whisper is about $0.19\ \mathrm{mm/s}$: the thermal motion outpaces the drift by a factor of several billion.

Why then is the lamp instant? Because you never wait for a particular electron. The moment the switch closes, the battery's push propagates along the wire as an electromagnetic signal — the field organises itself around the whole loop in nanoseconds, guided by the wire itself. Every electron already inside the lamp filament feels the field at essentially the same moment as every electron near the switch, and all of them begin to drift together. Think of a goods train a kilometre long: the locomotive tugs, and the last wagon jerks forward almost immediately, even though no wagon travels fast. The wire was already full; the order to march travels at nearly $c$; the marchers themselves crawl. One bookkeeping note before the algebra, inherited from Electric Current & Charge Flow: conventional current points the way positive charge would move. The electrons drift toward the battery's positive terminal, so the current arrow points the opposite way, out of the positive terminal and around the loop. Nothing in this article changes that convention — we simply compute how fast the crowd actually shuffles.

The bridge between the microscopic crawl and the ammeter reading is a counting argument. Before deriving it, meet the cast. The carrier number density $n$ counts mobile electrons per cubic metre — for copper, which donates about one free electron per atom, $n = 8.5 \times 10^{28}\ \mathrm{m^{-3}}$ (that number is worth memorising; JEE and NEET hand it to you, boards expect you to use it fluently). The wire's cross-section has area $A$, each carrier has charge $e = 1.6\times10^{-19}\ \mathrm{C}$, and the whole sea drifts at $v_d$. The claim:

where: $I$ — the current in amperes; $n$ — free-electron density in $\mathrm{m^{-3}}$; $A$ — cross-sectional area in $\mathrm{m^2}$; $v_d$ — drift speed in $\mathrm{m/s}$; $e$ — the elementary charge. Read the structure: current is how many ($nA$ per metre of wire) times how fast ($v_d$) times how much charge each carries ($e$). Because $n$ is astronomical, $v_d$ can be — must be — tiny for any household current. That inverse relationship is the whole personality of the equation.

Equation (1) rearranges into the form exams actually use — solving for the drift:

where: everything is as before, and the message lives in the denominator. For one metre of 1 mm² copper wire, $nAe \approx 1.4\times10^4$ coulombs of mobile charge — fourteen thousand coulombs already in place per metre. When the corridor holds that much charge, shipping 2 coulombs per second requires only a crawl. This is also your sanity check in any exam: a drift speed that comes out faster than walking pace means a slipped power of ten, almost always in $A$ (mm² to m² is $10^{-6}$, not $10^{-3}$).

Different materials respond to the same field with different drifts. The quantity that captures the response is mobility:

where: $\mu$ — mobility, in $\mathrm{m^2\,V^{-1}\,s^{-1}}$; $E$ — the electric field inside the conductor, in $\mathrm{V/m}$; $v_d$ — the drift speed the field produces. Mobility is "millimetres per second of drift purchased per volt-per-metre of push." Copper's electrons have $\mu \approx 4.3\times10^{-3}\ \mathrm{m^2V^{-1}s^{-1}}$; semiconductor electrons often beat that by orders of magnitude — but semiconductors have vastly fewer carriers, and it is the product $n\mu$ that decides conduction. Substituting $v_d = \mu E$ into equation (1) gives $I = nAe\mu E$: current proportional to field, at fixed temperature — which is exactly the microscopic seed of Ohm's law, harvested properly in Ohm's Law & Resistance. Why is the drift proportional to the field at all — why doesn't the electron just accelerate indefinitely? Here is the collision picture, in words, and it is worth telling well because it explains where resistance comes from. Between collisions, the field accelerates an electron smoothly, adding a little velocity along the wire. Then the electron slams into a vibrating lattice ion and the gained velocity is scrambled away — direction randomised, the ordered progress reset to zero. The field rebuilds; the next collision erases. The average time between collisions, written $\tau$ and called the relaxation time, is about $10^{-14}$ seconds in copper. The drift velocity is simply the velocity gained in one relaxation time — acceleration $eE/m$ sustained for time $\tau$:

where: $m$ — the electron mass, $9.11\times10^{-31}\ \mathrm{kg}$; $\tau$ — the average time between collisions. Read it physically: the drift is what a free-fall of duration $\tau$ under the electrical "gravity" $eE/m$ delivers before a collision wipes the slate. Double the field, double the drift — proportionality, hence mobility $\mu = e\tau/m$ is a material constant at fixed temperature. Heat the wire and the ions jiggle harder, collisions come sooner, $\tau$ shrinks — drift and mobility fall. One algebra step further (dividing the push by the response) shows the material's resistivity is $\rho = m/(n e^2 \tau)$ — few carriers or frequent collisions mean stubborn material. That formula is stated here for the story, not for calculation: your exams want the picture (more collisions → more resistance → why heating raises a metal's resistance) and the proportionalities, and the full macroscopic treatment belongs to Ohm's Law & Resistance and, later, Resistivity & Geometry.

The examiner's favourite ambush in this topic is to blur three utterly different speeds that coexist in the same piece of copper. Keep them filed separately:

Ten orders of magnitude separate the drift from the signal. "The electrons rush from the switch to the bulb" is wrong twice over: they crawl, and the bulb never waits for them anyway.

Board level — the classic drift-speed calculation (CBSE/ISC pattern, 3 marks)

A copper wire of diameter $1.0\ \mathrm{mm}$ carries a steady current of $2.0\ \mathrm{A}$. The free-electron density of copper is $8.5 \times 10^{28}\ \mathrm{m^{-3}}$. Calculate the drift velocity of the electrons, and the time an electron would take to drift one metre along the wire.

Step 1. Cross-sectional area from the diameter (radius $0.5\ \mathrm{mm} = 5\times10^{-4}\ \mathrm{m}$): $A = \pi r^2 = \pi (5\times10^{-4})^2 = 7.85\times10^{-7}\ \mathrm{m^2}$.

Step 2. Apply equation (2): $v_d = \dfrac{I}{nAe} = \dfrac{2.0}{(8.5\times10^{28})(7.85\times10^{-7})(1.6\times10^{-19})}$.

Step 3. Evaluate the denominator first: $(8.5\times10^{28})(7.85\times10^{-7})(1.6\times10^{-19}) = 1.068\times10^{4}$. Then $v_d = 2.0 / 1.068\times10^4 = \boxed{1.9\times10^{-4}\ \mathrm{m/s}}$ — about $0.19$ millimetres per second.

Step 4. Time for one metre at that crawl: $t = \dfrac{1}{v_d} = \dfrac{1}{1.9\times10^{-4}} \approx 5340\ \mathrm{s} \approx \boxed{89\ \text{minutes}}$. State the moral for the mark: the lamp lights instantly because the field, not any individual electron, carries the signal.

Competitive level — drift, mobility and the narrowing wire (JEE-Main/NEET pattern, ~3 minutes)

A copper wire of cross-section $2.0\ \mathrm{mm^2}$ carries $3.2\ \mathrm{A}$. Take $n = 8.5\times10^{28}\ \mathrm{m^{-3}}$, $e = 1.6\times10^{-19}\ \mathrm{C}$. (a) Find the drift speed. (b) The field inside the wire is $0.05\ \mathrm{V/m}$; find the electron mobility. (c) The wire then narrows to half its area while carrying the same current — what happens to $v_d$ there? (d) How many electrons cross a section each second?

Step 1. Part (a): $v_d = \dfrac{I}{nAe} = \dfrac{3.2}{(8.5\times10^{28})(2.0\times10^{-6})(1.6\times10^{-19})} = \dfrac{3.2}{2.72\times10^{4}} = \boxed{1.2\times10^{-4}\ \mathrm{m/s}}$.

Step 2. Part (b): mobility is drift per unit field, equation (3): $\mu = \dfrac{v_d}{E} = \dfrac{1.18\times10^{-4}}{0.05} = \boxed{2.4\times10^{-3}\ \mathrm{m^2V^{-1}s^{-1}}}$.

Step 3. Part (c): the current is the same everywhere along one wire (charge does not pile up), so $nAv_d e$ is constant. Halve $A$ and $v_d$ doubles to $2.4\times10^{-4}\ \mathrm{m/s}$ — same crowd flux through a tighter corridor. This inverse $v_d$–$A$ link inside a single wire is a favourite one-mark twist.

Step 4. Part (d): electrons per second $= \dfrac{I}{e} = \dfrac{3.2}{1.6\times10^{-19}} = \boxed{2.0\times10^{19}\ \text{electrons per second}}$ — twenty billion billion per second, delivered by a crawl. Timing note: budget three minutes. Parts (a) and (d) are pure substitution — the marks are decided by powers of ten (mm² → $\times10^{-6}$). Part (c) needs no calculation at all if you see that $I$ fixed means $v_d \propto 1/A$; spotting that saves a full minute.

The mm² conversion. $1\ \mathrm{mm^2} = 10^{-6}\ \mathrm{m^2}$, never $10^{-3}$. Nearly every wrong drift-velocity answer on record is this slip; it announces itself with a drift speed a thousand times too big. Sanity gate: metallic drift speeds land between $10^{-5}$ and $10^{-3}\ \mathrm{m/s}$ for ordinary currents.

Diameter given, radius needed. "Wire of diameter 1 mm" — halve it before squaring. A factor of 4 in $A$, and a distractor option is waiting for exactly that error.

Three speeds blurred. Thermal $\sim10^6$, drift $\sim10^{-4}$, signal $\sim10^8\ \mathrm{m/s}$. An MCQ offering "the electrons move at nearly the speed of light" is testing whether you'll grab the signal speed for the carriers. Never write that the electrons are fast.

Drift "increasing along a thin wire" confusion. In a single series wire, $I$ is identical at every cross-section; only $v_d$ adjusts when $A$ changes. If a question shows a tapered conductor, the current does not change along it — the drift does.

$n$ is per cubic metre. If a problem quotes $n$ in $\mathrm{cm^{-3}}$ (some older ISC papers do), multiply by $10^6$ to reach $\mathrm{m^{-3}}$ before substituting.

Constants and conventions. All boards here use $e = 1.6\times10^{-19}\ \mathrm{C}$; copper's $n = 8.5\times10^{28}\ \mathrm{m^{-3}}$ is supplied in JEE/NEET stems but expected as a known order of magnitude in board theory answers. IB and A-Level papers write charge carriers generally (electrons in metals, ions in electrolytes) — say "free electrons in a metal," not just "charges," in definition answers.

Texts. NCERT Physics Class 12, Ch. 3 (Current Electricity, drift velocity section) · HC Verma, Concepts of Physics Vol. 2, Ch. 32 · CIE 9702 Topic 9.2 (drift velocity as an explicit learning outcome) · IB Physics Theme B.5.

Provenance. All results are standard. A verifier should recompute: $n_{\text{Cu}} = 8960 \times 6.022\times10^{23}/0.0635 = 8.5\times10^{28}\ \mathrm{m^{-3}}$; Example 1's $A = 7.85\times10^{-7}\ \mathrm{m^2}$, $v_d = 1.9\times10^{-4}\ \mathrm{m/s}$, $t \approx 89$ min; Example 2's $v_d = 1.2\times10^{-4}\ \mathrm{m/s}$, $\mu = 2.4\times10^{-3}\ \mathrm{m^2V^{-1}s^{-1}}$, $2.0\times10^{19}$ electrons/s; copper's $\mu = 1/(ne\rho) = 4.3\times10^{-3}\ \mathrm{m^2V^{-1}s^{-1}}$ and $\tau = m/(ne^2\rho) = 2.5\times10^{-14}\ \mathrm{s}$ at $\rho = 1.7\times10^{-8}\ \Omega\mathrm{m}$.

Prerequisites. Electric Current & Charge Flow. Leads to. Ohm's Law & Resistance (where $v_d \propto E$ becomes $V = IR$) · Resistivity & Geometry (coming in E3) · Current, Resistance and Resistivity (UG deep dive).