Electron Emission & Work Function

The humble act of coaxing an electron out of a metal is still, a century after Thomson, the beating heart of an enormous range of technology. Every electron microscope begins with an electron source, either a hot thermionic filament or a cold field-emission tip sharpened to a few atoms. The X-ray tubes in every hospital and airport scanner boil electrons off a heated cathode and slam them into a metal target. Field emission, the coldest of the four routes, drives scanning tunnelling microscopes, and the same tunnelling-through-a-thin-barrier physics underlies flash memory. Photomultiplier tubes ride on secondary emission to count single photons from distant galaxies or the faint flashes of a particle detector. Even the choice of caesium and other low-work-function coatings for photocathodes — the light-sensitive surfaces in night-vision devices — is a direct application of the idea that a lower $\phi$ means an easier escape. And conceptually, this article is the gateway: fix in your mind that electrons are trapped behind a surface barrier of height $\phi$, and the photoelectric effect, Einstein's equation, and the whole quantum story that follows become a series of increasingly clever ways to pay that one bill.

Picture a block of copper at the level of its atoms. The atoms sit locked in a regular lattice, but each has surrendered roughly one of its outer electrons to the collective. Those surrendered electrons are no longer attached to any single atom; they wander through the whole crystal like molecules of a gas trapped in a box. This is the free-electron sea — the mental model that explains at a stroke why metals conduct electricity, conduct heat, and gleam. Count them and the numbers are staggering: in copper the density of these mobile conduction electrons is about $8.5\times10^{28}$ per cubic metre, close to one free electron for every atom. A pinhead of metal contains more free electrons than there are stars in the observable universe. So if a metal is brimming with electrons that are already free to move, why don't they simply spill out of the surface the way water spills over the rim of a full glass? Because "free to move inside" is not the same as "free to leave." An electron deep in the metal feels the pull of all the positive ions around it in every direction, and those pulls cancel — it drifts easily. But an electron right at the surface feels ions only on the inward side; there is nothing but vacuum outside. The net inward tug builds an invisible wall at the surface, a potential-energy barrier. To an electron, leaving the metal is like a ball trying to roll out of a bowl: it can slosh around the bottom freely, but to escape over the rim it needs a definite, minimum amount of extra energy. That minimum climb, measured from the most energetic electrons already present up to the "lip" of the barrier (the energy of a free electron just outside, which we call the vacuum level), is the work function $\phi$.

The height of that rim depends only on the metal and the state of its surface — not on how big the block is, not on how the electron is eventually kicked out. Caesium, an alkali metal that gives up its outer electron with barely a fight, has one of the lowest work functions known, $\phi = 2.14\ \mathrm{eV}$. Sodium is close behind at $2.28\ \mathrm{eV}$. Refractory tungsten, the stuff of lamp filaments, holds on much harder at $4.5\ \mathrm{eV}$, and platinum harder still at about $5.65\ \mathrm{eV}$. That single number — the ransom in energy — is the hero of this entire chapter. Everything that follows is a story about four different ways to pay it.

Before we can weigh these ransoms we need a sensible unit. A joule is hopelessly clumsy here: a work function of a couple of $10^{-19}$ joules is not a number anyone wants to juggle. Physics solves this with a unit tailored to single charged particles, the electronvolt (eV), defined as the kinetic energy an electron picks up when it is accelerated through a potential difference of exactly one volt. Since the work done on a charge $q$ moved through a voltage $V$ is $W = qV$, and the electron's charge magnitude is $e = 1.60\times10^{-19}\ \mathrm{C}$, one electronvolt is:

where: $e$ is the elementary charge and the "$1\ \mathrm{V}$" is the accelerating voltage. The physical reading is that the electronvolt is not some abstract conversion factor — it is a piece of laboratory apparatus made into a unit. Drop an electron across one volt and you have handed it exactly one electronvolt of energy. This is why atomic and particle physicists speak almost entirely in eV, keV, MeV and GeV, and it is why every work function in this chapter is quoted in electronvolts. Converting is one multiplication: caesium's $\phi = 2.14\ \mathrm{eV}$ is

a genuinely tiny energy — about the amount a single blue-light photon carries, which is the first hint of why light will turn out to be so good at freeing electrons.

An electron sitting just below the rim needs only a little extra energy — at most a few eV — to get over it. Nature offers four distinct suppliers of that energy, and every vacuum-electronics device ever built runs on one of them.

Thermionic emission — pay with heat. Heat the metal. The conduction electrons share in the thermal jostling, and the hotter the metal, the faster the fastest of them move. At everyday temperatures the average electron has nowhere near $\phi$ of thermal energy, but push the temperature high enough — a tungsten filament runs near $2500\ \mathrm{K}$ — and a small but useful fraction of electrons in the high-energy tail of the distribution finally have enough to clear the barrier and boil off. This is exactly how the electron gun in an old television tube, an X-ray machine, or an electron microscope works: a hot filament sweats electrons into the vacuum, ready to be steered.

Photoelectric emission — pay with light. Shine light of high enough frequency on the surface and each absorbed photon can hand a single electron a packet of energy in one instant. If that packet is at least $\phi$, the electron escapes — no heating required. This is the emission mechanism that broke classical physics wide open, and it gets its own full article next. Here we only note that it is one photon, one electron, and that the photon's energy must clear the work-function bar.

Field (cold) emission — pay with a fierce electric field. Apply an enormous electric field at the surface — of order $10^{9}\ \mathrm{V\,m^{-1}}$, a billion volts per metre — and the field tilts the barrier so steeply that it becomes extremely thin, only a nanometre or so wide. Electrons then leak straight through it by quantum tunnelling, cold, with no heating at all. This is how a scanning tunnelling microscope and a field-emission display pull electrons from a sharp tip at room temperature.

Secondary emission — pay with impact. Fire a fast electron (or ion) at the surface and it can knock several more electrons loose on impact, like a cue ball scattering a rack. A single incoming electron often ejects two, three, or more secondaries. Photomultiplier tubes exploit this to turn one photoelectron into a measurable avalanche, and it is the effect that lets a single photon of starlight be counted.

Notice the common thread. In every case the metal supplies the electrons and the barrier sets the price; the four mechanisms differ only in who pays the bill and how. That is why the work function $\phi$ — a single property of the surface — is the organising idea of the whole chapter. Once you can say how big $\phi$ is and where the energy to beat it comes from, the photoelectric effect, Einstein's equation, and everything after are refinements of this one accounting.

It is worth pausing on thermionic emission because the numbers reveal something surprising: heat is a shockingly inefficient way to free electrons, and understanding why explains the whole design of a vacuum tube. The natural energy scale of thermal motion is $k T$, where $k = 1.38\times10^{-23}\ \mathrm{J\,K^{-1}}$ is Boltzmann's constant. At room temperature ($T = 300\ \mathrm{K}$):

where: $k$ is Boltzmann's constant and $T$ the absolute temperature. The physical reading is stark: the thermal energy sloshing around at room temperature is about $0.026\ \mathrm{eV}$, while the barrier it must beat is $2$–$5\ \mathrm{eV}$ — roughly a hundred times larger. No wonder a warm wire emits essentially nothing; on average an electron is a hundredfold short of the fare. Crank the tungsten filament up to $T = 2500\ \mathrm{K}$ and the average thermal energy rises only to about $0.32\ \mathrm{eV}$ (its average kinetic energy $\tfrac{3}{2}kT$), still more than ten times below tungsten's $4.5\ \mathrm{eV}$ barrier. Emission works at all only because a few electrons in the far tail of the speed distribution vastly exceed the average — and the number in that tail climbs steeply with temperature. That steep climb is captured by Richardson's law, which we quote (its derivation needs statistical mechanics beyond this article):

where: $J$ is the emitted current per unit area, $A$ a material constant, and the exponential factor $e^{-\phi/kT}$ is the fraction of electrons energetic enough to escape. Everything hinges on that exponential. This is not a formula you derive or memorise for a school exam — it is here to show you why thermionic devices must run white-hot, and why a small temperature rise dramatically increases the current.

Board level — converting work functions and reading the escape condition (CBSE and ICSE pattern)

The work function of tungsten is $4.5\ \mathrm{eV}$. (a) Express this in joules. (b) A caesium surface has $\phi = 2.14\ \mathrm{eV}$; express it in joules. (c) A packet of energy of $2.5\ \mathrm{eV}$ is delivered to a single electron in each metal. From which metal, if either, can the electron escape? Take $1\ \mathrm{eV} = 1.60\times10^{-19}\ \mathrm{J}$.

Step 1. Convert tungsten's work function: $\phi_{\mathrm W} = 4.5 \times 1.60\times10^{-19} = \boxed{7.20\times10^{-19}\ \mathrm{J}}$. Multiplying the eV value by the number of joules in one eV is the whole operation.

Step 2. Convert caesium's: $\phi_{\mathrm{Cs}} = 2.14 \times 1.60\times10^{-19} = \boxed{3.42\times10^{-19}\ \mathrm{J}}$. Caesium's barrier is roughly half tungsten's, as its much lower eV value already told you.

Step 3. Apply the escape condition, "energy supplied must be at least $\phi$." The packet is $2.5\ \mathrm{eV}$. For caesium, $2.5\ \mathrm{eV} > 2.14\ \mathrm{eV}$, so the electron escapes (with $2.5 - 2.14 = 0.36\ \mathrm{eV}$ to spare as kinetic energy). For tungsten, $2.5\ \mathrm{eV} < 4.5\ \mathrm{eV}$, so the electron cannot escape; it falls back into the sea.

Step 4. State the conclusion cleanly: the same $2.5\ \mathrm{eV}$ frees an electron from caesium but not from tungsten, purely because the two metals present different barriers. The supplier of energy is irrelevant to this comparison — only $\phi$ and the size of the packet matter.

Competitive level — the free-electron count and the thermal shortfall (JEE Main and NEET pattern)

(a) Estimate the number density of free electrons in copper, given density $\rho = 8960\ \mathrm{kg\,m^{-3}}$, molar mass $M = 63.5\ \mathrm{g\,mol^{-1}}$, one free electron per atom, and Avogadro's number $N_A = 6.02\times10^{23}\ \mathrm{mol^{-1}}$. (b) Compare copper's $2.5$-eV-scale barrier against the average thermal energy of an electron at $300\ \mathrm{K}$ and state, in one line, why copper does not emit electrons at room temperature.

Step 1. Number density from one electron per atom: the atoms per cubic metre equal $\rho N_A / M$. Put $M$ in kilograms: $M = 63.5\times10^{-3}\ \mathrm{kg\,mol^{-1}}$. Then $n = \dfrac{\rho N_A}{M} = \dfrac{(8960)(6.02\times10^{23})}{63.5\times10^{-3}}$.

Step 2. Evaluate: $\rho/M = 8960 / 0.0635 = 1.411\times10^{5}\ \mathrm{mol\,m^{-3}}$; multiply by $N_A$: $n = 1.411\times10^{5}\times6.02\times10^{23} = \boxed{8.5\times10^{28}\ \mathrm{m^{-3}}}$. Copper carries about $8.5\times10^{28}$ free electrons per cubic metre — one of the reasons it is such a good conductor.

Step 3. Thermal energy at $300\ \mathrm{K}$: average kinetic energy $\tfrac32 kT = \tfrac32(1.38\times10^{-23})(300) = 6.21\times10^{-21}\ \mathrm{J} = 0.039\ \mathrm{eV}$. The barrier is of order $4$–$5\ \mathrm{eV}$ for copper, so the ratio is $\sim100$.

Step 4. One-line conclusion: with the average electron carrying only about $0.039\ \mathrm{eV}$ against a barrier around a hundred times higher, thermal emission at room temperature is utterly negligible — copper must be heated to well over $1000\ \mathrm{K}$ before any measurable current boils off. Timing note: about 90 seconds. In an MCQ you never finish the density arithmetic — recognising that $n \approx \rho N_A/M \sim 10^{29}\ \mathrm{m^{-3}}$ and that $\tfrac32 kT$ at room temperature is a few hundredths of an eV, a hundredfold below any metal's $\phi$, is enough to pick the answer and move on.

The work function belongs to the metal, full stop. Do not write it as a function of the incident light's frequency, the temperature, or the applied field. Those decide whether an electron escapes and how much energy it leaves with; they never change $\phi$ itself. A question that changes the light colour and asks for the new work function is testing exactly this — the answer is "unchanged."

Free inside is not free to leave. Metals conduct because their electrons roam freely, yet the same electrons are bound at the surface. State both facts; marks are lost by students who claim conduction electrons are "already free so no energy is needed to remove them."

Units: electronvolts, not joules. Work functions are a few eV, i.e. a few $\times10^{-19}\ \mathrm{J}$. When a problem gives $\phi$ in eV and asks for a joule quantity, convert with $1\ \mathrm{eV} = 1.60\times10^{-19}\ \mathrm{J}$ before combining it with SI quantities. Mixing eV and joules in one equation is the classic silent error.

No photoelectric equation here — yet. This article deliberately stops at "energy of at least $\phi$ frees an electron." The relation $KE_{\max} = hf - \phi$, the stopping potential, and the threshold frequency belong to the next two articles. If an exam question asks for maximum kinetic energy or stopping potential, it is testing Einstein's equation, not this one — do not invent it here.

Thermionic ≠ photoelectric. Both free electrons and both involve $\phi$, but the supplier differs: heat versus light. CBSE and NEET routinely ask you to name the mechanism from a one-line description ("electrons emitted from a heated filament" is thermionic; "from an illuminated surface" is photoelectric; "from a cold sharp tip in a strong field" is field emission). Learn the four names and their signatures.

Board scope. The work function is core descriptive material for CBSE, ICSE, JEE Main, NEET, and A-Level (CIE 9702) — every board that carries this topic. The four-way emission-types taxonomy (thermionic / photoelectric / field / secondary), however, is NCERT/Indian-board framing: it is a named exam objective only for CBSE, ICSE, JEE Main, and NEET (ex-NCERT Ch. 11). A-Level (CIE 9702), like AP Physics 2 and IB, does not itemise the four emission mechanisms as a named objective — only the work function appears there, inside the photoelectric context. IGCSE 0625 does not cover dual-nature physics at all. So do not spend A-Level, AP, or IB revision time memorising the four routes as an exam target, though the work-function idea itself reappears in each of those inside the photoelectric effect.

Texts. NCERT Physics Class 12, Chapter 11 ("Dual Nature of Radiation and Matter"), the opening sections on electron emission and work function · H. C. Verma, Concepts of Physics Vol. 2, "Photoelectric Effect and Wave–Particle Duality," the introductory discussion of emission · Halliday, Resnick and Walker, the photon chapter's treatment of the work function.

Provenance. Standard descriptive results with locked benchmark numbers. Recompute: $1\ \mathrm{eV} = 1.60\times10^{-19}\ \mathrm{J}$; $\phi$ in joules for Cs/Na/Ca/W/Pt $= 3.42/3.65/5.12/7.20/9.04\times10^{-19}\ \mathrm{J}$; Thomson's $e/m = 1.60\times10^{-19}/9.11\times10^{-31} = 1.76\times10^{11}\ \mathrm{C\,kg^{-1}}$; $kT = 0.026\ \mathrm{eV}$ and $\tfrac32 kT = 0.039\ \mathrm{eV}$ at $300\ \mathrm{K}$; $\tfrac32 kT = 0.32\ \mathrm{eV}$ at $2500\ \mathrm{K}$; barrier ratios $\approx116$ and $\approx13.9$ for tungsten; Boltzmann factor $e^{-20.9} \approx 8.6\times10^{-10}$; field-emission barrier width $d = \phi/(eE) = 4.5/(3\times10^{9}) = 1.5\times10^{-9}\ \mathrm{m}$; copper free-electron density $n = \rho N_A/M = 8.5\times10^{28}\ \mathrm{m^{-3}}$; $h = 4.14\times10^{-15}\ \mathrm{eV\,s}$ and $hc/e \approx 1240\ \mathrm{eV\,nm}$.

Prerequisites. The Nature of Electromagnetic Waves — light as an electromagnetic wave, the picture the photoelectric effect will overturn. Leads to. The Photoelectric Effect and Its Laws — where light of high enough frequency pays the work-function ransom, and classical wave theory fails to explain how. For the university seam, the statistical derivation of Richardson's law and the quantum theory of the metallic surface barrier are enrichment "go deeper" topics, not exam dependencies here.