Quantum Physics: From Glowing Metal to Entanglement
How a handful of puzzling experiments forced physicists to describe nature with waves of probability, and what that description can and cannot do.
Glowing hot objects
Start with something you have seen. Heat a metal rod in a flame and it glows dull red, then orange, then nearly white. The colour tells you the temperature, which is why astronomers can read the temperature of a star from its light. By the late eighteen hundreds physicists could measure this glow very precisely, as a curve of brightness against wavelength, for an ideal absorber and emitter called a blackbody. Then they tried to predict the curve using the best physics they had: Newton's mechanics, Maxwell's electromagnetism and the statistics of heat. The prediction failed badly. Classical theory said a hot object should pour out more and more energy at shorter and shorter wavelengths, without limit. That would mean infinite energy, which is plainly false. A warm oven does not blind you with ultraviolet light. This failure was called the ultraviolet catastrophe, and it is where quantum physics begins.
Planck's energy packets
In nineteen hundred Max Planck found a formula that matched the measured curve perfectly, but only by making a strange assumption. He supposed that the glowing material could give out or absorb light energy only in fixed lumps. Each lump has energy E equals h times f, where f is the frequency of the light and h is a tiny new constant, now called Planck's constant, about six point six two six times ten to the minus thirty four joule seconds. Why does this cure the catastrophe? High frequency lumps are expensive. At a given temperature the vibrating atoms rarely have enough thermal energy to pay for one, so short wavelengths are starved and the curve turns over. Think of a shop that only sells goods in whole boxes: with little money, you almost never buy the expensive ones. A fixed smallest lump like this is called a quantum.
Light knocks out electrons
Planck thought his lumps were a property of the material. In nineteen oh five Einstein argued that light itself comes in lumps, later called photons. The evidence is the photoelectric effect. Shine light on a clean metal surface and electrons fly out, but only if the frequency is above a threshold. Dim violet light ejects electrons immediately, while intensely bright red light ejects none at all. A pure wave picture cannot explain this, because a bright enough wave should eventually shake electrons loose. The photon picture explains it neatly. Each photon gives all of its energy, h f, to a single electron. The electron must spend a minimum amount, called the work function, phi, to escape the metal. Whatever is left over becomes kinetic energy. Brighter light means more photons and so more electrons, not faster ones. Higher frequency means faster electrons.
Worked example: photoelectric effect
Let us put numbers in. Violet light of wavelength four hundred nanometres falls on sodium, whose work function is about two point two eight electron volts. An electron volt is the energy an electron gains crossing one volt, about one point six times ten to the minus nineteen joules. A handy shortcut is that h times c, Planck's constant times the speed of light, is about one thousand two hundred and forty electron volt nanometres. So the photon energy is one thousand two hundred and forty divided by four hundred, which is three point one zero electron volts. Subtract the work function and the fastest electrons leave with about zero point eight two electron volts. To stop them you need a reverse voltage of about zero point eight two volts, called the stopping potential. Red light at seven hundred nanometres carries only one point seven seven electron volts: too little, so no electrons at all.
Atoms have energy levels
Next clue: heat hydrogen gas and pass its light through a prism. Instead of a smooth rainbow you see a few sharp coloured lines, the same lines every time. Each atom can only emit photons of particular energies, so it can only have particular energies itself. In nineteen thirteen Niels Bohr pictured the electron circling the proton only in certain allowed orbits, with energies minus thirteen point six electron volts divided by n squared, where n is a whole number. A jump from level three to level two releases one point eight nine electron volts, the red line at six hundred and fifty six nanometres. Be careful: Bohr's tiny planetary orbits are a simplified picture that we now know is wrong, because electrons do not follow definite paths. But his energy formula for hydrogen survives in the full theory, and the core lesson stands: bound systems have discrete energy levels.
Matter behaves like waves
If light, a wave, arrives in particle-like lumps, perhaps particles have wave-like behaviour. Louis de Broglie proposed in nineteen twenty four that any particle with momentum p has a wavelength lambda equals h divided by p. For a thrown ball the wavelength is around ten to the minus thirty four metres, far too small ever to notice, which is why everyday objects seem perfectly classical. For an electron accelerated through fifty four volts the wavelength is about zero point one seven nanometres, similar to the spacing of atoms in a crystal. So a crystal should act like a diffraction grating for electrons. In nineteen twenty seven Davisson and Germer fired electrons at a nickel crystal and saw exactly the predicted diffraction peaks. Today electron microscopes use this wave nature to image individual atoms, because the short wavelength gives very fine detail.
One electron, two slits
Now the experiment Feynman called the heart of quantum mechanics. Send electrons, one at a time, towards a wall with two narrow slits, and record where each lands on a screen behind. Each electron arrives as a single dot at a single place, like a particle. But after thousands of dots, a pattern of bright and dark bands appears, an interference pattern, just like water waves passing through two gaps. Close one slit and the bands disappear, leaving a single smooth hump. Here is the puzzle. With both slits open, some places on the screen receive fewer electrons than with only one slit open. Opening an extra route reduces arrivals there. No picture of a little ball going through one slit or the other can do that. Something about each single electron is sensitive to both slits at once.
Add amplitudes, then square
The rule that works is this. For each way an event can happen, quantum mechanics assigns a complex number called a probability amplitude. When the ways cannot be told apart, you add the amplitudes first, and then take the squared magnitude to get the probability. Squaring a sum produces an extra cross term, the interference term, which can be positive or negative. Where the two amplitudes point in opposite directions they cancel, giving the dark bands. Now suppose you place a detector at the slits that records which way each electron went. Then the routes are distinguishable, you add probabilities instead of amplitudes, the cross term vanishes, and the bands disappear. Interference is not caused by electrons bumping into each other, since it happens with one electron at a time. It depends on whether information about the path exists anywhere.
The wavefunction and probability
Collect the amplitudes for a particle to be found at every position and you get a function, psi of x, called the wavefunction. Max Born's rule says the probability of finding the particle between x and x plus d x is the absolute value of psi squared, times d x. So psi squared is a probability density, and integrated over all space it must equal one, because the particle is certainly somewhere. This requirement is called normalisation. Notice what the wavefunction is not. It is not a cloud of smeared-out electric charge, and it is not a physical wave in a medium, like sound in air. Its overall phase cannot be measured, though relative phases matter, as interference shows. Whether psi is a real physical thing or only a summary of our knowledge is still debated. What is not debated is that it predicts measured statistics extremely accurately.
Schrödinger's equation
How does the wavefunction change in time? Erwin Schrödinger proposed the equation in nineteen twenty six. It cannot be derived from classical physics; it is a postulate, trusted because its predictions match experiment. It says i h-bar times the rate of change of psi equals minus h-bar squared over two m times the second derivative of psi, plus V times psi. Here h-bar is h divided by two pi, m is the mass, and V is the potential energy. The first term on the right is kinetic energy, linked to how sharply psi curves; the second is potential energy. Together they form the Hamiltonian, the energy operator. When V does not change in time, we look for states of definite energy obeying H psi equals E psi. Assumptions: speeds far below light, a fixed number of particles, and a classical potential.
Observables and operators
In quantum mechanics each measurable quantity, called an observable, is represented by an operator, a rule that acts on the wavefunction. Position multiplies psi by x. Momentum is minus i h-bar times the derivative with respect to x. That is de Broglie's idea in mathematical form, since a wave with a short wavelength wiggles fast and has a large derivative. Each operator has special functions, called eigenfunctions, that it only multiplies by a number, the eigenvalue. The measurement postulate says that a single measurement can only give one of the eigenvalues, that the probability of each is the squared size of psi's component along that eigenfunction, and that straight afterwards the state is that eigenfunction. The average over many identical measurements is the expectation value. Observables are Hermitian operators, which guarantees real eigenvalues, as measured numbers must be.
The uncertainty principle
Try to make a wave that sits in one place. A pure sine wave has a perfectly sharp wavelength, and so a perfectly sharp momentum, but it stretches forever. To squeeze it into a short pulse you must add many wavelengths together, which spreads the momentum. This is true of all waves; it is why a very short click of sound contains many pitches. For quantum particles it becomes Heisenberg's uncertainty principle: the spread in position times the spread in momentum is at least h-bar over two. The spreads are standard deviations over many identically prepared particles. This limits what states can exist at all, not just how well we measure. It also explains why atoms do not collapse: squeezing the electron close to the nucleus raises its momentum spread, and so its kinetic energy, which balances the electrical attraction.
Particle in a box
The simplest bound system is a particle trapped between two perfectly hard walls a distance L apart, with zero potential energy inside. This is an idealisation, but it captures electrons in tiny semiconductor crystals and in long dye molecules. Since the particle cannot be outside, psi must be zero at both walls, just as a guitar string is fixed at both ends. Only waves that fit a whole number of half wavelengths are allowed. So the wavelength is two L over n, the momentum is n h over two L, and the energy is n squared h squared over eight m L squared. Energy is quantised purely because the wave must fit. Two consequences follow. Levels grow as n squared. And the lowest energy is not zero: a confined particle is never at rest, which is called zero-point energy, as the uncertainty principle demands.
Worked example: electron in box
Put an electron in a box one nanometre wide, about the size of a small molecule. Use E one equals h squared over eight m L squared. Planck's constant squared is about four point three nine times ten to the minus sixty seven. Eight times the electron mass, nine point one one times ten to the minus thirty one kilograms, times L squared, ten to the minus eighteen square metres, is about seven point two nine times ten to the minus forty eight. Dividing gives about six point zero times ten to the minus twenty joules, roughly zero point three eight electron volts. Level two is four times higher, so the jump from one to two needs three times E one, about one point one three electron volts, a photon near eleven hundred nanometres, in the infrared. Smaller boxes shift the light towards blue, the idea behind quantum dot displays.
The harmonic oscillator
The next model is a particle held by a spring-like force, with potential energy one half m omega squared x squared. Almost any system near a stable equilibrium looks like this for small displacements, so it describes the vibration of atoms in a molecule, atoms in a solid, and even each mode of the electromagnetic field. Solving Schrödinger's equation gives energies E n equals n plus one half, times h-bar omega, with n equal to zero, one, two and so on. Unlike the box, the levels are evenly spaced, like rungs on a ladder. That is why a molecule such as carbon monoxide absorbs infrared light mainly at one vibrational frequency. Again the lowest energy, one half h-bar omega, is not zero. Real chemical bonds are not perfect springs, so their levels crowd closer together at high energy until the bond breaks.
Tunnelling through barriers
Classically, a ball without enough energy to climb a hill is simply turned back. Quantum mechanically, if the barrier has finite height and width, the wavefunction does not stop dead at the wall. Inside the barrier it decays exponentially instead of oscillating, and if the barrier is thin, some of it survives on the far side. So there is a nonzero probability of finding the particle beyond, an effect called tunnelling. For a wide barrier the transmission probability is roughly e to the minus two kappa L, where kappa is the square root of two m times V minus E, divided by h-bar. Because of the exponential, tunnelling is extremely sensitive to barrier width and particle mass. It explains alpha decay of heavy nuclei, allows fusion in the Sun's core, lets the scanning tunnelling microscope image single atoms, and underlies flash memory.
The hydrogen atom
Solve Schrödinger's equation in three dimensions with the electrical attraction of a proton, and the hydrogen atom comes out. The energies are minus thirteen point six electron volts over n squared, the same as Bohr's, but now the states are three-dimensional standing waves called orbitals. Each is labelled by quantum numbers. The principal number n sets the energy. The orbital angular momentum number l runs from zero to n minus one and sets the shape: l equals zero is spherical, called s, and l equals one is dumbbell-shaped, called p. The magnetic number m l runs from minus l to plus l and sets the orientation. Pictures of orbitals show where psi squared is large; they are probability maps, not paths. Notice that s orbitals have zero angular momentum, something a planet-like orbit could never have.
Spin and Stern-Gerlach
In nineteen twenty two Otto Stern and Walther Gerlach sent silver atoms through a non-uniform magnetic field. Classically, tiny magnets pointing in random directions should spread into a continuous smear. Instead the beam split into just two spots. The cause is spin, an intrinsic angular momentum carried by the electron. Measured along any chosen axis, an electron's spin component is always either plus or minus h-bar over two, called spin up and spin down. It is tempting to picture a little spinning ball, but that is only a loose analogy: the electron has no known size, and a spinning surface would have to move faster than light. Spin is simply an extra quantum property. It adds a fourth quantum number, m s. Nuclei such as protons also have spin, which MRI scanners use, and spins are leading candidates for quantum bits.
Pauli exclusion and chemistry
Particles come in two families. Fermions, like electrons, protons and neutrons, have half-integer spin. Bosons, like photons, have integer spin. For identical fermions, the combined wavefunction must change sign when two particles are swapped. A direct result is the Pauli exclusion principle: no two electrons in an atom can share all four quantum numbers. So electrons cannot all crowd into the lowest level; they fill orbitals at most two at a time, with opposite spins. This filling, adjusted for repulsion between electrons, builds the periodic table: why helium is unreactive, why sodium easily gives up one outer electron, why carbon forms four bonds. Exclusion also makes matter take up space and resist compression, and it holds up white dwarf stars. Bosons do the opposite: they can pile into one state, which gives lasers and superfluids.
Superposition and measurement
Because Schrödinger's equation is linear, if two states are allowed, any weighted sum of them is allowed too. This is superposition. An electron can be in a superposition of spin up and spin down, or of passing through both slits. A measurement gives just one outcome, with probabilities set by the amplitudes. The standard textbook rule says the state then collapses to the measured outcome. But why, and when? Schrödinger mocked this with a cat in a box, linked to a decaying atom, apparently both alive and dead until someone looks. Part of the answer is decoherence: when a quantum system interacts with its large environment, the phase relations needed for interference leak away extremely fast, so big objects behave classically. Decoherence explains why we never see cats in superposition, but whether it fully solves the measurement problem is still debated.
Entanglement and Bell's test
When two particles interact, their joint state often cannot be written as one state for each separately. Such a state is called entangled. Prepare two electrons with total spin zero and send them far apart. Measure both along the same axis and you always find opposite results, although each single result looks completely random. Einstein suspected hidden instructions carried from the start. In nineteen sixty four John Bell showed that any such local hidden-variable theory obeys an inequality that quantum mechanics violates when the detectors are set at suitable different angles. Experiments, including loophole-free tests in twenty fifteen recognised by the twenty twenty two Nobel Prize, side with quantum mechanics. Yet entanglement cannot send messages faster than light, because each side's results alone are random. It is a key resource for quantum cryptography and quantum computing.
Common mistake
A very common mistake is to think that observation in quantum physics needs a conscious observer, or that the uncertainty principle only means our instruments are too clumsy. Neither is right. In physics, a measurement is any interaction that leaves a lasting record of the result in the environment: a detector, a photographic plate, even stray air molecules or heat. In the double slit, the bands vanish whenever which-path information exists, whether or not anyone ever reads it. And uncertainty is not just disturbance by a measuring device. It follows from the wave nature of quantum states: a state with a sharply defined position simply does not have a sharply defined momentum. A clumsy measurement can add extra disturbance on top, but even a perfect instrument cannot beat the limit. Nothing mystical is required, although the interpretation remains open.
Quantum physics in technology
Quantum mechanics is the foundation of modern technology. In a crystal, the levels of many neighbouring atoms merge into energy bands, which explains why copper conducts, glass insulates and silicon is a semiconductor, giving us transistors and every computer chip. Lasers rely on stimulated emission between discrete levels. Light-emitting diodes and solar cells exchange photons across a band gap. MRI scanners flip nuclear spins. Atomic clocks, which keep satellite navigation accurate, count the frequency of an atomic transition. Chemistry is applied quantum mechanics, since bonds come from shared electron wavefunctions. Neighbouring topics build directly on these ideas: statistical mechanics adds quantum counting of states, solid state physics adds periodic potentials, nuclear physics applies tunnelling and exclusion to nucleons, and quantum information treats superposition and entanglement as resources for computing and secure communication.
Limits of the model
Now the honest limits. The Schrödinger equation we used is non-relativistic: it assumes speeds well below light and a fixed number of particles, and it adds spin by hand. Paul Dirac's relativistic equation for the electron includes spin naturally and predicted antimatter. Combining quantum mechanics fully with special relativity requires quantum field theory, where particles are excitations of fields and can be created and destroyed. Quantum electrodynamics, the field theory of light and electrons, matches measurements of the electron's magnetic moment to better than one part in a billion. But deep questions remain open. We have no tested quantum theory of gravity, so black hole interiors and the earliest universe are beyond our reach. And physicists still disagree about what the wavefunction is and what happens in measurement; Copenhagen, many-worlds and pilot-wave views all fit current experiments.
Explain it back
Let us retrace the path. Glowing objects and the photoelectric effect showed that light energy is exchanged in quanta of h f. Atomic spectra showed that bound systems have discrete energy levels. Electron diffraction showed that matter has a wavelength, h over p. The double slit showed that we add amplitudes, then square, to get probabilities. The wavefunction obeys Schrödinger's equation, and confinement makes energy quantised, as in the box, the oscillator and the hydrogen atom. Uncertainty, tunnelling, spin, exclusion and entanglement all follow from these rules. Now try to explain to a friend, in your own words, why an electron in a box cannot have zero energy, and why opening a second slit can make fewer electrons land somewhere. If you can do that clearly, without notes, you truly understand the core of quantum physics.