Physics › Section U · Quantum and atomic physics › Topic U.3
Wave-particle duality and uncertainty
Light interferes like a wave yet arrives in lumps like particles, and electrons turn out to do exactly the same. This double nature sets a hard limit on what can be known about any particle, a limit that has nothing to do with how good our instruments are.
Light: wave or particle?
Light passing through two slits makes bands of bright and dark, an interference pattern that only waves produce. Yet in the photoelectric and Compton effects, light acts as a stream of separate photons, each with a definite energy and momentum. Both sets of evidence are solid, so light has both wave and particle properties. This is called wave-particle duality. In any single experiment you see one side or the other: the wave picture describes how light travels and interferes, and the particle picture describes how it is emitted, absorbed and detected. Neither picture alone is complete, and neither is wrong.
De Broglie's matter waves
In his doctoral thesis of 1924, Louis de Broglie argued that nature should be symmetric: if waves can act as particles, particles should act as waves. He proposed that every particle with momentum \(p\) has a wavelength, the de Broglie wavelength
This is the photon relation \(p = h/\lambda\) turned around and applied to matter. For a particle moving well below the speed of light, \(p = mv\), so \(\lambda = h/(mv)\). Because \(h\) is tiny, the wavelength of anything large is absurdly small. Wave effects become noticeable only when the wavelength is comparable to the size of the objects or gaps the particle meets.
Worked example: a baseball and an electron
Compare the de Broglie wavelength of a 0.145 kg baseball thrown at 40 m/s with that of an electron accelerated from rest through 100 V. 1. Baseball:
2. Electron energy: crossing 100 V gives \(KE = eV = 100\) eV, or \(1.602 \times 10^{-17}\) J. 3. Since \(KE = p^2/2m\), the momentum is
4. So \(\lambda = h/p = 1.23 \times 10^{-10}\) m, or 0.123 nm. 5. Check: the electron's wavelength is about the size of an atom, so it can diffract from a crystal; the baseball's is far smaller than a proton.
Electrons diffract
In 1927 Clinton Davisson and Lester Germer in the United States scattered slow electrons from a nickel crystal and found strong beams at particular angles, just as X-rays diffract from the regular rows of atoms in a crystal. The same year George Paget Thomson in Britain sent faster electrons through thin metal foils and recorded rings of diffraction. Both patterns matched \(\lambda = h/p\). Neutrons, whole atoms and even large molecules have since been diffracted. The electron microscope puts this to work: electrons accelerated through tens to hundreds of kilovolts have wavelengths tens of thousands of times shorter than visible light, so the best instruments can resolve details smaller than 0.1 nm and show individual atoms.
Waves of probability
Send electrons at two slits so slowly that only one is in the apparatus at a time. Each electron lands at one definite spot, like a particle. Spot by spot, though, the familiar interference pattern builds up, so each electron's wave must pass through both slits and interfere with itself. The wave does not say where an electron will land; it gives the probability of finding it at each place, and only the overall pattern can be predicted. Now add a detector that records which slit each electron uses. Every electron is then found at one slit or the other, and the interference pattern disappears. Measuring the system changes it.
Heisenberg's uncertainty principle
To locate an electron you must hit it with something, such as a photon, and the shorter the probe's wavelength, the more precisely it fixes the position but the bigger the kick it gives. In 1927 Werner Heisenberg showed that the uncertainties in a particle's position \(\Delta x\) and momentum \(\Delta p\), measured at the same time, obey
with \(h/(4\pi) = 5.27 \times 10^{-35}\ \text{J}\cdot\text{s}\). Neither uncertainty can be zero, and making one smaller forces the other to grow. This is a property of nature, following from matter's wave character, not a flaw in our instruments. One consequence is that an electron in an atom cannot follow a definite orbit that we could track.
Worked example: an electron and a dust grain
Find the smallest uncertainty in velocity for an electron located to within 0.20 nm, and for a 1.0 µg dust grain located to within 1.0 µm. 1. Electron:
2. Divide by the mass: \(\Delta v = \Delta p/m\), which is \(2.9 \times 10^5\) m/s. 3. Dust grain: the mass is \(1.0 \times 10^{-9}\) kg and \(\Delta x = 1.0 \times 10^{-6}\) m, so \(\Delta p = 5.3 \times 10^{-29}\ \text{kg}\cdot\text{m/s}\). 4. Then \(\Delta v = 5.3 \times 10^{-20}\) m/s. 5. Interpret: for the electron the uncertainty is huge, while for the dust grain it is far too small ever to measure.
Energy and time
A second form of the principle links energy and time:
Here \(\Delta E\) is the uncertainty in a system's energy and \(\Delta t\) is the time available to measure it, such as the lifetime of a state. An excited atom that lasts only about \(1.0 \times 10^{-8}\) s before emitting a photon has an energy uncertain by at least \(5.3 \times 10^{-27}\) J, or \(3.3 \times 10^{-8}\) eV. So even perfectly measured spectral lines are not infinitely sharp: each has a small natural width, and shorter-lived states give broader lines. A state that lasts a long time can have a sharply defined energy.
Duality reviewed
Both light and matter are quantum objects that are neither classical waves nor classical particles. Each travels as a wave, which sets the probability of where it will be found, and each is detected as a particle, at one place, carrying a definite amount of energy. Photons and electrons obey the same relations, \(\lambda = h/p\) and \(E = hf\); the main difference is that a photon has no mass. Large objects show neither quantisation nor wave effects that we could measure, because \(h\) is so small, which is the correspondence principle at work once more. The wave nature of electrons also explains why atoms have quantised energy levels, as later topics in this section show.
Common mistakes to avoid
First, using \(p = mv\) for a photon; photons have no mass, though \(\lambda = h/p\) still holds. Second, using the accelerating voltage as an energy in joules; an electron through 100 V gains 100 eV, which must be converted before using \(p = \sqrt{2mKE}\). Third, leaving mass in grams or distances in nanometres. Fourth, forgetting the \(4\pi\) in the uncertainty principle. Fifth, blaming uncertainty on clumsy instruments; it is a property of nature. Sixth, thinking an electron splits in two at a double slit; it is always detected whole, but its probability wave passes through both slits.
Key takeaways
Light and matter both show wave-particle duality: they travel as waves and are detected as particles. Every particle has a de Broglie wavelength λ = h/p, which is h/(mv) at ordinary speeds. Electron diffraction, first seen in 1927, confirmed it, and electron microscopes use it. The wave gives the probability of finding the particle; observing which slit an electron uses destroys the interference pattern. Heisenberg's uncertainty principle: Δx Δp ≥ h/4π and ΔE Δt ≥ h/4π, so position and momentum cannot both be known exactly, and short-lived states have broad energies. The effects are huge for electrons and negligible for everyday objects.
Flashcards
Flashcard
Light and matter show both wave and particle properties
Test your understanding
Question 1 of 5
An electron (mass 9.109 × 10⁻³¹ kg) moves at 1.0 × 10⁶ m/s. Find its de Broglie wavelength.
Question 2 of 5
Why does a thrown baseball show no measurable wave behaviour?
Question 3 of 5
An electron is located to within 1.0 nm. How uncertain must its velocity be, at the least? Use h/4π = 5.27 × 10⁻³⁵ J·s.
Question 4 of 5
Electrons are sent one at a time towards two slits, with detectors that record which slit each electron passes through. What appears on the screen?
Question 5 of 5
Which statement about the Heisenberg uncertainty principle is correct?
