First Year Engineering is coming soon

We're still writing and reviewing the notes for every subject, and we'll open it when every subject is ready. Meanwhile you can practise aptitude, ask Aria about any subject, or see the branches that are ready.

Heisenberg's uncertainty principle

Heisenberg’s uncertainty principle: limits on simultaneous precision of conjugate variables (position–momentum, energy–time), their derivations, formulas, examples, and exam-focused notes.

Drafted with Aria, reviewed by the AiCanCode.org team · Updated . How we write and check our notes. Spotted an error? Use Give Feedback at the bottom of the page.

Why it matters

Engineers use the uncertainty principle to understand limits in measuring and controlling microscopic systems: electron beams in electron microscopes, linewidths in lasers and LEDs, carrier behavior in semiconductors and nanomaterials, and tunnelling in devices like the scanning tunnelling microscope. It also underpins design trade-offs in quantum technologies (quantum sensors, qubits) and explains why classical intuition fails at nanoscale.

Key ideas

  • State (wavefunction): A complete description of a quantum system, given by a complex function ψ(x,t) normalized such that ∫|ψ|² dx = 1. Probability density is |ψ(x,t)|².
  • Expectation value: The average outcome of many measurements of an observable A in state ψ: ⟨A⟩ = ∫ψ* Â ψ dx.
  • Standard deviation (uncertainty): For observable A, the spread of measurement outcomes ΔA = sqrt(⟨A²⟩ − ⟨A⟩²). We use Δ to denote statistical spread, not instrument error.
  • Conjugate variables: Pairs of observables linked by Fourier transform structure or non-commuting operators, e.g., position x and momentum p, whose precise simultaneous determination is fundamentally limited.
  • Commutator: For operators  and 𝐵̂, [Â,𝐵̂] = Â𝐵̂ − 𝐵̂Â. If [Â,𝐵̂] ≠ 0, the observables cannot be simultaneously sharp in the same state.
  • Heisenberg’s uncertainty principle (exact statement): For any quantum state ψ, the standard deviations of position and momentum satisfy Δx · Δp ≥ ħ/2, where ħ = h/(2π). Equality holds for minimum-uncertainty (Gaussian) states.
  • Assumptions for the standard x–p form: (i) ψ is normalizable and differentiable; (ii) position and momentum operators on this domain satisfy [x̂, p̂] = iħ; (iii) uncertainties defined as standard deviations.

Typical derivations

A) From Fourier analysis and de Broglie waves (university-favourite, step-by-step):

  1. A localized particle is described by a wave packet: a superposition of de Broglie plane waves with wave numbers around k₀. Position-space ψ(x) and wave-number-space φ(k) are Fourier transform pairs.
  2. A basic result of Fourier analysis for square-integrable functions is the inequality Δx · Δk ≥ 1/2, with Δ defined as standard deviations, saturated by Gaussians.
  3. Using the de Broglie relation p = ħk, we have Δp = ħ·Δk.
  4. Therefore, Δx · Δp = Δx · ħΔk ≥ ħ/2. This is Heisenberg’s uncertainty relation.
  • What to draw and label (diagram for answer): A qualitative sketch of (i) a narrow wave packet ψ(x) in x (small Δx) and broad φ(k) in k (large Δk); and (ii) a broad ψ(x) (large Δx) and narrow φ(k) (small Δk). Label axes x, |ψ|², k, |φ|², and indicate Δx, Δk.

B) From the commutator (Robertson–Schrödinger inequality):

  1. For any two observables with operators  and 𝐵̂, Robertson’s inequality states ΔA · ΔB ≥ (1/2)|⟨[Â,𝐵̂]⟩|. This follows by applying the Cauchy–Schwarz inequality to the vectors ( − ⟨A⟩)ψ and (𝐵̂ − ⟨B⟩)ψ. The stronger Schrödinger form adds a covariance (anticommutator) term: ΔA²ΔB² ≥ (1/4)|⟨[Â,𝐵̂]⟩|² + (1/4)|⟨{Â−⟨A⟩, 𝐵̂−⟨B⟩}⟩|².
  2. For position and momentum, with the standard representation p̂ = −iħ d/dx, we have [x̂, p̂] = iħ (a constant operator).
  3. Hence, Δx · Δp ≥ (1/2)|⟨iħ⟩| = ħ/2.
  • Assumptions: Operators act on a common dense domain; ψ is normalized; commutator expectation is well-defined.
  • What to draw and label: A small box listing the commutator [x̂, p̂] = iħ, the general inequality ΔAΔB ≥ (1/2)|⟨[Â,𝐵̂]⟩|, and the substitution to get ΔxΔp ≥ ħ/2.

C) Heisenberg’s gamma-ray microscope thought experiment (physical intuition):

  1. To localize an electron with accuracy Δx, use a microscope with light of wavelength λ and semi-aperture angle α. Diffraction of light sets a limit Δx ≳ λ/(2 sin α).
  2. A high-energy photon scatters from the electron; due to the photon's momentum pγ = h/λ, the component imparted along x is uncertain within ±pγ sin α, giving electron momentum uncertainty Δp_x ≳ pγ sin α = (h/λ) sin α.
  3. Product: Δx · Δp_x ≳ [λ/(2 sin α)] · [(h/λ) sin α] = h/2 ≈ (2π)(ħ/2). With more careful quantum treatment the bound reduces to ħ/2.
  • What to draw and label: A schematic of a microscope: electron at focus, incident photon, scattered ray within cone of semi-angle α, objective lens, image screen. Label λ, α, Δx at object, and momentum components.
  • Note: This is a heuristic argument showing the trade-off; the exact inequality comes from wave mechanics and operator algebra.

Related uncertainty relations and caveats

  • Energy–time uncertainty: ΔE · Δt ≥ ħ/2. Here time t is not an operator in non-relativistic quantum mechanics; Δt is a characteristic time such as the lifetime of an excited state (natural linewidth), measurement duration, or the time over which a state appreciably changes. A useful corollary for spectral lines: Δf · Δt ≥ 1/(4π), since ΔE = hΔf.
  • Angular momentum–angle: A commonly quoted form is ΔL_z · Δφ ≥ ħ/2, but angle φ is a periodic variable (0 to 2π), so the rigorous operator form is subtle. In practice, for states well localized in angle and not spanning the periodic boundary, this relation is a good approximation.
  • Multiple dimensions: The uncertainty relations apply component-wise, e.g., ΔxΔp_x ≥ ħ/2, ΔyΔp_y ≥ ħ/2, ΔzΔp_z ≥ ħ/2. There is no lower bound directly coupling, say, Δx with Δp_y.
  • Minimum-uncertainty states: Gaussian wave packets and the ground state of the quantum harmonic oscillator saturate ΔxΔp = ħ/2.

Physical meaning

  • The principle is not about poor instruments; it is about quantum states. Preparing a state with small Δx forces large Δp, and vice versa.
  • Repeated measurements on identically prepared systems will have spreads Δx and Δp consistent with the inequality. A single system cannot have both sharp x and p simultaneously.
  • In nanoscale devices (see also: Quantum tunnelling and the scanning tunnelling microscope, Semiconductors and the Fermi level), confinement increases momentum spread and thus kinetic energy, affecting design and operation.

Formulas

Δx · Δp ≥ ħ/2

  • Δx: standard deviation of position (m). Δp: standard deviation of momentum (kg·m·s⁻¹). ħ: reduced Planck’s constant (J·s). Applies to any quantum state for one spatial component.

[x̂, p̂] = i·ħ

  • x̂: position operator (m). p̂: momentum operator (kg·m·s⁻¹). i: imaginary unit. Fundamental commutator underlying ΔxΔp ≥ ħ/2.

ΔA · ΔB ≥ (1/2) · |⟨[Â, 𝐵̂]⟩|

  • ΔA, ΔB: standard deviations of observables A and B. ⟨…⟩: expectation value in the state. General Robertson uncertainty relation.

p = ħ·k = h/λ

  • p: particle momentum (kg·m·s⁻¹). k: wave number (m⁻¹). λ: wavelength (m). h: Planck’s constant (J·s). Links wave and particle pictures.

Δk = Δp/ħ

  • Δk: standard deviation of wave number (m⁻¹). Valid whenever p = ħk (de Broglie relation) applies—for massive or massless particles; not restricted to the non-relativistic regime.

ΔE · Δt ≥ ħ/2

  • ΔE: uncertainty (natural spread) in energy (J). Δt: characteristic time (s), e.g., lifetime or measurement duration. Applies to processes with finite observation time or decaying states.

Δf · Δt ≥ 1/(4·π)

  • Δf: frequency uncertainty (Hz). Derived from ΔE = hΔf and ΔEΔt ≥ ħ/2. Useful for estimating spectral linewidths from lifetimes.

a · sinθ = λ (first minimum in single-slit diffraction)

  • a: slit width (m). θ: diffraction angle (radians). λ: wavelength (m). Used to estimate momentum spread via p_x ≈ p·sinθ.

K_min ≈ (Δp)²/(2·m)

  • K_min: minimum kinetic energy implied by momentum spread (J). m: particle mass (kg). Applies non-relativistically when Δp ≪ mc.

Constants for use (SI):

  • h ≈ 6.626 × 10⁻³⁴ J·s; ħ = h/(2π) ≈ 1.054 × 10⁻³⁴ J·s; c ≈ 3.00 × 10⁸ m·s⁻¹; m_e (electron) ≈ 9.109 × 10⁻³¹ kg; m_p (proton) ≈ 1.673 × 10⁻²⁷ kg; 1 eV = 1.602 × 10⁻¹⁹ J.

Worked examples

Example 1: Electron localized in an atom-sized region — find the minimum velocity spread

  • Given: Δx = 0.10 nm = 1.0 × 10⁻¹⁰ m; m_e = 9.109 × 10⁻³¹ kg; ħ = 1.054 × 10⁻³⁴ J·s.
  1. Use position–momentum uncertainty: Δp ≥ ħ/(2Δx).
  2. Compute Δp: Δp ≥ (1.054 × 10⁻³⁴) / (2 × 1.0 × 10⁻¹⁰) = 5.27 × 10⁻²⁵ kg·m·s⁻¹.
  3. Convert to velocity spread using non-relativistic relation: Δv ≥ Δp/m_e.
  4. Compute Δv: Δv ≥ (5.27 × 10⁻²⁵) / (9.109 × 10⁻³¹) = 5.79 × 10⁵ m·s⁻¹.
  • Final: Minimum velocity spread Δv ≈ 5.8 × 10⁵ m·s⁻¹.
  • Comment: This is a sizable fraction of electron speeds in atoms; the estimate is non-relativistic and adequate here because Δv ≪ c.

Example 2: Natural linewidth from excited-state lifetime

  • Given: Excited-state lifetime τ = 10 ns = 1.0 × 10⁻⁸ s; ħ = 1.054 × 10⁻³⁴ J·s; h = 6.626 × 10⁻³⁴ J·s; c = 3.00 × 10⁸ m·s⁻¹; central wavelength λ₀ = 600 nm = 6.00 × 10⁻⁷ m.
  1. Energy–time uncertainty: ΔE ≥ ħ/(2τ).
  2. Compute ΔE: ΔE ≥ (1.054 × 10⁻³⁴) / (2 × 1.0 × 10⁻⁸) = 5.27 × 10⁻²⁷ J.
  3. Convert to frequency width using ΔE = hΔf: Δf ≥ ΔE/h = (5.27 × 10⁻²⁷) / (6.626 × 10⁻³⁴) = 7.95 × 10⁶ Hz.
  4. Convert to wavelength width (small-width approximation): Δλ ≈ (λ₀² / c) · Δf.
  5. Compute λ₀² / c: (6.00 × 10⁻⁷)² / (3.00 × 10⁸) = (3.60 × 10⁻¹³) / (3.00 × 10⁸) = 1.20 × 10⁻²¹ s·m.
  6. Compute Δλ: Δλ ≈ (1.20 × 10⁻²¹) × (7.95 × 10⁶) = 9.54 × 10⁻¹⁵ m.
  • Final: Minimum natural linewidth Δf ≈ 8.0 × 10⁶ Hz; corresponding Δλ ≈ 9.5 × 10⁻¹⁵ m at 600 nm.
  • Comment: Very narrow in wavelength units; this sets a lower limit for ideal laser/LED linewidths (see also: Lasers: principle and types; Applications of lasers).

Example 3: Confinement of a proton in a nucleus — estimate minimum kinetic energy

  • Given: Confinement length Δx = 1.0 fm = 1.0 × 10⁻¹⁵ m; ħ = 1.054 × 10⁻³⁴ J·s; m_p = 1.673 × 10⁻²⁷ kg.
  1. Momentum uncertainty: Δp ≥ ħ/(2Δx) = (1.054 × 10⁻³⁴) / (2 × 1.0 × 10⁻¹⁵) = 5.27 × 10⁻²⁰ kg·m·s⁻¹.
  2. Minimum kinetic energy from spread: K_min ≈ (Δp)²/(2m_p).
  3. Compute (Δp)²: (5.27 × 10⁻²⁰)² = 2.78 × 10⁻³⁹ (kg²·m²·s⁻²).
  4. Compute denominator: 2m_p = 3.346 × 10⁻²⁷ kg.
  5. Compute K_min: 2.78 × 10⁻³⁹ / 3.346 × 10⁻²⁷ = 8.31 × 10⁻¹³ J.
  6. Convert to electronvolts: K_min (eV) = (8.31 × 10⁻¹³) / (1.602 × 10⁻¹⁹) = 5.19 × 10⁶ eV = 5.19 MeV.
  • Final: K_min ≈ 5.2 MeV.
  • Comment: This order of magnitude matches typical nuclear binding energies, showing why nucleons are energetic when confined to femtometre scales (see also: Band theory of solids for contrast with electron confinement in crystals).

Model long-answer outline (theory): State and derive ΔxΔp ≥ ħ/2

  • Start with the statement of the principle with symbols defined.
  • Show the general Robertson inequality and substitute [x̂,p̂] = iħ to get ΔxΔp ≥ ħ/2, with assumptions.
  • Alternatively, outline the Fourier-based derivation: wave packet, ΔxΔk ≥ 1/2, p = ħk ⇒ ΔxΔp ≥ ħ/2.
  • Add physical interpretation and the conditions for equality (Gaussian states). Conclude with relevance to measurement limits and examples (gamma-ray microscope; natural linewidth).

Common mistakes

  • Dropping the factor 1/2 and writing ΔxΔp ≥ ħ by mistake; examiners expect the correct coefficient.
  • Mixing up h and ħ (use ħ in ΔxΔp ≥ ħ/2, use h in p = h/λ and ΔE = hΔf).
  • Using time as an operator and applying the Robertson form blindly to time; remember ΔEΔt ≥ ħ/2 is of a different character and Δt must be defined contextually.
  • Forgetting component-wise nature: writing ΔxΔp ≥ ħ/2 for 3D magnitudes; correct is ΔxΔp_x ≥ ħ/2, etc.
  • Using relativistic particles with non-relativistic formulas (e.g., converting Δp to Δv when Δv would approach c). Check Δp ≪ mc for non-relativistic use.
  • Confusing diffraction formula a sinθ = λ (optical) with de Broglie wavelength for matter waves; ensure the correct λ is used (de Broglie for particles, optical for photons).
  • Claiming the principle arises from measurement disturbance only; while disturbance provides intuition, the exact inequality is a property of the quantum state and operator algebra.

For your exam

  • Short notes: Gamma-ray microscope, minimum-uncertainty states, physical meaning of ΔEΔt. Include a labeled diagram description for the microscope.
  • Derivations: (i) ΔxΔp ≥ ħ/2 via commutator method; (ii) via Fourier analysis and de Broglie relation; (iii) general Robertson inequality. State assumptions and steps; show how equality is achieved for Gaussians.
  • Numericals: Confinement length to momentum/velocity spread; lifetime to linewidth (ΔE, Δf, Δλ); nuclear confinement to kinetic energy. Carry units throughout, use SI constants, and check order of magnitude.
  • Diagrams: Wave packet vs k-space spread; gamma-ray microscope geometry. Draw clean axes, label Δx, Δk, λ, α, and momentum components.
  • Full-marks answers: Correct statement with coefficient 1/2, correct constants (h vs ħ), explicit formulas before substitution, stepwise calculation with units, clear concluding sentence interpreting the result.

Quick check

  1. Write the Heisenberg uncertainty relation for position and momentum along x, and state when equality holds.
  2. An excited state has lifetime 5 ns. What is the minimum frequency linewidth Δf implied by ΔEΔt ≥ ħ/2?
  3. If a particle’s Δp is halved, what happens to the minimum possible Δx?

Answers: 1) ΔxΔp_x ≥ ħ/2; equality holds for Gaussian (minimum-uncertainty) states. 2) Δf ≥ 1/(4πΔt) = 1/(4π × 5 × 10⁻⁹) ≈ 1.59 × 10⁷ Hz. 3) Minimum Δx doubles (since Δx ∝ 1/Δp at the bound).

Finished this topic? Mark it so your progress, study plan and readiness keep up.

Stuck on something here?