How does this differ from a Michelson interferometer?

Both split and recombine beams, but the Mach–Zehnder uses two separate paths that meet at a second beamsplitter. That geometry can make it easier to send each arm through different samples or modulators without sending light back toward the source, and it offers distinct output ports corresponding to complementary interference phases.

Why is the intensity periodic in Δ with period λ?

Each wavelength of OPD adds 2π to the phase difference between the arms. Because intensity depends on cos² of half the phase (for equal amplitudes), a full 2π phase cycle returns the same intensity; equivalently, the pattern repeats when Δ changes by one wavelength.

What happens if the amplitudes are not perfectly balanced?

The fringe contrast (visibility) drops. The general two-beam intensity is not symmetric about zero when one arm is weaker; minima are no longer exactly zero. This simulator fixes equal amplitudes to isolate the phase dependence.

Does the schematic include dispersion in the beamsplitters?

No. Real cube beamsplitters and coatings introduce wavelength-dependent phase; broadband sources then wash out fringes unless compensated. Here λ is a single design wavelength and materials are ideal.

Mach–Zehnder Interferometer

Two-beam recombination; I ∝ cos²(πΔ/λ) vs one-arm OPD; schematic + fringe plot.

Live graphs

About this model

The Mach–Zehnder interferometer splits a coherent beam into two arms, delays one relative to the other, and recombines them so interference can be read at an output port. This page uses the standard two-beam intensity law for balanced amplitudes: I/I_max = cos²(πΔ/λ), where Δ is the optical path difference (OPD) introduced in one arm and λ is the vacuum wavelength. A schematic shows beam splitters and mirrors; the graph plots fringe contrast versus Δ. The layout highlights how small length or index changes in one arm translate directly into intensity swings — the basis of optical sensing, some quantum optics demonstrations, and photonic modulators. Assumptions include ideal 50/50 splitters, perfect spatial mode overlap, no dispersion or polarization filtering, and a point detector. Real devices must manage vibration, thermal drift, and wavelength-dependent phase; those effects are not modeled here.

Who it's for: Undergraduate optics and photonics students learning two-beam interference, metrology, and interferometer topologies beyond the Michelson layout.

Key terms

Mach–Zehnder
Optical path difference
Interference
Beamsplitter
Coherence
Fringe visibility
Phase sensitivity
Two-beam model

How it works

Two-beam interferometer: recombining amplitudes with a controllable phase delay in one arm shifts output power — basis for sensing and photonics.

Key equations

I/I_max = cos²(π Δ / λ) (balanced, equal amplitudes)

Frequently asked questions

How does this differ from a Michelson interferometer?: Both split and recombine beams, but the Mach–Zehnder uses two separate paths that meet at a second beamsplitter. That geometry can make it easier to send each arm through different samples or modulators without sending light back toward the source, and it offers distinct output ports corresponding to complementary interference phases.
Why is the intensity periodic in Δ with period λ?: Each wavelength of OPD adds 2π to the phase difference between the arms. Because intensity depends on cos² of half the phase (for equal amplitudes), a full 2π phase cycle returns the same intensity; equivalently, the pattern repeats when Δ changes by one wavelength.
What happens if the amplitudes are not perfectly balanced?: The fringe contrast (visibility) drops. The general two-beam intensity is not symmetric about zero when one arm is weaker; minima are no longer exactly zero. This simulator fixes equal amplitudes to isolate the phase dependence.
Does the schematic include dispersion in the beamsplitters?: No. Real cube beamsplitters and coatings introduce wavelength-dependent phase; broadband sources then wash out fringes unless compensated. Here λ is a single design wavelength and materials are ideal.

Other simulators in this category — or see all 50.

View category →

NewUniversity / research

Fabry–Pérot Cavity / Etalon

Launch Simulator

Airy transmission T(L), finesse F, free spectral range FSR = λ/(2n), linewidth ΔL ≈ FSR/F; animated cavity-length scan.

NewUniversity / research

Gaussian Beam & q-Parameter

Launch Simulator

Waist w₀, Rayleigh range z_R, Gouy phase ζ(z), and ABCD propagation through a thin lens; w(z) and q at observation plane.

NewUniversity / research

Fourier Optics 4f System

Launch Simulator

Object → lens → Fourier plane → spatial filter (low/high-pass) → lens → image; 2D FFT with circular aperture in spectrum.

NewUniversity / research

Zernike Wavefront Aberrations

Launch Simulator

Defocus, astigmatism, coma, spherical Zernike modes on a circular pupil; PSF via FFT, Strehl ratio and Maréchal estimate.

NewUniversity / research

Optical Tweezers (Gradient Trap)

Launch Simulator

Gaussian trap U(r), stiffness k ∝ P/w₀², overdamped Langevin bead with Stokes drag and thermal Brownian jitter; ⟨r²⟩ trace.

NewUniversity / research

1D Photonic Crystal Bandgap

Launch Simulator

Alternating n₁/n₂ bilayers; TE transfer-matrix T(λ), R(λ); stop bands where |Tr(M_period)/2|>1 vs period Λ.

Mach–Zehnder Interferometer

Live graphs

About this model

How it works

Key equations

Frequently asked questions

More from Optics & Light

Fabry–Pérot Cavity / Etalon

Gaussian Beam & q-Parameter

Fourier Optics 4f System

Zernike Wavefront Aberrations

Optical Tweezers (Gradient Trap)

1D Photonic Crystal Bandgap

Mach–Zehnder Interferometer

Live graphs

About this model

How it works

Key equations

Frequently asked questions