Pseudo-Thermal Light Source

• Pseudo-thermal light sources are engineered systems that mimic the statistical properties of thermal light by imposing rapid, random phase and amplitude fluctuations on coherent lasers.
• They are typically realized through rotating ground-glass diffusers or mechanically perturbed fibers, producing Gaussian speckle fields with characteristic thermal bunching (g²(0) ≈ 2).
• These sources are vital in quantum optics, enabling techniques such as ghost imaging, optical coherence tomography, and advanced coherence engineering in precision sensing.

A pseudo-thermal light source is an engineered optical system that replicates the photon statistical and coherence properties of true thermal radiation while offering control, directionality, and convenience for laboratory and technological applications. The canonical pseudo-thermal source is realized by taking a coherent, narrow-band laser and imposing on it rapid, stochastic spatial and temporal phase and amplitude fluctuations, typically via a rotating ground-glass diffuser or, alternatively, a mechanically perturbed multimode fiber. This process produces a speckle ensemble with Gaussian field statistics, thereby emulating the bunching, higher-order correlations, and intensity-fluctuation properties characteristic of thermal light. Pseudo-thermal light sources are foundational tools in quantum optics, intensity interferometry, ghost imaging, phase microscopy, and optical coherence tomography.

1. Physical Principles and Standard Implementations

Pseudo-thermal light is generated by scattering a stable laser (e.g., He–Ne or diode, λ = 500–800 nm) from a dynamic, randomizing medium such as a rotating ground-glass (RG) disk, milk-in-water cell, or a mechanically agitated multimode fiber. The classical configuration directs the laser beam onto the RG at a controlled incident spot size; the scattered light forms a rapidly evolving speckle field, ensuring spatially and temporally stochastic intensities (Rai et al., 2024, Mehringer et al., 2016, Barge et al., 2022). The field statistics approach zero-mean circular complex Gaussianity at each point due to the central-limit effect from many scatterers or fiber modes.

The mutual coherence function is governed by the illuminated area (spatial coherence) and the speckle decorrelation time (temporal coherence). Explicitly, for a rotating RG source, the spatial coherence length at a propagation distance z_eff is

$l_c \approx \frac{\lambda z_\text{eff}}{D},$

where D is the illuminated spot diameter (Rogalski et al., 2022). Temporal coherence time $\tau_c$ is set by the decorrelation rate of the RG (i.e., its tangential velocity), typically tunable from microseconds to hundreds of milliseconds (Claveria et al., 9 Jan 2025, Barge et al., 2022). Pseudo-thermal speckles exhibit an exponential (thermal) intensity distribution,

$$P(I) = \frac{1}{\langle I\rangle} \exp\left(-\frac{I}{\langle I\rangle}\right),$$

leading to the central Hanbury Brown–Twiss effect, $g^{(2)}(0)=2$ (Rai et al., 2024, Barge et al., 2022).

Alternative configurations replace the RG with a multimode fiber whose modal phases are randomized by gentle flexure, achieving identical Gaussian random field statistics and photon-number (Bose–Einstein) distributions (Mehringer et al., 2016).

2. Theoretical Framework: Statistics and Coherence Functions

Pseudo-thermal light retains the narrow spectral bandwidth of the parent laser but reinstates the stochasticity of thermal fields. The field correlation function takes the form

$g^{(1)}(\tau) = e^{-|\tau|/\tau_c},$

where $\tau_c$ is the coherence time (Tan et al., 2023, Rai et al., 2024). The pivotal normalized second-order (intensity) correlation function follows the Siegert relation:

$$g^{(2)}(\tau) = 1 + |g^{(1)}(\tau)|^2 = 1 + e^{-2|\tau|/\tau_c},$$

with $g^{(2)}(0) = 2$ for a pure pseudo-thermal regime (Rai et al., 2024, Zhou et al., 2018).

For higher-order correlations (classically and in Glauber’s quantum formalism), the n-th order bunching scales with $n!$; for example, $g^{(3)}(0,0,0)=6$ in ideal conditions (0909.3512, Chen et al., 2013).

The two-photon (and higher-photon) interference effects following a pseudo-thermal source can be interpreted via Feynman path integrals or classical Gaussian moment analysis. Each detection event sums over all indistinguishable photon permutations; the enhancement in multiphoton detection probability is a direct consequence of this symmetrization (0909.3512, Zhou et al., 2017).

3. Superbunching and Engineered Intensity Statistics

By externally modulating the laser intensity before the RG, the degree of second- and higher-order coherence can be greatly increased above the thermal bounds—a phenomenon termed superbunching. Modulation can be deterministic (binary distribution, sinusoidal, or white noise) via an electro-optic modulator (EOM). For a binary intensity distribution, if the intensity takes $I_1$ (probability $p$) and $I_2$ (probability $1-p$), the second-order coherence at zero delay is

$$g^{(2)}(0) = \frac{p I_1^2 + (1-p) I_2^2}{[p I_1 + (1-p) I_2]^2} \, .$$

As the contrast $R=I_1/I_2$ increases and $p \rightarrow 0$ or $1$, $g^{(2)}(0)$ becomes arbitrarily large (Liu et al., 2021).

Experimentally, values as high as $g^{(2)}(0) = 20.45 \pm 0.10$ and $g^{(3)}(0) = 227.07 \pm 2.0$ were achieved with binary modulation at $p=0.05$ and $R\approx 60.8$ (Liu et al., 2021). By contrast, classical pseudothermal light yields $g^{(2)}(0) = 2$ and $g^{(3)}(0) = 6$ (0909.3512).

Classically, cascades of $N$ rotating ground glasses or intensity modulator stages yield $g^{(2)}(0)=2^N$ and $g^{(n)}(0) \propto (n!)^N$ (Zhou et al., 2017). Modulator-based superbunching produces high-order bunching without compromising statistical stability, in contrast to stochastic uncertainties inherent to cascaded RGs (Liu et al., 2021, Zhou et al., 2018).

4. Experimental Realizations and Characterization

A prototypical setup comprises:

• Laser Source: Single-mode, CW (λ = 632–795 nm), with variable attenuator (Zhou et al., 2018).
• Intensity Modulation (optional): EOM plus polarizers, driven by waveform generator (Liu et al., 2021, Zhou et al., 2018).
• Diffuser: Rotating ground-glass (typical grit 5–1500 μm) mounted on a motorized spindle (Barge et al., 2022, Rogalski et al., 2022).
• Collection and Filtering: Multi-mode fiber, iris, and lens for spatial homogenization (Ahmad et al., 2018).
• Detection: Single-photon avalanche diodes (timing jitter ≲100 ps); CMOS/CCD cameras for spatial statistics (0909.3512, Mehringer et al., 2016, Barge et al., 2022).

Speckle decorrelation time $\tau_c$ is measured via temporal autocorrelation of the intensity ($g^{(2)}(\tau)$) on a single pixel or detector. For rotating RG systems, $\tau_c$ can be tuned via rotational speed and beam spot size from μs to 10–200 ms or more (Claveria et al., 9 Jan 2025, Rogalski et al., 2022). The spatial coherence length, set by the spot size on the RG or the fiber core diameter, determines the transverse speckle size (Mehringer et al., 2016). Negative exponential intensity and Bose–Einstein photon counts in a speckle are routinely experimentally verified (Barge et al., 2022).

5. Applications in Imaging, Interferometry, and Sensing

Pseudo-thermal light sources are foundational in a broad suite of quantum and classical optical methodologies:

• Intensity Interferometry and Optical Coherence Tomography (OCT): Pseudo-thermal sources provide spatially extended, temporally coherent fields with tunable longitudinal coherence. In high-NA OCT, the longitudinal coherence length can be made independent of the parent laser's coherence, enabling axial resolutions down to 650 nm; pseudo-thermal sources thus combine monochromaticity (avoiding chromatic dispersion) with short longitudinal coherence gates (Ahmad et al., 2018).
• Ghost Imaging: Both spatial and temporal ghost imaging schemes exploit the correlation structure of pseudo-thermal light. Superbunching enhances image visibility—for example, $V = (g^{(2)}(0) - 1)/(g^{(2)}(0) + 1)$ can approach unity for modulator-based superbunching, compared to $V=1/3$ for standard pseudothermal light (Zhou et al., 2017, Liu et al., 2021). Temporal ghost imaging using deterministic pseudo-thermal speckle patterns enables single-shot recovery of high-bandwidth non-reproducible signals (Devaux et al., 2016).
• Quantum Sensing: Ultrabright pseudo-thermal sources (e.g., sub-threshold laser diode and spectral filtering) have been used for kilometer-scale optical time-of-flight range finding based on Hanbury Brown–Twiss photon bunching (Tan et al., 2023).
• Phase and Shadow Imaging: In phase microscopy, pseudo-thermal illumination suppresses coherent artifacts and enables high signal-to-noise, single-shot quantitative phase mapping (Hilbert phase microscopy) (Rogalski et al., 2022). Quadrature-noise shadow imaging leverages the super-Poissonian photon statistics for sensitive detection in the few-photon regime (Barge et al., 2022).
• Coherence Engineering: Source parameters (spot size, modulation, diffuser speed) allow the systematic control of coherence functions and photon statistics for tailored experiments in quantum optics, multiphoton interference, and intensity fluctuation studies (Chen et al., 2013, Claveria et al., 9 Jan 2025).

6. Variations, Constraints, and Comparative Advantages

Pseudo-thermal sources offer high brightness, laboratory accessibility, and tunability unattainable with true thermal sources (e.g., blackbody), and are straightforward to implement at any wavelength via an appropriate diffuser (Barge et al., 2022). Fiber-based pseudo-thermal sources offer inherent directionality, simplicity, and reconfigurability with comparable Gaussian statistics (Mehringer et al., 2016).

Constraints include mechanical complexity for moving parts (diffuser, fiber agitation), finite speckle lifetime for high-speed applications, and for some schemes, limited total power per solid angle compared to superluminescent diodes (Ahmad et al., 2018). In high-photon-number or rapid-frame regimes, detector noise and dark counts may degrade the measured statistics (e.g., reducing the observed $g^{(2)}(0)$ below 2) (Tan et al., 2023, Barge et al., 2022). For imaging, the achievable spatial and temporal resolutions are set by the speckle scale, decorrelation rates, and detector integration times.

7. Outlook: Quantum-Classical Transition and Advanced Coherence Engineering

Pseudo-thermal sources bridge the classical-quantum boundary in light-matter interactions. Their tunable bunching statistics have enabled direct studies of multiphoton interference, the role of higher-order correlations, and the experimental discrimination between classical and quantum interpretations of coherence (0909.3512, Zhou et al., 2017, Zhou et al., 2018). The ability to realize arbitrary photon statistics via premodulation opens a path to engineered light sources for advanced quantum metrology, sensing, and imaging modalities where contrast, background suppression, and multiphoton enhancement are critical (Liu et al., 2021, Claveria et al., 9 Jan 2025).

Pseudo-thermal light sources remain a central experimental platform for developing and benchmarking both foundational and applied quantum-optical protocols, as well as for exploring the boundary regimes between classical statistical optics and quantum coherence phenomena.