Photon statistics of superbunching pseudothermal light

This paper investigates the photon statistics of superbunching pseudothermal light using single-photon detectors, revealing that higher degrees of second-order coherence lead to significant deviations from thermal light distributions in the tail of the photon distribution, thereby offering insights into two-photon superbunching and potential applications in generating non-Rayleigh temporal speckles.

The Gist

Imagine light not as a smooth, continuous beam, but as a stream of tiny, invisible marbles called photons. Usually, when we talk about "thermal light" (like the light from a lightbulb or the sun), these marbles arrive in a somewhat predictable, random pattern. Scientists call this a "geometric distribution." It's like rain falling on a roof: sometimes you get a few drops close together, sometimes a gap, but generally, they follow a standard statistical rule.

This paper is about a special, "supercharged" version of this light called superbunching pseudothermal light. Here is a simple breakdown of what the researchers did and found:

1. The Setup: Making Light "Bunch"

The researchers started with a standard laser beam (which is usually very orderly, like soldiers marching in a straight line). To make it act like chaotic thermal light, they shone it through a rotating ground glass (think of it like a spinning, rough piece of sandpaper). This scrambles the light, making the photons arrive in random clusters. This is called "pseudothermal light."

But they didn't stop there. They added a twist: they modulated the intensity of the laser before it hit the spinning glass. Imagine turning the volume knob on a speaker up and down very quickly while the sound is being scrambled. This created a "superbunching" effect. The photons didn't just cluster randomly; they formed massive, explosive groups.

2. The Discovery: Breaking the Rules

The team wanted to know: What does the "marble count" look like for this super-clustered light?

• Normal Light: If you count the photons in a tiny slice of time, most of the time you get zero or one. Occasionally, you get two. The probability of getting a huge number (like 10 or 20) drops off very quickly. It's like a bell curve that tapers off fast.
• Superbunching Light: The researchers found that this light behaves differently. While it still mostly has zero or one photon, the "tail" of the distribution is much heavier. This means there is a much higher chance of catching a huge burst of photons all at once compared to normal light.

The Analogy:
Imagine two people rolling dice.

• Person A (Normal Light): Rolls mostly 1s and 2s. Occasionally a 3. It is extremely rare to roll a 6.
• Person B (Superbunching Light): Rolls mostly 1s and 2s too, but every now and then, they roll a 6, a 7, or even a 10. The "long tail" of high numbers is much more common than it should be.

3. The Measurement: Counting the Marbles

To prove this, they used a special setup (a Hanbury Brown-Twiss interferometer) with two detectors. They measured two things:

1. How "bunched" the light is: They measured the "degree of second-order coherence." For normal light, this number is 2. For their super-light, it went up to over 3. This confirmed the light was "superbunching."
2. The exact count: They counted exactly how many photons arrived in tiny time windows.

They found that as the "bunching" number got higher, the distribution of photon counts got "fatter" at the high end. The more the light wanted to bunch, the more likely it was to deliver a massive packet of photons in a single moment.

4. The Conclusion: New Types of "Speckles"

When light hits a rough surface, it creates a glittery pattern called a "speckle" (like the grainy look on a laser pointer dot on a wall).

• Normal Light: Creates "Rayleigh speckles," which follow the standard rules of randomness.
• Superbunching Light: The paper suggests this light creates "Non-Rayleigh temporal speckles."

Think of it this way: If you look at the glittery pattern on the wall over time, normal light flickers in a standard way. But this super-light flickers with explosive, unpredictable bursts of brightness that don't follow the usual rules.

Summary

The paper doesn't claim this will cure diseases or power new engines. Instead, it simply says: "We made a special kind of light that clusters photons in a way we haven't fully mapped out before. We measured it, and we found that it breaks the standard rules of how photons usually distribute themselves, creating a 'heavy tail' where massive photon bursts are more common than expected."

This helps scientists better understand the physics of how light particles interact and cluster, which is useful for advanced experiments in quantum optics and imaging.

Technical Summary

1. Problem Statement

• Context: Thermal and pseudothermal light are fundamental to quantum optics, particularly for studying second- and higher-order interference (e.g., the Hanbury Brown-Twiss effect). Standard pseudothermal light follows a geometric photon distribution (Rayleigh speckles), characterized by a degree of second-order coherence, $g^{(2)}(0)$, of approximately 2.
• The Gap: Recent advancements have introduced "superbunching" pseudothermal light sources (using multiple rotating ground glasses or intensity modulation) capable of achieving $g^{(2)}(0) \gg 2$. While the coherence properties of this light are known, its photon statistics (the full probability distribution of photon numbers, $P(n)$) had not been experimentally characterized.
• The Challenge: A high $g^{(2)}(0)$ value alone is insufficient to distinguish between different types of light (classical vs. non-classical, or different classical distributions). Understanding the full photon distribution is necessary to differentiate superbunching light from standard thermal light and to understand the underlying physics of multi-photon bunching.

2. Methodology

The authors employed an experimental setup combining intensity modulation and single-photon detection to characterize the light source.

• Light Source Generation:
  • A single-mode continuous-wave laser (780 nm) was passed through an Electro-Optical Modulator (EOM) driven by white noise from a signal generator.
  • The modulated beam was focused onto a rotating ground glass (RG) to generate pseudothermal light.
  • By varying the peak-to-peak voltage ($V_{pp}$) of the noise signal, the authors controlled the intensity fluctuations, transitioning the source from standard pseudothermal light ($V_{pp}=0$) to superbunching pseudothermal light ($V_{pp} > 0$).
• Measurement Systems:
  1. Hanbury Brown-Twiss (HBT) Interferometer: Used to directly measure the degree of second-order coherence, $g^{(2)}_m(0)$, via two-photon coincidence counting.
  2. Single-Photon Detection: One arm of the HBT setup was used to measure the photon number distribution ($P(n)$) by counting photons within specific time windows ($T = 5 \mu s$).
• Data Analysis:
  • The measured $P(n)$ was used to calculate the theoretical degree of second-order coherence, $g^{(2)}_c(0)$, using the formula:
    $$g^{(2)}_c(0) = \frac{\langle n(n-1) \rangle}{\langle n \rangle^2} = \frac{\sum P(n) \cdot n(n-1)}{(\sum P(n) \cdot n)^2}$$
  • The experimental $P(n)$ was compared against the theoretical geometric distribution (characteristic of Rayleigh speckles) with the same mean photon number.

3. Key Contributions

• First Characterization of Photon Statistics: This paper provides the first experimental measurement of the full photon number distribution for superbunching pseudothermal light.
• Quantitative Correlation: It establishes a direct link between the magnitude of $g^{(2)}(0)$ and the deviation of the photon distribution from the geometric (Rayleigh) model.
• Validation of Non-Rayleigh Speckles: The study confirms that superbunching light generates non-Rayleigh temporal speckles, distinguishing them from standard spatial non-Rayleigh speckles found in previous literature.

4. Key Results

• Photon Distribution Deviation:
  • For standard pseudothermal light ($V_{pp}=0$), the photon distribution follows a geometric distribution with $g^{(2)}_c(0) \approx 1.89$.
  • For superbunching light ($V_{pp} = 5, 7.5, 10$ V), the distribution deviates significantly. As $g^{(2)}_c(0)$ increases (up to $\approx 3.12$), the probability of detecting large numbers of photons (the "tail" of the distribution) increases substantially compared to the geometric prediction.
• Consistency of Measurements:
  • The $g^{(2)}(0)$ values calculated from the photon distribution ($g^{(2)}_c$) were consistent with the values directly measured by the HBT interferometer ($g^{(2)}_m$), with minor discrepancies attributed to the collection time window ($5 \mu s$) being slightly longer than the coherence time ($1.28 - 4.63 \mu s$).
• Temporal vs. Spatial Behavior:
  • The superbunching effect creates non-Rayleigh temporal speckles because the intensity of the input laser fluctuates rapidly in time, correlating the phases of the superposed fields.
  • However, at any single instant, the spatial distribution remains Rayleigh because there are no correlations between intensities at different points on the rotating ground glass.

5. Significance and Implications

• Fundamental Physics: The results deepen the understanding of two-photon superbunching with classical light, demonstrating that high-order coherence is accompanied by specific, non-geometric photon statistics.
• Application in Quantum Optics: The ability to generate non-Rayleigh temporal speckles opens new avenues for:
  • Ghost Imaging: Specifically high-visibility temporal ghost imaging.
  • Multi-photon Interference: Providing a classical light source with tunable, high-order correlation properties for studying multi-photon interference phenomena without requiring complex non-classical sources.
• Methodological Utility: The study suggests that analyzing the full photon distribution is a more robust method for characterizing complex light sources than relying solely on the second-order coherence parameter.