Erasing Photon-Number Correlations through Hong-Ou-Mandel Interference

Light behaves in two distinct ways that have long shaped how scientists understand the quantum world. On one hand, it acts like a continuous wave, where properties such as brightness and phase can be measured with extreme precision, revealing subtle effects like squeezing, where the uncertainty of a measurement is pushed below the standard limit in one direction while increasing in another. On the other hand, light behaves as a stream of individual particles called photons, where researchers count them one by one to study discrete events. For decades, these two perspectives were treated as separate fields of study, each with its own tools and theories. However, modern quantum technologies, which aim to build powerful new computers and sensors, are beginning to require a blend of both approaches. To make these hybrid systems work, scientists need to understand exactly how light behaves when it is manipulated in ways that involve both its wave-like and particle-like nature, particularly when dealing with sources that generate pairs of photons.

In a recent study, researchers set out to resolve a specific confusion about what happens when a special type of light source is split in half. This source, known as a parametric down-conversion generator, creates a state of light where two beams are deeply linked, or entangled, meaning the properties of one beam instantly reveal the properties of the other. When this entangled light is sent into a beam splitter—a device that divides a beam of light into two paths—the outcome depends on how you look at it. If you view the light as a continuous wave, the splitting process transforms the entangled pair into two separate, independent beams of squeezed light. But if you view the light as a stream of particles, the same process is often described as creating a highly entangled state where the photons are forced to travel together. The researchers wanted to know which description was actually correct when looking at the raw numbers of photons in the output beams, without making any assumptions or discarding data.

To answer this, the team built an experiment using a crystal that generates pairs of photons and sent them into a fiber-optic beam splitter. They carefully controlled the timing of the light pulses so that they arrived at the splitter at the exact same moment, a condition known as perfect overlap. They then used highly sensitive detectors capable of counting the exact number of photons in each output beam, rather than just detecting whether a photon was present or absent. By measuring millions of these events, they were able to map out the joint probability of finding specific numbers of photons in both beams simultaneously. The results were clear: when the light pulses overlapped perfectly, the photon numbers in the two output beams became completely independent. The data showed that the number of photons in one beam had no bearing on the number in the other, effectively erasing the correlations that existed before the split.

This finding confirms that the two seemingly different descriptions of the system are actually consistent, provided that one looks at the full picture without filtering out the data. The apparent entanglement seen in some particle-based experiments arises only when researchers ignore the vacuum state—the moments when no photons are detected—and focus solely on the instances where photons are found. When the entire statistical distribution is considered, including the empty moments, the output is revealed to be two separate, uncorrelated sources of squeezed light. The researchers verified this by comparing their experimental data with computer simulations that included the specific losses and imperfections of their setup. The match between the real-world measurements and the theoretical model was remarkably close, confirming that the source they used produced a high-quality, single-mode state of light.

The study also explored what happens when the timing is slightly off. When the pulses arrive at the beam splitter with a delay, the interference is incomplete, and the photon numbers in the two beams begin to show correlations again. The researchers measured this transition by varying the delay in tiny increments, observing how the statistical link between the beams grew stronger as the overlap decreased. They found that even a delay as small as a few picoseconds was enough to break the perfect independence of the output beams. This sensitivity to timing highlights the precision required in quantum experiments and demonstrates that the "erasure" of correlations is a delicate balance that depends entirely on the perfect synchronization of the light waves.

Beyond settling a theoretical debate, this work provides a practical tool for characterizing the light sources needed for advanced quantum computing. Many of the most promising quantum algorithms, such as Gaussian boson sampling, rely on having independent, high-quality sources of squeezed light. The methods developed in this study allow scientists to directly assess whether a light source is truly producing independent beams or if hidden correlations remain. By measuring the full distribution of photon numbers, rather than just looking for specific patterns, researchers can now verify the quality of their resources with greater confidence. The ability to confirm that a source is truly separable in the photon number picture is a crucial step toward building larger, more complex quantum networks that can perform tasks impossible for classical computers.

The researchers also noted that their results align with the idea that the two descriptions of light are not contradictory but rather complementary views of the same physical reality. The confusion in the past arose because different measurement techniques naturally emphasized different aspects of the light. By using detectors that can count photons with high resolution and by analyzing the complete data set, the team was able to show that the output is indeed separable. This clarity is essential for the future of hybrid quantum systems, where the interplay between continuous and discrete variables is the key to unlocking new capabilities. The experiment serves as a demonstration that with the right tools, the complex behavior of quantum light can be understood in a unified way, bridging the gap between the wave and particle pictures.

In the end, the work confirms that when two beams of squeezed light interfere perfectly, they emerge as two independent streams, each carrying its own quantum properties without being tied to the other. This independence is not just a theoretical ideal but a measurable reality that can be observed and quantified. The study provides a robust method for checking the quality of these light sources, ensuring that the building blocks of future quantum technologies are exactly what they are supposed to be. As the field moves toward more complex networks, the ability to verify the separability of light sources will become increasingly important, allowing scientists to construct systems that operate with the precision and reliability required for the next generation of quantum devices.