⚛️ quantum physics

High-Dimensional Maximally Entangled OAM States by Joint Pump and Modal-Space Engineering

This paper demonstrates that by jointly engineering the pump beam profile and the selected orbital angular momentum (OAM) subspace, it is possible to generate maximally entangled states of arbitrary dimensionality $d$ in spontaneous parametric down-conversion, overcoming the typical non-uniform mode distribution that limits entanglement quality.

Original author: Richard Bernecker, Baghdasar Baghdasaryan, Fabian Steinlechner, Stephan Fritzsche

Published 2026-11-10

📖 4 min read🧠 Deep dive

CC BY 4.0

Original author: Richard Bernecker, Baghdasar Baghdasaryan, Fabian Steinlechner, Stephan Fritzsche

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the author. For technical accuracy, refer to the original paper. Read full disclaimer

In the realm of quantum information, light is more than just a carrier of images; it is a versatile tool for encoding data. While traditional communication often relies on simple on-off switches to represent zeros and ones, quantum physicists have long sought to pack more information into single particles of light. One promising way to do this involves twisting the light itself. Imagine a beam of light not just moving forward, but spiraling as it travels, much like a corkscrew. This twisting motion carries a property called orbital angular momentum. Because this twist can take on many different strengths and directions, it offers a vast, almost infinite library of distinct states. By using these twisted states, scientists can create pairs of light particles that are linked in a profound way, known as entanglement. When two particles are maximally entangled, the state of one instantly reveals the state of the other, no matter how far apart they are. This connection is the backbone of future quantum networks, promising communication that is not only faster but also fundamentally secure against eavesdropping.

However, creating these perfect links is surprisingly difficult. When scientists generate these twisted light particles using a standard process involving special crystals, the resulting pairs are rarely perfectly balanced. Instead of producing an equal mix of all possible twists, the process tends to favor certain configurations while leaving others weak or missing. It is as if a factory intended to produce a perfect set of matching keys ended up making mostly one size, with a few others scattered around. This unevenness means the connection between the particles is not as strong as it could be, limiting the amount of information they can carry and the reliability of the system. For years, researchers have tried to fix this by carefully shaping the light that enters the crystal, but previous attempts were often limited to small, simple sets of twists or required complex, trial-and-error adjustments that did not work for larger, more powerful systems.

A team of researchers has now developed a new method to overcome these limitations, allowing for the creation of perfectly balanced, high-dimensional entangled states with a level of control previously thought difficult to achieve. Their approach involves a dual strategy: they carefully design the shape of the incoming light beam and simultaneously select a specific, tailored group of twisted states to work with. By doing both at once, they can force the system to produce exactly the right mix of particle pairs, ensuring that every possible state in their chosen group appears with equal strength. This balance is crucial because it creates what is known as a maximally entangled state, where the connection between the particles is as strong and uniform as physics allows.

The researchers demonstrated that this technique works for systems of varying sizes, from small groups of three states up to much larger groups containing twelve or more. They found that for smaller groups, they could arrange the states in a continuous sequence, but as they tried to build larger groups, the natural spread of the light's properties began to cause unwanted overlaps. To solve this, they introduced the concept of a "disassembled" group. Instead of trying to use every twist in a continuous range, they skipped over certain intermediate twists, leaving gaps in the sequence. By carefully choosing which twists to include and which to skip, and by adjusting the strength of the different components of the incoming light, they were able to isolate the perfect pairs and eliminate the noise that usually plagues these experiments.

This method is significant because it does not rely on guessing or complex computer optimizations to find the right settings. Instead, the researchers established a clear set of rules for how to construct the light beam and select the states, allowing them to predict exactly what would happen before they even turned on the equipment. They showed that by following these rules, they could generate pure, high-quality entangled states with a high degree of certainty. While the efficiency of the process naturally decreases as the size of the group increases, the ability to reach dimensions of twelve and beyond with a clean, predictable outcome represents a major step forward. The work proves that by engineering both the source of the light and the specific states being used, scientists can now reliably create the complex, high-capacity quantum links needed for the next generation of communication technologies.

Technical Summary: High-Dimensional Maximally Entangled OAM States by Joint Pump and Modal-Space Engineering

Problem Statement
Spontaneous parametric down-conversion (SPDC) inherently generates correlated photon pairs with high-dimensional spatial entanglement in orbital angular momentum (OAM) modes. However, the resulting two-photon states are typically not maximally entangled (MES) due to a highly non-uniform distribution of generated OAM modes. While many quantum information protocols require MESs (the high-dimensional analogues of Bell states) for optimal fidelity and error resilience, existing methods face significant limitations. Crystal engineering requires tailored nonlinear media optimized for fixed dimensions, and post-generation schemes like entanglement concentration add complexity without achieving genuine at-source engineering. Furthermore, while pump-beam engineering via spatial light modulators (SLMs) offers programmable control, prior works have largely been restricted to low-dimensional subspaces or relied on technically demanding numerical optimization without establishing general conditions for maximal entanglement in arbitrary dimensions ( $d > 2$ ).

Methodology
The authors propose a systematic approach to generate MESs in $d \times d$ OAM subspaces by jointly engineering the pump beam structure and the local OAM subspace. The method relies on the following theoretical framework:

Pump Engineering: The pump beam is modeled as an arbitrary coherent superposition of Laguerre-Gaussian (LG) modes, $V_p = \sum a_{\ell_p} \text{LG}(\ell_p, w_p)$ , where the coefficients $a_{\ell_p}$ and the waist $w_p$ are design parameters. In the low-gain regime, the generated two-photon state is a coherent sum of contributions from each pump OAM component.
Spectral Structure: Each pump component with OAM index $\ell_p$ populates a specific anti-diagonal in the OAM spectrum (where $\ell_s + \ell_i = \ell_p$ ). The relative weights $a_{\ell_p}$ control the amplitude of these anti-diagonals.
The "Sudoku Condition": To achieve an MES in a $d \times d$ subspace, the authors define a specific criterion: the state must be a coherent superposition of exactly $d$ orthogonal mode pairs, each with an amplitude of $1/\sqrt{d}$ , such that there is exactly one populated mode in every row and column of the subspace. All other mode amplitudes within the subspace must vanish.
Subspace Selection: The authors distinguish between continuous OAM subspaces (modes in a continuous interval) and disassembled OAM subspaces (modes in a discontinuous range). They demonstrate that by selecting specific pump modes and carefully choosing which signal and idler modes to include in the measurement subspace, unwanted contributions from the finite "spiral bandwidth" of the pump can be excluded.

Key Results
The paper presents analytical constructions and numerical simulations demonstrating the generation of MESs for arbitrary dimensions:

Continuous Subspaces: For continuous subspaces, the finite spiral bandwidth imposes a dimensionality limit. The authors show that MESs with high global fidelity can be constructed up to $d = 6$ . Beyond this, the finite bandwidth of the pump modes inevitably introduces non-negligible contributions from neighboring modes within the target subspace, violating the Sudoku condition and reducing the Schmidt number.
Disassembled Subspaces: To overcome the $d=6$ $d = 6$ limit, the authors introduce the concept of disassembled subspaces. By omitting specific pump components (e.g., removing the $a_0$ $a_{0}$ component) and selecting only the OAM modes that support the resulting degenerate central amplitudes, they can construct MESs for higher dimensions (demonstrated up to $d=12$ $d = 12$ ).
- For example, using pump components $a_{\pm 3}$ and $a_{\pm 7}$ allows for the construction of an $8 \times 8$ MES by selecting a subspace that excludes the modes generated by the "gap" between these pump indices.
- The separation between pump OAM indices must be sufficient to ensure that the target mode pairs do not overlap with unwanted neighboring modes.
Fidelity Trade-offs: While the method enables arbitrary dimensionality, the global fidelity ( $F_{\text{glob}}$ ) generally decreases as $d$ increases. This is because larger dimensions require larger separations between pump modes, which results in wider spiral bandwidths and more non-vanishing modes outside the target subspace. However, the subspace fidelity ( $F_{\text{sub}}$ ) remains high, ensuring the state is maximally entangled within the selected subspace.
Purity: Unlike approaches using higher-order radial modes to flatten spectra (which may introduce correlations with unresolved degrees of freedom), this method ensures spatial-spectral separability by selecting target mode pairs that share the same spectral dependence.

Significance and Claims
The paper claims to provide a "broadly applicable approach" for generating high-dimensional MESs without the need for numerical optimization of the pump profile. The primary contribution is the identification of the interplay between the spatial pump structure and the choice of OAM subspace, which allows for the direct analytical determination of the required pump profile for a target MES.

The authors emphasize that their method:

Is not restricted to specific phase-matching types or nonlinear media.
Ensures the spectral and spatial purity of the generated state.
Provides a systematic construction for generating pure MESs in flexibly chosen OAM subspaces of, in principle, arbitrary dimensionality.
Avoids the complexity of post-generation entanglement concentration or the fabrication of tailored crystals.

The work establishes that maximal entanglement with OAM requires a careful, nontrivial design of the effective high-dimensional Hilbert subspace, moving beyond simple pump shaping to a joint optimization of the pump and the measurement basis.

More like this