Constructing and characterising anharmonic potentials in a harmonic oscillator

Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.

Kai: We're looking at the paper: "Constructing and characterising anharmonic potentials in a harmonic oscillator".

Mira: Here's the gist of the paper. The gist The researchers demonstrate a systematic framework to implement programmable non-Gaussian phase gates on a superconducting harmonic oscillator using bosonic quantum signal processing, which provides a practical and reconfigurable route towards continuous-variable quantum information processing and analogue quantum simulation of molecular and chemical dynamics<ref:2609.02405#pg2> How it works The framework leverages bosonic quantum signal processing (QSP) to engineer the Fourier series of a target phase gate as a circuit of interleaved qubit rotations and qubit-conditioned oscillator displacements<ref:2609.02405#pg14>.

Kai: This approach allows reprogramming the target by only changing a different set of qubit rotations within the same circuit template, which is more efficient than constructing higher-order non-Gaussian operations with nontrivial commutation relations<ref:2609.02405#pg2> The compiled circuit is then represented as UˆM(Xˆ) = (Y M j=1 Rϕj (θj) CD(γCD)) Rϕ0 (θ0), where CD represents the qubit-conditioned displacement and Rϕj are single-qubit rotations<ref:2609.02405#pg12>.

Mira: Gate Implementation and Characterization The method is used to experimentally demonstrate representative cubic, double-well, and Morse potentials by varying the qubit rotation angles in a fixed calibrated circuit architecture<ref:2609.02405#pg2> The cubic phase gate, a key ingredient for universal quantum information processing, was implemented with order M = 11 using the potential Vcubic(Xˆ) = 0.6Xˆ 3<ref:2609.02405#pg13>.

Kai: These potentials are characterized by sampling their force profile using a pointwise force reconstruction protocol that leverages the Ehrenfest theorem, where the change in momentum enacted by the gate on a given input state is a sample of the force −V′ at the position of that state<ref:2609.02405#pg16>.

Title and authors: Kai: So this paper is all about building a way to create these kinds of anharmonic potentials on a superconducting harmonic oscillator using bosonic quantum signal processing. It’s about making these non-Gaussian operations that are usually really hard to engineer by just changing some qubit rotations in a fixed circuit template.

Mira: Exactly, it moves away from having to design entirely new circuits for every potential you want, and instead uses this Fourier series representation of the target phase gate as a structure for reprogramming. It suggests a more systematic way to implement these things.

Lev: From an error correction standpoint, what I’m hearing is that we can define the operation in terms of qubit rotations and qubit-conditioned displacements. That means the complexity is localized within those specific components, which might simplify how you think about running it on real hardware.

Kai: Right, and they show that this approach lets you test different types of potentials experimentally, like cubic, double-well, and Morse potentials by just adjusting those qubit rotation angles in the circuit architecture. It’s a practical way to see what these things look like in action.

Mira: And for the cubic phase gate specifically, they implemented it with an order M equals eleven using a potential defined as Vcubic(Xˆ) equals zero point sixXˆ cubed. That’s the core ingredient for universal quantum information processing, so getting that right is important.

Lev: If we take that cubic term, we need to worry about how noisy our measurements are translating into the true force profile of that potential. The paper says they use a pointwise force reconstruction protocol based on the Ehrenfest theorem to sample the force minus V prime at each input state's position, which is a very specific way to get that data.

Kai: And what they show with that reconstruction is pretty solid—they recovered the cubic coefficient within one standard deviation, and they generated high quality non-Gaussian states with fidelity greater than zero point eight four within a truncation dimension D equals twenty-five and Wigner negativity greater than or equal to zero point one one <ref:2609.02405#pg2,high quality non-Gaussian states with fidelity>.

Mira: That fidelity number, over that truncation dimension of twenty-five, is what makes it usable for generating complex quantum resources on this platform. But they also looked at double-well potentials and Morse potentials to see how well the reconstruction works in those more complicated cases.

Lev: For the double-well potentials, they reproduced the expected topology for both symmetric and asymmetric cases, but when it came to broken wells, none of their bootstrap samples recovered a double well topology. That’s a real limitation we have to consider when building things that need that specific structure.

Kai: They validated the asymmetry of those potentials using three tests: a statistically significant depth difference between wells, a parity null test on the raw measured force, and a characteristic function witness showing parity breaking in the asymmetric and broken cases. That gives us some concrete evidence for those features.

Title and authors: Mira: They also talked about the bandwidth limitations imposed by the probe state width, which sets a band limit of X less than or equal to two minus one over square root two for that reconstruction protocol <ref:2609.02405#pg1>. So, what you measure is limited by how wide your initial probe is.

Lev: And for the Morse potential, they ran into another constraint because using coherent probes resulted in a two-harmonic approximation due to the nature of those states and finite measurement shot noise. That means getting the full exponential form requires more squeezing than might be immediately available in this architecture.

Kai: The paper also quantifies the errors we are dealing with by separating them into amplitude error, compiled phase error, force error, and operator distance. They set a fidelity bound based on the worst case of that operator distance epsilon K being below two percent for every gate and both circuits they tested.

Mira: From a theorist's view, that two percent bound on the operator distance is what sets the limit on how accurately we can map the ideal operation onto what we actually build. It’s a good starting point for thinking about resource generation.

Lev: The decoherence part is also addressed—qubit decoherence tends to reduce the purity of the states they generate, while cavity loss introduces a thermal state that drives the oscillator back toward equilibrium. We need to account for both of those effects when we talk about actual performance on hardware.

Kai: Looking forward, they show that this gate can be used as an impulsive potential step, but by interleaving it with free harmonic evolution, you get access to Trotterized time dynamics under anharmonic Hamiltonians. This opens the door for simulating tunneling through barriers and lattice scalar fields.

Mira: That’s significant because it establishes a building block for programmable simulation of anharmonic potential energy surfaces that can be applied across different qubit-oscillator hardwares. It moves us toward simulating actual molecular dynamics.

Lev: If we want to run this on real hardware, we need to consider how much overhead that Trotterization adds when you’re trying to simulate something like tunneling. The complexity scales with the time step you choose for the simulation of those barriers.

Kai: Overall, this paper is laying groundwork for future demonstrations of genuine non-Gaussian quantum dynamics in a bosonic quantum simulator. It’s establishing a practical and reconfigurable route to programmable anharmonic potentials on a bosonic mode.

Mira: It shows that we can use these tools to generate the resources needed for bosonic quantum information processing and analogue simulation of molecular physics. It’s about making the hard parts of non-Gaussian operations more accessible.

Title and authors: Lev: For us in error correction, it means having a clear idea of how errors like amplitude error or force error land within that compiled circuit structure before you even start designing the full fault-tolerant scheme.

Kai: So, to wrap up on this paper "Constructing and characterising anharmonic potentials in a harmonic oscillator", they’ve shown a systematic framework for implementing programmable non-Gaussian phase gates by engineering qubit rotations and displacements. They characterized cubic, double-well, and Morse potentials with fidelities over zero point eight four for the cubic gate within twenty-five dimensions <ref:2609.02405#pg2>.

Mira: And they demonstrated how to reconstruct those potentials using a specific force protocol that recovers the cubic coefficient within one standard deviation, while also resolving the topology of double wells and asymmetric potentials through multiple tests.

Lev: The main caveat they point out is that reconstruction is limited by the finite width of the probe state, setting a band limit where coherent probes only support harmonics up to n equals two.

Kai: And they also noted that for the Morse potential, using coherent probes leads to a two-harmonic approximation because of measurement shot noise.

Mira: But they showed you can use squeezed states as probes to get more accurate reconstructions, even for potentials like the Morse potential, showing how adaptive probe optimization could work there.

Lev: The error analysis shows that the fidelity bound is set by the worst case of operator distance epsilon K being below two percent for both circuits they tested.

Kai: This means we have a solid benchmark for what we can expect when implementing these gates on actual hardware, considering decoherence and loss effects too.

Mira: It opens a path toward analogue quantum simulation of molecular and chemical dynamics, which is where the real potential application lies.

Lev: For running this on real hardware, you have to factor in how much overhead that Trotterization adds when you’re simulating something like tunneling through those barriers they showed.

Kai: So, the paper "Constructing and characterising anharmonic potentials in a harmonic oscillator" gives us a practical way to engineer these anharmonic potentials on a bosonic mode by reprogramming qubit rotations.

Mira: It provides high quality non-Gaussian states, fidelity over zero point eight four for the cubic gate, and tools to explore more complex shapes like double wells and Morse potentials through force reconstruction <ref:2609.02405#pg2,high quality non-Gaussian states>.

Lev: The main limitation they state is that the reconstruction protocol is limited by the finite width of the probe state, which sets a band limit where coherent probes only support harmonics up to n equals two.

Kai: And for the Morse potential, using coherent probes results in a two-harmonic approximation due to measurement shot noise.

Kai: So, the paper "Constructing and characterising anharmonic potentials in a harmonic oscillator" gives us a practical way to engineer these anharmonic potentials on a bosonic mode by reprogramming qubit rotations.

The paper's summary: Kai: So, this paper lays out how you can take those complex non-Gaussian shapes—the cubic or double-well potentials—and actually make them happen on a physical system, specifically a superconducting harmonic oscillator, by just tweaking some qubit rotations and displacements in a fixed circuit template.

Mira: It’s about building this systematic framework for programmable gates. Basically, you don't have to design an entirely new circuit every time you want to simulate a different chemical vibration; you just change the specific rotation angles inside the existing structure.

Lev: From my side, that means we are treating the target phase gate as a Fourier series of these basic qubit rotations and those qubit-conditioned displacements, which simplifies things because it keeps the complexity localized in those defined parts of the circuit.

Kai: Exactly. And they show how this works by using that structure to implement representative potentials, like cubic, double-well, and Morse potentials. They use these physical systems to actually test if their framework is working with different kinds of shapes.

Mira: For the cubic gate specifically, they implemented it with an order M equals eleven using a potential like zero point six times X cubed. That’s the basic building block for universal quantum computing, so getting that right on this platform is what we need to see.

Lev: And they show how to check if that construction is accurate by using a specific method called pointwise force reconstruction, which samples the force profile based on the Ehrenfest theorem—basically looking at how much momentum changes given a certain state.

Kai: The results are pretty solid for the cubic case; they recovered that cubic coefficient within one standard deviation and generated non-Gaussian states with fidelity greater than zero point eight four when they used a twenty-five dimension truncation.

Mira: That fidelity number, over that twenty-five dimension truncation, is what makes the state useful for creating complex quantum information resources on this hardware. But they also looked at double-well potentials and Morse potentials to see if it held up in harder cases.

Lev: For the double-well potentials, they managed to reproduce the expected topology for both symmetric and asymmetric shapes, but when it came to broken wells, none of their bootstrap samples recovered that specific double well shape. That’s a real limitation we have to consider when building things that need that exact structure.

Kai: They also validated the asymmetry of those potentials using three different tests: a significant depth difference between the wells, a parity null test on the raw measured force, and a characteristic function witness showing parity breaking in the asymmetric and broken cases. That gives us some concrete evidence for those features.

Mira: They also talked about how much information you can actually extract, like bandwidth limitations from the probe state width. For instance, coherent probes only support harmonics up to n equals two because of that width constraint.

Lev: And for the Morse potential, using coherent probes led to a two-harmonic approximation because of measurement shot noise, which means they need more squeezing than might be immediately available in this architecture to get the full exponential form right.

Kai: The error analysis breaks down errors into amplitude error, compiled phase error, force error, and operator distance. They set the fidelity bound based on the worst case of that operator distance being below two percent for every gate and both circuits they tested.

Mira: That two percent bound on the operator distance is what sets the limit on how accurately we can map their ideal mathematical operation onto what we actually build in practice. It’s a good number to work with for resource generation, I think.

Lev: And you gotta factor in decoherence too, right? Qubit decoherence reduces the purity of the states they generate, while cavity loss introduces a thermal state that drives the oscillator back toward equilibrium. We need to account for both those effects when we talk about actual performance on hardware.

Kai: Looking ahead, they show that this gate can be an impulsive step, but by interleaving it with free harmonic evolution, you get access to Trotterized time dynamics under anharmonic Hamiltonians. This opens the door for simulating tunneling through barriers and lattice scalar fields.

Mira: That’s significant because it establishes a building block for programmable simulation of anharmonic potential energy surfaces that can be applied across different qubit-oscillator hardwares, which is exactly where we want to go with molecular dynamics.

Lev: If you want to run that on real hardware, you have to consider how much overhead that Trotterization adds when you’re trying to simulate something like tunneling through those barriers they showed; the complexity scales with the time step you choose for the simulation of those barriers.

Kai: So, this paper gives us a practical way to engineer these anharmonic potentials on a bosonic mode by reprogramming qubit rotations, and it shows we can get high-quality states for universal gates.

The paper's improvements: Kai: So, they didn't just stop at demonstrating the gates; they laid out how to make this framework even better for real hardware, which is what I’m focused on—making it work reliably in a lab setting.

Mira: Right, they point out that their current method uses a Fourier truncation for the target gate, but you can design it so that you only need to change a smaller set of qubit rotations to reprogram the operation.

Lev: That’s good because fewer parameters mean fewer things we have to tune on the physical hardware, which reduces errors in calibration and makes implementing those non-Gaussian gates much more practical for error correction schemes.

Kai: And they suggest using that structure not just for gates, but for simulating time dynamics—interleaving the impulsive step with free harmonic evolution. That lets you simulate things like tunneling through barriers under anharmonic Hamiltonians, which is a big step toward molecular physics simulations.

Mira: It moves the capability from just generating static states to actually simulating how molecules move and tunnel, which is what we really want to see in this area.

Lev: But that simulation comes with overhead; you gotta think about how much time step you need for the Trotterization when you’re trying to model something like a barrier crossing, which adds complexity to the error budget.

Kai: They also talk about adaptive probe optimization, suggesting that by using squeezing strength derived from required amplification ceiling G, you can recover features of complex potentials like the Morse potential even when coherent probes fail.

Mira: That's interesting because it shows a way to overcome those bandwidth limitations we talked about earlier; using squeezed states lets you push the reconstruction to higher harmonics, like n equals seven or higher.

Lev: From an error correction standpoint, having that adaptive probe optimization is valuable because it helps mitigate some of the noise inherent in the measurement process before you even get to the main gate operation.

Kai: So they're suggesting a more robust experimental setup where you can dynamically adjust your probing based on what potential shape you’re trying to reconstruct.

Mira: It’s about making the reconstruction protocol more flexible, moving beyond a fixed probe width limitation that just caps your accuracy at n equals two.

Lev: If we can improve the fidelity of state generation by optimizing those probes, it directly impacts how much noise is fed into the overall system during gate implementation.

Kai: So, it's about taking this systematic framework and making the experimental realization more flexible and robust for simulating complex dynamics.

Conclusion: Kai: So, to wrap up on "Constructing and characterising anharmonic potentials in a harmonic oscillator," they’ve shown this systematic framework for building programmable non-Gaussian phase gates by engineering those qubit rotations and displacements.

Mira: It’s a solid result because they successfully characterized cubic, double-well, and Morse potentials using that same structure, generating high fidelity states for the cubic gate.

Lev: And the most important thing is that they proved this method works experimentally on superconducting hardware, giving us a blueprint for what to actually build and measure in the lab.

Kai: They also gave us a clear path forward by showing how you can use this as a building block for simulating time dynamics, which means we can finally simulate molecular tunneling under anharmonic Hamiltonians.

Mira: That moves the field beyond just generating static states to actually simulating how molecules move and tunnel, which is exactly the kind of complex behavior we need to study.

Lev: But remember that running those simulations requires careful consideration of the Trotterization overhead, because you're dealing with time steps that make things computationally expensive on real quantum processors.

Kai: They also showed how you can adapt your probe states using squeezing to recover features of potentials like the Morse potential even when standard coherent probes just don't cut it.

Mira: It means we don't have to be limited by those fixed probe widths anymore; adaptive optimization lets us push the reconstruction accuracy higher, even for more complicated shapes.

Lev: From an error correction viewpoint, this gives us a clearer idea of the errors—amplitude error, compiled phase error—so we can start designing better fault-tolerant schemes before we even build the full system.

Kai: Overall, this paper gives us a practical way to engineer these anharmonic potentials on a bosonic mode by reprogramming qubit rotations and it shows we can get high-quality states for universal gates.

Mira: It provides high quality non-Gaussian states, fidelity over zero point eight four for the cubic gate, and tools to explore more complex shapes like double wells and Morse potentials through force reconstruction.

Lev: The main limitation they state is that the reconstruction protocol is limited by the finite width of the probe state, which sets a band limit where coherent probes only support harmonics up to n equals two.

Kai: That means we know exactly what our current measurement setup stops supporting without using more advanced techniques like squeezing.

Mira: And for the Morse potential, using coherent probes results in a two-harmonic approximation because of measurement shot noise, which is why adaptive probing is so necessary.

Lev: So we’ve got a clear picture now of the experimental constraints and how those constraints translate into errors we need to manage on real hardware.

Kai: This work on "Constructing and characterising anharmonic potentials in a harmonic oscillator" lays down the groundwork for programmable simulation of anharmonic potential energy surfaces across qubit-oscillator hardwares.