Tuning long-range interactions in Sr Rydberg atoms: the effect of series perturbations

arXiv:1505.07152 · physics.atom-ph · Submitted 2015-05-26 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Physics. The day in general physics: instruments, methods and results across the field.

Kai: We're looking at the paper: "Tuning long-range interactions in Sr Rydberg atoms".

Title and authors: Kai: So, we’re talking about this paper now called "Tuning long-range interactions in Sr Rydberg atoms: the effect of series perturbations." It's really about how you can change the strength and even the sign of these interactions between strontium atoms just by choosing a specific Rydberg state.

Mira: Exactly. The authors are looking at second-order dipole-dipole interactions, specifically between strontium in the five plus n five plus one states, like the 5sns one and the 5snp one <ref:1505.07152#pg1,second-order dipole-dipole interactions>. They want to see if you can use series perturbation to engineer these long-range forces in a new way.

Lev: So it sounds like they're trying to find a way to make these interactions stronger or weaker than what you get just by looking at the simple hydrogenic scaling, which is that n plus one over n squared relationship.

Kai: Right. And they found something really interesting. They saw strong peaks in the van der Waals interaction between two strontium atoms in the five plus one states outside of that expected hydrogenic scaling for low principal quantum numbers, specifically when n is less than twenty.

Mira: That's because they identified the source of those peaks: it comes from how intermediate states, specifically the five plus three d two series, get perturbed by other nearby states like the four d two and four d n prime s states.

Lev: It means that these divalent atoms give you a unique advantage over alkali metals because their energy spectra are so much more messed up, which allows for these kinds of strong deviations in the interaction strength.

Kai: They used a one-dimensional optical lattice geometry to calculate those C six coefficients, and they found pronounced nonmonotonic behavior in them for the five plus one states even when n is under twenty.

Mira: The reason they saw that nonmonotonic behavior is traced back to unusually small energy denominators in the second order energy expression. This happens because of the strong variation in atomic energies caused by all these series interactions.

Lev: So, what does this mean for us trying to build real hardware? If you try to implement this on a real system, you have to be careful because that tiny denominator is what drives the big changes in the interaction strength.

Kai: They also looked at different interaction channels, like comparing the five plus n plus one states with other intermediate channels, such as going from five plus three p one states up to five plus two s and another state.

Mira: And they pinpointed a resonance-like peak at n equals eighteen in the C six coefficients for the five plus one states that is really due to that highly perturbed nature of the five plus three d two series.

Lev: That resonance around n equals eighteen suggests you could potentially tune these interactions to be attractive or repulsive just by picking the right principal quantum number, which is a huge engineering possibility.

Title and authors: Kai: And they showed how this affects the angular dependence too, expressing the C six coefficients using Legendre polynomials for three different channels involved in the calculation.

Mira: For instance, they found that at n equals eighteen, the total C six coefficient is entirely determined by the two plus three d two plus five s n minus one d two channel because that series is strongly perturbed.

Lev: That's telling because it shows that for low n, the van der Waals interaction actually dominates over quadrupole quadrupole interactions when you look at the length scales they considered.

Kai: They also did an error analysis, and even accounting for experimental uncertainties doesn't change the picture for n under twenty. They noted that needing a very large uncertainty in the quantum defect for n equals eighteen to make the error bar look like its smooth counterpart is too much, suggesting they should use a model potential to evaluate radial wave functions.

Mira: The overall conclusion is that for n under twenty, those C six coefficients show strong nonmonotonic deviations from the simple hydrogenic scaling because of the highly perturbed nature of the five plus three d two series.

Lev: It suggests that by choosing an appropriate n, you can make these interactions attractive or repulsive. That's a direct engineering outcome they identified from this calculation.

Kai: And they mentioned that for applications, people can fit those scaled C six coefficients to rational functions of n, like something such as an plus b over n plus d.

Mira: They use those fits to try and approach a finite limit in the high principal quantum number limit of the interaction.

Lev: So what this means for real hardware is that if you want to build these systems, you need to be aware of this resonance-like structure around n equals eighteen because that's where things get really interesting.

Kai: So, we've seen how they can engineer long-range interactions in strontium Rydberg atoms by tuning the principal quantum number n and understanding the series perturbations. What a lot to take in about these divalent atoms.

Mira: It really shows how crucial it is to look at those intermediate states, like the five plus three d two series, when you're trying to understand these interactions beyond simple scaling laws.

Lev: Yeah, knowing where those resonances are helps us know where we might find unexpected behavior when we try to run this on actual experimental platforms.

Kai: We’ll keep an eye on how the AI systems can help predict these strengths and angular dependencies for pairs of atoms in the five plus one states as they explore these series perturbation effects.

Mira: And that will be important for figuring out how to actually engineer those asymmetric long-range interactions we're looking for.

Lev: We’ll see what the next paper has to offer in terms of translating this theoretical tuning into something we can build and measure on a real scale.

The paper's summary: Kai: So, to wrap up this piece, they found that for strontium atoms in these five plus one Rydberg states, you can really tune their long-range interactions just by changing the principal quantum number n and understanding those series perturbations.

Mira: Exactly. They’re showing that the simple scaling you expect from hydrogenic systems doesn't hold up for low n values, because of how those intermediate states get mixed up by other nearby states like the five plus three d two series.

Lev: From an error correction standpoint, that means if you were building a system based on this, you have to account for those small energy denominators causing big swings in the interaction strength when you try to run it on real hardware.

Kai: It’s about engineering asymmetric interactions. They found that at n equals eighteen, the interaction strength can swing really large or really small depending on what state you pick, which is a huge deal for building controlled quantum systems.

Mira: That resonance-like behavior around n equals eighteen means you could potentially make the interaction either attractive or repulsive by just picking that specific number n.

Lev: So, if we were to build a system to test this, we wouldn't just look at one state; we’d need to map out these regions where the interaction flips sign, and that tells us how much control you actually have over the long-range force.

Kai: And they even looked at how things change with direction too, using Legendre polynomials for the angular dependence of those C six coefficients.

Mira: That’s interesting because it connects the atomic structure to the actual geometry of where you place your atoms in that optical lattice setup.

Lev: It also shows that for low n, this van der Waals interaction is dominant over quadrupole quadrupole interactions, which is good because if you're trying to control these forces over short distances, you want the van der Waals term to be what's actually dictating the physics.

Kai: So it boils down to this: series perturbation in strontium Rydberg atoms gives us a way to engineer strong or weak long-range interactions that we couldn't get with simpler alkali metals.

Mira: It’s about moving beyond the simple n plus one over n squared scaling and using these subtle energy shifts to create highly controllable forces.

Lev: The next step would be figuring out the overhead for error correction when you have these highly state-dependent interaction potentials that change sign so quickly across small changes in n.

The paper's improvements: Tom: So, to recap the main finding, the paper shows that you can use series perturbation to tune those long-range interactions in strontium Rydberg atoms by picking a specific principal quantum number n and using that as a resonance point.

Kai: Right. It’s about finding those specific n values, like eighteen, where things get really big or really small because of how the intermediate states are perturbed.

Mira: The authors suggest that this means we can engineer asymmetric interactions by choosing different n values to control whether the force is attractive or repulsive.

Lev: If you want to build a system to test this, you need those precise tuning parameters for n, and that requires knowing exactly how large the uncertainty in your quantum defect has to be around that resonance point.

Kai: They suggest fitting these C six coefficients to rational functions of n, like something with an plus b over n plus d, which lets us approach a finite limit when we look at very high principal quantum numbers.

Mira: That fits the theory well because it gives us a way to model what happens as n gets really big, which is important for predicting behavior in larger systems.

Lev: For error correction, that means we need to know if those rational function fits hold up when we try to implement them on a real superconducting circuit or something similar; the overhead will depend entirely on how well that fit captures the true physics.

Kai: They also pointed out that for low n, you have to be very careful because of that series perturbation effect, and you can't just use simple scaling laws there.

Mira: It’s a warning against assuming simple hydrogenic behavior holds across the whole range; those intermediate states are interfering so much at low n that it breaks the expected pattern.

Lev: This tells me that for any real hardware design, you can't just rely on a single potential model; you have to account for this complex interplay of different Rydberg series.

Kai: And they also touched on angular dependence, showing how the direction of your magnetic field influences those C six coefficients using Legendre polynomials like P0 and P2.

Mira: That’s useful because it means the interaction strength isn't just a single number; it depends on the angle you set up in your optical lattice geometry.

Lev: So, if we’re looking at building a device to measure this, we need to design our experiment to be sensitive to those angular variations as well as the n variation.

Kai: It’s about making sure our experimental setup can actually probe these specific tuning regions where the interaction strength changes sign.

Conclusion: Kai: So we’re wrapping up on "Tuning long-range interactions in Sr Rydberg atoms: the effect of series perturbations." The big picture here is that we can use these small energy shifts to make the interaction strength either attractive or repulsive by just choosing the principal quantum number n.

Mira: It shows that even in these highly excited Rydberg states, which usually follow simple scaling rules, you have to look at those intermediate states—like the five plus three d two series—to understand why things deviate.

Lev: From a hardware side, that means we need precise control over the n value and we need to know exactly how much error correction overhead is needed for those resonant regions where the interaction flips sign.

Kai: The implication is that this opens up a new way to engineer asymmetric forces in quantum systems, which could be useful for designing specific types of quantum gates or traps.

Mira: It’s about moving beyond just knowing the basic van der Waals scaling and using these subtle perturbation effects to create much more nuanced control over how atoms interact across space.

Lev: If we are building an experiment on a real system, we have to make sure our experimental parameters can actually probe those specific n values where the C six coefficient is changing sign.

Kai: And they also showed that the angular dependence matters, so you’re not just looking at a strength number but how that strength changes based on the direction of your magnetic field.

Mira: That adds another layer of complexity, showing that geometry plays a big role in determining the final interaction we observe.

Lev: The paper also flagged that for certain low n values, the uncertainty in the quantum defect is too large to ignore, meaning we can't just assume it’s negligible without checking.

Kai: So it’s a reminder that when dealing with nonhydrogenic atoms, you can’t just rely on hydrogenic scaling; those residual deviations matter a lot.

Mira: This work really solidifies the idea that detailed analysis of the Rydberg spectrum is essential for predicting how these systems will behave in practice.

Lev: The next challenge is taking this theoretical tuning and figuring out if we can actually implement it with low enough error rates to get reliable results on our current quantum hardware.

Turker Topcu, Andrei Derevianko

Department of Mathematics, Virginia Tech · Department of Physics, University of Nevada, Reno

physics.atom-ph

Submitted: 2015-05-26

Updated: 2026-11-13

Journal ref: T Topcu and A Derevianko 2026 J. Phys. B: At. Mol. Opt. Phys. 59 215002

DOI: 10.1088/1361-6455/aebb7c

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 83/100

The gist: Tuning long-range interactions in Sr Rydberg atoms: the effect of series perturbations investigates how series perturbation modifies second-order dipole-dipole interactions between strontium atoms in…

Key concepts

Series Perturbation
This occurs when two Rydberg energy levels with identical total angular momenta and parities are very close in energy. The off-diagonal Coulomb interaction mixes these states, leading to strong deviations from the standard van der Waals scaling ($R^{-6}$) due to accidental near-degeneracies.
C6 Coefficients
These coefficients describe the strength of the second-order dipole-dipole interaction between two atoms. The study analyzes how these C6 values change with the principal quantum number (n) and atomic state, revealing nonmonotonic behavior that can be tuned by selecting a specific n.
5snd(1D2) Series Perturbation
The 5snd(1D2) Rydberg series in Strontium acts as a key perturber. Its strong mixing with other states causes unusually small energy denominators in the second-order energy expression, which is the primary cause for the observed resonance-like peaks and nonmonotonic features at low n.

Terminology

Summary

Tuning long-range interactions in Sr Rydberg atoms: the effect of series perturbations investigates how series perturbation modifies second-order dipole-dipole interactions between strontium atoms in specific Rydberg states to engineer long-range interactions

The Core Phenomenon

The study focuses on modifying the strength and sign of second-order dipole-dipole interactions between strontium atoms in the 5sns(1S0) and 5snp(1P1) Rydberg states using series perturbation The van der Waals interaction between two strontium atoms in the 5snp(1P1) states shows strong peaks outside the anticipated hydrogenic n 1/n scaling, which is identified as a result of the perturbation of 5snd(1D2) intermediate states by the 4d2(1D2) and 4dn's(1D2) states in the n < 20 range. This demonstrates that divalent atoms offer a unique advantage for generating substantially stronger or weaker inter-atomic interactions than those achievable using alkali metal atoms due to their highly perturbed spectra. p. 0

Series Perturbation Mechanism

The series interaction arises when two Rydberg series with identical total angular momenta and parities are nearly degenerate in energy, causing the levels to be mixed by the off-diagonal Coulomb interaction The van der Waals interaction scales as R-6, but accidental near-degeneracies between target and intermediate states can lead to strong deviations from the n 1/n scaling behavior. This mechanism is the cause of the irregularities in the van der Waals interaction strength reported in this work. p. 1

Computational Framework and Results

The authors investigate these interactions within a one-dimensional optical-lattice geometry, deriving C6 coefficients for pairs of atoms separated by different lattice constants. They find pronounced nonmonotonic behavior in the C6 coefficients for pairs of singlet 5snp(1P1) states even in the low-n, n < 20 range. p. 1 The origin of this behavior is traced to unusually small energy denominators in the second-order energy expression resulting from strong variation in atomic energies due to series interaction. p. 1

Interaction Channel Analysis

The analysis involves comparing contributions from different intermediate channels, such as nP + nP → n2S + n'2S, and nP + nP → nD + (n-1)D. The resonance-like peak at n = 18 in the C6 coefficients for the 5snp(1P1,Mx=1) states is attributed to the highly perturbed nature of the 5snd(1D2) Rydberg series. p. 9 This is because the energy difference 2EnP − (EnD + E(n-1)D) is unusually small around n = 17. p. 8

Conclusion and Engineering Potential

The results show that the C6 coefficients display highly non-monotonic behavior and change sign in the small n-region, suggesting that the interaction can be made attractive or repulsive by choosing appropriate n. Particularly at n = 18, Ce6 is much larger than it is in any other state for the entire n-range considered, providing a possibility for engineering strongly asymmetric long-range interactions by contrasting it with the van der Waals interaction between two Sr atoms in 5sns(1S0) states. p. 2

Angular Dependence and Comparison

The angular dependence of the C6 coefficients is expressed in terms of Legendre polynomials P0(cos θ), P2(cos θ), and P4(cos θ) for the three channels involved. The total Ce6 is entirely determined by the P2(cos θ) character of the 5snd(1D2) + 5s(n-1)d(1D2) channel at n = 18, where the series is strongly perturbed. p. 2 The van der Waals interaction dominates over quadrupole-quadrupole interactions over the length scales considered for two 5s18p(1P1) states. p. 3

Error Analysis and Model Potential Justification

The analysis of error bars suggests that taking experimental uncertainties into account does not change the qualitative picture for n < 20. The required uncertainty in the quantum defect for n = 18 to make the error bar large enough to encapsulate its monotonic counterpart is too large even for a model potential, suggesting that using a model potential for evaluating radial wave functions is justified. p. 6

Summary of Key Findings

The scaled Ce6 = C6/n 1/1 coefficients for the van der Waals interaction between pairs of 5sns(1S0) and 5snp(1P1,Mx=1) atoms are plotted as a function of n in Figs. 3 and 4. For n < 20, the C6 an der Waals coefficients for the 5snp(1P1) states show strong non-monotonic deviations from the hydrogenic n 1/n scaling due to the highly perturbed nature of the 5snd(1D2) series as discussed in Ref. p. 2 [57].

C6 for the 5sns(1S0) states

In Fig. 3, open blue diamonds show the Ce6 coefficients calculated using Eq. p. 7 (7) in which both the energy denominator and the one-electron orbitals in the reduced matrix elements are calculated using the model potential (12). p. 7 The parameters in this model potential are adjusted from those listed in Ref. p. 7 [67] to better match the experimental energy denominators. p. 7 However, around n ≈ 12, the Ce6 coefficients become extremely large due to an almost vanishing energy denominator. p. 7

C6 for the 5snp(1P1,1x) states

Unlike the 5sns(1S0) + 5sns(1S0) interaction, numerically calculated values of Ce6 using the model potential (12) between 5snp(1P1,Mx=1) states do not display the same pathological behavior at low n. We observe an unusual feature in the low-n region below n = 20: the Ce6 coefficients display a non-monotonic behavior. p. 8

Where the gains come from

The deviation of the nearest-neighbor energy differences from the hydrogenic 1/n 1/3 scaling is a consequence of series perturbation. The energies of the 5sns(1S0) states almost perfectly scale as 1/n 1/2 for all n 1/, whereas 5snp(1P1) and 5snd(1D2) states increasingly deviate from this scaling as n 1/ gets below ∼20. p. 8

Fitting Ce6 to rational functions of n

For use in applications, the scaled Ce6 coefficients are fitted to rational functions of n, such as (an + b)/(n + d) and (an squared + bn + c)/(n squared + en + f). These fits are used to approach a finite limit in the high-n limit. p. 9

Angular Dependence of the van der Waals Interaction

The dependence of the Ce6 coefficients for the 1P1,1x + 1P1,1x van der Waals interaction on the angle θ that the quantizing B-field makes with the internuclear axis can be expressed in term of Legendre polynomials P0(cos θ), P2(cos θ) and P4(cos θ). The total Ce6 is entirely determined by the P2(cos θ) character of the 5snd(1D2) + 5s(n-1)d(1D2) channel at n = 18. p. 6

Conclusion

We find that for n < 20, the C6 an der Waals coefficients for the 5snp(1P1) states show strong non-monotonic deviations from the hydrogenic n 1/n scaling due to the highly perturbed nature of the 5snd(1D2) series as discussed in Ref. [57]. As a result, the C6 coefficients display highly non-monotonic behavior and change sign in the small n-region suggesting that the interaction can be made attractive or repulsive by choosing appropriate n. p. 1

Appendix A: Derivation of van der Waals expressions

The second-order energy shift is given by δEvdWγ1[L1S1J1]M1(I)γ'1L'S'J'M'1= 1/R6(A4) × Xγ2,γ'2 J2, J'2 M2, M'2 Dγ1[L1S1J1]M1(I) γ′1L'S'J'M'1 Vdd γ2[L2S2J2]M2(I) γ′ 2L'S'J'M'2 Eγ1[L1S1J1] + Eγ'+Eγ'' - Eγ3[L3S3J3] + Eγ'' - Eγ4[L4S4J4].

Appendix B: Introducing singlet-triplet mixing into the 1P1 + 1 P1 interaction

The authors include triplet admixture in the nominal singlet D2 intermediate states by making the replacement 1D2 → α1 3D2. The matrix elements connecting the 1P1 − 3D∗2 states are relativistically suppressed by a factor of α squared ∼ 10-4, so they are ignored in this work. p. 4

Appendix C: Derivations of the Quadrupole-Quadrupole interactions

The quadrupole-quadrupole interaction EQQ can be expressed in terms of the individual two-atom matrix elements, and for the singlet P1 states, it is shown that the quadrupole-quadrupole interaction can be neglected for the low-n range where the van der Waals interaction exhibits pronounced nonmonotonic behavior.

Appendix D: Error estimates for the dipole matrix elements

The fractional error in C6 is given by δC6 squared = 2 δdn,n2 dn,n2 squared + δdn,n'2 dn,n'2 2. For n = 18, the uncertainty in the quantum defect is estimated to be around 3. p. 035.

Table I

The C6 coefficient is expected to scale as n 1/1. This scaling is exact for hydrogen, but nonhydrogenic atoms exhibit residual deviations because the Rydberg electron interacts with a polarizable, penetrable ionic core. p. 7 p. 7 p. 7 Two types of residual behavior are relevant: a smooth monotonic trend, associated with core polarization, core penetration, and exchange effects, and resonance-like structure. p. 7 p. 7 p. 7

Table II

The Ce6 coefficient is expected to scale as n 1/1. This scaling is exact for hydrogen, but nonhydrogenic atoms exhibit residual deviations because the Rydberg electron interacts with a polarizable, penetrable ionic core. Two types of residual behavior are relevant: a smooth monotonic trend, associated with core polarization, core penetration, and exchange effects, and resonance-like structure.

Table III

The parameters in Table III indicate in the limit n → ∞ the van der Waals interactions between pairs of 5sns(1S0) atoms is roughly a factor of 3 stronger than those between pairs of 5snp(1P1,1x) atoms.

Table IV

The Ce6 coefficient for the 5snp(1P1,1x) + 5snp(1P1,1x) interaction shows a non-monotonic behavior in the low-n region below n = 20. The large error bar at n = 18 stems from near degeneracy in the energy denominator. p. 8

Table V

The scaled energy denominator in the second order perturbation theory expression for the Ce6 coefficients in the long-range interaction between two 5snp(1P1,1x) Sr atoms is flat above n ∼ 20, but it is far from regular below n ∼ 20 in the nP + nP → nD + (n-1)D channel. This “resonance-like” structure is facilitated by an unusually small ∆E around n ∼ 17 and is possible because of the highly perturbed nature of the 5snd(1D2) series in Sr. p. 9

Table VI

The total Ce6 is entirely determined by the nP +nP → n'D +n''D channel at n = 18. This is because the resonance-like peak at n = 18 in Fig. p. 9 4 is due to the highly perturbed nature of the 5snd(1D2) Rydberg series which results in an unusually small energy denominator for n = 18 in the nP + nP → n'D +n''D channel.

Table VII

The C6 coefficient is expected to scale as n 1/1. This scaling is exact for hydrogen, but nonhydrogenic atoms exhibit residual deviations because the Rydberg electron interacts with a polarizable, penetrable ionic core. Two types of residual behavior are relevant: a smooth monotonic trend, associated with core polarization, core penetration, and exchange effects, and resonance-like structure.

Acknowledgements

This work was supported by the National Science Foundation (NSF) Grant No. PHY-1212482. A.D. was also supported by the Simons Foundation as a Simons fellow in theoretical physics, and by NSF Grant No. p. 0 PHY-2207546. T.T. and A.D. would like to thank the Institute for Theoretical Atomic, Molecular, and Optical Physics (ITAMP) and the Harvard University Physics Department for their hospitality. p. 0 The authors would like to thank P. p. 6 K´om´ar and M. D. Lukin for valuable discussions.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved AI systems can accurately predict long-range van der Waals interaction strength for pairs of strontium atoms in 5snp(1P1) states across a wide range of principal quantum numbers n by incorporating series perturbation effects. The improved system will be able to distinguish between the smooth hydrogenic n to the power 1/1 scaling and the nonmonotonic behavior caused by intermediate state perturbations, such as those from 5snd(1D2) states, especially for low n values.

  2. The AI can perform asymmetric Rydberg blockade engineering by explicitly accounting for both weak and strong interactions simultaneously. This capability stems from understanding how series perturbation leads to the presence of both relatively small and large van der Waals interactions in the 5snp(1P1) states with n less than 20.

  3. The system can predict the angular dependence of van der Waals coefficients, calculating Ce6 as a function of angle theta for different Rydberg state pairs. This allows for precise engineering of interaction strengths based on the geometry of the optical lattice, using functions like fss(theta), fdd(theta), and fsd(theta).

  4. The AI can assess whether quadrupole-quadrupole interactions are relevant by comparing them to van der Waals interactions across different length scales, such as within a single lattice site versus between adjacent sites. It can determine when the quadrupole-quadrupole interaction is negligible compared to the van der Waals interaction for low-n singlet P1 states.

  5. The system can identify the specific perturber states responsible for nonmonotonic features in C6 coefficients, such as 4d2(1D2), 5p23/2(1D2), and 4d6s(1D2) perturbing the 5snd(1D2) series. This allows for targeted experimental searches to confirm these specific atomic configurations.

  6. The improved system can utilize experimental energy denominators to calculate fractional uncertainties in C6 coefficients, specifically quantifying how much the error bars need to be increased to capture the nonmonotonic resonance-like behavior at n = 18.

Sources

Related papers