Nonperturbative regime of low-order harmonic generation in intense low-frequency laser field

arXiv:2512.06816 · physics.optics, physics.atom-ph · Submitted 2025-12-07 · 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: "Nonperturbative regime of low-order harmonic generation in intense low-frequency laser field".

Mira: Here's the gist of the paper. We find the atomic response to the intense femtosecond laser pulse via solving numerically the three-dimensional non-stationary Schr¨odinger equation (TDSE) for a model atom and calculating its dipole moment The accuracy of this description is unsatisfactory, regardless of the order of non-linearity taken into account, for intensities higher than about 0.6 1014 W/cm2 Limitations of Perturbative Descriptions For weak quasi-static fields, the response is well described by a perturbation approach, but for intensities higher than about 0.6 1014 W/cm2 the accuracy of this description is unsatisfactory, regardless of the order of non-linearity taken into account.

Kai: The perturbation series does not converge when the requirement (3) is not fulfilled, meaning pronounced deviation from linear dependence on intensity can occur.

Mira: The authors suggest fitting the numerical TDSE solution results with a Pad´e expansion, which describes the response well both in the perturbative regime and beyond it for intensities approximately up to 1.4 1014 W/cm2.

Kai: Alternative Approximations for Response The authors propose an alternative approximation for the dipole moment that describes well the numerical results both for low and high intensities including those for which the series (4) is inapplicable.

Paper summary: Kai: So, thinking about the title of this paper and what it all boils down to, it’s about mapping out exactly where the standard tools for calculating harmonic generation stop being reliable when you use strong low frequency lasers >

Mira: They found that the TDSE calculations don't show a non perturbative increase in the effective first order polarizability for ultraviolet fields because photoelectrons affect it there, but they do see this non perturbative increase confirmed by TDSE calculations for infrared fields >

Lev: And they describe those third and fifth order nonlinearities well in the infrared field, including that non perturbative growth with laser intensity, although they admit the applicability of these models is limited by photoionization and resonant contribution for high intensities and frequencies >

Kai: So, even though the TDSE doesn't show a non perturbative effect for UV fields, those models are useful tools when you are working in the infrared where those effects actually happen >

Mira: The paper’s main point seems to be that we need these alternative approximations like the Padé expansion or the nonlinear oscillator model when we go into that non perturbative domain, because they describe what happens better than a pure perturbation theory does >

Lev: It really tells us that for real experiments involving intense fields, you can't just rely on the simplest models; you have to use these more complex tools depending on whether you’re dealing with UV or infrared light and how intense the field is >

Conclusion: Kai: So, this paper is about where low-order harmonic generation gets weird when you crank up the laser intensity and keep it at those lower frequencies >

Mira: Yeah, they looked at how things behave in that non perturbative regime for these intense low-frequency fields >

Lev: The authors used a few different ways to look at the math, from solving the full time dependent schrodinger equation numerically to using some oscillator models >

Kai: The core finding is that the standard methods break down above about zero point six times one ten to fourteen watts per centimeter squared when you try to get an accurate picture of the atomic response >

Mira: That's because the simple math series they start with just doesn't converge when you push it into those higher intensity ranges, meaning you can see big deviations from what you expect linearly >

Lev: They suggest fitting the numerical solutions with something called a padé expansion which works pretty well up to about one point four times one ten to fourteen watts per centimeter squared >

Kai: It means even when we use the most detailed calculation, like solving the time dependent schrodinger equation, it's not giving us a clean answer without some extra math >

Mira: But they also found that for infrared fields, which are lower frequency, these models actually describe the growth of efficiency quite well in those non perturbative areas >

Lev: That's important because it shows where we can trust our simplified models for other nonlinear stuff, like third and fifth harmonic generation in the infrared field >

Kai: The limitation they point out is that this whole picture breaks down when you start looking at ultraviolet fields, because photoelectrons get involved in a way that changes the polarizability >

Mira: So it’s a clear split: these models work for IR fields where you see those non perturbative effects, but not for UV fields where things get more complicated due to photoionization and resonances >

Lev: It really frames the problem as knowing which physics—the simple perturbation or the full numerical solution—you actually need to use depending on the frequency and intensity of your setup >

Kai: If we want to build real hardware that handles these intense fields, we have to keep this split in mind when choosing our approximation methods >

P.N. Lebedev Physical Institute of the Russian Academy of Sciences · National Research Nuclear University MEPhI

physics.optics, physics.atom-ph

Submitted: 2025-12-07

Updated: 2026-11-05

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 76/100

The gist: We find the atomic response to the intense femtosecond laser pulse via solving numerically the three-dimensional non-stationary Schr¨odinger equation (TDSE) for a model atom and calculating its…

Key concepts

Time-Dependent Schrödinger Equation (TDSE)
This is a mathematical equation used to model how an atom reacts when hit by a strong laser pulse. The researchers solved this equation numerically to get the atomic response, which showed that simple models are inaccurate for very intense light.
Padé Expansion
This is a mathematical tool used to improve approximations, especially when standard series expansions don't work well. It was used here to fit the numerical TDSE results across a wide range of laser intensities, making the description more reliable beyond weak field limits.
Nonlinear Oscillator Model (Duffing Oscillator)
This model treats the atomic response like a physical spring system that is being pushed by an external force. It was used to describe how the polarization changes under intense fields, particularly for processes like third and fifth harmonic generation in infrared light.
Nonperturbative Regime
This refers to situations where the laser field is so strong that standard mathematical approximations (like perturbation theory) break down. The paper explores how atomic responses change significantly in this regime, especially when dealing with higher-order nonlinear effects.

Terminology

Summary

We find the atomic response to the intense femtosecond laser pulse via solving numerically the three-dimensional non-stationary Schr¨odinger equation (TDSE) for a model atom and calculating its dipole moment The accuracy of this description is unsatisfactory, regardless of the order of non-linearity taken into account, for intensities higher than about 0. p. 0

Where the paper says it

com (Dated: May 21, 2026) We find the atomic response to the intense femtosecond laser pulse via solving numerically the three-dimensional non-stationary Schr¨odinger equation (TDSE) for a model atom and calculating its dipole moment. For weak quasi-static fields, the response is well described by a perturbation approach, but for intensities higher than about 0.

6 1014 W/cm2

Limitations of Perturbative Descriptions

For weak quasi-static fields, the response is well described by a perturbation approach, but for intensities higher than about 0.6 1014 W/cm2 the accuracy of this description is unsatisfactory, regardless of the order of non-linearity taken into account. p. 0

Where the paper says it

For weak quasi-static fields, the response is well described by a perturbation approach, but for intensities higher than about 0. 6 1014 W/cm2 the accuracy of this description is unsatisfactory, regardless of the order of non-linearity taken into account.

The perturbation series does not converge when the requirement (3) is not fulfilled, meaning pronounced deviation from linear dependence on intensity can occur. p. 0
Where the paper says it

Note that as the intensity increases, pronounced deviation from the linear dependence can take place. This means that requirement (3) is not fulfilled, so the perturbation series does not converge.

The authors suggest fitting the numerical TDSE solution results with a Pad´e expansion, which describes the response well both in the perturbative regime and beyond it for intensities approximately up to 1. p. 0
Where the paper says it

6 1014 W/cm2 the accuracy of this description is unsatisfactory, regardless of the order of non-linearity taken into account. We suggest fitting the numerical TDSE solution results with a Pad´e expansion, and show that this approximation describes the response well both in the perturbative regime and beyond it for intensities approximately up to 1.

4 1014 W/cm2. p. 0
Where the paper says it

For weak quasi-static fields, the response is well described by a perturbation approach, but for intensities higher than about 0. 6 1014 W/cm2 the accuracy of this description is unsatisfactory, regardless of the order of non-linearity taken into account.

Alternative Approximations for Response

The authors propose an alternative approximation for the dipole moment that describes well the numerical results both for low and high intensities including those for which the series (4) is inapplicable. This approximation's applicability range is limited by the intensity at which photoionization starts to essentially contribute to the dipole moment of the system. The Pad´e approximation, given by Eq. p. 1

Where the paper says it

TDSE in the single-active electron approximation is written in cylindrical coordinates using atomic units as: i ∂ ∂tΨ(t, ρ, z) = Hˆ (t)Ψ(t, ρ, z), (8) where Ψ(t, ρ, z) is the time-dependent wave function of the electron and Hˆ (t) is the Hamiltonian of the system. The Hamiltonian is given by: Hˆ (t) = − 1 2 ∆ + U(r) + zE(t) − iW(ρ, z), (9) where r 2 = ρ 2 + z 2 , U(r) is the effective one-electron potential which models the argon atom and −iW(ρ, z) is an imaginary (absorbing) potential (see Appendix A for more details).

(16), agrees well with TDSE result over the whole region of the field strengths, including the non-perturbative one. p. 3
Where the paper says it

This can be a natural perspective for our future studies. 1 shows that the approximation (16) (solid black line) agrees well with TDSE result over the whole region of the field strengths, including the non-perturbative one.

Nonlinear Oscillator Model Application

The authors develop a nonlinear oscillator (known as Duffing oscillator for a cubic nonlinearity) model to describe the dispersion of the nonlinear polarization including the region of intensities where it behaves non-perturbatively. This model is defined by assuming that in the quasi-static limit, the displacement and external force are linked by Eq. p. 3

Where the paper says it

Under the conditions employed in our calculations, the actual atomic emission time 1/Γ is much longer than the laser pulse duration, so we neglect damping in further calculations. Our model is defined by the assumption that in the quasi-static limit the displacement and the external force are linked by Eq.

(16). The model fails to predict the behaviour of the nonlinear refractive index in the nonperturbative domain, but it describes well the nonperturbative growth of efficiency with laser intensity for other nonlinear optical processes, namely, the third and fifth harmonic generation in the IR field and optical rectification in a two-color field. p. 0
Where the paper says it

To consider the non-perturbative nonlinearity beyond the quasi-static limit we use the model of nonlinear oscillator with the restoring force defined by the found Pad´e expression. Our model fails to predict the behaviour of the nonlinear refractive index in the nonperturbative domain, but it describes well the nonperturbative growth of the efficiency with the laser intensity for other nonlinear optical processes, namely, the third and fifth harmonic generation in the IR field and the optical rectification in a two-color field.

Comparison of Nonlinear Polarizabilities

The authors compare the nonlinear polarizabilities obtained by different methods: within the numerical TDSE solution, within the quasistatic approach using the Pad´e approximation, and within the nonlinear oscillator model. The results show that for low intensity and low frequency, TDSE result agrees with those of the quasi-static and nonlinear oscillator. p. 3

Where the paper says it

3), and fifth (Fig. 2 we see that for low intensity and low frequency (1 eV) the TDSE result agrees with those of the quasi-static and nonlinear oscillator.

For higher frequencies, the quasi-static model’s applicability is worse, but the nonlinear oscillator model is still applicable, thus it adequately describes the dispersion. p. 3
Where the paper says it

2 we see that for low intensity and low frequency (1 eV) the TDSE result agrees with those of the quasi-static and nonlinear oscillator. For higher frequencies the quasi-static model’s applicability is worse, but the nonlinear oscillator model is still applicable, thus it adequately describes the dispersion.

Harmonic Generation and Phase Behavior

The analysis of the harmonic phase shows that for low intensities, the atomic response adiabatically follows the electric field so harmonic phases are zero. The non-zero harmonic phase is found only in the numerical TDSE solution. p. 4

Where the paper says it

This delay leads to the non-zero harmonic phase for high intensities in Figs. The non-zero harmonic phase is found only in the numerical TDSE solution.

The quasi-static model naturally predicts zero harmonic phases because it supposes instantaneous response to the field. p. 4
Where the paper says it

The non-zero harmonic phase is found only in the numerical TDSE solution. The quasi-static model supposes instantaneous response to the field and so naturally predicts zero harmonic phases.

Conclusion on Non-Perturbative Effects

The nonperturbative increase of the effective first-order polarizability predicted by our models is not observed in the TDSE calculation for UV fields, as photoelectrons affect the polarizability for IR fields and resonant processes affect it for UV fields. However, the non-perturbative increase of these polarizabilities is confirmed by the TDSE calculations for IR field. p. 4

Where the paper says it

For the third- and fifth-order nonlinearities the agreement for the UV field is even worse, so these results are not shown in Figs. However, for IR field the non-perturbative increase of these polarizabilities predicted by the models is confirmed by the TDSE calculations, see Figs.

The models describe well the third and fifth order nonlinearities in the IR field, including the non-perturbative growth of efficiency with laser intensity. p. 6
Where the paper says it

Our model fails to describe the nonlinear refractive index in the nonperturbative domain, in terms of both amplitude and phase, because it doesn’t take into account the contribution of either the photoelectrons (dominating for the IR fundamental) or resonances (dominating for the UV fundamental). In contrast, the third and fifth harmonic generation in the IR field, as well as optical rectification in a two-color field, are well described by our model, including the non-perturbative growth of the efficiency with the laser intensity.

The applicability of these models is limited by photoionization and resonant contribution for high intensities and frequencies. p. 6
Where the paper says it

In contrast, the third and fifth harmonic generation in the IR field, as well as optical rectification in a two-color field, are well described by our model, including the non-perturbative growth of the efficiency with the laser intensity. For these processes as well, the applicability of the model for high intensities and frequencies is limited by photoionization and resonant contribution.

Appendix Details

The TDSE is solved numerically by the Crank–Nicolson method, see more details in [38]. The imaginary absorbing potential in Eq. p. 1

Where the paper says it

TDSE in the single-active electron approximation is written in cylindrical coordinates using atomic units as: i ∂ ∂tΨ(t, ρ, z) = Hˆ (t)Ψ(t, ρ, z), (8) where Ψ(t, ρ, z) is the time-dependent wave function of the electron and Hˆ (t) is the Hamiltonian of the system. The Hamiltonian is given by: Hˆ (t) = − 1 2 ∆ + U(r) + zE(t) − iW(ρ, z), (9) where r 2 = ρ 2 + z 2 , U(r) is the effective one-electron potential which models the argon atom and −iW(ρ, z) is an imaginary (absorbing) potential (see Appendix A for more details).

(9) is used to prevent unphysical reflections of the wave packet from the boundaries of the numerical box. p. 6
Where the paper says it

The imaginary absorbing potential −iW(ρ, z) in Eq. (9) is used to prevent unphysical reflections of the wave packet from the boundaries of the numerical box.

The Chebyshev approximation allows decomposition of the dipole response to quasi-static field into a series up to the seventh order in E = [−Emax, Emax]. p. 3
Where the paper says it

The atomic dipole moment in the quasi-static field calculated via numerical TDSE solution (solid green curve), its approximation with Pad´e expansion (solid black curve), and with the polynomial series (see the legend). The black arrow shows the boundary of the approximation interval Emax.

The Duffing oscillator model is expanded up to the third order for the restoring force F(d) = −omega 2 0d(1 + γd 2). p. 7
Where the paper says it

Duffing oscillator model We expand the restoring force in Eq. (17) up to the third order: F(d) = −Ω 2 0d(1 + γd2 ), (32) where Ω2 0 = 1/κ(1) , γ = −κ (3)/(κ (1)) 3 .

The resulting equations (37) and (39) are solved numerically to find d1(ε) and d3(ε), which are then used to obtain the effective polarizabilities. p. 7
Where the paper says it

Let us assume that the response contains only the first and third harmonics: d(τ ) = d1 cos(Ωτ + φ1) + d3 cos(3Ωτ + φ3), (36) where d1, d3 are the amplitudes of the first and third harmonics and φ1, φ3 are the phases of these harmonics. Substituting (36) into (33), neglecting components at multiple excitation frequencies (except the third one) and collecting coefficients at the first and the third harmonic frequencies, we find:    (1 − Ω 2 )d1 + 3d 3 1 γ 4 + 3d1d 2 3 γ 2 + 3d 2 1 d3γ 4 cos(φ3 − 3φ1) = ε cos φ1 (1 − 9Ω2 )d3 + 3d 3 3 γ 4 + d 3 1 γ 4 cos(φ3 − 3φ1) + 3d 2 1 d3γ 2 cos(φ3 − φ1) = 0 d 3 1 γ 4 sin(φ3 − 3φ1) + 3d 2 1 d3γ 2 sin(φ3 − φ1) = 0 3d 2 1 d3γ 4 sin(3φ1 − φ3) = ε sin φ1 (37) From the latter two equations we obtain: sin φ1 = − 9d1d 2 3γ 2ε sin(φ1 − φ3).

The numerical values for the coefficients κ(m) were found using Eq. p. 7
Where the paper says it

(28) we find coefficients βm using numerically calculated d(E) and Emax = 0. (30) we find the following values of κ (m) : κ (1) ≈ 3.

(30). The results show that these models take into account neither photoionization nor resonances for the third- and fifth-order nonlinearities. p. 4
Where the paper says it

our models take into account neither photoionization nor resonances. For the third- and fifth-order nonlinearities the agreement for the UV field is even worse, so these results are not shown in Figs.

The effective third-order nonlinear polarizability associated with this process is given by Eq. p. 5
Where the paper says it

The fundamental frequency is ω0 = 1 eV, the ratio of the generating fields’ amplitudes are ν = 1 (a) and ν = 1/ √ 2 (b). The effective third-order nonlinear polarizability associated with this process is: α˜ (3)(ω = 2ω0 − ω0 − ω0) = ¯d(ω = 2ω0 − ω0 − ω0) E¯(2ω0) E¯(ω0) 2 .

(20). The results of the TDSE numerical integration (solid curves), of the quasi-static model using Pad´e approximation (black dashed curve with symbols), and of the nonlinear oscillator model (dashed curves without symbols) are compared in Fig. p. 4
Where the paper says it

The dependence of the effective polarizability of the first order ˜α (1)(ω0) on the intensity for different frequencies (a) and for different pulse durations for ω0 = 3. The results of the TDSE numerical integration (solid curves), of the quasi-static model using Pad´e approximation (black dashed curve with symbols), of the nonlinear oscillator model (dashed curves without symbols), of the perturbation approach (solid grey curve).

8. The effective polarizability depends on phase difference ∆ϕ. p. 5
Where the paper says it

The effective third-order nonlinear polarizability associated with this process is: α˜ (3)(ω = 2ω0 − ω0 − ω0) = ¯d(ω = 2ω0 − ω0 − ω0) E¯(2ω0) E¯(ω0) 2 . (20) Note that this effective third-order nonlinear polarizability depends on phase difference ∆ϕ.

The applicability of our models is limited by the maximal strength of total field, achieved already at moderate intensities of the fundamental. p. 5
Where the paper says it

Note that the intensity shown along the horizontal axis is the intensity of the first harmonic only. The applicability of our models is limited by the maximal strength of total field, achieved already at moderate intensities of the fundamental

Acknowledgements

This study was funded by RSF through Grant No. 24-12-00461. p. 6

Where the paper says it

ACKNOWLEDGMENT This study was funded by RSF through Grant No. 24- 12-00461.

The paper is arXiv: 2512. p. 0
Where the paper says it

When the field’s frequency ω0 is low enough so that the microscopic response d depends only on the instantaneous strength of the field E, the response as a function of the field can be expanded in the Taylor series: d(E) = X m=1,3. κ (m)E m (4) In this case using the binomial expansion [E(t)]m = Em 0 [(exp(−iω0t) + exp(iω0t)) /2]m = arXiv:2512.

06816v2 [physics. p. 0
Where the paper says it

κ (m)E m (4) In this case using the binomial expansion [E(t)]m = Em 0 [(exp(−iω0t) + exp(iω0t)) /2]m = arXiv:2512. 06816v2 [physics.

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 can accurately predict atomic dipole moment in quasi-static electric fields across a wider intensity range by using Padé expansion instead of unstable polynomial series for high intensities up to 1.4 x 10 14 W/cm2, as this approximation describes well the numerical results both for low and high intensities including those for which the series (4) is inapplicable.

  2. Improved AI can model nonlinear optical processes beyond the perturbative regime by using a nonlinear oscillator model defined by Eq. (18), which "describes well the nonperturbative growth of the efficiency with the laser intensity for other nonlinear optical processes, namely, the third and fifth harmonic generation in the IR field and the optical rectification in a two-color field."

  3. Improved AI can analyze spectral dependence of nonlinear polarizabilities by solving coupled equations derived from Eq. (37) and Eq. (39) within a Duffing oscillator framework to find solutions for first and third harmonics, which agree well with TDSE numerical results when considering IR fields and non-perturbative increase of these polarizabilities predicted by the models is confirmed by the TDSE calculations.

Abstract

We find the atomic response to the intense femtosecond laser pulse via solving numerically the three-dimensional non-stationary Schrödinger equation (TDSE) for a model atom and calculating its dipole moment. For weak quasi-static fields, the response is well described by a perturbation approach, but for intensities higher than about 0.6,, 10 14 W/cm squared the accuracy of this description is unsatisfactory, regardless of the order of non-linearity taken into account. We suggest fitting the numerical TDSE solution results with a Padé expansion, and show that this approximation describes the response well both in the perturbative regime and beyond it for intensities approximately up to 1.4,, 10 14 W/cm squared. To consider the non-perturbative nonlinearity beyond the quasi-static limit we use the model of nonlinear oscillator with the restoring force defined by the found Padé expression. Our model fails to predict the behaviour of the nonlinear refractive index in the nonperturbative domain, but it describes well the nonperturbative growth of the efficiency with the laser intensity for other nonlinear optical processes, namely, the third and fifth harmonic generation in the IR field and the optical rectification in a two-color field.

Related papers