Quantum Tunneling from Perturbation Theory Revisited

arXiv:2609.10134 · hep-th, hep-lat · Submitted 2026-09-09 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: High Energy Physics: Theory. The day in high-energy theory: gravity, strings, fields and the questions left open.

Kai: I'm Kai.

Mira: And I'm Mira, here with Lev.

Kai: We're looking at the paper: "Quantum Tunneling from Perturbation Theory Revisited".

Mira: Here's the gist of the paper. Quantum tunneling from perturbation theory revisited reexamines Suzuki and Yasuta's compact formula for extracting vacuum decay rates from perturbative coefficients in D-dimensional O(N) theories, finding that this approach converges in an intermediate coupling regime and yields results close to semi-classical approximations.

Kai: The Problem with Conventional Perturbation Theory The study investigates the interplay between perturbation theory and non-perturbative phenomena, noting that the convergence of a perturbative series is linked to vacuum instability under a sign flip of the coupling constant, and tunneling phenomena obstruct Borel summability.

Mira: The Euclidean action considered is for the D-dimensional O(N) λ(ϕ2)2 theory with an unbounded potential, where the classical vacuum is metastable and decays via quantum tunneling.

Kai: The decay width per unit volume Γ is defined as the imaginary part of the ground state energy density E, given by Γ = −2 Im E/ħ.

Title and authors: Kai: So let's go over the title and the authors of this paper, "Quantum Tunneling from Perturbation Theory Revisited." It’s a revisiting of an older idea.

Mira: They are taking Suzuki and Yasuta’s compact formula for extracting vacuum decay rates from conventional perturbative coefficients in D-dimensional O(N) lambda phi squared theory.

Lev: So the core idea is using those vacuum bubble diagrams, which are the conventional perturbative coefficients, to find the imaginary part that comes from the Borel integral along a discontinuity of that transform.

Kai: The paper notes that this formula worked okay for D equal to one, but it wasn't clear for D greater than or equal to two because they only had limited perturbative coefficients available there.

Mira: This new work is about fixing that convergence problem by using a finite-order truncated Borel transform with those two tunable parameters, b and s.

Lev: So what does this mean for the listener? It means they are trying to make sure you can actually get a meaningful physical rate when the math gets really complicated in higher dimensions.

Kai: Right, it’s about making sure that when we look at these coefficients, we aren't just seeing something that blows up or doesn't converge.

Mira: They introduce a conformal mapping technique for the integration variable to map it onto a disk, which helps keep the integration contour confined as you deform it.

Lev: That sounds like they’re trying to tame the divergence by changing the geometry of the math itself before they even start calculating anything concrete.

Kai: It's about controlling where you are in your mathematical space so you can isolate that physical imaginary part.

Mira: And this setup lets them identify that specific imaginary part when they deform it as z moves toward z plus i epsilon.

The paper's summary: Kai: Now, let’s look at the main summary of "Quantum Tunneling from Perturbation Theory Revisited." It explains that the interplay between perturbation theory and non-perturbative physics is actually quite intriguing.

Mira: They organize this whole field using concepts like resurgence and trans-series to describe how these two things interact.

Lev: Basically, they’re saying that tunneling phenomena mediated by instantons obstruct the Borel summability of the perturbation series, so you can't just sum up all the perturbative terms easily.

Kai: But they argue that this interplay is organized under these notions, and this paper uses them to extract the imaginary part.

Mira: They consider a specific theory where the Euclidean action is S equals Z d x one two partial over mu phi partial over mu phi plus one two m squared phi squared minus one four factorial g of mu phi squared squared.

Lev: They set m square greater than zero and g greater than zero so that the potential energy is unbounded from below, making the classical vacuum at phi equal to zero metastable.

Kai: And this leads to a decay width per unit volume, which is defined as the imaginary part of the ground state energy density E.

Mira: The key result they are pushing is that while one-loop determinants behave like e to negative S zero over ge raised to the power of one, the power series expansion of that behaves like it vanishes to all orders in ge.

Lev: So this means in perturbation theory alone, the tunneling is effectively invisible; it doesn't show up at all.

Kai: But they argue that because the Borel summation can be done after analytic continuation from g less than zero to g greater than zero, we can use that process to yield the imaginary part.

Mira: They say this process should give the same answer whether you look at the unique real saddle point at phi equal to zero when g is negative, or when you sum over infinitely many saddle points when g is positive.

The paper's improvements: Kai: So what are the actual improvements they suggest in this paper? They focus on using that Borel resummation approach we just talked about.

Mira: The key improvement is introducing those two tunable parameters, b and s into a finite-order truncated Borel transform to extract the non-perturbative information.

Lev: This is where the AI system can now perform adaptive Borel resummation for non-perturbative physics, allowing it to extract vacuum decay widths from conventional perturbative coefficients.

Kai: And because of this, we can dynamically tune parameter s to minimize that quantity number three point three, which mimics a criterion they used previously.

Mira: This tuning leads to convergence of the imaginary part with respect to perturbative order P, even in regions where previous methods showed no sign of convergence.

Lev: That’s useful for real hardware because it means you don't have to trust that the result will be stable just because you picked a certain coupling constant.

Kai: And they compare the resummed result against both the leading order semi-classical approximation with that one-loop determinant and higher loop corrections.

Mira: They found that in the intermediate coupling region, these results are close to the semi-classical result but about ten to twenty percent larger than what you get with just the two-loop correction.

Lev: So they’re saying that those higher orders around the bounce solution are actually quite large in this region and almost cancel out the two loop correction.

Kai: This suggests that using these parameters b and s is a way to get results that are comparable to leading semi-classical approximations.

Conclusion: Mira: So, let’s wrap up with the conclusion of "Quantum Tunneling from Perturbation Theory Revisited." They summarize that this approach works well in the intermediate coupling regime and gives results close to leading semi-classical approximations.

Kai: The main implication is that they proved you can extract this non-perturbative quantity from conventional perturbation theory for D equal to two and N equal to one.

Lev: It validates the strategy of using tunable parameters b and s to improve convergence in these kinds of problems.

Mira: It also shows it works in both the field theory case they are studying and that analogous quantum mechanical check, which really supports this method.

Kai: The method’s success in both cases points toward a robust strategy for extracting these quantities using conventional perturbation theory.

Lev: So we've seen how parameter tuning helps bring the discussion to a point where it yields results comparable to what we expect from semi-classical calculations.

Hiroshi Suzuki, Yasuta

Department of Physics, Kyushu University · Quantum and Spacetime Research Institute (QuaSR), Kyushu University

hep-th, hep-lat

Submitted: 2026-09-09

Updated: 2026-10-31

Comments: 30 pages, 9 figures, comparison with a continuum extrapolation of the linked-cluster expansion in a quantum $ϕ^4$ chain is given; the final version to appear in PTEP

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

Importance score: 79/100

The gist: Quantum tunneling from perturbation theory revisited reexamines Suzuki and Yasuta's compact formula for extracting vacuum decay rates from perturbative coefficients in D-dimensional O(N) theories…

Key concepts

Quantum Tunneling
This refers to a non-perturbative process where a system can pass through an energy barrier even if it classically lacks the necessary energy. In this context, it describes how the metastable vacuum of a field theory decays over time via quantum mechanical fluctuations.
Borel Resummation
This is a technique used in quantum field theory to sum up an infinite series of perturbative calculations that otherwise diverge. The method involves transforming the series into a Borel transform, which allows for the extraction of non-perturbative information like decay rates from the coefficients.
Intermediate Coupling Regime
This is a specific range for the coupling constant (ge) where standard perturbation theory often fails to converge. The paper shows that by tuning parameters b and s, this method achieves convergence in this regime, producing reliable results that are close to semi-classical predictions.

Terminology

Summary

Quantum tunneling from perturbation theory revisited reexamines Suzuki and Yasuta's compact formula for extracting vacuum decay rates from perturbative coefficients in D-dimensional O(N) theories, finding that this approach converges in an intermediate coupling regime and yields results close to semi-classical approximations.

The Problem with Conventional Perturbation Theory

The study investigates the interplay between perturbation theory and non-perturbative phenomena, noting that the convergence of a perturbative series is linked to vacuum instability under a sign flip of the coupling constant, and tunneling phenomena obstruct Borel summability. The Euclidean action considered is for the D-dimensional O(N) λ(ϕ2)2 theory with an unbounded potential, where the classical vacuum is metastable and decays via quantum tunneling. p. 1 The decay width per unit volume Γ is defined as the imaginary part of the ground state energy density E, given by Γ = −2 Im E/ħ. p. 1

Revisiting the Borel Resummation Approach

The authors reexamine Suzuki and Yasuta's proposal to extract the decay width from conventional perturbative coefficients using a finite-order truncated Borel transform with two tunable parameters, b and s. They introduced a conformal mapping for the integration variable z to map the complex z-plane onto a disk, which confines the integration contour of the Borel integral. p. 5 This mapping allows for the identification of an imaginary part that corresponds to the physical decay width by deforming the integration path as z → z + iε. p. 4

Convergence and Parameter Tuning

The convergence of the imaginary part with respect to perturbative order is shown when tuning parameters b and s, particularly in the intermediate coupling range 3 ≲ ge ≲ 8. The optimal choice for b is set to b = (D + N)/2, matching the power of ge in Eq. p. 3 (1.2). The parameter s is tuned to minimize a specific quantity (Eq. p. 3 3.3), which mimics the criterion used in previous work. p. 3

Comparison with Semi-Classical Results

The resulting imaginary part in the intermediate coupling region is found to be rather close to the leading-order semi-classical approximation with the one-loop determinant (1.2). However, this result is 10–20% larger than the semi-classical result computed with the two-loop correction. p. 0 This suggests that higher perturbative corrections around the bounce are significant in this region and almost compensate the two-loop correction. p. 3

Quantum Mechanical Consistency

The strategy is validated by an analogous numerical experiment in the quantum mechanical case, D = 1 and N = 1, where an exact result can be obtained to check the method's validity. In this case, for P ≥ 7, the optimal value of s is found to be close to 1/3, which aligns with the strong coupling behavior Im E ∼e ge1/3. p. 3 The paper concludes that it is possible to extract this non-perturbative quantity from conventional perturbation theory for D = 2 and N = 1. p. 2

Conclusion

The present approach successfully extracts the imaginary part of the vacuum energy density from conventional perturbative coefficients, demonstrating convergence in the intermediate coupling region and yielding results comparable to leading semi-classical approximations. The method's success in both field theory and quantum mechanics supports the strategy of using tunable parameters b and s to improve convergence.

References

[1] F. J. Dyson, “The Uses of Instantons,” Subnucl. p. 1 Ser. 15, 805 (1979) HUTP-78-A004

[2] G. ’t Hooft, “Can We Make Sense Out of Quantum Chromodynamics? p. 7,” Subnucl. Ser. 15, 943 (1979) PRINT77-0723 (UTRECHT)

[3] E. B. Bogomolny, Phys. Lett. B 91, 431-435 (1980) doi:10.1016/0370-2693(80)91014-X

[4] J. Zinn-Justin, Nucl. Phys. B 192, 125-140 (1981) doi:10.1063/1.524919

[5] I. Aniceto, G. Ba¸sar and R. Schiappa, Phys. Rept. 809, 1-135 (2019) doi:10.1016/j.physrep.2019.02.003

[6] S. R. Coleman, “The Uses of Instantons,” Subnucl. p. 1 Ser. 15, 805 (1979) HUTP-78-A004

[7] J. J. Loeffel, A. Martin, B. Simon and A. S. Wightman, Phys. Lett. B 30, 656-658 (1969) doi:10.1016/0370-2693(69)90087-2

[8] B. Simon, Annals Phys. 58, 76-136 (1970) doi:10.1016/0003-4916(70)90240-X

[9] S. Graffi, V. Grecchi and B. Simon, Phys. Lett. B 32, 631-634 (1970) doi:10.1016/0370-2693(70)90564-2

[10] J. P. Eckmann, J. Magnen and R. S´en´eor, Commun. Math. Phys. 39, no.4, 251-271 (1975) doi:10.1007/BF01705374

[11] J. Magnen and R. S´en´eor, Commun. Math. Phys. 56, 237 (1977) doi:10.1007/BF0161421

[12] E. Witten, AMS/IP Stud. Adv. Math. 50, 347-446 (2011) [arXiv:1001. p. 02933 [hep-th]]

[13] E. Witten, [arXiv:1009. p. 06032 [hep-th]]

[14] T. Fujimori, M. Honda, S. Kamata, T. Misumi and N. Sakai, PTEP 2018, no.12, 123B03 (2018) doi:10.93/ptep/pty118 [arXiv:KYUSHU-HET-371>

[15] M. Serone, G. Spada and G. p. 0 p. 0 Villadoro, JHEP 05, 056 (2017) doi:10. p. 01007/JHEP05(2017)056 [arXiv:KYUSHU-HET-371>

[16] E. Br´ezin, and G. Parisi, J. Stat. Phys. 19, 269-292 (1978) doi:10.1007/BF0101

[17] E. M. Malatesta, G. p. 0 Parisi, and T. p. 0 Rizzo, Nucl. p. 0 Phys. B 922, 293-318 (2017) [arXiv:KYUSHU-HET-371>

[18] H. Kleinert, Phys. Lett. B 300, 261-270 (1993) doi:10.1016/0370-2693(93)90364-N

[19] R. Karrlein and H. Kleinert, Phys. Lett. A 187, 133 (1994) doi:10.1016/0375-9601(94)90050-7

[20] H. Kleinert and I. Mustapic, Int. J. Mod. Phys. A 11, 4383-4400 (1996) doi:10.1142/S0217751X96002029

[21] R. Seznec and J. Zinn-Justin, J. Math. Phys. 20, 1398 (1979) doi:10.63/1.524247

[22] J. C. Le Guillou and J. Zinn-Justin, Annals Phys. 147, 57–84 (1983) doi:10.1016/0003-4916(83)90067-2

[23] R. Guida, K. Konishi and H. Suzuki, Annals Phys. p. 0 p. 0 241, 152-184 (1995) doi:10.006/aphy.1995.1059 [arXiv:KYUSHU-HET-371>

[24] R. Guida, K. Konishi and H. Suzuki, Annals Phys. 249, 109-145 (1996) doi:10.1006/aphy.1996.0066 [arXiv:KYUSHU-HET-371>

[25] H. Kleinert and W. Janke, Phys. Lett. A 206, 283 (1995) doi:10.1016/0375-9601(95)00521-4 [arXiv:KYUSHU-HET-371>

[26] H. Suzuki and H. p. 0 p. 0 Yasuta, Phys. p. 0 p. 0 p. 0 Lett. B 400, 341-345 (1997) doi:10.1016/S0370-2693(97)00368-7 [arXiv:KYUSHU-HET-371>

[27] H. Suzuki and H. Yasuta, Phys. Rev. D 57, 2500-2506 (1998) doi:10.1103/PhysRevD.57.2500 [arXiv:KYUSHU-HET-371>

[28] H. Yasuta, Phys. Lett. B 418, 145-148 (1998) doi:10.1016/S0370-2693(97)01493-7

[29] J. J. Loeffel, Saclay Report DPh-T/76-20 (1976), unpublished, reprinted in Ref. [31]

[30] M. Serone, G. Spada and G. Villadoro, JHEP 08, 148 (2018) doi:10. p. 01007/JHEP08(2018)148 [arXiv:KYUSHU-HET-371>

[31] J. C. Le Guillou and J. Zinn-Justin, “Large order behavior of perturbation theory,” (North-Holland, Amsterdam, 1990)

[32] C. M. Bender and T. T. Wu, Phys. Rev. 184, 1231-1260 (1969) doi:10.1103/PhysRev.85.631

[33] J. Zinn-Justin, J. Math. Phys. 22, 511 (1981) doi:10.63/1.524919

Appendix A: Quantum Mechanics Check

The quantum mechanical case D = 1 and N = 1 provides a consistency check for the strategy, where the exact result is known. In this case, for P ≥ 7, the optimal value of s is found to be close to 1/3. p. 3 The approach works well in this quantum mechanical case as well. p. 2

Table A1: Quantum Mechanical Results

The ratio [Im E˜]P (b, s0)/[Im E˜]bounce with b = 1. s0 is fixed to the value which minimizes Eq. p. 8 p. 8 p. 8 (3.3) at ge = 2. For P ≥ 7, the deviation from the exact result never exceeds the error estimated by Eq. p. 3 p. 3 p. 3 (3.4). The ratio of the former to the latter is at most 0. p. 6 p. 6 p. 69 and is typically below 0.8.

Table A2: Quantum Mechanical Parameter Tuning

For D = 1, N = 1, Eq.(3.3) at ge=2, b=1, s2 [s0 ¡¢s; s0 + ¢s], where ¢s = 1=2. For P ≥ 7, the optimal value of s is found to be 0. p. 3 p. 32–0.4 and this is close to 1/3, the value expected from the strong coupling behavior Im E ∼e ge1/3. p. 3 p. 3 The ratio [Im Ee(ge)]P (b, s0) (2. p. 7 p. 712) and the leading bounce result (1. p. 7 p. 72) as a function of ge for approximation orders P = 5–10 is shown in Fig. p. 7 p. 7 A3. The exact ratio is tabulated in Table A1. p. 3 p. 3 The final results are close to the leading semi-classical approximation for the intermediate coupling region 3 ≲ ge ≲ 8. p. 2 p. 2 p. 2 p. 2

Table A3: Quantum Mechanical Ratios

The ratio [Im E˜]P (b, s0)/[Im E˜]bounce with b = 1. s0 is fixed to the value which minimizes Eq. (3.3) at ge = 2. For P ≥ 7, the deviation from the exact result never exceeds the error estimated by Eq. (3.4). The ratio of the former to the latter is at most 0.9 and is typically below 0.8. The final results are close to the leading semi-classical approximation for the intermediate coupling region 3 ≲ ge ≲ 8.

Table A4: Quantum Mechanical Parameter Tuning

For D = 1, N = 1, Eq.(3.3) at g =2, b=1, s2 [s0 ¡¢s; s0 + ¢s], where ¢s = 1=2. For P ≥ 7, the optimal value of s is found to be 0.2–0.4 and this is close to 1/3, the value expected from the strong coupling behavior Im E ∼e ge1/3. The ratio [Im Ee(ge)]P (b, s0) (2.12) and the leading bounce result (1.2) as a function of ge for approximation orders P = 5–10 is shown in Fig. A3. The exact ratio is tabulated in Table A1. The final results are close to the leading semi-classical approximation for the intermediate coupling region 3 ≲ ge ≲ 8.

Table A5: Quantum Mechanical Ratios

The ratio [Im E˜]P (b, s0)/[Im E˜]bounce with b = 1. s0 is fixed to the value which minimizes Eq. (3.3) at ge = 2. For P ≥ 7, the deviation from the exact result never exceeds the error estimated by Eq. (3.4). The ratio of the former to the latter is at most 0.9 and is typically below 0.8. The final results are close to the leading semi-classical approximation for the intermediate coupling region 3 ≲ ge ≲ 8.

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. Adaptive Borel Resummation for Non-perturbative Physics: The AI system can now extract non-perturbative quantities like vacuum decay widths from conventional perturbative coefficients by introducing two tunable parameters b and s in a finite-order truncated Borel transform, as detailed in Eq. (2.12).

  2. Convergence Optimization via Parameter Tuning: The system can dynamically tune the parameter s to minimize the quantity (3.3), which imitates the criterion in Ref. [30], leading to convergence of the imaginary part with respect to perturbative order P, even in regions where previous methods showed no sign of convergence.

  3. Semi-classical Approximation Comparison: The improved system can compare its resummed result against both the leading-order semi-classical approximation (1.2) and higher-loop corrections, quantifying how much higher-order corrections in the semi-classical approximation are rather large in the intermediate region.

  4. Coupling Constant Regime Prediction: The AI can predict whether a given coupling constant ge falls into a range where the imaginary part converges with respect to the perturbative order, specifically identifying this as occurring in the intermediate range of 3 <= ge <= 8 for D=2, N=1.

  5. Asymptotic Behavior Modeling: By analyzing the asymptotic behavior derived from Eq. (2.14), the system can model how "the parameter b controls the behavior of the approximation at the weak coupling ge -> 0, whereas s controls the behavior at the strong coupling limit."

Sources

Related papers