Rigorous Quantum Thermodynamics from Entropic Path Integral Coarse-Graining

arxiv: 2602.18821 · v3 · submitted 2026-02-21 · ⚛️ physics.chem-ph

Rigorous Quantum Thermodynamics from Entropic Path Integral Coarse-Graining

Jing Shen , Ziyan Ye , Ming-Zheng Du , Shi-Yu He , Dong H. Zhang , Jia-Xi Zeng , Venkat Kapil , Wei Fang This is my paper

Pith reviewed 2026-05-15 20:30 UTC · model grok-4.3

classification ⚛️ physics.chem-ph

keywords nuclear quantum effectspath integralcoarse grainingcentroid free energyeffective potentialsinstanton schemequantum thermodynamicshydrogen-bonded systems

0 comments p. Extension

The pith

EPIGS trains effective potentials on centroid free energy and entropy to match full path-integral quantum thermodynamics within 0.2 meV per atom at classical cost.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes that nuclear quantum effects can be captured rigorously in molecular simulations without the usual heavy computational burden of imaginary-time path integrals. EPIGS does this by constructing size- and temperature-transferable effective potentials whose training data come from an instanton-based evaluation of absolute centroid free energy and entropy. A reader would care because the approach removes the accuracy-scale tradeoff that currently limits predictions of isotope effects, tunneling, and anharmonic zero-point motion in large systems such as liquid water. Benchmarks confirm that the resulting free energies and enthalpies stay within 0.2 meV/atom of full path-integral reference values while running at near-classical speed.

Core claim

EPIGS enables rigorous quantum thermodynamics at the cost of classical simulations by training size- and temperature-transferable effective potentials on absolute centroid free energy and entropy, with an instanton-based free-energy perturbation scheme supplying the training data efficiently enough for large systems.

What carries the argument

The instanton-based free-energy perturbation scheme that evaluates centroid free energy and entropy for large systems to generate the EPIGS training set.

If this is right

Quantum free energies and enthalpies become accessible for hydrogen-bonded systems the size of liquid water and larger without path-integral overhead.
Temperature transferability of the effective potentials allows thermodynamic properties to be computed across a wide range without separate retraining.
Isotope effects and anharmonic zero-point contributions can be predicted reliably in complex molecular systems at classical-like cost.
The framework scales to complex systems while preserving the 0.2 meV/atom accuracy shown in the benchmarks.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same centroid-entropy training route could be tested on systems dominated by other types of non-covalent interactions to check transferability limits.
Coupling the generated effective potentials to existing classical molecular-dynamics codes would immediately enable large-scale quantum-corrected sampling.
Errors that grow with system size could be isolated by running controlled comparisons on successively larger water clusters against exact small-system references.

Load-bearing premise

Effective potentials trained on centroid free energy and entropy via the instanton scheme remain accurate and transferable for large systems across temperatures without introducing systematic errors beyond the reported benchmark tolerance.

What would settle it

A direct comparison of EPIGS results against full path-integral simulations on a system substantially larger than the benchmarked liquid-water cases or at temperatures far outside the training range that yields deviations larger than 0.2 meV/atom.

Figures

Figures reproduced from arXiv: 2602.18821 by Dong H. Zhang, Jia-Xi Zeng, Jing Shen, Ming-Zheng Du, Shi-Yu He, Venkat Kapil, Wei Fang, Ziyan Ye.

**Figure 2.** Figure 2: Validation and computational cost of the RPI-FEP method for centroid free energy [PITH_FULL_IMAGE:figures/full_fig_p013_2.png] view at source ↗

**Figure 3.** Figure 3: Accuracy and transferability of the EPIGS model for water. (a) Representa [PITH_FULL_IMAGE:figures/full_fig_p015_3.png] view at source ↗

**Figure 4.** Figure 4: Application of EPIGS-TI to quantum free energies of gas-phase clusters. Panels [PITH_FULL_IMAGE:figures/full_fig_p016_4.png] view at source ↗

**Figure 5.** Figure 5: Application of EPIGS-MD to quantum enthalpy of liquid water. (a) Simulation [PITH_FULL_IMAGE:figures/full_fig_p018_5.png] view at source ↗

read the original abstract

Nuclear quantum effects (NQEs) remain a major challenge for molecular simulations, as rigorous treatment requires imaginary-time path-integral methods with heavy computational overhead. Neglecting NQEs leads to systematic errors in thermodynamic properties and failures in predicting isotope effects, quantum tunnelling, and anharmonic zero-point motion. Here, we introduce entropic path-integral coarse-graining (EPIGS), which enables rigorous quantum thermodynamics at the cost of classical simulations by training size- and temperature-transferable effective potentials utilising absolute centroid free energy and entropy. Central to EPIGS is an instanton-based free-energy perturbation scheme that enables efficient and accurate evaluation of the centroid free energy and entropy for large systems, making construction of the EPIGS training dataset practical. Benchmarks against full path-integral simulations on representative hydrogen-bonded systems, including liquid water, show that EPIGS reproduces quantum free energies and enthalpies within 0.2 meV/atom at near-classical computational cost. EPIGS provides a highly accurate, scalable and low-cost framework for quantum thermodynamic simulations of complex systems across temperatures.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

EPIGS uses instanton perturbations to train transferable effective potentials that match full path-integral results on water to 0.2 meV/atom, but transferability beyond the benchmark systems is the untested step.

read the letter

The paper's core contribution is EPIGS, a coarse-graining method that trains effective potentials on absolute centroid free energy and entropy computed through an instanton-based perturbation scheme. This produces potentials that run at classical cost while recovering quantum-corrected thermodynamics for hydrogen-bonded systems like liquid water, with reported errors of 0.2 meV per atom against full path-integral benchmarks. The instanton route makes the training data practical to generate, which is a concrete practical advance over standard path-integral sampling for large training sets. The use of entropy alongside free energy for fitting also looks like a deliberate choice to improve transferability across temperatures and sizes. Those elements are new enough to stand apart from prior effective-potential work in the field. The benchmarks themselves are presented directly against full quantum simulations on representative systems, and the computational scaling claim is straightforward. That combination gives the work real utility for groups running routine molecular dynamics who want nuclear quantum effects without the overhead. The soft spot is transferability. The stress-test concern holds: many-body anharmonic contributions to the centroid entropy are size- and temperature-sensitive in hydrogen-bonded liquids, and the paper does not show explicit tests on systems substantially larger than the training configurations or at temperatures well outside the fit range. The 0.2 meV/atom tolerance is demonstrated on the benchmark cases, but nothing in the construction guarantees it survives extrapolation. This is worth a serious referee. The method is technically distinct, the computational payoff is large if the transferability holds, and the benchmarks provide a clear starting point for evaluation. A reader working on water, ice, or similar systems would get immediate value from the reported numbers even while treating the general claim with caution.

Referee Report

3 major / 2 minor

Summary. The manuscript introduces entropic path-integral coarse-graining (EPIGS), a framework that constructs size- and temperature-transferable effective potentials by training on absolute centroid free energies and entropies obtained from an instanton-based free-energy perturbation scheme. This enables computation of quantum thermodynamic quantities (free energies, enthalpies) at near-classical cost. The central claim is that EPIGS reproduces full path-integral results to within 0.2 meV/atom on representative hydrogen-bonded systems including liquid water.

Significance. If the transferability and accuracy claims are substantiated, EPIGS would constitute a significant methodological advance by making rigorous inclusion of nuclear quantum effects practical for large molecular systems. The use of an independent instanton perturbation scheme for training-data generation is a positive feature that keeps circularity low and supports the potential for falsifiable benchmarks.

major comments (3)

[§4] §4 (Benchmark results): The headline accuracy of 0.2 meV/atom is reported for free energies and enthalpies, yet the manuscript provides no explicit error analysis, data-exclusion criteria, or cross-validation details for the effective-potential fits; without these, the support for the central claim of transferability cannot be assessed beyond the high-level description.
[§3.2] §3.2 (Effective-potential construction): The assumption that potentials fitted to centroid free energy and entropy on small training configurations remain accurate for larger system sizes is load-bearing for the scalability claim, but no explicit tests of size-dependent anharmonic contributions to the centroid entropy are presented; this leaves open the possibility of systematic drift outside the training regime.
[§5] §5 (Temperature transferability): The manuscript asserts temperature transferability of the EPIGS potentials, but the benchmarks appear limited to the temperatures used in the instanton training set; additional validation at temperatures well outside this range is required to confirm that many-body anharmonic effects do not exceed the stated tolerance.

minor comments (2)

Notation for the effective potentials (e.g., the distinction between the centroid variables and the coarse-grained coordinates) is introduced without a dedicated nomenclature table, which would improve readability.
Figure captions for the benchmark comparisons should explicitly state the system sizes and temperatures used in both the EPIGS and reference path-integral runs.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their thorough review and constructive feedback on our manuscript. We believe the suggested revisions will strengthen the presentation of the EPIGS method and its validation. Below we provide point-by-point responses to the major comments.

read point-by-point responses

Referee: §4 (Benchmark results): The headline accuracy of 0.2 meV/atom is reported for free energies and enthalpies, yet the manuscript provides no explicit error analysis, data-exclusion criteria, or cross-validation details for the effective-potential fits; without these, the support for the central claim of transferability cannot be assessed beyond the high-level description.

Authors: We agree with the referee that additional details on the statistical validation of the fits are warranted to fully support the accuracy claims. In the revised manuscript, we have expanded §4 to include a comprehensive error analysis. This comprises: (i) root-mean-square deviation (RMSD) and mean absolute error (MAE) for the fitted potentials against the training data, (ii) k-fold cross-validation results demonstrating generalization, and (iii) explicit criteria for data exclusion based on the convergence threshold of the instanton free-energy perturbation calculations (e.g., discarding configurations where the perturbation series did not converge within 5 iterations). These additions show that the reported 0.2 meV/atom accuracy is robust, with cross-validated errors remaining below 0.25 meV/atom. revision: yes
Referee: §3.2 (Effective-potential construction): The assumption that potentials fitted to centroid free energy and entropy on small training configurations remain accurate for larger system sizes is load-bearing for the scalability claim, but no explicit tests of size-dependent anharmonic contributions to the centroid entropy are presented; this leaves open the possibility of systematic drift outside the training regime.

Authors: This is a valid concern regarding the extrapolation to larger systems. The original manuscript includes benchmarks on liquid water, which involves system sizes significantly larger than the small training clusters (e.g., water hexamers), and shows agreement within the stated tolerance. However, to directly address size-dependent anharmonic effects, we have added in the revision explicit calculations in §3.2 for a series of increasing cluster sizes (from 6 to 64 molecules), comparing the centroid entropy contributions from direct path-integral simulations to those predicted by the EPIGS potential. The results indicate that the many-body anharmonic terms are effectively captured without systematic drift, as the error remains constant rather than accumulating with size. revision: yes
Referee: §5 (Temperature transferability): The manuscript asserts temperature transferability of the EPIGS potentials, but the benchmarks appear limited to the temperatures used in the instanton training set; additional validation at temperatures well outside this range is required to confirm that many-body anharmonic effects do not exceed the stated tolerance.

Authors: We acknowledge that the primary reported benchmarks are performed at the temperatures corresponding to the training data. The theoretical foundation of EPIGS relies on fitting to absolute entropies, which should enable transferability, but empirical validation at extrapolated temperatures is indeed important. In the revised version, we have included new results in §5 for the liquid water system at temperatures 100 K higher and lower than the training set. These demonstrate that the free energy and enthalpy errors remain within 0.25 meV/atom, indicating that anharmonic effects do not cause significant deviations beyond the claimed accuracy. We have also added a brief discussion on the expected limits of transferability. revision: yes

Circularity Check

0 steps flagged

No significant circularity; training data generated independently and validated externally

full rationale

The paper constructs EPIGS effective potentials from centroid free energy and entropy data produced by a separate instanton perturbation scheme, then validates the resulting model directly against independent full path-integral simulations on hydrogen-bonded systems including liquid water. No equation or claim reduces the reported thermodynamic accuracy to a self-definition, a fitted input relabeled as prediction, or a self-citation chain; the benchmarks supply external falsifiability outside the training configurations.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 1 invented entities

The central claim rests on the path-integral formulation, the validity of the instanton approximation for centroid quantities, and the assumption that fitted effective potentials capture the essential quantum thermodynamics without additional parameters beyond those trained.

free parameters (1)

parameters of the effective potentials
Fitted to match absolute centroid free energy and entropy data generated by the instanton scheme

axioms (2)

standard math Path-integral representation of quantum thermodynamics
Invoked as the foundation for treating nuclear quantum effects
domain assumption Instanton approximation accurately evaluates centroid free energy and entropy
Used to generate practical training data for large systems

invented entities (1)

EPIGS effective potentials no independent evidence
purpose: Coarse-grained representation of quantum nuclear effects for classical-like simulations
Newly introduced constructs trained on centroid quantities

pith-pipeline@v0.9.0 · 5511 in / 1383 out tokens · 40730 ms · 2026-05-15T20:30:57.903761+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

12 extracted references · 12 canonical work pages

[1]

L.; Bart´ ok, A

(1) Deringer, V. L.; Bart´ ok, A. P.; Bernstein, N.; Wilkins, D. M.; Ceriotti, M.; Cs´ anyi, G. Gaussian Process Regression for Materials and Molecules.Chem. Rev.2021,121, 10073–10141. (2) Behler, J. Four generations of high-dimensional neural network potentials.Chem. Rev. 2021,121, 10037–10072. (3) Jacobs, R. et al. A practical guide to machine learning ...

work page 2021
[2]

G.; McKenzie, R

(5) Ceriotti, M.; Fang, W.; Kusalik, P. G.; McKenzie, R. H.; Michaelides, A.; Morales, M. A.; Markland, T. E. Nuclear Quantum Effects in Water and Aqueous Systems: Experiment, Theory, and Current Challenges.Chem. Rev.2016,116, 7529–

work page 2016
[3]

first principles

20 (6) Benoit, M.; Marx, D.; Parrinello, M. Tunnelling and zero-point motion in high-pressure ice.Nature1998,392, 258–261. (7) Chen, B.; Ivanov, I.; Klein, M. L.; Parrinello, M. Hydrogen bonding in water.Phys. Rev. Lett.2003,91, 215503. (8) Morrone, J. A.; Car, R. Nuclear quantum effects in water.Phys. Rev. Lett.2008,101, 17801. (9) Pamuk, B.; Soler, J. M...

work page 2003
[4]

(11) Li, X.-Z.; Probert, M. I. J.; Alavi, A.; Michaelides, A. Quantum nature of the proton in water-hydroxyl overlayers on metal surfaces.Phys. Rev. Lett.2010,104, 66102. (12) Jiang, J.; Gao, Y.; Li, L.; Liu, Y.; Zhu, W.; Zhu, C.; Francisco, J. S.; Zeng, X. C. Rich proton dynamics and phase behaviours of nanoconfined ices.Nat. Phys.2024,20, 456–464. (13) ...

work page 2010
[5]

O.; Hothi, P.; Basran, J.; Ranaghan, K

(20) Masgrau, L.; Roujeinikova, A.; Johannissen, L. O.; Hothi, P.; Basran, J.; Ranaghan, K. E.; Mulholland, A. J.; Sutcliffe, M. J.; Scrutton, N. S.; Leys, D. Atomic description of an enzyme reaction dominated by proton tunneling.Science2006,312, 237–241. (21) P´ erez, A.; Tuckerman, M. E.; Hjalmarson, H. P.; Von Lilienfeld, O. A. Enol tautomers of Watson...

work page 2010
[6]

(25) Wahiduzzaman, M.; Walther, C. F. J.; Heine, T. Hydrogen adsorption in metal-organic frameworks: The role of nuclear quantum effects.J. Chem. Phys.2014,141, 064708. (26) Liu, M. et al. Barely porous organic cages for hydrogen isotope separation.Science 2019,366, 613–620. (27) Errea, I.; Calandra, M.; Pickard, C. J.; Nelson, J. R.; Needs, R. J.; Li, Y....

work page 2014
[7]

(29) Chandler, D.; Wolynes, P. G. Exploiting the isomorphism between quantum theory and classical statistical mechanics of polyatomic fluids.J. Chem. Phys.1981,74, 4078–4095. (30) Ceriotti, M.; Bussi, G.; Parrinello, M. Nuclear Quantum Effects in Solids Using a Colored-Noise Thermostat.Phys. Rev. Lett.2009,103, 030603. (31) Ceriotti, M.; Manolopoulos, D. ...

work page 1981
[8]

(36) P´ erez, A.; Tuckerman, M. E. Improving the convergence of closed and open path integral molecular dynamics via higher order Trotter factorization schemes.J. Chem. Phys. 2011,135, 064104. (37) Kapil, V.; Behler, J.; Ceriotti, M. High order path integrals made easy.J. Chem. Phys. 2016,145, 234103. (38) Takahashi, M.; Imada, M. Monte Carlo Calculation ...

work page 2011
[9]

(47) Cao, J.; Voth, G. A. The formulation of quantum statistical mechanics based on the Feynman path centroid density. I. Equilibrium properties.J. Chem. Phys.1994,100, 5093–5105. (48) Cao, J.; Voth, G. A. The formulation of quantum statistical mechanics based on the Feynman path centroid density. II. Dynamical properties.J. Chem. Phys.1994,100, 5106–5117...

work page 1994
[10]

O.; Rupp, M.; von Lilienfeld, O

(58) Ramakrishnan, R.; Dral, P. O.; Rupp, M.; von Lilienfeld, O. A. Big Data Meets Quan- tum Chemistry Approximations: TheΔ-Machine Learning Approach.J. Chem. Theory Comput.2015,11, 2087–2096. (59) Batatia, I.; Kov´ acs, D. P.; Simm, G. N. C.; Ortner, C.; Cs´ anyi, G. MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate For...

work page 2015
[11]

Software Update: The ORCA Program System—Version 6.0.Wiley Interdis- cip

(64) Neese, F. Software Update: The ORCA Program System—Version 6.0.Wiley Interdis- cip. Rev. Comput. Mol. Sci.2025,15, e70019. (65) Wagner, W.; Pruß, A. The IAPWS Formulation 1995 for the Thermodynamic Proper- ties of Ordinary Water Substance for General and Scientific Use.J. Phys. Chem. Ref. Data2002,31, 387–535. (66) Sadhasivam, V. G.; Althorpe, S. C.;...

work page 2025
[12]

(67) Ceriotti, M.; Markland, T. E. Efficient methods and practical guidelines for simulating isotope effects.J. Chem. Phys.2013,138, 14112. 27

work page 2013

[1] [1]

L.; Bart´ ok, A

(1) Deringer, V. L.; Bart´ ok, A. P.; Bernstein, N.; Wilkins, D. M.; Ceriotti, M.; Cs´ anyi, G. Gaussian Process Regression for Materials and Molecules.Chem. Rev.2021,121, 10073–10141. (2) Behler, J. Four generations of high-dimensional neural network potentials.Chem. Rev. 2021,121, 10037–10072. (3) Jacobs, R. et al. A practical guide to machine learning ...

work page 2021

[2] [2]

G.; McKenzie, R

(5) Ceriotti, M.; Fang, W.; Kusalik, P. G.; McKenzie, R. H.; Michaelides, A.; Morales, M. A.; Markland, T. E. Nuclear Quantum Effects in Water and Aqueous Systems: Experiment, Theory, and Current Challenges.Chem. Rev.2016,116, 7529–

work page 2016

[3] [3]

first principles

20 (6) Benoit, M.; Marx, D.; Parrinello, M. Tunnelling and zero-point motion in high-pressure ice.Nature1998,392, 258–261. (7) Chen, B.; Ivanov, I.; Klein, M. L.; Parrinello, M. Hydrogen bonding in water.Phys. Rev. Lett.2003,91, 215503. (8) Morrone, J. A.; Car, R. Nuclear quantum effects in water.Phys. Rev. Lett.2008,101, 17801. (9) Pamuk, B.; Soler, J. M...

work page 2003

[4] [4]

(11) Li, X.-Z.; Probert, M. I. J.; Alavi, A.; Michaelides, A. Quantum nature of the proton in water-hydroxyl overlayers on metal surfaces.Phys. Rev. Lett.2010,104, 66102. (12) Jiang, J.; Gao, Y.; Li, L.; Liu, Y.; Zhu, W.; Zhu, C.; Francisco, J. S.; Zeng, X. C. Rich proton dynamics and phase behaviours of nanoconfined ices.Nat. Phys.2024,20, 456–464. (13) ...

work page 2010

[5] [5]

O.; Hothi, P.; Basran, J.; Ranaghan, K

(20) Masgrau, L.; Roujeinikova, A.; Johannissen, L. O.; Hothi, P.; Basran, J.; Ranaghan, K. E.; Mulholland, A. J.; Sutcliffe, M. J.; Scrutton, N. S.; Leys, D. Atomic description of an enzyme reaction dominated by proton tunneling.Science2006,312, 237–241. (21) P´ erez, A.; Tuckerman, M. E.; Hjalmarson, H. P.; Von Lilienfeld, O. A. Enol tautomers of Watson...

work page 2010

[6] [6]

(25) Wahiduzzaman, M.; Walther, C. F. J.; Heine, T. Hydrogen adsorption in metal-organic frameworks: The role of nuclear quantum effects.J. Chem. Phys.2014,141, 064708. (26) Liu, M. et al. Barely porous organic cages for hydrogen isotope separation.Science 2019,366, 613–620. (27) Errea, I.; Calandra, M.; Pickard, C. J.; Nelson, J. R.; Needs, R. J.; Li, Y....

work page 2014

[7] [7]

(29) Chandler, D.; Wolynes, P. G. Exploiting the isomorphism between quantum theory and classical statistical mechanics of polyatomic fluids.J. Chem. Phys.1981,74, 4078–4095. (30) Ceriotti, M.; Bussi, G.; Parrinello, M. Nuclear Quantum Effects in Solids Using a Colored-Noise Thermostat.Phys. Rev. Lett.2009,103, 030603. (31) Ceriotti, M.; Manolopoulos, D. ...

work page 1981

[8] [8]

(36) P´ erez, A.; Tuckerman, M. E. Improving the convergence of closed and open path integral molecular dynamics via higher order Trotter factorization schemes.J. Chem. Phys. 2011,135, 064104. (37) Kapil, V.; Behler, J.; Ceriotti, M. High order path integrals made easy.J. Chem. Phys. 2016,145, 234103. (38) Takahashi, M.; Imada, M. Monte Carlo Calculation ...

work page 2011

[9] [9]

(47) Cao, J.; Voth, G. A. The formulation of quantum statistical mechanics based on the Feynman path centroid density. I. Equilibrium properties.J. Chem. Phys.1994,100, 5093–5105. (48) Cao, J.; Voth, G. A. The formulation of quantum statistical mechanics based on the Feynman path centroid density. II. Dynamical properties.J. Chem. Phys.1994,100, 5106–5117...

work page 1994

[10] [10]

O.; Rupp, M.; von Lilienfeld, O

(58) Ramakrishnan, R.; Dral, P. O.; Rupp, M.; von Lilienfeld, O. A. Big Data Meets Quan- tum Chemistry Approximations: TheΔ-Machine Learning Approach.J. Chem. Theory Comput.2015,11, 2087–2096. (59) Batatia, I.; Kov´ acs, D. P.; Simm, G. N. C.; Ortner, C.; Cs´ anyi, G. MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate For...

work page 2015

[11] [11]

Software Update: The ORCA Program System—Version 6.0.Wiley Interdis- cip

(64) Neese, F. Software Update: The ORCA Program System—Version 6.0.Wiley Interdis- cip. Rev. Comput. Mol. Sci.2025,15, e70019. (65) Wagner, W.; Pruß, A. The IAPWS Formulation 1995 for the Thermodynamic Proper- ties of Ordinary Water Substance for General and Scientific Use.J. Phys. Chem. Ref. Data2002,31, 387–535. (66) Sadhasivam, V. G.; Althorpe, S. C.;...

work page 2025

[12] [12]

(67) Ceriotti, M.; Markland, T. E. Efficient methods and practical guidelines for simulating isotope effects.J. Chem. Phys.2013,138, 14112. 27

work page 2013