arxiv: 2605.03422 · v1 · submitted 2026-05-05 · ⚛️ physics.plasm-ph

Recognition: unknown

Hybrid simulation of the energy cost of O(¹D) and O(³P) generation in a capacitive Ar/O₂ discharge driven by sawtooth-type voltage waveforms

Jun-Xi Guo , Wan Dong , Yuan-Hong Song

Authors on Pith no claims yet

Pith reviewed 2026-05-07 13:09 UTC · model grok-4.3

classification ⚛️ physics.plasm-ph

keywords capacitively coupled plasmaAr/O2 mixtureenergy costO(1D)O(3P)sawtooth waveformdischarge modehybrid simulation

0 comments

The pith

O(³P) generation costs less energy than O(¹D) in Ar/O2 plasmas, and N=2 harmonics at 10% O2 further reduce cost via mode shift.

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

This study simulates how sawtooth voltage waveforms with varying harmonic numbers affect the energy cost of producing O(¹D) and O(³P) in low-pressure Ar/O2 capacitive plasmas. O(³P) generation proves more efficient than O(¹D) under all tested conditions, and both species become cheaper to produce as the oxygen fraction rises. In most cases higher harmonic numbers increase the energy cost, but at 10% O2 the shift from one to two harmonics triggers a change from drift-ambipolar to alpha-drift-ambipolar hybrid mode. This mode change broadens the ionization zone, raises electron density, and boosts generation rates enough to lower the overall energy cost. The findings point to waveform tuning as a way to improve efficiency in processes that use these reactive oxygen atoms for thin-film oxidation.

Core claim

The paper establishes that O(³P) generation is consistently more energy-efficient than O(¹D) generation. The energy cost decreases with higher O2 ratios but increases with harmonic number N, except at 10% O2 where N=2 causes a transition to the α-DA hybrid mode. This transition expands the spatio-temporal range of the ionization rate, enhances its peak value, increases electron density, and thereby significantly enhances generation rates, reducing the energy cost for medium-energy electrons.

What carries the argument

The discharge mode transition from DA to α-DA hybrid mode controlled by the harmonic number N of sawtooth voltage waveforms, which alters the ionization rate profile and electron density to affect generation energy costs.

If this is right

Increasing O2 ratio lowers energy costs for both O(¹D) and O(³P).
Monotonically increasing N does not minimize energy cost; optimal N sustains the hybrid mode.
The hybrid mode at N=2 for 10% O2 reduces cost by increasing electron density and generation rates.
Generation of these species is driven by medium-energy electrons in the 8-20 eV range.

Where Pith is reading between the lines

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

Waveform optimization may extend to other plasma chemistries for better efficiency in semiconductor manufacturing.
Experimental validation in 3D setups could confirm if the 1D-predicted mode shift holds in practical reactors.
Adjusting N could provide a method to tune species generation without altering gas pressure or composition.

Load-bearing premise

The one-dimensional fluid and electron Monte Carlo hybrid model, using standard cross-sections, accurately captures the energy costs and the mode transition in physical three-dimensional discharges.

What would settle it

Direct measurement of O(¹D) and O(³P) densities or production rates in an experimental Ar/O2 discharge with sawtooth waveforms at N=1 and N=2 for 10% O2, to check if the energy cost drops at N=2.

read the original abstract

Low-pressure radio-frequency capacitively coupled plasmas operated in Ar/O$_2$ gas mixtures are widely adopted in critical semiconductor manufacturing processes. O($^3$P) and O($^1$D) are key highly reactive species for oxidation or as oxygen sources for deposited thin films. Optimizing external parameters to realize efficient generation of these species under limited energy deposition is essential for improving process yield.Based on a one-dimensional (1D) fluid/electron Monte Carlo (EMC) hybrid model, this study investigates the energy cost of O($^1$D) and O($^3$P) generation driven by sawtooth up-type voltage waveforms at a fixed peak-to-peak voltage, focusing on the effects of the harmonic number ($N$) and the O$_2$ ratio. The results show that O($^3$P) generation is consistently more efficient than that of O($^1$D). The generation energy cost decreases with increasing O$_2$ ratio, yet increases as $N$ increases. However, in the specific scenario of 10% O$_2$, an inflection point can be observed at $N = 2$. As $N$ increases from 1 to 2, the discharge mode shifts from the DA mode to the $\alpha$-DA hybrid mode, expanding the effective spatio-temporal range of the ionization rate and enhancing its peak, which increases electron density. Consequently, the generation rates are significantly enhanced, leading to a reduction in the generation energy cost.Moreover, as discussed above, monotonically increasing the harmonic number $N$ does not reduce the generation energy cost of O($^1$D) and O($^3$P) associated with medium-energy (8-20 eV) electrons. Only by selecting the appropriate $N$ to sustain the discharge in the hybrid $\alpha$-DA mode, thereby increasing the electron density and promoting the generation of these species, can the generation energy cost be reduced.

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

This 1D hybrid simulation finds that at 10% O2 a sawtooth waveform with N=2 lowers energy cost for O(3P) and O(1D) generation via a DA-to-alpha-DA mode shift that raises electron density, while O(3P) remains cheaper than O(1D) overall.

read the letter

The main result is that energy cost per O atom drops when N goes from 1 to 2 at 10% O2 because the discharge enters a hybrid alpha-DA mode. That mode widens the region where ionization occurs and raises its peak value, which lifts electron density and therefore the generation rates of both O(3P) and O(1D). Beyond N=2 the cost climbs again. O(3P) production stays more efficient than O(1D) at every point they checked, and raising the O2 fraction generally reduces cost except for the N=2 dip at low oxygen content. The authors tie the inflection directly to changes in the spatio-temporal ionization profile rather than to any change in the underlying cross sections.

Referee Report

3 major / 2 minor

Summary. The manuscript employs a one-dimensional fluid/electron Monte Carlo hybrid model to compute the energy costs of O(¹D) and O(³P) generation in low-pressure Ar/O₂ capacitive discharges driven by sawtooth-type voltage waveforms at fixed peak-to-peak voltage. It reports that O(³P) generation is consistently more efficient than O(¹D), that energy costs decrease with rising O₂ fraction but increase with harmonic number N, and that an inflection occurs at N=2 for 10% O₂ because the discharge transitions from DA to α-DA hybrid mode, expanding the ionization-rate region, raising electron density, and thereby lowering the energy cost per generated atom.

Significance. If the modeled trends and mode-transition mechanism hold, the work identifies a concrete waveform-design lever (choice of N to sustain the hybrid mode) for reducing energy expenditure on key reactive oxygen species in semiconductor plasma processing. The explicit linkage between discharge mode, electron-density enhancement, and generation efficiency is a useful physical insight that could guide further optimization studies.

major comments (3)

[Model description and results discussion of mode transition] The explanation for the N=2 inflection at 10% O₂ (abstract and results) rests on the 1D hybrid model correctly predicting the spatio-temporal expansion of the ionization rate and the consequent electron-density increase during the DA-to-α-DA transition. The manuscript does not discuss or test the limitations of the 1D approximation in real three-dimensional CCPs, where radial non-uniformities, edge effects, or standing-wave phenomena could suppress or shift the hybrid mode and alter the claimed generation-rate enhancement.
[Methods and results sections] No experimental benchmarking, validation data, or sensitivity checks against measured electron densities or species production rates are presented for sawtooth waveforms in Ar/O₂. Because the quantitative energy-cost values and the inflection magnitude depend directly on the accuracy of the simulated generation rates and input power, the absence of such checks leaves the central claims vulnerable to model artifacts.
[Results (energy-cost plots and tables)] The energy-cost figures lack reported uncertainties, convergence tests with respect to grid resolution or Monte Carlo particle number, or sensitivity to the chosen literature cross-sections. This is especially relevant for the 10% O₂, N=2 case where the reduction is attributed to a narrow mode-transition window.

minor comments (2)

[Methods] Clarify in the methods how the sawtooth waveform is constructed for each integer N and how the fixed peak-to-peak voltage is maintained while N is varied.
[Figures] Figure captions and axis labels should explicitly indicate which curves correspond to the DA versus α-DA regimes so that the mode-transition argument is visually traceable.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive comments and positive evaluation of the significance of our work. We address each major comment point by point below and have revised the manuscript where appropriate to strengthen the presentation and address limitations.

read point-by-point responses

Referee: [Model description and results discussion of mode transition] The explanation for the N=2 inflection at 10% O₂ (abstract and results) rests on the 1D hybrid model correctly predicting the spatio-temporal expansion of the ionization rate and the consequent electron-density increase during the DA-to-α-DA transition. The manuscript does not discuss or test the limitations of the 1D approximation in real three-dimensional CCPs, where radial non-uniformities, edge effects, or standing-wave phenomena could suppress or shift the hybrid mode and alter the claimed generation-rate enhancement.

Authors: We agree that the one-dimensional model cannot fully capture three-dimensional effects such as radial non-uniformities, edge effects, or standing-wave phenomena that may influence the hybrid mode in real CCP reactors. In the revised manuscript, we have added an explicit discussion of these limitations in the Conclusions section, noting that while the 1D axial treatment isolates the waveform effects and reveals the underlying mechanism, future multidimensional simulations would be required to assess any quantitative shifts in the mode transition or generation-rate enhancement. The physical insight into the role of the hybrid α-DA mode remains a useful guide for waveform design. revision: yes
Referee: [Methods and results sections] No experimental benchmarking, validation data, or sensitivity checks against measured electron densities or species production rates are presented for sawtooth waveforms in Ar/O₂. Because the quantitative energy-cost values and the inflection magnitude depend directly on the accuracy of the simulated generation rates and input power, the absence of such checks leaves the central claims vulnerable to model artifacts.

Authors: This is a computational study using an established hybrid fluid/electron Monte Carlo model. No new experimental data for sawtooth waveforms are presented because matching measurements are not available in the literature and performing dedicated experiments lies outside the scope of the present work. We have added citations to prior experimental validations of the same hybrid model in Ar/O₂ capacitive discharges and included computational sensitivity checks on key input parameters (e.g., cross-section variations and power deposition) in the revised Methods section. The reported trends and mode-transition mechanism are therefore presented as model-derived insights rather than absolute quantitative predictions. revision: partial
Referee: [Results (energy-cost plots and tables)] The energy-cost figures lack reported uncertainties, convergence tests with respect to grid resolution or Monte Carlo particle number, or sensitivity to the chosen literature cross-sections. This is especially relevant for the 10% O₂, N=2 case where the reduction is attributed to a narrow mode-transition window.

Authors: We thank the referee for highlighting this point. In the revised manuscript we now report statistical uncertainties on the energy-cost values arising from the Monte Carlo sampling. Convergence tests with respect to spatial grid resolution and the number of Monte Carlo particles have been performed and summarized in the text, confirming that the results, including the inflection at N=2 for 10% O₂, are converged. A sensitivity analysis to the literature electron-impact cross sections for oxygen species has also been added, demonstrating that the mode-transition-related reduction remains robust within the range of accepted cross-section uncertainties. revision: yes

Circularity Check

0 steps flagged

No circularity: energy costs and mode shifts are direct simulation outputs

full rationale

The paper's central results (O(³P) vs O(¹D) efficiency, dependence on O₂ ratio and N, inflection at N=2 for 10% O₂ due to DA-to-α-DA transition) are obtained by running a 1D fluid/EMC hybrid model, computing generation rates from the simulated electron density and reaction rates, then dividing by input power. These quantities are not defined in terms of each other, nor are any fitted parameters renamed as predictions. The model uses external literature cross-sections; no self-citation chain or ansatz is invoked to force the reported trends. The derivation chain is therefore self-contained and does not reduce to its inputs by construction.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 0 invented entities

The simulation draws on standard low-pressure plasma modeling assumptions and tabulated reaction data; no new physical entities are introduced. The control variables N and O2 ratio are scanned rather than fitted to the reported costs.

free parameters (2)

harmonic number N
Scanned as the primary independent variable while holding peak-to-peak voltage fixed
O2 gas fraction
Scanned as the second independent variable

axioms (2)

domain assumption The 1D fluid/EMC hybrid model assumptions remain valid across the scanned N and O2 range
Invoked to justify quantitative extraction of energy costs and mode identification
domain assumption Literature electron-impact cross sections and reaction rates are sufficiently accurate for energy-cost calculations
Required to compute generation rates and power deposition

pith-pipeline@v0.9.0 · 5681 in / 1797 out tokens · 97132 ms · 2026-05-07T13:09:36.605084+00:00 · methodology

Review history (2 revisions) →

discussion (0)

Reference graph

Works this paper leans on

62 extracted references

[1]

2006 Plasma Electronics: Applications in Microelectronic Device Fabrication (London: Taylor and Francis) pp3–9

Makabe T., Petrović Z. 2006 Plasma Electronics: Applications in Microelectronic Device Fabrication (London: Taylor and Francis) pp3–9

2006
[2]

and Braithwaite N

Chabert P. and Braithwaite N. 2011 Physics of Radio-Frequency Plasmas (Cambridge: Cambridge University Press)

2011
[3]

A., Lichtenberg A

Lieberman M. A., Lichtenberg A. J. 2005 Principles of Plasma Discharges and Materials Processing (New York: Wiley) pp1–750

2005
[4]

N., Cao Y

Fang G.Y., Xu L. N., Cao Y. Q., et al. 2015 Chem. Commun. 51 1341 18

2015
[5]

1998 IEDM 593-596

Nagamine M., Itoh H., Satake H., et al. 1998 IEDM 593-596

1998
[6]

2023 Coatings 13 279

Izdebska-Podsiadły J. 2023 Coatings 13 279

2023
[7]

A., Hess D

Hartney M. A., Hess D. W., Soane D. S. 1989 J. Vac. Sci. Technol. B 7 1

1989
[8]

Huang S., Huard C., Shim S. et al. 2019 J. Vac. Sci. Technol. A 37 031304

2019
[9]

2007 Plasma Sources Sci

Gudmundsson J.T., Thorsteinsson E.G. 2007 Plasma Sources Sci. Technol. 16 399

2007
[10]

and Kushner M

Qu C., Sakiyama Y., Agarwal P. and Kushner M. J. 2021 J. Vac. Sci. Technol. A 39 052403

2021
[11]

Kaspar T., Tuan A., Tonkyn R., et al. 2003 J. Vac. Sci. Technol. B 21 895

2003
[12]

Tanimura T., Watanabe Y., Sato Y., et al. 2013 J. Appl. Phys. 113 064102

2013
[13]

Pilli A., Lee V., Jones J., et al. 2020 J. Phys. Chem. C 124 25846

2020
[14]

Sankaran A., Kushner M.J. 2004 J. Vac. Sci. Technol. A 22 1242

2004
[15]

Hartney M.A., Hess D.W., Soane D.S. 1989 J. Vac. Sci. Technol. B 7 1

1989
[16]

2024 Plasma Sources Sci

Derzsi A., Vass M., Masheyeva R., et al. 2024 Plasma Sources Sci. Technol. 33 025005

2024
[17]

2020 Plasma Sources Sci

Wang L., Wen D.Q., Hartmann P., et al. 2020 Plasma Sources Sci. Technol. 29 105004

2020
[18]

Gudmundsson J.T., Snorrason D.I. 2017 J. Appl. Phys. 122 193302

2017
[19]

2006 Phys

Lee S.H., Iza F., Lee J.K. 2006 Phys. Plasmas 13 057102

2006
[20]

Rauf S., Kushner M.J. 1997 J. Appl. Phys. 82 2805

1997
[21]

Liu J., Zhang Q.Z., Liu Y.X., et al. 2013 J. Phys. D: Appl. Phys. 46 235202

2013
[22]

Nikolić M., Sepulveda I., Gonzalez C., et al. 2021 J. Phys. D: Appl. Phys. 54 275203

2021
[23]

Schulze J., Heil B., Luggenhölscher D., et al. 2008 J. Phys. D: Appl. Phys. 41 042003

2008
[24]

1991 Phys

Surendra M., Graves D. 1991 Phys. Rev. Lett. 66 1469

1991
[25]

Derzsi A., Horváth B., Korolov I., et al. 2019 J. Appl. Phys. 126 043303

2019
[26]

Lafleur T., Chabert P., Booth J.P. 2013 J. Phys. D: Appl. Phys. 46 135201

2013
[27]

2013 Appl

Korolov I., Derzsi A., Donkó Z., et al. 2013 Appl. Phys. Lett. 103 064102

2013
[28]

1986 IEEE Trans

Godyak V., Khanneh A. 1986 IEEE Trans. Plasma Sci. 14 112

1986
[29]

2011 Phys

Schulze J., Derzsi A., Dittmann K., et al. 2011 Phys. Rev. Lett. 107 275001

2011
[30]

2010 Plasma Sources Sci

Proshina O., Rakhimova T., Rakhimov A., et al. 2010 Plasma Sources Sci. Technol. 19 065013

2010
[31]

2016 Phys

Liu Y.X., Schüngel E., Korolov I., et al. 2016 Phys. Rev. Lett. 116 255002

2016
[32]

2024 Chin

Gao Z.Y., Dong W., Tian C.B., et al. 2024 Chin. Phys. B 33 095203

2024
[33]

2025 Plasma Sources Sci

Dong W., Gao Z.Y., Wang L., et al. 2025 Plasma Sources Sci. Technol. 34 025008

2025
[34]

Goto H.H., Löwe H.D., Ohmi T. 1992 J. Vac. Sci. Technol. A 10 3048

1992
[35]

2004 IEEE Trans

Lee J.K., Babaeva N.Y., Kim H.C., et al. 2004 IEEE Trans. Plasma Sci. 32 47

2004
[36]

Zhang Y., Kushner M.J., Sriraman S., et al. 2015 J. Vac. Sci. Technol. A 33 031302

2015
[37]

Schulze J., Gans T., O'Connell D., et al. 2007 J. Phys. D: Appl. Phys. 40 7008

2007
[38]

2013 Plasma Sources Sci

Bi Z.H., Dai Z.L., Zhang Y.R., et al. 2013 Plasma Sources Sci. Technol. 22 055007

2013
[39]

2011 Plasma Sources Sci

Schulze J., Derzsi A., Donko Z., et al. 2011 Plasma Sources Sci. Technol. 20 045008

2011
[40]

2021 Plasma Sources Sci

Zhao K., Su Z.X., Liu J.R., et al. 2021 Plasma Sources Sci. Technol. 29 124001

2021
[41]

2021 Plasma Sources Sci

Krüger F., Lee H., Nam S.K., et al. 2021 Plasma Sources Sci. Technol. 30 085002

2021
[42]

2015 Phys

Bruneau B., Gans T., O’Connell D., et al. 2015 Phys. Rev. Lett. 114 125002

2015
[43]

2015 Plasma Sources Sci

Bruneau B., Lafleur T., Gans T., et al. 2015 Plasma Sources Sci. Technol. 25 01LT02

2015
[44]

2018 Plasma Sources Sci

Donkó Z., Derzsi A., Vass M., et al. 2018 Plasma Sources Sci. Technol. 27 104008

2018
[45]

Saikia P., Bhuyan H., Escalona M., et al. 2018 J. Appl. Phys. 123 183303

2018
[46]

2017 Plasma Process Polym 14 1600117

Schungel E., Donko Z., Schulze J. 2017 Plasma Process Polym 14 1600117

2017
[47]

2024 Phys

Sharma S., Sirse N., Turner M.M., et al. 2024 Phys. Lett. A 498 129346

2024
[48]

Sharma S., Sirse N., Turner M. M. 2023 Phys. Plasmas 30 073506

2023
[49]

2021 Plasma Sources Sci

Wang L., Hartmann P., Donkó Z., et al. 2021 Plasma Sources Sci. Technol. 30 054001

2021
[50]

2024 Plasma Sources Sci

Wang X.K., Masheyeva R., Liu Y.X., et al. 2024 Plasma Sources Sci. Technol. 33 085006

2024
[51]

F., Dai Z

Dong W., Zhang Y. F., Dai Z. L., et al. 2022 Plasma Sources Sci. Technol. 31 025006

2022
[52]

W., Dai Z

Dong W., Xu H. W., Dai Z. L., et al. 2021 Acta Phys. Sin. 70 095213

2021
[53]

F., Schulze J., et al

Dong W., Zhang Y. F., Schulze J., et al. 2024 Plasma Sources Sci. Technol. 33 025020

2024
[54]

1990 Appl

Surendra M., Graves D.B. 1990 Appl. Phys. Lett. 56 1022

1990
[55]

Nitschke T.E., Graves D.B. 1994 J. Appl. Phys. 76 5646

1994
[56]

Hammond E.P., Mahesh K., Moin P. 2002 J. Comput. Phys. 176 402

2002
[57]

Q., Zhang Q

Wang L., Wen D. Q., Zhang Q. Z., et al. 2019 Plasma Sources Sci. Technol. 28 055007

2019
[58]

2019 AIP Adv

Wen Y.Y., Zhang Y.R., Jiang G., et al. 2019 AIP Adv. 9 055019

2019
[59]

J. O. Hirschfelder, C. F. Curtiss, and R. B. Bird, Molecular Theory of Gases and Liquids (Wiley, New-York, 1954) 19

1954
[60]

2018 Phys

Jia W.Z., Liu R.Q., Wang X.F., et al. 2018 Phys. Plasmas 25 093501

2018
[61]

2017 Phys

Wang X.F., Jia W.Z., Song Y.H., et al. 2017 Phys. Plasmas 24 113503

2017
[62]

Zhang Y.F., Dong W., Jia W.Z., et al. 2024 J. Phys. D: Appl. Phys. 57 415205

2024