X-ray mirror figure correction using differential deposition

A. Vivo; Ch. Morawe; F. Perrin; R. Barrett; S. Labour\'e

REVIEW 2 major objections 5 minor 26 references

X-ray mirror figure correction using differential deposition

T0 review · 2 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Differential deposition of tungsten silicide films routinely reduces x-ray mirror figure errors by a factor of 20–30, reaching residuals below 0.5 nm RMS.

desk verdict A credible engineering milestone—differential deposition now reaches 0.1–0.2 nm RMS on long X-ray mirrors—but the abstract overstates the improvement and the static-to-dynamic kernel assumption deserves an explicit test. read the letter →

arxiv 2506.01409 v1 pith:WKHB5LIX submitted 2025-06-02 physics.optics physics.app-ph

Ch. Morawe , S. Labouré , F. Perrin , A. Vivo , R. Barrett This is my paper

classification physics.opticsphysics.app-ph

keywords x-rayopticsmirrorsdifferentialdepositionfigurecorrectionmagnetronsputteringthinfilmsinterferometricstitchingmetrology

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

X-ray mirrors for synchrotron beamlines must be flat to well below a nanometer to preserve coherent wavefronts, and polishing long substrates to that level is slow. This paper reports a coating-based alternative: differential deposition, in which a tungsten silicide ($\mathrm{WSi_2}$) film with a deliberately non-uniform thickness cancels the mirror's measured height errors. The authors show on four 300 mm silicon mirrors that three corrective iterations, using successively narrower deposition apertures (24 mm, 2 mm, 1 mm), routinely reduce figure errors by a factor of 20–30, to below 0.5 nm RMS. On the best mirror the residual error reached 0.1 nm RMS and 0.5 nm peak-to-valley over a 260 mm clear aperture, down to spatial periods of 2 mm. If this result holds, differential deposition becomes a practical finishing step for long x-ray mirrors, complementing ion-beam figuring and elastic emission machining.

What carries the argument

The carrying object is the convolution of Eq. (1), $$t(x_s)=R\int_{-S}^{S}\frac{f(x_m-x_s)}{v(x_m)}\,\mathrm{d}x_m,$$ where $t(x_s)$ is the deposited thickness at substrate position $x_s$, $f(x_m-x_s)$ is the normalized static particle flux profile seen through a beam-defining aperture, $v(x_m)$ is the substrate motion speed, and $R$ is the central growth rate. Because the desired thickness is known and the speed profile is unknown, the correction is a deconvolution: the measured static flux profile is the kernel, and the algorithm solves a discretized linear system by matrix inversion, using numerical routines originally developed for astronomical image deconvolution. The second carrying element is the aperture cascade (24 mm, 2 mm, and 1 mm slits), which successively corrects long, medium, and short spatial periods, pushing the corrected spatial frequency range down to about $0.5\,\mathrm{mm}^{-1}$ (2 mm periods).

What would settle it

Deposit a known test profile in dynamic mode using a velocity profile computed from the static flux kernel, then measure the resulting thickness with the same stitched interferometry; if the residual difference from the predicted profile exceeds the stated metrology repeatability of 0.05 nm RMS, the static-kernel assumption is wrong.

Watch

Extended reading notes

Core claim

The central claim is that differential deposition is a deterministic, repeatable figure-correction method for long x-ray mirrors at the sub-nanometer level. The paper demonstrates this on four 300 mm flat silicon mirrors coated with tungsten silicide: after an initial uniform 50 nm layer, each mirror received two corrective coatings on a large coating system and one on a compact system, with the deposition aperture narrowed from 24 mm to 2 mm to 1 mm. The measured figure errors decreased nearly exponentially with iteration, with total improvement factors between 10x and 36x depending on mirror and metric. The best residual levels were 0.1 nm RMS and 0.5 nm PV over 260 mm, with the frequency analysis showing that each aperture width suppresses height errors down to spatial periods roughly matching that width, ending at about 2 mm. The authors attribute the remaining uncorrected features to localized surface defects, short-period spikes, and hardware limits, and note that one mirror's progress was limited by surface pollution.

Load-bearing premise

The correction loop assumes that the static WSi2 flux profile measured on stationary substrates is the effective deposition kernel while the substrate is moving, so the deconvolution's velocity profile deposits exactly the inverse height profile; the paper does not experimentally validate this equivalence.

Editorial extensions

If this is right

Four 300 mm mirrors reached sub-nanometer residuals with only three corrective iterations, so differential deposition can serve as a final finishing step before multilayer coating of synchrotron mirrors.
The aperture-by-aperture cascade gives wavelength-selective correction: each narrower slit cleans up shorter spatial periods without disturbing the already corrected long periods.
Metrology repeatability below 0.05 nm RMS is sufficient to guide the correction loop, meaning off-line interferometric stitching can reliably drive figure errors to the 0.1 nm RMS scale.
The demonstrated roughly tenfold improvement in achievable RMS errors over six years, at about a doubling every 1.8 years, suggests the process still has headroom before hitting fundamental limits.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Inference: if the static flux profile remains the deposition kernel during motion, the same deconvolution should transfer to curved or actively bent mirrors, since only the measured height profile and kernel change.
Inference: the correction acts only along the central trace, so a two-dimensional extension would require lateral aperture shaping or a second motion axis; the paper itself notes that off-trace errors worsen.
Inference: closed-loop correction using in-situ metrology, which the authors list as future work, could remove contamination during transfer and speed up convergence.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

This is a solid engineering result. The ESRF group has pushed differential deposition to 0.1–0.2 nm RMS figure errors on 300 mm flat Si mirrors using the new compact coating system and a 1 mm aperture. The measured residuals are real, the PSD analysis shows frequency bands dropping exactly where the aperture widths predict, and the four-mirror series gives a useful sense of repeatability. The genuinely new content is the CMCS coater, the narrower aperture, and the demonstrated sub-0.2 nm level—previous work from this group was at the sub-nanometer scale. That is a legitimate milestone for 4th-generation synchrotron optics.

Now the soft spots, in order of importance. First, the abstract's "20–30x" improvement is not what Table 3 shows: PV improvements are 10–16x, and RMS improvements range from ~17x to ~36x across the four mirrors. The word "routinely" overstates a best-case result—only WP#37 reaches 0.1 nm RMS; the other WP mirrors sit at 0.20–0.23 nm, and CO#7 stalls at 0.37 nm due to surface pollution. Second, the 0.1 nm value is close to the 0.05 nm RMS repeatability of the Fizeau stitching, so the true figure could be anywhere in that range. No error bars are given.

The deeper methodological concern is the untested assumption that the static flux profile, measured on stationary substrates, is the effective deposition kernel during dynamic motion. The stress-test note is right: a kernel mismatch would enter the deconvolution as a systematic bias, not as noise, and iterative convergence does not by itself prove the kernel is unbiased at the 0.1 nm scale. I weigh this as a request for additional evidence rather than a fatal flaw, because the final residuals are empirical outcomes independent of the model, and the PSD-band agreement gives indirect validation. Still, a direct open-loop test—deposit a known target profile and compare measured versus predicted thickness—would settle the question cleanly. The authors should also publish the algorithm specification or code and the raw metrology data; the paper currently references an IDL astronomy library but gives no details of the deconvolution implementation.

For whom is this paper? Anyone working on synchrotron mirror fabrication or adaptive optics for coherent X-ray sources will want to know that differential deposition has reached this level. It deserves a serious referee, not a desk reject—the engineering result is credible and useful, and the authors have a track record of careful work. My recommendation: send it to review, but ask the referee to insist on reconciling the abstract with Table 3 and to request at least one open-loop validation of the kernel equivalence if the authors can do it without much extra effort.

Referee Report

2 major / 5 minor

Summary. The paper describes an experimental development of differential deposition (DD) for correcting the figure of 300 mm flat silicon mirrors intended for x-ray beamlines. A deconvolution algorithm, built on Eq. (1), converts a measured height-error profile into a substrate velocity profile using the static deposition-flux kernel measured in Section 3.1. Corrections are applied iteratively on two DC magnetron sputter systems with apertures from 24 mm down to 1 mm, and the resulting surfaces are characterized by Fizeau stitching metrology. The authors report residual errors down to 0.1 nm RMS and 0.5 nm PV over a 260 mm clear aperture, PSD reductions in frequency bands corresponding to the aperture widths, and no significant roughness degradation. They conclude that DD is a practical sub-nanometer finishing method for long x-ray mirrors.

Significance. If the headline residual levels are taken at face value, the paper is a significant experimental contribution: it demonstrates iterative convergence of DD into the sub-nanometer regime on 300 mm mirrors, with cross-checks (LTP comparison, PSD analysis, roughness tracking) that strengthen the qualitative conclusions. The per-iteration data in Table 3 and Figure 5 give a transparent record of convergence and spread, and Figure 6 provides a clear link between aperture choice and corrected spatial-frequency band. The central technical caveat is that the deposition kernel is measured statically but applied dynamically, so the absolute accuracy of the 0.1 nm RMS claim rests on an assumption that is not directly tested.

major comments (2)

[Sections 2.1 and 3.1] The deconvolution in Eq. (1) uses the normalized static flux profile f determined from stationary-substrate depositions (Section 3.1, Fig. 1) as the deposition kernel during dynamic substrate motion. The manuscript does not provide a direct experimental test that this kernel is identical under dynamic conditions; plasma drift, motion-dependent shadowing, and the stated sample position accuracy of ±0.1–0.3 mm with aperture-substrate distances of 1–3 mm can all change the effective flux profile seen by a moving substrate. A kernel mismatch enters the computed velocity profile as a systematic bias, not as noise, and the same biased profile is reused in subsequent iterations, so the observed monotone convergence does not by itself validate the calculation at the 0.1 nm RMS scale claimed in Section 4. I recommend adding a validation experiment in which a known dynamic deposition profile is predicted from Eq. (1) and compared with the measured profile, or an equivalent test of the static-to-dynamic kernel equivalence.
[Abstract, Section 4, Table 3] The abstract states that shape errors were 'routinely reduced by a factor of 20-30', and Section 4 says 'about 30x (RMS) and 20x (PV)'. Table 3 gives improvement factors over 280 mm of 16.8–35.6x for RMS and 10.2–16.0x for PV; over 260 mm the PV factors are 8.1–24.7x. These tabulated values do not support a uniform '20-30x' claim and never support '20x PV'. The wording should be revised to report the measured ranges separately for RMS and PV and to note that one mirror (CO#7) was affected by contamination, which weakens the 'routine' characterization.

minor comments (5)

[Eq. (1)] Equation (1) is rendered with a formatting artifact that makes the division by v(x_m) illegible; please provide a cleanly typeset equation and define the integration variable explicitly.
[Table 3 and Section 3.2] The improvement factors in Table 3 are computed with respect to iteration #1, which is the state after the initial uniform WSi2 coating, not the bare substrate. Although Section 3.2 explains that iteration #0 was omitted because the uniform coating did not change the figure, the table caption should state this baseline explicitly.
[Section 2.4 and Table 3] The Fizeau repeatability is quoted as better than 0.05 nm RMS, which is non-negligible compared with the best residual values in Table 3 (e.g., 0.095 nm RMS for WP#37 over 260 mm). Please include an uncertainty estimate or at least state whether the quoted residuals are repeatability-limited.
[Section 3.4 and Fig. 7] The roughness discussion would benefit from a quantitative statement of the measurement uncertainty for the Wyko RMS values, since the claimed 'virtually constant' behavior and the WP#32 increase rest on differences of order 0.1 nm.
[Section 3.2 and Fig. 4] Residual profiles in Fig. 4 are shown on a common scale, but the spatial-frequency content that dominates each residual is not quantified; a short discussion or a mini-PSD panel would help support the statement that long-period undulations have weaker wavefront impact.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the deconvolution uses independently measured kernel inputs, and final residuals are measured after physical deposition with an external LTP cross-check.

full rationale

The derivation chain is Eq. (1), a forward convolution model where the deposited thickness is computed from the measured static flux profile f, the central growth rate R, and the unknown speed profile v. The inputs R and f are experimentally characterized in Sections 2.2 and 3.1, and the target thickness is the measured height error profile. The speed profile is obtained by matrix inversion; no term of the final residual is used as an input or fitted parameter. The reported 0.1 nm RMS / 0.5 nm PV residuals are Fizeau measurements taken after physical deposition, not outputs of the deconvolution. The same Fizeau instrument provides both the error map used for correction and the verification map, which could in principle close the metrology loop, but the paper documents an external Long Trace Profiler cross-comparison with sub-nanometer accuracy and stitching repeatability better than 0.05 nm RMS, so the verification is not solely self-referential. The static-to-dynamic kernel equivalence is an untested modeling assumption and a correctness risk, not a circularity: a wrong kernel would bias the correction but would not make the result equivalent to its input by construction. Self-citations [12-15,18] describe prior applications and the convolution formula, but the present result does not depend on accepting an unverified claim from those papers. No circular step can be quoted; the central claim is an experimental demonstration with independent inputs.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The correction procedure uses measured experimental inputs (growth rate R and static flux profile f) rather than fitted parameters. The central claim depends on the convolution model, on the equivalence of static and dynamic flux profiles, on additive film behavior without stress effects, on the accuracy of the Fizeau stitching metrology, and on the numerical stability of the deconvolution. No new entities are postulated.

assumptions (5)

domain assumption Deposited thickness is related to the speed profile by the convolution in Eq. (1), with a normalized static flux profile f and constant growth rate R.
This linear, shift-invariant model is the basis of the deconvolution algorithm in Section 2.1.
domain assumption The static flux profile measured on a stationary substrate is identical to the flux profile during dynamic motion through the aperture.
Used to compute speed profiles in Section 2.2 and characterized in Section 3.1; dynamic effects are not separately validated.
domain assumption Thin WSi2 film addition does not change the substrate figure except by the intended thickness profile, with no stress, oxidation, or interface effects altering the shape.
The correction protocol in Table 2 and Section 3.2 assumes added layers contribute additively to height error.
domain assumption Fizeau stitching metrology gives sub-nanometer accuracy and 0.05 nm RMS repeatability as stated in Section 2.4.
All before/after height errors and improvement factors depend on this instrument performance.
standard math The matrix inversion deconvolution is numerically stable and the resulting speed profile is physically realizable.
Section 2.1 describes matrix inversion without discussing conditioning or regularization; the successful iterations support but do not prove this.

how reviews work

0 comments

Cite this review

Pith. "Pith review of X-ray mirror figure correction using differential deposition." pith.science (2026). https://pith.science/paper/WKHB5LIX

@misc{pith2026250601409,
  author       = {Pith},
  title        = {Pith review of: X-ray mirror figure correction using differential deposition},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WKHB5LIX}},
  note         = {Machine review of arXiv:2506.01409}
}

read the original abstract

The surface figure of x-ray mirrors can be improved by differential deposition of thin films. To achieve the required corrections, WSi2 layers of variable thickness were deposited through beam-defining apertures of different openings. The substrates were moved in front of the particle source with specific velocity profiles that were calculated with a deconvolution algorithm. Two different DC magnetron sputter systems were used to investigate the correction process. Height errors were evaluated before and after each iteration using off-line visible light surface metrology. Four 300 mm long flat Si mirrors were used to study the impact of the initial shape errors on the performance of the correction approach. The shape errors were routinely reduced by a factor of 20-30 down to levels below 0.5 nm RMS.

Figures

Figures reproduced from arXiv: 2506.01409 by the authors.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

26 extracted references · 26 canonical work pages

[1]

Raimondi, P., ”ESRF -EBS: The Extremely Brilliant Source Project,” Synchrotron Radiation News 29, 6 (2016)

work page 2016
[2]

Single-nanometer focusing of hard X -rays by Kirkpatrick -Baez mirrors ,

Yamauchi, K., Mimura, H., Kimura, T., Yumoto, H., Handa, S., Matsuyama, S., Arima, K., Sano, Y., Yamamura, K., Inagaki, K., Nakamori, H., Kim, J., Tamasaku, K., Nishino, Y., Yabashi, M., Ishikawa, T., “Single-nanometer focusing of hard X -rays by Kirkpatrick -Baez mirrors ,” J. Phys.: Condens. Matter 23, 394206 (2011)

work page 2011
[3]

Focusing with subnanometer -level accuracy by numerically controlled elastic emission machining,

Yamauchi, K., Mimura, H., Inagaki, K., Mori, Y., “Focusing with subnanometer -level accuracy by numerically controlled elastic emission machining,” Rev. Sci. Instrum. 73, 4028-4033 (2002)

work page 2002
[4]

Ultra- precision surface f inishing by ion beam and plasma jet techniques - status and outlook,

Arnold, T., Böhm, G., Fechner, R., Meister, J., Nickel, A., Frost, F., Hänsel, T., Schindler, A., “Ultra- precision surface f inishing by ion beam and plasma jet techniques - status and outlook,” Nucl. Instrum. Methods Phys. A 616, 147-156 (2010)

work page 2010
[5]

Development of a position-velocity-time-modulated two-dimensional ion beam figuring system for synchrotron x-ray mirror fabrication,

Wang, T., Huang, L., Zhu, Y., Vescovi, M., Kuhne, D., Kang, H., Choi, H., Kim, D. W., Tayabaly, K., Bouet, N., Idir, M., “Development of a position-velocity-time-modulated two-dimensional ion beam figuring system for synchrotron x-ray mirror fabrication,” Appl. Optics 59, 3306-3314 (2020)

work page 2020
[6]

Ion beam figuring for X-ray mirrors: history, state-of-the-art and future prospects

Shurvinton, R., Wang, H., Pradhan, P., Nistea, I-T., Alcock, S., Bazan Da Silva, M., Majhi, A., Sawhney, K., “Ion beam figuring for X-ray mirrors: history, state-of-the-art and future prospects”, J. Synchrotron Rad. 31, 655-669 (2024)

work page 2024
[7]

Differential deposition technique for figure corrections in grazing-incidence X-ray optics,

Kilaru, K., Ramsey, B. D., Gubarev, M. V., Gregory, D. A., “Differential deposition technique for figure corrections in grazing-incidence X-ray optics,” Opt. Eng. 50, 106501 (2011)

work page 2011
[8]

Highly accurate differential deposition for X-ray reflective optics,

Handa, S., Mimura, H., Yumoto, H., Kimura, T., Matsuyama, S., Sano, Y., Yamauchi, K., “Highly accurate differential deposition for X-ray reflective optics,” Surf. Interface Anal. 40, 1019-1022 (2008)

work page 2008

Show all 26 references

[9]

A preferential coating t echnique for fabricating large, high quality optics ,

Alcock, S. G., Cockerton, S., “A preferential coating t echnique for fabricating large, high quality optics ,” Nucl.Instrum. Methods Phys. A 616, 110-114 (2010)

2010
[10]

Nanofocusing of X -ray free -electron laser using wavefront-corrected multilayer focusing mirrors,

Matsuyama, S., Inoue, T., Yamada, J., Kim, J., Yumoto, H., Inubushi, Y., Osaka, T., Inoue, I., Koyama, T., Tono, K., Ohashi, H., Yabashi, M., Yamauchi, K., “Nanofocusing of X -ray free -electron laser using wavefront-corrected multilayer focusing mirrors,” Scientific Reports 8...

2018
[11]

Development of a one -dimensional differential deposition system for x -ray mirror figure correction ,

Kim, J., Kim, J. S., Kim, H -Y., Ryu, C. K., Jeong, D., Chae, B., Lim, J., Kim, J. H., Rah, S., “Development of a one -dimensional differential deposition system for x -ray mirror figure correction ,” Prec. Eng. 71, 1 -6 (2021)

2021
[12]

X-ray mirror figure correction by differential deposition and off-line metrology

Morawe, Ch., Labouré, S., Peffen, J-Ch., Perrin, F., Vivo, A., Barrett, R., “X-ray mirror figure correction by differential deposition and off-line metrology”, J. Synchrotron Rad. 26, 1872-1878 (2019)

2019
[13]

Mirror figure correction on variable length scales,

Morawe, Ch., Bras, P., Labouré, S., Perrin, F., Tran -Caliste, T., Vivo, A., “Mirror figure correction on variable length scales,” Proc. SPIE 11837, (2021)

2021
[14]

X -ray mirrors with sub -nanometer figure errors obtained by differential deposition of thin WSi2 films,

Bras, P., Morawe, Ch., Labouré, S., Perrin, F., Vivo, A., Barrett, R., “X -ray mirrors with sub -nanometer figure errors obtained by differential deposition of thin WSi2 films,” J. Synchrotron Rad. 30, 708-716 (2023)

2023
[15]

Surface figure correction using differential deposition of WSi2

Morawe, Ch., Bras, P., Labouré, S., Perrin, F., Vivo, A., “Surface figure correction using differential deposition of WSi2”, Proc. SPIE 12694, (2023)

2023
[16]

The new ESRF multilayer deposition facility,

Morawe, Ch., Borel, Ch., Peffen, J -Ch., “The new ESRF multilayer deposition facility,” Proc. SPIE 6705, (2007)

2007
[17]

The new ESRF compact multilayer coating system

Morawe, Ch., Bras, P., Guilloud, C., Labouré, S., Peffen, J-Ch., “The new ESRF compact multilayer coating system”, Proc. SPIE 12240, (2022)

2022
[18]

Thickness control of large area X-ray multilayers,

Morawe, Ch., Peffen, J-Ch., “Thickness control of large area X-ray multilayers,” Proc. SPIE 7448, (2009)

2009
[19]

Varosi, F., NASA/GSFC, The IDL Astronomy User’s Library, https://idlastro.gsfc.nasa.gov (1992)

1992
[20]

The new ESRF thin film x-ray reflectometer,

Morawe, Ch., Peffen, J-Ch., Pakawanit, P., “The new ESRF thin film x-ray reflectometer,” Proc. SPIE 10760, (2018)

2018
[21]

Surface studies of solids by total reflection of x-rays,

Parratt, L. G., “Surface studies of solids by total reflection of x-rays,” Phys. Rev. 45, 359 (1954)

1954
[22]

Fizeau stitching at the European Synchrotron Radiation Facility (ESRF),

Vivo, A., Barrett, R., “Fizeau stitching at the European Synchrotron Radiation Facility (ESRF),” Proc. SPIE 10385, (2017)

2017
[23]

Stitching techniques for measuring X-ray synchrotron mirror topography,

Vivo, A., Barrett, R., Perrin, F., “Stitching techniques for measuring X-ray synchrotron mirror topography,” Review of Scientific Instruments 90, 021710 (2019)

2019
[24]

The application of wavefront sensing methods to optical surface metrology,

Adapa, B. R. “The application of wavefront sensing methods to optical surface metrology,” Doctoral dissertation. Université Grenoble Alpes (France)
[25]

SciPy1.0: fundamental algorithms for scientific computing in Python,

Virtanen, P. et al., “SciPy1.0: fundamental algorithms for scientific computing in Python,” Nature Methods 17, 261–272 (2020)

2020
[26]

Numerical recipes: the art of scientific computing,

Press, W. H. , Teukolsky, S. A., Vetterling, W. T., Flannery, B. P., “Numerical recipes: the art of scientific computing,” Cambridge, UK ; New York: Cambridge University Press (2007)

2007

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

Raimondi, P., ”ESRF -EBS: The Extremely Brilliant Source Project,” Synchrotron Radiation News 29, 6 (2016)

work page 2016

[2] [2]

Single-nanometer focusing of hard X -rays by Kirkpatrick -Baez mirrors ,

Yamauchi, K., Mimura, H., Kimura, T., Yumoto, H., Handa, S., Matsuyama, S., Arima, K., Sano, Y., Yamamura, K., Inagaki, K., Nakamori, H., Kim, J., Tamasaku, K., Nishino, Y., Yabashi, M., Ishikawa, T., “Single-nanometer focusing of hard X -rays by Kirkpatrick -Baez mirrors ,” J. Phys.: Condens. Matter 23, 394206 (2011)

work page 2011

[3] [3]

Focusing with subnanometer -level accuracy by numerically controlled elastic emission machining,

Yamauchi, K., Mimura, H., Inagaki, K., Mori, Y., “Focusing with subnanometer -level accuracy by numerically controlled elastic emission machining,” Rev. Sci. Instrum. 73, 4028-4033 (2002)

work page 2002

[4] [4]

Ultra- precision surface f inishing by ion beam and plasma jet techniques - status and outlook,

Arnold, T., Böhm, G., Fechner, R., Meister, J., Nickel, A., Frost, F., Hänsel, T., Schindler, A., “Ultra- precision surface f inishing by ion beam and plasma jet techniques - status and outlook,” Nucl. Instrum. Methods Phys. A 616, 147-156 (2010)

work page 2010

[5] [5]

Development of a position-velocity-time-modulated two-dimensional ion beam figuring system for synchrotron x-ray mirror fabrication,

Wang, T., Huang, L., Zhu, Y., Vescovi, M., Kuhne, D., Kang, H., Choi, H., Kim, D. W., Tayabaly, K., Bouet, N., Idir, M., “Development of a position-velocity-time-modulated two-dimensional ion beam figuring system for synchrotron x-ray mirror fabrication,” Appl. Optics 59, 3306-3314 (2020)

work page 2020

[6] [6]

Ion beam figuring for X-ray mirrors: history, state-of-the-art and future prospects

Shurvinton, R., Wang, H., Pradhan, P., Nistea, I-T., Alcock, S., Bazan Da Silva, M., Majhi, A., Sawhney, K., “Ion beam figuring for X-ray mirrors: history, state-of-the-art and future prospects”, J. Synchrotron Rad. 31, 655-669 (2024)

work page 2024

[7] [7]

Differential deposition technique for figure corrections in grazing-incidence X-ray optics,

Kilaru, K., Ramsey, B. D., Gubarev, M. V., Gregory, D. A., “Differential deposition technique for figure corrections in grazing-incidence X-ray optics,” Opt. Eng. 50, 106501 (2011)

work page 2011

[8] [8]

Highly accurate differential deposition for X-ray reflective optics,

Handa, S., Mimura, H., Yumoto, H., Kimura, T., Matsuyama, S., Sano, Y., Yamauchi, K., “Highly accurate differential deposition for X-ray reflective optics,” Surf. Interface Anal. 40, 1019-1022 (2008)

work page 2008