Pith. sign in

REVIEW 4 major objections 6 minor 21 references

Scanning Thermal Microscopy in Air and Vacuum: A Comparison

T0 review · 4 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read A side-by-side air-vacuum comparison of the same thermocouple tip shows air signals 2.5–40 times larger but edge widths about 39% wider after correcting for simulated heat spreading.

desk verdict Useful same-tip air/vacuum SThM comparison with a genuinely interesting meniscus-persistence observation, but the 39% resolution claim is about one sigma in the raw widths and should not be taken at face value. read the letter →

arxiv 2411.17507 v1 pith:LWRQZDFI submitted 2024-11-26 physics.ins-det cond-mat.mes-hallcond-mat.mtrl-sci

Jabez J. McClelland , Evgheni Strelcov , Ami Chand This is my paper · ORCID
classification physics.ins-detcond-mat.mes-hallcond-mat.mtrl-sci
keywords scanningthermalmicroscopythermocoupleprobeairvsvacuumcomparisonwatermeniscusresolutionsiliconeoilcontaminationnanoscaletransportedge-widthanalysis
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

This paper reports a side-by-side comparison of scanning thermal microscopy (SThM) performed in ambient air and in vacuum with the same thermocouple probe and the same heated sample, a silicon substrate patterned with silver squares buried under a thin flat layer of silicon dioxide. The central finding is that air boosts the measured thermocouple signal by a factor of 2.5 to 40 over vacuum and makes the signal reproducible, while vacuum gives smaller signals that drift and depend on the probe's history. The paper also finds that edges appear about 39% wider in air after correcting for the simulated spreading of heat through the sample's 58 nm silica cap. The practical interest is that most SThM is done in air, and these results quantify the trade-off between air's higher, more stable signal and vacuum's sharper, more physically interpretable contact signal.

What carries the argument

The load-bearing object is a single commercial thermocouple SThM probe used with the same nanofabricated sample under both environments, so tip-to-tip and sample-to-sample variations are removed. The resolution comparison rests on fitting line scans across the Ag square edge to an empirical double-error-function form and extracting the 25–75% edge width; the air-vacuum difference is then sharpened by subtracting in quadrature a 72 nm edge-spreading width obtained from a finite-element simulation of a 20 nm rod scanning the same buried-edge geometry. The physical mechanism invoked is a water-related meniscus, possibly stabilized by silicone oil contaminating the probe, that increases heat flow into the tip and widens the effective contact area.

What would settle it

Measure raw 25–75% edge widths on the same embedded-Ag sample in air, in dry nitrogen, and in vacuum without applying the quadrature correction; if the raw widths in air and vacuum overlap (as the reported (180 ± 18) nm and (139 ± 36) nm do within combined uncertainty), then the claimed 39% resolution penalty is not directly observable and the correction's validity becomes the crux.

Watch

Extended reading notes

Core claim

The paper's central claim is that for a single thermocouple SThM tip scanning a hot, topographically flat sample at 101 °C, the ambient environment changes both the magnitude and lateral resolution of the thermal signal. In air the tip-sample junction carries heat through a persistent water-related meniscus plus conduction and convection through the surrounding gas, giving an average thermocouple voltage of (0.97 ± 0.04) mV across nine landings, 2.5 to 40 times the (0.025–0.37) mV spread seen in vacuum. Air signals remained steady, whereas vacuum signals were highest immediately after air exposure and decayed by a factor of 5–10 during an hour or two of scanning, an effect attributed to a silicone-oil-contaminated meniscus that is rubbed off the tip. The 25–75% edge widths from line-scan fits were (180 ± 18) nm in air and (139 ± 36) nm in vacuum; subtracting a finite-element-simulated 72 nm heat-spreading width in quadrature turns those into 165 nm versus 119 nm, i.e. 39% larger in air. The authors present this as evidence that the air environment increases signal and stability at a measurable but moderate cost in resolution.

Load-bearing premise

The claimed 39% resolution difference rests on the assumption that the measured edge profile is the true thermal response convolved with the tip response, so that the simulated 72 nm heat-spreading width can be subtracted in quadrature; if that assumption fails, the raw air and vacuum edge widths are not statistically different.

Editorial extensions

If this is right

  • Air SThM on flat, uniformly heated samples can be read as a stable relative thermal map, with a slowly decaying background from conduction and convection that can largely be subtracted.
  • Vacuum SThM removes that background, but the contact signal may be small and history-dependent: after air exposure, a meniscus or contamination can inflate the signal for hours before scanning erodes it.
  • For samples with nearby hot spots or significant topography, the air background is not necessarily flat and can distort thermal images.
  • For the tested tip geometry, air resolution is measurably worse than vacuum resolution, but the difference is modest rather than drastic.
  • The reported data cannot separate meniscus effects from conduction and convection; dry-air or inert-gas measurements would be needed to separate them.

Reading between the lines

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

  • An implication the paper leaves implicit is that quantitative SThM users should not treat a single air calibration as transferable: the meniscus state changes with scan history, so signal-to-temperature conversion may drift in vacuum and stabilize in air only after the meniscus is fully formed.
  • The silicone-oil mechanism predicts a testable storage effect: probes kept in gel-lined containers should show larger vacuum signal decay and longer-lived high-signal periods than probes stored in sealed dry packaging.
  • If a dry-nitrogen comparison, suggested in the paper's conclusions, shows the edge width shrinking to the vacuum value while the signal stays elevated, that would isolate conduction and convection as the resolution limiter and the meniscus as the signal amplifier; if instead the edge width stays large, the meniscus must be the dominant blur source.
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.

Referee Report

4 major / 6 minor

Summary. The manuscript reports a direct, single-tip comparison of scanning thermal microscopy (SThM) in air and vacuum on a heated Si/SiO2 sample with nanofabricated 2.5 µm Ag squares embedded below a 58 nm SiO2 cap. In air the thermocouple signal is larger and more stable, while in vacuum the signal is lower and strongly dependent on the recent exposure/scanning history, decaying after one to two hours of scanning. Resolution is assessed by fitting line scans across Ag edges with an empirical error-function form, yielding 25-75% edge widths of (180±18) nm in air (n=9) and (139±36) nm in vacuum (n=10). After subtracting an estimated 72 nm thermal-spreading width from a finite-element simulation in quadrature, the authors report that the air edge width is about 39% larger than the vacuum value. The observations are interpreted in terms of a water-related meniscus and air conduction/convection.

Significance. The qualitative findings are solid and useful: the lift-height curves in Fig. 2 directly demonstrate long-range air-mediated heat transfer, the signal stability comparison is well documented, and the observation that the high-conductivity meniscus persists in vacuum for hours and is removed by scanning is an interesting, well-supported contribution to the SThM literature. The manuscript is also careful to note that the finite-element model is used only as an interpretive tool and that the empirical fit function in Eq. (1) is chosen for convenience. The main quantitative claim, however, is not yet established at the level of certainty implied by the abstract, because the 39% figure depends on an unvalidated deconvolution assumption and on a statistical reporting choice that is left ambiguous.

major comments (4)
  1. [Results, edge-width fits (Eq. (1))] Please clarify the statistical meaning of the quoted uncertainties and provide an explicit comparison of the air and vacuum widths. The text reports '(180±18) nm' and '(139±36) nm' as 'one standard deviation about the mean'; if these are standard deviations of the individual fit widths, the standard error of the difference is approximately 13 nm and the raw difference of 41 nm is significant, but the paper does not state this. If they are instead intended as uncertainties of the means, the difference is only about one combined standard deviation and is not significant. The abstract's firm '39% larger' claim cannot be evaluated without this distinction being resolved.
  2. [Results, finite-element thermal-spreading correction (Fig. 5)] The 39% claim rests on subtracting the simulated 72 nm width in quadrature from the measured 25-75% widths. The text justifies this only with 'it is reasonable to subtract it in quadrature,' but for a 25-75% width this operation is not generally valid: widths of convolved signals add in quadrature only when the kernels are Gaussian (or otherwise variance-additive), and no evidence is given that the tip response and the simulated spreading are Gaussian. The simulation also uses a 20 nm Cr rod with its distal end fixed at 20 °C; the resulting 72 nm value is a free-parameter-dependent estimate, and no uncertainty is assigned to it. Without an independent validation of the convolution model, the corrected values of 165 nm and 119 nm should be presented only as a scenario, not as the primary result.
  3. [Results, vacuum signal variability and edge widths] The vacuum measurements are pooled over a signal that varies from 0.025 mV to 0.37 mV depending on the recent history of the tip, with the high value decaying after one to two hours of scanning. Because the meniscus state is expected to affect not only signal level but also resolution, the ten vacuum edge-width values may mix different physical tip states. The authors should report the individual widths, or at least correlate them with the signal level at the time of each image, before using the pooled vacuum mean as the baseline for the resolution comparison.
  4. [Abstract and Discussion/Conclusions] The abstract states that edge widths in air are 'approximately 39 % larger' as a definite result, whereas the Discussion says the resolution is 'somewhat lower' and the Conclusions say 'somewhat poorer resolution.' This inconsistency should be removed by either supporting the quantitative claim with the full statistical and uncertainty analysis or by softening the abstract to reflect the current level of support.
minor comments (6)
  1. [Introduction] 'Currently thinking postulates' should read 'Current thinking postulates.'
  2. [Introduction] 'may also is spurious signals' should read 'may also produce spurious signals.'
  3. [Discussion] 'still be larger is regions' should read 'still be larger in regions.'
  4. [References] Reference 9 should read 'J. Vac. Sci. & Technol.' rather than 'J. V ac. Sci. & Technol.', and reference 20 should read 'Electrical' rather than 'Eletrical.'
  5. [Methods] The finite-element thermal analysis that supports the statement that the sample surface is within (1 to 2) °C of the RTD temperature is not described; please add a brief description or a citation to the simulation details.
  6. [Results] The signal-amplification factors (2000 in air, 10,000 in vacuum) are given in the figure captions but the text speaks of 'thermocouple voltage'; please state explicitly whether the reported millivolt values are raw thermocouple voltages after dividing out the amplifier gains, so that the 2.5 to 40 ratio is unambiguous.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the paper is an experimental comparison; the finite-element correction is an independent model, not fitted to the measured widths.

full rationale

This paper reports direct SThM measurements and compares air/vacuum signals; it does not derive a target result from a fitted parameter. The only quantitative comparison that could look circular is the 39% edge-width claim: raw 25-75 widths of (180±18) nm (air, n=9) and (139±36) nm (vacuum, n=10) are corrected by subtracting a 72 nm finite-element thermal-spreading width in quadrature. This correction is not fitted to the measured widths; it comes from an independent thermal simulation of a 20 nm rod scanning an Ag edge in SiO2, with the rod diameter chosen small enough to represent an essentially unconvoluted spread. The quadrature subtraction is an assumption ('it is reasonable to subtract it in quadrature'), and for non-Gaussian kernels widths generally do not add in quadrature, so the assumption is a correctness risk. But it is not circular: the simulation's 72 nm value does not depend on the air/vacuum edge-width data, and the authors do not rename the simulation output as a measurement. Likewise, Eq. (1) is an empirical fit used only to extract widths, not to predict a target quantity. There are no load-bearing self-citations: refs [19]-[21] are external contamination/hydrolysis literature used to interpret the meniscus persistence. The paper explicitly admits that it cannot differentiate conduction/convection from meniscus effects on resolution ('The present studies do not allow differentiation between conduction/convection or meniscus effects on the resolution'), which further limits the interpretation but does not make the comparison circular. Hence no circularity.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

No new physical entities are introduced. The key model inputs are the rod diameter in the finite-element simulation and the assumption that quadrature subtraction is valid.

free parameters (1)
  • finite-element tip rod diameter = 20 nm
    Chosen by hand to represent the SThM tip in the thermal spreading simulation; the simulated 25-75 edge width (72 nm) depends on this value and is subtracted in quadrature from measured widths.
assumptions (4)
  • domain assumption The thermocouple voltage is proportional to the local tip temperature.
    The paper interprets amplified thermocouple voltage as a thermal signal without calibrating to absolute temperature.
  • domain assumption In vacuum, conduction and convection through air are negligible.
    The paper assumes the vacuum signal comes only from direct tip-sample contact.
  • ad hoc to paper The measured edge profile is a convolution of the true thermal spread with the tip response, so the simulated spread can be subtracted in quadrature.
    This is a stated assumption ('it is reasonable to subtract it in quadrature') without experimental validation; it directly affects the 39% resolution difference.
  • ad hoc to paper A 20 nm Cr rod with distal end at 20 C adequately represents the thermocouple tip.
    The paper acknowledges this is an approximation and states it 'provides a reasonable approximation' of the thermal spread.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Scanning Thermal Microscopy in Air and Vacuum: A Comparison." pith.science (2026). https://pith.science/paper/LWRQZDFI

@misc{pith2026241117507,
  author       = {Pith},
  title        = {Pith review of: Scanning Thermal Microscopy in Air and Vacuum: A Comparison},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LWRQZDFI}},
  note         = {Machine review of arXiv:2411.17507}
}
read the original abstract

We present measurements comparing scanning thermal microscopy in air and vacuum. Signal levels are compared and resolution is probed by scanning over the edge of a nanofabricated Ag square embedded in SiO2. Signals measured in air were seen to be 2.5 to 40 times larger than in vacuum. Furthermore, the air signals were stable while the vacuum signals varied significantly. Edge widths measured in air were approximately 39 % larger than those measured in vacuum. Our observations are consistent with the air measurements experiencing heat transfer from the surrounding sample via conduction and convection as well as the formation of a water-related meniscus at the tip-sample junction. These results contribute to the understanding of the complex heat exchange effects that can occur in scanning thermal microscopy when it is conducted in an ambient atmosphere.

Figures

Figures reproduced from arXiv: 2411.17507 by the authors.

Figure 1
Figure 1. (a) Photograph of the SThM tip and sample mounted on the resistance temperature detector (RTD) and thermoelectric heater. (b) SEM of the SThM tip landed on the sample. measured with the SThM. Experiments were conducted in steady state, with constant temperature applied to the bottom of the sample via a thermoelectric heater monitored and controlled by a temperature sensor; see [PITH_FULL_IMAGE:figures/full_fig_p002… view at source ↗
Figure 2
Figure 2. Thermal signal as a function of tip height above the sample. (a) Air measurement, showing a drop of 37 % when the tip leaves the sample, followed by a slow decline as the tip moves farther away. (b) Vacuum measurement, showing an abrupt drop at the first lift step when the tip loses contact with the sample. The thermal signal is the thermocouple voltage amplified by a factor of 2000 in the case of air, or 10,000 in … view at source ↗
Figure 3
Figure 3. (a) Thermal image of region over 2.5 µm Ag square recorded in air. (b) Horizontal line scan across image in (a). (c) Topography signal corresponding to (a). (d) Thermal image of region over 2.5 µm Ag square recorded in vacuum. (e) Horizontal line scan across image in (d). (f) Topography signal corresponding to (d). Thermal signal is thermocouple voltage after amplification by a factor of 2000 (air) or 10,000 (vacuum… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: SEM images of sample. (a) Top down view, showing Ag squares. The SEM beam energy was 5 keV, enabling imaging of the Ag through the thin SiO2 covering. (b) Focused ion beam-milled cross section of edge of Ag square, viewed at an angle of 54◦ . The observation of a 39 % …
Figure 5
Figure 5. Figure 5: Finite-element calculation of signal spread due to thermal diffusion between Ag square edge and surface of SiO2. (a) Geometry of calculation, showing Si substrate, Ag square, SiO2 coating, and Cr rod used to represent the tip. (b) Plot of temperature measured at a poin…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

21 extracted references · 15 canonical work pages

  1. [1]

    Pollock, H. M. & Hammiche, A. Micro-thermal analysis: techniques and applications. J. Phys. D: Appl. Phys. 34, R23, DOI: 10.1088/0022-3727/34/9/201 (2001)

    work page doi:10.1088/0022-3727/34/9/201 2001

  2. [2]

    & Reddy, P

    Jeong, W., Hur, S., Meyhofer, E. & Reddy, P. Scanning Probe Microscopy for Thermal Transport Measurements.Nanoscale Microscale Thermophys. Eng. 19, 279–302, DOI: 10.1080/15567265.2015.1109740 (2015). Place: Philadelphia Publisher: Taylor & Francis Inc WOS:000366663600004

    work page arXiv 2015

  3. [3]

    & Chapuis, P.-O

    Gomès, S., Assy, A. & Chapuis, P.-O. Scanning thermal microscopy: A review. physica status solidi (a) 212, 477–494, DOI: 10.1002/pssa.201400360 (2015). _eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1002/pssa.201400360

    doi:10.1002/pssa.201400360 2015

  4. [4]

    Zhang, Y . et al. A Review on Principles and Applications of Scanning Thermal Microscopy (SThM). Adv. Funct. Mater. 30, 1900892, DOI: 10.1002/adfm.201900892 (2020). _eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1002/adfm.201900892

    doi:10.1002/adfm.201900892 2020

  5. [5]

    & Ka´ zmierczak-Bałata, A

    Bodzenta, J. & Ka´ zmierczak-Bałata, A. Scanning thermal microscopy and its applications for quantitative thermal measurements. J. Appl. Phys. 132, 140902, DOI: 10.1063/5.0091494 (2022)

    work page doi:10.1063/5.0091494 2022

  6. [6]

    Majumdar, A. et al. Thermal imaging by atomic force microscopy using thermocouple cantilever probes. Rev. Sci. Instruments 66, 3584–3592, DOI: 10.1063/1.1145474 (1995)

    work page doi:10.1063/1.1145474 1995

  7. [7]

    & Reddy, P

    Kim, K., Jeong, W., Lee, W. & Reddy, P. Ultra-High Vacuum Scanning Thermal Microscopy for Nanometer Resolution Quantitative Thermometry. ACS Nano 6, 4248–4257, DOI: 10.1021/nn300774n (2012). Publisher: American Chemical Society

    doi:10.1021/nn300774n 2012

  8. [8]

    Shekhawat, G. S. et al. Micromachined Chip Scale Thermal Sensor for Thermal Imaging. ACS Nano 12, 1760–1767, DOI: 10.1021/acsnano.7b08504 (2018). Publisher: American Chemical Society

    work page doi:10.1021/acsnano.7b08504 2018

Show all 21 references
  1. [9]

    & Rangelow, I

    Edinger, K., Gotszalk, T. & Rangelow, I. W. Novel high resolution scanning thermal probe. J. V ac. Sci. & Technol. B: Microelectron. Nanometer Struct. Process. Meas. Phenom. 19, 2856–2860, DOI: 10.1116/1.1420580 (2001)

    2001 doi

  2. [10]

    & Kolosov, O

    Tovee, P., Pumarol, M., Zeze, D., Kjoller, K. & Kolosov, O. Nanoscale spatial resolution probes for scanning thermal microscopy of solid state materials. J. Appl. Phys. 112, 114317, DOI: 10.1063/1.4767923 (2012)

    2012 doi

  3. [11]

    Menges, F. et al. Temperature mapping of operating nanoscale devices by scanning probe thermometry. Nat. Commun. 7, 10874, DOI: 10.1038/ncomms10874 (2016). Publisher: Nature Publishing Group

    2016 doi

  4. [12]

    & Esashi, M

    Kim, S.-J., Ono, T. & Esashi, M. Thermal imaging with tapping mode using a bimetal oscillator formed at the end of a cantilever. Rev. Sci. Instruments 80, 033703, DOI: 10.1063/1.3095680 (2009)

    2009 doi

  5. [13]

    & Bergaud, C

    Aigouy, L., Lalouat, L., Mortier, M., Löw, P. & Bergaud, C. Note: A scanning thermal probe microscope that operates in liquids. Rev. Sci. Instruments 82, 036106, DOI: 10.1063/1.3567794 (2011)

    2011 doi

  6. [14]

    Tovee, P. D. & Kolosov, O. V . Mapping nanoscale thermal transfer in-liquid environment—immersion scanning thermal microscopy. Nanotechnology 24, 465706, DOI: 10.1088/0957-4484/24/46/465706 (2013). Publisher: IOP Publishing

    2013 doi

  7. [15]

    Hinz, M., Marti, O., Gotsmann, B., Lantz, M. A. & Duerig, U. High resolution vacuum scanning thermal microscopy of HfO2 and SiO2. Appl. Phys. Lett. 92, 043122, DOI: 10.1063/1.2840186 (2008). Place: Melville Publisher: Amer Inst Physics WOS:000252860400097

    2008 doi

  8. [16]

    Meng, J. et al. Temperature Distribution in TaOx Resistive Switching Devices Assessed In Operando by Scanning Thermal Microscopy. ACS Appl. Electron. Mater. 5, 2414–2421, DOI: 10.1021/acsaelm.3c00229 (2023). Publisher: American Chemical Society

    2023 doi

  9. [17]

    & Majumdar, A

    Shi, L. & Majumdar, A. Thermal Transport Mechanisms at Nanoscale Point Contacts. J. Heat Transf. 124, 329–337, DOI: 10.1115/1.1447939 (2001)

    2001 doi

  10. [18]

    Deshmukh, S. et al. Direct measurement of nanoscale filamentary hot spots in resistive memory devices. Sci. Adv. 8, eabk1514, DOI: 10.1126/sciadv.abk1514 (2022). Publisher: American Association for the Advancement of Science

    2022 doi

  11. [19]

    V .et al

    Ievlev, A. V .et al. Chemical Phenomena of Atomic Force Microscopy Scanning. Anal. Chem. 90, 3475–3481, DOI: 10.1021/acs.analchem.7b05225 (2018). Publisher: American Chemical Society

    2018 doi

  12. [20]

    & Crine, J.-P

    Garbay, H., Grob, R., Casanovas, J. & Crine, J.-P. Measurements and influence of water content in silicone oil. In 1984 IEEE International Conference on Eletrical Insulation , 297–300, DOI: 10.1109/EIC.1984.7465202 (1984)

    1984

  13. [21]

    D., Zhu, J., Xu, F., Hong, S

    Piner, R. D., Zhu, J., Xu, F., Hong, S. & Mirkin, C. A. "Dip-Pen" Nanolithography. Science 283, 661–663, DOI: 10.1126/science.283.5402.661 (1999). Publisher: American Association for the Advancement of Science. 7/8 Acknowledgments The authors would like to thank Daron Westly A...

    1999 doi

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.