Ludwig-Soret microscopy with vibrational photothermal effect

Keiichiro Toda; Takuro Ideguchi

REVIEW 3 major objections 4 minor 31 references

Ludwig-Soret microscopy with vibrational photothermal effect

T0 review · 3 major / 4 minor · reviewed 2026-08-08 · deepseek-v4-flash

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

Pith's one-line read Time-resolved vibrational photothermal optical diffraction tomography separates thermal-expansion refractive index changes from concentration-driven changes, enabling label-free measurement of thermophoretic transport of biomolecules…

desk verdict A genuinely new label-free way to image intracellular thermophoresis with simultaneous temperature mapping; the headline D and S_T values rest on a model assumption the authors state but do not validate. read the letter →

arxiv 2502.04578 v1 pith:4HNYTSKR submitted 2025-02-07 physics.optics

Keiichiro Toda , Takuro Ideguchi This is my paper · ORCID

classification physics.optics

keywords thermophoresisSoreteffectlabel-freeimagingopticaldiffractiontomographyvibrationalphotothermalmicroscopylivingcellsdiffusiophoresisdrymasstransport

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

The paper introduces ViPS imaging, which uses a focused infrared beam to heat water inside a living cell and an optical diffraction tomography microscope to watch how the refractive index changes over time. By separating the fast refractive index change from thermal expansion from the slower change caused by molecules physically moving along the temperature gradient, the method extracts the thermophoretic (Soret) transport directly. In living COS7 cells the authors measure a nuclear diffusion coefficient of about 4.6 μm² s⁻¹ and a Soret coefficient of about 0.01 K⁻¹, implying a 1% dry-mass redistribution per kelvin. They also report reversed transport in the cytoplasm, which they attribute to thermophoresis-induced diffusiophoresis, and a loss of thermophoretic activity in dying cells under CO₂ depletion, consistent with glass-like aggregation.

What carries the argument

The central object is the decomposition of the photothermal refractive-index change Δn into a fast thermal-expansion part Δn_ρ and a slow dry-mass-concentration part Δn_σ (Eq. 1), enabled by time-resolved ODT with 20 ms resolution. The steady-state temperature map ΔT_s is recovered from the instantaneous Δn_ρ image via the thermo-optic coefficient (Eq. 3), and the Δn_σ images are integrated over depth to give dry-mass concentration change Δσ (Eq. 2). These data are then fitted with a two-dimensional diffusion-drift simulation (Eq. 7) in which the Soret coefficient S_T and diffusion coefficient D are the free parameters.

What would settle it

Compare ViPS fits on the same nucleus before and after disrupting the cytoskeleton or otherwise altering the steady-state dry-mass distribution while keeping the temperature map unchanged; if the recovered D and S_T shift systematically, the excluded ∇²σ term in Eq. 7 is not actually balanced. Alternatively, run the Eq. 7 simulation with and without the steady-state dry-mass gradient term on measured σ images and test whether the parameter search still returns the same D and S_T.

Watch

Extended reading notes

Core claim

This paper claims that a single time-resolved refractive-index dataset from a vibrational photothermal optical diffraction tomography microscope can separate the thermal-expansion contribution from the concentration contribution to the photothermal signal, because the two evolve on very different time scales (microseconds vs. milliseconds to seconds). Exploiting this separation allows the temperature-rise map to be recovered from the fast component and the dry-mass transport to be followed in the slow component, so that both the temperature gradient driving thermophoresis and the molecular response to it are measured in the same label-free experiment. Using this approach the authors determine diffusion and Soret coefficients in the nucleus of living COS7 cells (D ≈ 4.6 μm² s⁻¹, S_T ≈ 0.01 K⁻¹), observe a negative Soret effect in the cytoplasm consistent with diffusiophoresis, and document a marked decline of thermophoretic mobility in dying cells.

Load-bearing premise

The simulation assumes that the cell's pre-existing uneven dry-mass distribution is already balanced by forces unrelated to the temperature gradient, so any baseline concentration gradient does not relax on its own during the measurement; if that balance does not hold, the fitted diffusion and Soret coefficients would absorb the cell's baseline inhomogeneity instead of pure thermophoresis.

Editorial extensions

If this is right

Intracellular thermophoresis can be quantified label-free in a single measurement, giving both the driving temperature field and the molecular response.
A 1 K intracellular temperature gradient redistributes roughly 1% of nuclear dry mass within seconds for molecules near 100 kDa.
RNA contributes an appreciable fraction of the observed nuclear thermophoretic signal, based on the Actinomycin D depletion experiment.
The reversal of transport direction in the cytoplasm indicates that diffusiophoresis can dominate thermophoresis in crowded intracellular environments.
Time-lapse ViPS imaging can track the loss of intracellular fluidity during cell death, offering a direct assay of glass-like transitions.

Reading between the lines

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

Because the method records all molecular species simultaneously rather than a single fluorescent label, it could in principle be extended to survey thermophoretic mobility across the whole proteome, provided the diffusion-and-Soret fit is generalized to multi-component mixtures.
If the approximation in Eq. 7 is relaxed to include the pre-existing dry-mass gradient, the same dataset could yield a map of intracellular forces that balance baseline diffusion, turning the current assumption into a measurable quantity.
The reported disappearance of thermophoretic mobility in dying cells could be developed into a label-free viability assay, with the time constant of mobility loss potentially predicting the onset of irreversible aggregation.
Molecular-species-specific ViPS, using bond-selective mid-infrared excitation instead of the water overtone band, could separate the thermophoretic contributions of proteins, lipids, and nucleic acids and directly test the diffusiophoresis interpretation.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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

You should know two things about this paper. First, it is a real technical advance: combining vibrational photothermal microscopy with optical diffraction tomography to image label-free Soret transport inside living cells, while deriving the temperature map from the fast thermal-refractive-index component. That temperature map is independent of the transport model, and the 1064-nm control cleanly rules out optical tweezing as the cause of the slow concentration changes. Second, the headline numbers (D ≈ 4.6 μm²/s, S_T ≈ 9.3×10⁻³ K⁻¹) are less solid than they look. They come from fitting a 2D advection–diffusion model (Eq. 7) to what appears to be a single representative cell's Δσ curve, and the model explicitly excludes ∇²σ, the diffusion of the pre-existing dry-mass distribution, on a force-balance assumption. The authors state this assumption openly, but they give no evidence that the baseline dry-mass gradients are actually balanced by non-thermal forces. If the assumption fails, the fitted D and S_T absorb the cell's baseline inhomogeneity rather than pure thermophoresis. The reader's stress-test note quantifies this risk: over 2.4 s, D∇²σ t can be comparable to the observed signal near steep RI gradients. That is a legitimate model-error concern, not a refutation—it is testable by showing the magnitude of ∇²σ in the analyzed region or by fitting multiple cells and checking stability.

What else is good: the RNA-reduction experiment (Actinomycin D) is a nice perturbation, giving a chemically specific handle on what is migrating. The observed opposite direction in cytoplasm versus nucleus is qualitatively compelling, and the time-lapse decline in thermophoresis over hours is suggestive, though only four cells and no significance testing. The nucleus–cytoplasm Soret values in Fig. 3c are single numbers without error bars, and the fitted D and S_T come without cell-to-cell statistics. No code or data are shipped; "available upon request" is weaker than it should be.

Overall, the core physical separation of timescales is sound, the method is new, and the phenomena are interesting. The quantitative fragility is real but addressable, and the paper's own discussion shows the authors know some limits. This deserves peer review; a good referee should push for validation of the force-balance assumption, multi-cell statistics, and a look at the baseline ∇²σ magnitude. I would cite it if I worked on thermophoresis or label-free intracellular imaging, and it is worth a reading-group slot.

Referee Report

3 major / 4 minor

Summary. The manuscript introduces ViPS (vibrational photothermal Soret) microscopy, which combines continuous mid-IR heating with time-resolved optical diffraction tomography (ODT) to separate the fast thermal-expansion refractive-index change (Δn_ρ) from the slower dry-mass-concentration change (Δn_σ) driven by thermophoresis. The authors report intracellular diffusion and Soret coefficients in the nucleus of a living COS7 cell, D ≈ 4.6 μm² s⁻¹ and S_T ≈ 0.01 K⁻¹, obtained by fitting a 2D advection-diffusion model to the measured Δσ(t) curve. They also observe a reversed transport direction in the cytoplasm, which they attribute to diffusiophoresis, and a time-lapse reduction of thermophoretic activity in dying cells, which they interpret as glass formation. The central quantitative claims rest on the model in Eq. 7 and on a single-cell fit.

Significance. If the quantitative extraction is valid, ViPS imaging would be a significant methodological advance: it is label-free, provides simultaneous temperature and concentration maps, and extends thermophoresis measurements from in vitro systems to living cells. The time-scale separation is physically motivated, the 1064-nm off-resonance control addresses an important artifact, and the RNA-depletion experiment is a reasonable first step toward molecular attribution. However, the two headline coefficients come from a single representative cell, and the governing equation explicitly excludes a term that may be comparable to the signal; these issues must be resolved before the quantitative claims can be accepted.

major comments (3)

[Methods, Eq. 7] The model excludes the diffusion term D∇²σ(x) of the pre-existing dry-mass distribution, with the stated assumption that the forces maintaining this distribution are balanced by factors other than the temperature gradient. This assumption is load-bearing: the measured Δσ signal is at the percent level, and the RI maps show substantial intranuclear inhomogeneity, so D∇²σ(x) over 2.4 s could easily contribute a signal comparable to the observed Δσ at the edge of the heating spot. The manuscript provides no direct evidence that σ is stationary over the measurement window in the absence of heating, nor a robustness test of the fitted D and S_T when the term is included or when a residual force term is added. Please provide such a control (e.g., a no-heating time series of σ from the same cells) or a sensitivity analysis of the fit; otherwise the reported D and S_T are effective parameters that may absorb baseline cell inhomogeneity.
[Results, Fig. 2e] The reported D = 4.49 ± 0.10 μm²/s and S_T = 9.28 ± 0.10 × 10⁻³ K⁻¹ are obtained from a single representative cell's Δσ(t) curve, and the quoted uncertainties are residual-based confidence intervals, not cell-to-cell variability. The abstract presents these values as the method's quantitative output and uses them to infer a molecular weight around 100 kDa. The later 12-cell measurement (Fig. 3c) uses a different, more approximate estimator (−Δσ/σ/ΔT_s) and does not provide a complementary D distribution. Please report fitted D and S_T values across a population of cells, or explicitly state that the values are a single-cell demonstration and provide population-based estimates using a consistent analysis pipeline.
[Eq. 7 and Fig. 2b] The simulation is two-dimensional, but the measured quantity is a depth-integrated projection of a three-dimensional transport process, and the driving temperature field is the depth-averaged ΔT_s from a 4.2-μm-thick slab. The no-flux boundary condition (Eq. 8) constrains only the lateral nuclear boundary; axial thermophoretic and diffusive fluxes through the top and bottom of the integration volume are not modeled or bounded. This collapse from 3D to 2D may introduce systematic errors in D and S_T that are not reflected in the fit residuals. The authors should quantify this error, for example by performing a 3D simulation on the same cell geometry or by showing that the axial temperature gradient is negligible over the integration thickness.

minor comments (4)

[Results] The statement that S_T ≈ 0.01 K⁻¹ means '~1% of the molecules migrate after 2.4 seconds of heating with a 1 K temperature difference' is imprecise; S_T relates the steady-state concentration gradient to the temperature gradient, not a finite-time fractional migration. Consider rephrasing to 'a steady-state relative concentration change of ~1% per kelvin'.
[Results] The diffusion coefficient is compared with FRAP measurements that are typically performed at 37°C, while the present measurements were performed at 24°C. A brief comment on the expected temperature dependence of D would help the reader assess the molecular-weight inference.
[Supplementary Note 1] The cooling duration of 10.2 s is stated to ensure that the cell returns to its initial state, but no return-to-baseline data are shown. A supplementary figure of the temporal recovery of σ after heating would support the repeatability of the measurement cycle.
[Fig. 4] The interpretation of the time-lapse data as 'glass formation during the dying process' is presented as a conclusion from four cells and indirect RI evidence. Consider softening the claim or adding a quantitative metric of mobility change (e.g., a time-dependent apparent diffusion coefficient) to strengthen the connection.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the reported D and S_T are explicit inverse-fit results, not predictions, and the key model assumptions are stated rather than definitionally forced.

full rationale

The paper's central quantitative claim is the measurement of intracellular diffusion and Soret coefficients by fitting a numerical model to time-resolved dry-mass-change images. The paper explicitly states that the values were obtained by parameter search: "Through a systematic parameter search, we found that S_T = 9.28 (± 0.10) × 10⁻³ K⁻¹ and D = 4.49 ± 0.10 μm²/s best reproduced the experimental results." This is an inverse estimation procedure, not a claim that the coefficients are predicted from first principles, so the fitted-input-called-prediction pattern does not apply. The temperature map used in the model is measured independently from the initial fast Δnρ image via Eq. 3, and the slow Δnσ signal is obtained from the same dataset only after the stated timescale separation; this is an experimental assumption, not a definitional equivalence. The exclusion of the pre-existing dry-mass diffusion term ∇²σ(x) in Eq. 7 is acknowledged explicitly as an assumption: "the diffusion term associated with the steady-state inhomogeneous dry mass distribution prior to heating (i.e., ∇²(σ(x))) is excluded, based on the assumption that the diffusion forces from this steady-state distribution are balanced by forces generated by factors other than the temperature gradient." That is a model-robustness limitation that could bias the fitted parameters if violated, but it is not a circular step because the model is not constructed from the measured Δσ curve by definition; it is a stated physical approximation. The 5 ms thermalization timescale is supported by the authors' prior work (Ref. 12), but that prior work is an externally falsifiable experimental measurement, not a uniqueness theorem or ansatz smuggling in the current result. The manuscript also includes independent checks against external benchmarks (FRAP diffusion coefficients from other groups), an off-resonance 1064-nm control for the optical-tweezer effect, and an Actinomycin D RNA-depletion comparison. No step in the derivation reduces to its inputs by construction, so no circularity is found.

Assumptions & free parameters 2 free parameters · 6 assumptions · 0 invented entities

The ledger shows that the quantitative coefficients rest on a fitted two-parameter model and several domain assumptions about the cellular environment, not on a first-principles derivation. No new physical entities are introduced.

free parameters (2)

Diffusion coefficient D = 4.49 ± 0.10 μm²/s
Fitted by grid search (step 0.02 μm²/s) to minimize squared error between simulated and measured Δσ at the center of the heating spot. This is a core quantitative result.
Soret coefficient S_T = 9.28 ± 0.10 × 10⁻³ K⁻¹
Fitted simultaneously with D by the same parameter search. It is not independently predicted.

assumptions (6)

domain assumption Thermophoretic flux follows J = -D∇c - D_T c∇T with S_T = D_T/D (Eqs. 4-5).
Adopted from ref [20]; standard model for dilute solutions, applied here to the crowded intracellular environment.
domain assumption The steady-state temperature rise ΔT_s can be recovered from the early Δnρ image using the bulk water thermo-optic coefficient dn/dT = -1.0 × 10⁻⁴ K⁻¹ (Eq. 3).
Assumes cell water has the same thermo-optic response as bulk water and that thermal expansion reaches steady state within 5 ms (Ref. 12).
ad hoc to paper The diffusion term from the pre-existing inhomogeneous dry-mass distribution ∇²σ(x) is excluded from the simulation (Eq. 7).
Stated in Methods: assumes the steady-state gradients are balanced by forces other than the temperature gradient. If not, fitted D and S_T include baseline inhomogeneity.
domain assumption A 2D depth-integrated model in σ (fg/μm²) captures the 3D transport relevant to the ODT measurement.
The simulation solves Eq. 7 in-plane and ignores axial concentration gradients; the ODT axial resolution is about 5 μm.
domain assumption Nuclear boundary is impermeable to the migrating molecules (Eq. 8 no-flux condition).
Based on the observation that 85% of expelled molecules remain in the nucleus; applied as a boundary constraint.
domain assumption Actinomycin D depletes most intracellular RNA, so the 19% signal drop can be attributed to RNA.
Relies on literature values (refs 22, 25) for RNA mass fraction and drug response.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Ludwig-Soret microscopy with vibrational photothermal effect." pith.science (2026). https://pith.science/paper/4HNYTSKR

@misc{pith2026250204578,
  author       = {Pith},
  title        = {Pith review of: Ludwig-Soret microscopy with vibrational photothermal effect},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4HNYTSKR}},
  note         = {Machine review of arXiv:2502.04578}
}

read the original abstract

Vibrational microscopy provides label-free, bond-selective chemical contrast by detecting molecular vibrations, making it invaluable for biomedical research. While conventional methods rely on the direct detection of Raman scattering or infrared absorption, recently developed vibrational photothermal (ViP) microscopy achieves chemical contrast indirectly through refractive index (RI) changes. This indirect approach enables unique imaging capabilities beyond traditional chemical imaging. Here, we introduce a novel application of ViP microscopy: label-free intracellular thermophoretic (Soret) imaging, which visualizes biomolecular transport driven by temperature gradients. ViP-induced Soret (ViPS) imaging leverages a steady-state temperature distribution generated by optical heating through vibrational photothermal effect, combined with time-resolved RI imaging via optical diffraction tomography (ODT). Using ViPS imaging, we measured thermophoretic behavior in living COS7 cells, determining intracellular diffusion and Soret coefficients. Notably, we observed a reversed direction of molecular transport (negative Soret effect) in the cytoplasm compared to the nucleus, possibly driven by thermophoresis-induced diffusiophoresis. Furthermore, time-lapse imaging under CO2-depleted conditions revealed a remarkable reduction in thermophoretic activity, suggesting glass formation during the dying process, likely due to polymer aggregation. ViPS imaging represents a new frontier in intracellular thermophoretic studies, expanding the capabilities of vibrational microscopy.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

31 extracted references · 29 canonical work pages

[1]

Wrobel, T. P et al. Infrared Spectroscopic Imaging Advances as an Analytical Technology for Biomedical Sciences. Anal. Chem. 90, 1444–1463 (2018)

work page 2018
[2]

Shipp, D. W. et al. Raman spectroscopy: techniques and applications in the life sciences. Adv. Opt. Photonics 9, 315 (2017)

work page 2017
[3]

Bai, Y. et al. Bond-selective imaging by optically sensing the mid-infrared photothermal effect. Sci. Adv. 7: eabg1559 (2021)

work page 2021
[4]

Zhang, D. et al. Depth-resolved mid-infrared photothermal imaging of living cells and organisms at submicrometer spatial resolution. Sci. Adv. 2: e1600521 (2016)

work page 2016
[5]

Fu, P. et al. Super-resolution imaging of non-fluorescent molecules by photothermal relaxation localization microscopy. Nat. Photonics 17, 330–337 (2023)

work page 2023
[6]

Yin, J. et al. Video-rate mid-infrared photothermal imaging by single-pulse photothermal detection per pixel. Sci. Adv. 9, eadg8814 (2023)

work page 2023
[7]

Ishigane, G. et al. Label-free mid-infrared photothermal live-cell imaging beyond video rate. Light Sci. Appl. 12, 174 (2023)

work page 2023
[8]

Tamamitsu, M. et al. Mid-infrared wide-field nanoscopy. Nat. Photonics 18, 738-743 (2024)

work page 2024

Show all 31 references

[9]

Freudiger, C. W. et al. Label-Free Biomedical Imaging with High Sensitivity by Stimulated Raman Scattering Microscopy. Science 322, 1857–1861 (2008)

2008
[10]

Samolis, P. D. et al. Label-free imaging of fibroblast membrane interfaces and protein signatures with vibrational infrared photothermal and phase signals. Biomed. Opt. Express 12, 5400 (2021)

2021
[11]

Yin, J. et al. Nanosecond-resolution photothermal dynamic imaging via MHz digitization and match filtering. Nat. Commun. 12, 1–11 (2021)

2021
[12]

Toda, K. et al. Label-free mid-infrared photothermal microscopy revisits intracellular thermal dynamics : what do fluorescent nanothermometers measure ? arXiv 2406.16265 (2024)

2024 arXiv
[13]

Duhr, S. et al. Why molecules move along a temperature gradient. PNAS 103, 19678–19682 (2006)

2006
[14]

Baaske, P. et al. Optical thermophoresis for quantifying the buffer dependence of aptamer binding. Angew. Chem., Int. Ed. 49, 2238–2241 (2010)

2010
[15]

Wienken, C. J. et al. Protein-binding assays in biological liquids using microscale thermophoresis. Nat. Commun. 1, 100 (2010)

2010
[16]

Chen, Z. et al. Heat-Mediated Optical Manipulation. Chem. Rev. 122, 3122–3179 (2022)

2022
[17]

Kollipara, P. S. et al. Hypothermal opto-thermophoretic tweezers. Nat. Commun. 14, 1–9 (2023)

2023
[18]

Popescu, G. et al. Optical imaging of cell mass and growth dynamics. Am. J. Physiol. - Cell Physiol. 295, 538– 544 (2008)

2008
[19]

Park, Y. K. et al. Quantitative phase imaging in biomedicine. Nat. Photonics 12, 578–589 (2018)

2018
[20]

Fukuyama, T. et al. Opto-thermal diffusiophoresis of soft biological matter: from physical principle to molecular manipulation. Biophys. Rev. 12, 309–315 (2020)

2020
[21]

Arrio-Dupont, M. et al. Translational diffusion of globular proteins in the cytoplasm of cultured muscle cells. Biophys. J. 78, 901–907 (2000)

2000
[22]

Cell Biology by the Numbers

Milo, R. Cell Biology by the Numbers. 1st edn. (New York: Garland Science, 2015)

2015
[23]

Shav-Tal, Y. et al. Dynamics of single mRNPs in nuclei of living cells. Science 304, 1797–1800 (2004)

2004
[24]

Dirk, G. et al. Transport between the cell nucleus and the cytoplasm. Annu. Rev. Cell Dev. Biol. 15, 607–660 (1999)

1999
[25]

Perry, R. P. et al. Inhibition of RNA synthesis by actinomycin D: Characteristic dose‐response of different RNA species. J. Cell. Physiol. 76, 127–139 (1970)

1970
[26]

Maeda, Y. et al. Thermal separation: Interplay between the soret effect and entropic force gradient. Phys. Rev. Lett. 107, 1–4 (2011)

2011
[27]

Nishizawa, K. et al. Universal glass-forming behavior of in vitro and living cytoplasm. Sci. Rep. 7, 1–12 (2017)

2017
[28]

Reichl, M. et al. Thermophoretic manipulation of molecules inside living cells. J. Am. Chem. Soc. 136, 15955– 15960 (2014)

2014
[29]

Umeda, K. et al. Activity-dependent glassy cell mechanics II: Nonthermal fluctuations under metabolic activity. Biophys. J. 122, 4395-4413 (2023)

2023
[30]

Hugonnet, H. et al. Improving Specificity and Axial Spatial Resolution of Refractive Index Imaging by Exploiting Uncorrelated Subcellular Dynamics. ACS Photonics 11, 257–266 (2024)

2024
[31]

Maeda, Y. et al. Effects of long DNA folding and small RNA stem-loop in thermophoresis. PNAS 109, 17972– 17977 (2012). Methods Thermophoresis (Soret effect) and diffusiophoresis. A spatial temperature gradient in a solution induces unidirectional molecular transport. The total...

2012

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.

[1] [1]

Wrobel, T. P et al. Infrared Spectroscopic Imaging Advances as an Analytical Technology for Biomedical Sciences. Anal. Chem. 90, 1444–1463 (2018)

work page 2018

[2] [2]

Shipp, D. W. et al. Raman spectroscopy: techniques and applications in the life sciences. Adv. Opt. Photonics 9, 315 (2017)

work page 2017

[3] [3]

Bai, Y. et al. Bond-selective imaging by optically sensing the mid-infrared photothermal effect. Sci. Adv. 7: eabg1559 (2021)

work page 2021

[4] [4]

Zhang, D. et al. Depth-resolved mid-infrared photothermal imaging of living cells and organisms at submicrometer spatial resolution. Sci. Adv. 2: e1600521 (2016)

work page 2016

[5] [5]

Fu, P. et al. Super-resolution imaging of non-fluorescent molecules by photothermal relaxation localization microscopy. Nat. Photonics 17, 330–337 (2023)

work page 2023

[6] [6]

Yin, J. et al. Video-rate mid-infrared photothermal imaging by single-pulse photothermal detection per pixel. Sci. Adv. 9, eadg8814 (2023)

work page 2023

[7] [7]

Ishigane, G. et al. Label-free mid-infrared photothermal live-cell imaging beyond video rate. Light Sci. Appl. 12, 174 (2023)

work page 2023

[8] [8]

Tamamitsu, M. et al. Mid-infrared wide-field nanoscopy. Nat. Photonics 18, 738-743 (2024)

work page 2024