Influence of surfactant, particle size and dispersion medium on surface plasmon resonance of silver nanoparticles

Divya Verma; Gunadhor Singh Okram; Vikash Sharma

REVIEW 5 major objections 6 minor 1 cited by

Influence of surfactant, particle size and dispersion medium on surface plasmon resonance of silver nanoparticles

T0 review · 5 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read This paper reports that dispersing silver nanoparticles in common solvents splits their single ultraviolet surface plasmon resonance peak into up to seven distinct peaks, with solvent polarity, refractive index, and surfactant coverage…

desk verdict The paper's real asset is a systematic synthesis/DLS dataset; its 'first time' claim of solvent-induced multi-peak SPR splitting is not reproducible from the evidence shown. read the letter →

arxiv 1908.03064 v1 pith:GLYBGJT2 submitted 2019-08-08 physics.app-ph cond-mat.mtrl-sci

Vikash Sharma , Divya Verma , Gunadhor Singh Okram This is my paper · ORCID

classification physics.app-phcond-mat.mtrl-sci

keywords silvernanoparticlessurfaceplasmonresonancemulti-peakSPRultravioletplasmonicssolventrefractiveindexchargetransferzetapotentialDrudemodel

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 is trying to establish that the single surface-plasmon-resonance peak seen for silver nanoparticles in air is not fixed: dispersing the same particles in ordinary solvents splits that peak into several distinct ultraviolet resonances, with up to seven peaks in n-hexane. The paper argues that solvent polarity and refractive index, together with the surfactant coating and particle size, control where these peaks sit and how many appear, through charge transfer and interface dipoles between the nanoparticle and adsorbed solvent molecules. A sympathetic reader would care because this would make dispersion chemistry a practical lever for producing and tuning multiple UV plasmon resonances for photocatalysts, UV sensors, and detectors, without changing the particle material.

What carries the argument

The central mechanism is interface charge transfer between the nanoparticle and adsorbed solvent molecules, summarized in an energy-level diagram of the metal Fermi level ($E_F \approx 5.49$ eV for Ag) and the solvent HOMO and LUMO levels. Depending on the direction of the interface dipole barrier ($V_\text{dipole}$), the electron and hole injection barriers ($\Delta_e$, $\Delta_h$) change, so transferred charge forms dipoles whose oscillations add new frequencies to the plasmon response. The paper also uses the Drude relation $\lambda^2 = \lambda_P^2(\epsilon_\alpha + 2\epsilon_m)$, with $\lambda_P$ the bulk plasmon wavelength and $\epsilon_\alpha$ the high-frequency dielectric constant, as the test of whether refractive index alone accounts for the shift. Dynamic light scattering supplies zeta potential and hydrodynamic diameter as the measure of surface charge density and agglomeration that the argument connects to the number and splitting of the peaks.

What would settle it

Measure the UV-visible absorption of the same dispersion after removing large agglomerates by centrifugation or filtration and at several concentrations; if the number and positions of the split peaks survive while the hydrodynamic diameter distribution collapses to a single narrow peak, the splitting is intrinsic, whereas if the extra peaks vanish or move with the agglomerate fraction, the charge-transfer assignment is not supported.

Watch

Extended reading notes

Core claim

The authors synthesize phase-pure fcc silver nanoparticles (Scherrer sizes 15.1 to 33.4 nm) capped with oleylamine, trioctylphosphine, or polyvinylpyrrolidone. In air, diffuse reflectance shows a main SPR dip at 316 nm for 31.5 nm particles that shifts to 320 nm for 15.1 nm particles: a red shift with decreasing size that they attribute to increasing adsorbate coverage rather than the usual size-induced blue shift. Dispersing 15.1 nm particles in methanol, deionized water, ethanol, n-hexane, benzene, and toluene moves the main peak to 259, 261, 277, 255, 275, and 282 nm, respectively, and the single air peak splits into five, four, three, and seven resolvable peaks in methanol, water, ethanol, and n-hexane. The paper attributes the splitting to charge transfer across the particle–solvent interface: adsorption of solvent molecules modifies the metal Fermi level relative to the solvent HOMO and LUMO, forming interface dipoles that oscillate at slightly different frequencies and producing new resonances. It further reports that the Drude relation $\lambda^2 = \lambda_P^2(\epsilon_\alpha + 2\epsilon_m)$ is followed by the nonpolar solvents but not by the polar ones, indicating that refractive index alone does not describe polar solvents.

Load-bearing premise

The load-bearing premise is that the extra ultraviolet peaks are genuine resonances at the particle–solvent interface rather than artifacts of particle clumping, light scattering, or the switch from measuring powders by reflection to measuring liquids by transmission.

Editorial extensions

If this is right

Choosing the dispersion solvent becomes a tuning knob: simply moving the same 15 nm Ag particles from air into methanol, water, ethanol, or n-hexane moves the main ultraviolet resonance from about 320 nm to 259, 261, 277, or 255 nm, respectively, and changes how many resonances appear.
A single dispersion can present several distinct ultraviolet resonances at once (up to seven in n-hexane), which is useful for multi-wavelength photocatalysis, UV sensing, and detection without fabricating new nanostructures.
Increasing the trioctylphosphine coverage red-shifts the main SPR even as particle size decreases, so surfactant chemistry can be used to compensate or override the usual size-dependent blue shift.
The Drude-model test implies that empirical predictions of SPR position in polar solvents must include polarity or charge-transfer effects, not just solvent refractive index.
Because the weak split peaks grow with concentration and ultrasonication time, dispersion preparation conditions (concentration, sonication, stirring) need to be controlled to get reproducible multi-peak spectra.

Reading between the lines

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

The authors' qualitative Fermi-level/HOMO-LUMO picture predicts a quantitative trend that the paper does not compute: the energetic spacing of the split peaks should correlate with the electron-donor or electron-acceptor character of the solvent, so computing charge-transfer energies could turn the diagram into a predictive rule.
If the splitting is truly at the single particle–solvent interface, then single-particle spectroscopy of one Ag nanoparticle in each solvent should show the multi-peak structure without any agglomeration; the paper's ensemble spectra leave this untested.
The failure of the Drude relation specifically in polar solvents suggests that solvent polarity parameters, rather than refractive index alone, should enter empirical models of UV SPR shifts; this could be tested with solvent pairs matched in refractive index but differing in polarity.
A confirmed multi-peak UV resonance in simple colloidal dispersions would be a low-cost route to multi-wavelength UV plasmonic substrates, replacing top-down patterning with a beaker-scale dispersion step.

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

The thing to know: this is a real experimental dataset on Ag NPs—seven syntheses, XRD/TEM/XPS/FTIR, DLS zeta and hydrodynamic diameters in six solvents, UV-Vis in air and in dispersion—and the directional trends (size red-shift with TOP coverage, refractive-index red-shift, multipolar modes in air) are credible and consistent with prior work. But the paper's headline claim—that a single SPR peak in air splits into three, four, five and seven peaks in ethanol, DIW, methanol and n-hexane, observed 'for the first time'—is not supported by the evidence actually shown. I agree with the stress-test note: Fig. 14 only lists two features per solvent, Fig. 17 gives different counts (five clear plus two feeble in ethanol, two plus one in n-hexane), and Fig. 12(d) already shows six dips in the air reflectance. So the 'single peak in air' premise is not literally correct, and the peak-count inventory is not reproducible from the paper. The Drude-model claim inherits the same problem: which sub-peak was selected as λ_max after splitting? Three points per solvent class, no fit or residuals, and the nonpolar-linearity conclusion is visually asserted rather than quantified.

What the paper does well: the synthesis matrix is systematic; the DLS/zeta measurements are at least reported (though the second-run variability argues for more replicates); the spectra show a clear solvent-dependent shift of the main UV feature that should be straightforward to reproduce; and the idea that the Drude relation fails in polar solvents due to interface charge transfer is a testable hypothesis, even if the model in Fig. 15 is only a qualitative energy-level diagram with no calculation connecting the observed peak positions to the proposed dipoles.

There are also smaller soft spots: no error bars on SPR peak positions, no table of all claimed sub-peak positions/amplitudes, and the 'first time' phrasing is broader than the literature warrants (multipolar SPR and solvent-index shifts are already known; the new bit is the specific multi-peak set in these solvents). The charge-transfer explanation is post hoc, but the authors do flag it as a model; that's okay as long as the phenomenon itself is secure. Right now it isn't.

Who is this for? Someone working on Ag NP synthesis and UV plasmonics might mine the dataset, but they'd need to re-measure the peak counts. It deserves peer review—the editor should send it out—but a serious referee should demand a reproducible peak inventory, error bars, replicate spectra, and a quantitative or at least clearly falsifiable model before the splitting claim is accepted.

Recommendation: engage with it, but with the counts as the make-or-break issue.

Referee Report

5 major / 6 minor

Summary. The manuscript reports a systematic study of surface plasmon resonance (SPR) in thermally decomposed silver nanoparticles with different surfactants (oleylamine, trioctylphosphine, PVP) and sizes, dispersed in polar and nonpolar solvents. The main claims are: (i) the main SPR peak in air shifts with particle size, surfactant coverage, and solvent refractive index; (ii) a single SPR peak in air splits into as many as seven peaks in certain solvents, reported 'for the first time'; (iii) the Drude model for the SPR peak shift applies only in nonpolar solvents, not polar solvents; and (iv) the splitting is attributed to charge transfer and interface-dipole formation. The paper includes structural and surface characterization (XRD, TEM, XPS, FTIR), DLS-based zeta potential and hydrodynamic diameter measurements, and UV-Vis spectra in reflectance and transmission modes.

Significance. If the peak-splitting phenomenon and the solvent-dependent Drude behavior are confirmed, the paper would offer a practical route to generate and tune multiple UV plasmon resonances in Ag nanoparticle dispersions, which is relevant for photocatalysis and UV sensing. The materials characterization is generally careful, and the Drude-model comparison is a falsifiable test. However, the evidence for the central claim is currently insufficient: the peak counts are internally inconsistent, no reproducible peak inventory is provided, and the charge-transfer explanation is not quantitatively constrained.

major comments (5)

[Influence of refractive index of solvent on SPR] The claimed 3/4/5/7-peak splitting is not reproducible from the presented data. In Figure 14 only two absorption features per solvent are labelled (e.g., 209 and 255 nm in n-hexane; 214 and 259 nm in methanol), and no table lists the positions and amplitudes of all claimed components. The probe-ultrasonication experiment in Figure 17 states 'Five (two) peaks can be clearly seen, while two (one) peaks are feeble in ethanol (n-hexane)', which implies seven peaks in ethanol and three in n-hexane—the reverse of the abstract's 'three ... in ethanol ... and seven ... in n-hexane'. Please reconcile the counts with labelled spectra and provide a complete peak inventory.
[Influence of particle size and adsorbate coverage on UV-Vis spectroscopic response] The reference state 'a single SPR peak exhibited in air' is contradicted by Figure 12(d), which shows six reflectance dips at 220, 320, 378, 454, 498 and 554 nm for Ag4, with the text assigning P2–P6 to SPR modes. The claim that solvent dispersion splits a single SPR peak is therefore ambiguous; the authors should state explicitly which air feature is being split and how the pre-existing multipolar modes in air are accounted for.
[Influence of refractive index of solvent on SPR] Figure 16 is constructed from only three solvent points per class, with no linear fit, residuals, or goodness-of-fit statistic. Because the manuscript reports multiple sub-peaks per solvent, the criterion for selecting lambda_max for the Drude plot is not defined. As a result, the conclusion that the Drude model applies only in nonpolar solvents is not quantitatively supported.
[Charge-transfer mechanism discussion (after Figure 15)] The charge-transfer/interface-dipole explanation is post hoc: it does not predict the number of split peaks or their positions, and the HOMO/LUMO values (e.g., 'LUMO ≈ 3.50 eV of methanol', 'HOMO ≈ 5.39 eV of ethanol') are stated without references or calculations. The model also does not explain why n-hexane, with seven peaks, differs from ethanol, with three. As it stands, the proposed mechanism is a qualitative cartoon rather than a falsifiable explanation.
[Dynamic Light Scattering Study] The DLS data in Figures 9 and 10 show poor run-to-run reproducibility (e.g., two HD peaks in methanol in the first run change to different positions in the second run; the zeta-potential distribution in methanol changes from five peaks to three). Since the FESEM images indicate agglomeration, particularly in n-hexane, the instability of the dispersion state makes it difficult to exclude agglomeration or scattering artifacts as the origin of the extra UV peaks. The paper's charge-transfer interpretation requires a more stable dispersion characterization.

minor comments (6)

[Figure 12(b)] Figure 12(b) axis label contains the typo 'Reflactance'; it should be 'Reflectance'.
[Figure 16] The axis label in Figure 16 says '2εm' but the caption says 'εm'; clarify the quantity plotted.
[Drude model paragraph] The text says the graph is 'λ²/1000 versus 2εm' while the caption says 'λ²/1000 versus εm'; make these consistent.
[Conclusion] The conclusion lists only one SPR position per solvent (e.g., 259 nm for methanol, 277 nm for ethanol) without reconciling it with the 'splitting' claim; clarify which of the split components is designated as the main SPR peak.
[Figure S1] The Gaussian fitting shown in Figure S1 should include the number of peaks fitted, the baseline model, and residuals; the current figure is not sufficient to judge the quality of the fit used to determine peak positions.
[References] References 38 and 44 have inconsistent formatting (Ref. 38 has a parenthetical year; Ref. 44 is incomplete).

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's claims are empirical observations compared against an external model, with no parameter fitted to the target conclusion.

full rationale

I walked the derivation chain: the central claims are (i) the position and number of UV-Vis absorption features of Ag nanoparticles in air and in six solvents, determined from measured reflectance/absorbance spectra and Gaussian peak fitting; (ii) the observation that a feature near 316-320 nm in air shifts to ~255-282 nm in solvents; and (iii) a comparison of the solvent peak positions against the Drude relation lambda^2 = lambda_P^2(epsilon_alpha + 2 epsilon_m). None of these is self-definitional. The Drude plot (Fig. 16) is an external benchmark: lambda_P and epsilon_alpha are not fitted from the same data to reproduce the solvent shifts; the paper simply plots observed lambda^2 against 2 epsilon_m and notes linearity for nonpolar and non-linearity for polar solvents. The charge-transfer/HOMO-LUMO picture (Fig. 15) is an interpretive explanation offered after the spectra are recorded; it is not used to define or generate the peak positions, and no parameter of the model is fitted to force agreement. The few self-citations (Refs. 16, 38) are used only for comparison with prior Ni NP results or damping behavior; they are not load-bearing premises from which the present claims are derived. Possible concerns about peak-count reproducibility or about attributing UV features to SPR are correctness/evidence issues, not circularity. Hence score 0.

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

No free parameters are fitted in the central argument; the physical input comes from standard Drude/Mie theory and from assumed molecular orbital energies. The main risk is not circularity but ad hoc explanation: the interface-charge-transfer model is used to rationalize the observed splitting without independent quantitative support.

assumptions (5)

domain assumption The Drude/Mie relation lambda^2 = lambda_P^2 (epsilon_alpha + 2 epsilon_m) with constant lambda_P and epsilon_alpha applies to these coated, agglomerated Ag nanoparticles.
Used in figure 16 to test solvent dependence; the model itself is standard but its applicability to these specific particles is assumed without correction for adsorbates or aggregation.
domain assumption The quoted HOMO and LUMO energies for methanol and ethanol are correct and relevant to the charge-transfer explanation.
Quoted in the charge-transfer discussion (e.g., LUMO ~3.50 eV for methanol, HOMO ~5.39 eV for ethanol) without a citation; the peak-splitting explanation depends on these numbers.
domain assumption DLS intensity peaks correspond to distinct particle populations that map onto the multiple SPR peaks.
The connection between multiple HD/zeta peaks and multiple SPR peaks relies on this assumption; the DLS data are run-dependent and highly variable.
domain assumption The SPR peak assignments (P1 interband, P2 dipole SPR, P3-P6 multipolar or coupled modes) are correct for these shapes and coatings.
Assigned by analogy to refs 1,4,18,17,19,20,21,35; no electrodynamics calculation is performed for the specific particle shapes, sizes, or surfactant shells.
ad hoc to paper Interface charge transfer and dipole formation between solvent molecules and the nanoparticle surface are the physical cause of the observed peak splitting.
Invoked solely to rationalize the observed splitting; the paper provides a schematic energy-level diagram (Fig. 15) but no quantitative derivation or independent test.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Influence of surfactant, particle size and dispersion medium on surface plasmon resonance of silver nanoparticles." pith.science (2026). https://pith.science/paper/GLYBGJT2

@misc{pith2026190803064,
  author       = {Pith},
  title        = {Pith review of: Influence of surfactant, particle size and dispersion medium on surface plasmon resonance of silver nanoparticles},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GLYBGJT2}},
  note         = {Machine review of arXiv:1908.03064}
}

read the original abstract

Clear influence of particle size, surfactants and dispersion medium on surface plasmon resonance (SPR) features of Ag nanoparticles (NPs), synthesized in thermal decomposition method, in the broad range of ultraviolet (UV) radiation, critical for many potential applications such as a photocatalyst, UV-sensor and detector, has been demonstrated here. It involves adsorbate coverage, interparticle distance or agglomeration, surface charge density and solvent refractive index ({\mu}). NP agglomeration and surface charge density in solvents of varying {\mu} have been studied systematically through zeta-potential ({\zeta}) and hydrodynamic diameter (HD) using dynamic light scattering (DLS). The main SPR feature found at 316 nm in 31.5 nm NPs shifts to 320 nm in 15.1 nm NPs. The peak at 320 nm in air shifts to 259, 261 and 277 nm in polar solvent methanol, deionized water and ethanol, respectively and to 255, 275 and 282 nm in non-polar solvent n-hexane, benzene and toluene, respectively. In general, the decrease in particle size and increase in {\mu} of solvents show red-shift. Curiously, a number of peaks up to seven in these solvents that are attributed to charge-transfer mechanism and change in inter-particle interaction of the NPs turning from a single peak of SPR in air has been observed for the first time. A model for re-adjustment of Fermi level (E_F) of Ag NP and the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) to explain them has also been used. Moreover, the Drude model for shift in the position of SPR in these NPs is only applicable in non-polar solvents, not in polar solvents. Such novel features will be potential candidates for various applications.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

Electrical transport and thermoelectric properties of silver nanoparticles
cond-mat.mes-hall 2019-08 conditional novelty 5.0 of 10

Silver nanoparticles 15 to 33 nm show a fitted 36% drop in Debye temperature and size-dependent phonon-drag minima in the Seebeck coefficient, but the electron-phonon coupling trend reverses for TOP-only samples.

Reference graph

Works this paper leans on

1 extracted references · 1 canonical work pages · cited by 1 Pith paper

[1]

1 S. Peng, J. M. McMahon, G. C. Schatz, S. K. Gray, and Y. Sun, Proc. Natl. Acad. Sci. 107, 14530 (2010). 2 W. L. Barnes, A. Dereux, and T. W. Ebbesen, Nature 424, 824 (2003). 3 A. H. Alshehri, M. Jakubowska, A. Młożniak, M. Horaczek, D. Rudka, C. Free, and J. D. Carey, ACS Appl. Mater. Interfaces 4, 7007 (2012). 4 J. Gong, R. Dai, Z. Wang, and Z. Zhang, ...

work page 2010

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

1 S. Peng, J. M. McMahon, G. C. Schatz, S. K. Gray, and Y. Sun, Proc. Natl. Acad. Sci. 107, 14530 (2010). 2 W. L. Barnes, A. Dereux, and T. W. Ebbesen, Nature 424, 824 (2003). 3 A. H. Alshehri, M. Jakubowska, A. Młożniak, M. Horaczek, D. Rudka, C. Free, and J. D. Carey, ACS Appl. Mater. Interfaces 4, 7007 (2012). 4 J. Gong, R. Dai, Z. Wang, and Z. Zhang, ...

work page 2010