REVIEW 4 major objections 4 minor 4 references
Lithographically defined site-selective growth of Fe filled multi-walled carbon nanotubes using a modified photoresist
T0 review · 4 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash
Pith's one-line read Catalyst-doped photoresist grows iron-filled nanotubes only where patterned
desk verdict A practical and genuinely new catalyst-patterning trick, but the Fe-filled claim rests on an explicitly withheld composition analysis—send to referees with a request for EDX and controls. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing object is the modified photoresist (Mod-PR): a 0.2 M solution of iron(III) acetylacetonate in conventional HPR 504 photoresist, spin-coated at 4000 rpm onto Si(1 1 1). This layer does double duty: it is patterned by ordinary photolithography, and it stores the iron that later forms catalyst nanoparticles. The mechanism chain is decomposition of Fe(acac)$_3$ during the 900 °C hydrogen/argon anneal to form Fe nanoparticles (mean 70 nm), followed by propane pyrolysis at 850 °C, with carbon diffusing through or over the nanoparticle and liquid-like Fe being drawn into the growing tube to leave elongated fillings; tube diameter is set by the particle diameter, and particle-at-tip images identify tip growth.
What would settle it
After the usual development step but before CVD, map iron on the substrate with EDX or XPS: if iron is absent from the developed patterns, or if tubes later appear on unpatterned regions, the claim of site-selective catalyst retention and growth would be falsified.
Extended reading notes
Core claim
The paper's central claim is that a single modified photoresist layer can serve simultaneously as the patterning mask and the catalyst source for site-selective growth of partially iron-filled multi-walled carbon nanotubes. On the paper's account, annealing the patterned Fe(acac)$_3$-doped photoresist at 900 °C decomposes the iron precursor into catalytic nanoparticles with a mean diameter of 70 nm; subsequent propane CVD at 850 °C grows MWCNTs with outer diameters of 30–130 nm whose distribution tracks the nanoparticle size distribution, with catalyst particles observed at the tube tips. The iron inclusions inside the tubes are attributed to capillary drawing of liquid-like Fe particles during nucleation, following a growth model from the paper's earlier work. A schematic and SEM images show CNTs confined to the lithographically defined pattern, which the paper presents as a simple, reproducible, IC-compatible route to positioned magnetic-metal-filled nanotubes.
Load-bearing premise
The patterned photoresist retains enough Fe(acac)$_3$ through exposure and development to form catalytic Fe nanoparticles during the 900 °C anneal, and those particles—not background contamination—are the ones nucleating tubes in the patterned areas.
Editorial extensions
If this is right
- Site-selective growth of iron-filled MWCNTs can be achieved with only a doped photoresist and one lithography step, so no separate catalyst deposition is needed.
- Because the tube diameter distribution follows the catalyst nanoparticle distribution, controlling particle size through Fe concentration, anneal conditions, or resist processing should control tube diameter.
- Patterned, partially Fe-filled tubes are positioned by the lithographic mask, so arrays suitable for magnetic force microscopy, recording media, or bio-device integration can in principle be laid out directly.
- The tip-growth mode leaves the catalyst at the top of each tube, which may matter for electrical contacting or for catalyst removal depending on the application.
- The route is claimed to be reproducible and IC-compatible because it relies on standard exposure and development equipment.
Reading between the lines
- The paper does not quantify how much Fe(acac)$_3$ survives exposure and development; measuring that retention, for example by EDX or XPS on patterned versus unpatterned resist, would identify the margin by which the selectivity holds.
- Nothing in the method is specific to iron: the same resist-doping trick could plausibly be tried with other metal acetylacetonates to grow site-selected Ni- or Co-catalyzed or metal-filled nanostructures, though the paper does not claim this.
- A testable refinement would be shrinking the 70 nm mean particle size by lowering the precursor concentration or using a size-filtered resist layer; if diameter control follows, thinner tubes should result.
- The demonstrated outcome is morphological; future work would need to show that the site-selected filled tubes retain the magnetic behavior expected of Fe-filled CNTs.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript reports a CVD process in which a conventional photoresist (HPR 504) is doped with Fe(acac)3, spin-coated on Si(111), annealed, and used both as a catalyst source and as a lithographic patterning layer for the growth of multi-walled carbon nanotubes (MWCNTs). SEM shows catalyst nanoparticles with a reported mean diameter of 70 nm and spaghetti-like CNT growth; TEM shows multi-walled structures with dark elongated inclusions, which the authors identify as Fe without displaying the supporting composition analysis. A photolithographic version is presented in which CNTs grow from lithographically defined patterns. The central claim is that this simple, IC-compatible route yields site-selective growth of partially Fe-filled MWCNTs.
Significance. If fully substantiated, the approach would be attractive because it combines catalyst deposition and patterning in a single spin-coating step using standard photoresist technology, potentially simplifying the integration of metal-filled CNTs into microfabricated devices. The manuscript does not provide machine-checked proofs or digitally available data, but the experimental design is straightforward and the claims are falsifiable. The main value lies in the proposed combination of materials and process steps; however, the evidence presented is predominantly qualitative and omits the key compositional measurement, so the significance of the result as stated is currently not established.
major comments (4)
- [Experimental and Fig. 3] The sentence 'Chemical composition analysis confirms that the elongated particles are of Fe (not shown here)' withholds the decisive measurement for the paper's central claim. Without EDX, EELS, XPS, or another composition analysis, the high-contrast elongated particles in Fig. 3(c-f) could be iron oxide, carbonaceous residue, or imaging artifacts. Consequently, the title, abstract, and conclusion assertions of 'partially Fe filled MWCNTs' are not supported by any displayed data. The authors should include the composition analysis or, failing that, materially weaken the claims to 'particles whose composition was not determined.'
- [Experimental section, catalyst control] No control experiment using unmodified HPR 504 photoresist is reported. Such a control is necessary to show that the observed CNT growth and the encapsulated particles are caused by the Fe(acac)3 additive rather than by resist residue, substrate contamination, or the photoresist itself. This gap also affects the site-selectivity claim, because attributing growth on the patterned areas to the Fe-doped resist requires knowing that Fe is actually the active component.
- [Fig. 5 and Conclusion] The site-selective growth claim that CNTs 'were grown only on a pre-defined surface' is not quantitatively supported. Fig. 5 shows SEM images of a pattern before and after growth, but it lacks scale bars, a comparison of CNT density on patterned versus unpatterned regions, and a demonstration over multiple patterns or larger areas. A control with resist-only patterns or a statistical analysis of growth localization is needed to exclude random nucleation on the substrate.
- [Fig. 1 and Fig. 2 insets] The size distribution histograms for catalyst nanoparticles and CNT diameters are presented without sample sizes or error bars, and no statistical test is used to support the claim that the CNT diameter distribution mirrors the catalyst particle size distribution. The Gaussian-profile claim for the catalyst particles is therefore qualitative, and the comparison between the two distributions is not established quantitatively.
minor comments (4)
- [All figures] Several SEM and TEM images (Figs. 1, 2, 3, and 5) appear to lack visible scale bars or scale-bar annotations in the captions as printed; scale bars should be added or explicitly described so that the reported dimensions (e.g., 70 nm particles, CNT diameter range 30-130 nm) can be verified from the images.
- [Experimental section] The developer solution used after UV exposure is not specified (type, concentration, or manufacturer). For reproducibility of the photolithographic step, the developer composition and development conditions should be stated.
- [Reference [3]] Reference [3] lists the author as 'Palen EB'; the proper surname is likely 'Borowiak-Palen' (E. Borowiak-Palen). Please correct the citation.
- [Growth model, Section 3] The capillary-action growth model is invoked by citing the authors' earlier paper [4] but is not summarized in the present manuscript. A two-sentence description of the model would make the discussion self-contained and help the reader understand why liquid-like Fe particles are expected to fill the tubes.
Circularity Check
No significant circularity: the central claims are supported by direct experimental observations; the single self-citation is explanatory, not load-bearing.
full rationale
The paper is an experimental report with no equations, fitted parameters, or derived predictions. The central claims are: (1) a mixture of Fe(acac)3 and photoresist acts as a catalyst source for CNT growth, (2) the catalyst nanoparticle size governs CNT diameter, (3) the CNTs are partially Fe-filled, and (4) photolithographic patterning gives site-selective growth. These are supported by SEM and TEM images, size-distribution histograms, and comparison of the pattern before and after growth. The only self-citation is reference [4], the authors' 2010 paper, invoked for the 'growth model of partially Fe filled CNT' via capillary action of liquid-like Fe particles. This model is used as a post-hoc explanation of the observed filling, not as a premise from which the experimental outcome is deduced, so it is not load-bearing. The manuscript also states that 'Chemical composition analysis confirms that the elongated particles are of Fe (not shown here)'; this is an omitted proof and a genuine evidentiary gap, but it is a matter of experimental verification, not circular reasoning. No step in the paper reduces by definition or by self-citation to its own inputs. The derivation chain is therefore self-contained with respect to circularity, though the missing composition data raises correctness risk.
Assumptions & free parameters
assumptions (3)
- domain assumption Fe(acac)3 decomposes during the 900 degree Celsius anneal to form Fe catalytic nanoparticles on Si(111).
- domain assumption The developed Mod-PR pattern retains enough iron to nucleate CNT growth only in the patterned areas.
- ad hoc to paper Filling occurs by capillary action of liquid-like Fe particles during CNT nucleation, as described in the authors' prior paper [4].
Cite this review
Pith. "Pith review of Lithographically defined site-selective growth of Fe filled multi-walled carbon nanotubes using a modified photoresist." pith.science (2026). https://pith.science/paper/VN7QEZGM
@misc{pith2026190802450,
author = {Pith},
title = {Pith review of: Lithographically defined site-selective growth of Fe filled multi-walled carbon nanotubes using a modified photoresist},
year = {2026},
howpublished = {\url{https://pith.science/paper/VN7QEZGM}},
note = {Machine review of arXiv:1908.02450}
}
read the original abstract
Partially Fe filled multi-walled carbon nanotubes (MWCNTs) were grown by chemical vapor deposition with propane at 850 {\deg}C using a simple mixture of iron (III) acetylacetonate (Fe(acac)3) powder and conventional photoresist. Scanning electron microscopy revealed that catalytic nanoparticles with an average diameter of 70 nm are formed on the Si substrate which governs the diameter of the MWCNTs. Transmission electron microscopy shows that the nanotubes have a multi-walled structure with partial Fe filling. A site-selective growth of partially Fe filled MWCNTs is achieved by a simple photolithographic route.
Reference graph
Works this paper leans on
-
[1]
Magnetic force microscopy sensors using iron-filled carbon nanotubes
Winkler A, Mu¨ hl T, Menzel S, Koseva RK, Hampel S, Leonhardt A, et al. Magnetic force microscopy sensors using iron-filled carbon nanotubes. J Appl Phys 2006;99:1049051–5
work page 2006
-
[2]
Kuo CT, Lin CH, Lo AY. Feasibility studies of magnetic particle- embedded carbon nanotubes for perpendicular recording media. Diamond Relat Mater 2003;12:799–805
work page 2003
-
[3]
Iron filled carbon nanotubes for bio-applications
Palen EB. Iron filled carbon nanotubes for bio-applications. Mater Sci (Poland) 2008;26:413–8
work page 2008
-
[4]
Sengupta J, Jacob C. The effect of Fe and Ni catalysts on the growth of multiwalled carbon nanotubes using chemical vapor deposition. J Nanopart Res. 2010;12:457–65
work page 2010
Reviewed August 14, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.