Electron Beam Radiolysis-Assisted Growth of Rutile TiO2 Thin Films

Bharat Jalan; K. Andre Mkhoyan; Nitin Sathish Kumar; Silu Guo; Sreejith Nair; Supriya Ghosh

REVIEW 2 major objections 4 minor 6 references

Electron Beam Radiolysis-Assisted Growth of Rutile TiO2 Thin Films

T0 review · 2 major / 4 minor · reviewed 2026-08-02 · deepseek-v4-flash

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

Pith's one-line read Electron-beam radiolysis can crystallize TiO2 films at substrate temperatures that normally produce amorphous growth.

desk verdict The stripe-crystallization result is real and practically useful, but the radiolysis mechanism is not uniquely established — worth refereeing with a request for a measured beam current or a heating control. read the letter →

arxiv 2607.13851 v1 pith:ISQ5G3TN submitted 2026-07-15 cond-mat.mtrl-sci

Silu Guo , Nitin Sathish Kumar , Sreejith Nair , Supriya Ghosh , Bharat Jalan , K. Andre Mkhoyan This is my paper · ORCID

classification cond-mat.mtrl-sci

keywords electronbeamradiolysisTiO2thinfilmsrutilemolecularepitaxyRHEEDlow-temperaturecrystallizationamorphous-to-crystallinetransformationfilmgrowth

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 sets out to show that electron-beam radiolysis—the beam-driven excitation and rearrangement of atoms, as opposed to knock-on collision damage—can be a constructive tool during thin-film growth. The authors grow rutile TiO2 on TiO2(001) by hybrid molecular-beam epitaxy at substrate temperatures of 100–150 °C, where films normally grow amorphous, while a 14-keV reflection high-energy electron diffraction (RHEED) beam continuously irradiates the surface. In films grown at 130–150 °C, the region struck by the beam is crystalline and the surrounding film is amorphous, with crystallinity increasing toward the beam center and with substrate temperature. The paper concludes that radiolysis supplies enough local energy to crystallize TiO2 at these low temperatures, and that dose and beam energy control the effect. If correct, this adds a low-temperature crystallization mechanism that could apply to other radiolysis-susceptible materials and other growth methods.

What carries the argument

The mechanism is electron-beam radiolysis, quantified by the cross-section of Eq. (1), which grows as the beam energy E0 drops; this is why the 14-keV RHEED beam, a low-energy reflection high-energy electron diffraction beam used for growth monitoring, has roughly ten times the radiolysis cross-section of a 200–300-keV STEM beam. The diagnostic that carries the argument is the degree of crystallinity S defined by Eq. (2), the average integrated intensity of Bragg spots in the Fourier transform of a HAADF-STEM cross-section image, normalized against the same spots in the crystalline substrate and by the zero-frequency spot. S turns atomic-resolution images into a number that tracks electron d

What would settle it

Measure the actual surface temperature rise under the 14-keV RHEED beam with a calibrated sensor during growth, or sweep beam energy at fixed dose; if crystallization tracks the heating profile instead of the inverse-energy radiolysis cross-section of Eq. (1), the central claim fails.

Watch

Extended reading notes

Core claim

The central claim is that when the electron beam irradiates the surface of a growing TiO2 film inside the MBE chamber, radiolysis effects are strong enough to transform otherwise amorphous TiO2 into a crystalline film. The evidence is a set of MBE-grown films: at 130, 140, and 150 °C a roughly 200-µm-wide stripe under the RHEED beam is crystalline while the off-beam film is amorphous; at 100 °C even the beam-exposed area remains amorphous, and at 300 °C the film is crystalline everywhere. In the intermediate-temperature films, the degree of crystallinity rises with electron dose and substrate temperature, following the RHEED beam profile, and a room-temperature STEM beam crystallizes an amor

Load-bearing premise

The load-bearing premise is that the observed crystallization is caused by radiolysis; the experiments do not independently measure or rule out local electron-beam heating, knock-on atom displacement, or beam-induced contamination as alternative ordering forces.

Editorial extensions

If this is right

If the central claim is right, crystalline TiO2 films can be grown at 130–150 °C instead of several hundred degrees, opening up low-thermal-budget substrates.
The dose-dependence of crystallinity means the crystalline quality is graded and controllable in situ, not just a binary amorphous/crystalline outcome.
The spatial confinement of crystallization to the beam footprint implies that crystalline patterns can be written into an amorphous film during growth, without lithography or post-processing.
The authors state the mechanism should transfer to other film-growth techniques and to other materials that undergo radiolysis, broadening the method's reach.
Lowering the electron beam energy directly increases the radiolysis cross-section in Eq. (1), so the temperature or dose required should fall further at lower beam energies.

Reading between the lines

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

Editorial inference: the dose–temperature trade-off implied by the data could be mapped as an equivalence curve, letting growers choose how much thermal budget to trade for beam dose; the paper does not draw this map.
Editorial inference: the same stripe mechanism might enable wafer-scale write-once patterning of crystallinity for oxide devices, but this would require demonstrating that beam writing does not damage already-crystalline areas.
Editorial inference: a decisive test of the mechanism would be a growth-chamber experiment with a tunable low-energy beam at constant absorbed dose; Eq. (1) predicts more crystallization at lower energy, whereas local-heating explanations would not.

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

Bottom line: the stripe-crystallization result is real. Films grown at 130–150 °C are amorphous outside the RHEED footprint and crystalline inside, with crystallinity tracking the beam profile and substrate temperature. The room-temperature STEM dose series adds a second, independent demonstration that electron exposure alone can crystallize amorphous TiO2. That is a genuinely useful observation with immediate practical implications for low-temperature oxide growth.

The paper also does some things right. The cross-section controls are clean: same film, same growth, exposed vs. unexposed regions. The FFT-based crystallinity metric is reasonable, though it has no error bars and the disk-radius choice is not justified in detail. The citation to their own earlier work on TiO2 radiolysis mending cracks (ref 25) is appropriate; it's a published, independent-looking basis, not a circular hook.

The soft spot is the mechanism attribution. The text says the results "directly confirm" radiolysis is the driving force, but the study does not actually separate radiolysis from beam heating or contamination-driven ordering. The Methods admit the RHEED beam current is assumed at 100 nA, not measured; if the real current is 10–100 times higher, local heating could be substantial. The STEM experiment at 200 keV is a partial control — at that energy and in a thin lamella, beam heating is small — but it is not the grazing 14 keV beam on a bulk substrate during growth. The paper never tests the energy-scaling prediction of Eq. (1), which would be the cleanest way to distinguish radiolysis from heating. So the mechanism label is not yet nailed down. That said, the empirical phenomenon does not depend on the label: something about the beam crystallizes the film, with a dose-dependent response.

Minor: the crystallinity vs dose plot could use error bars; the 100 nA assumption should be flagged as a real uncertainty, not a footnote.

Who is this for? Anyone growing oxide films by MBE or PLD who cares about low-temperature crystallization. It deserves refereeing, with the request that the authors either measure the RHEED beam current, add a heating estimate, or run the energy-dependence experiment. I'd accept it into review.

Referee Report

2 major / 4 minor

Summary. The paper reports a new approach to low-temperature thin-film growth in which a RHEED electron beam irradiates the substrate during hybrid MBE growth of TiO2. Films grown at 130–150 °C are crystalline only in the RHEED-beam-exposed stripe, while unexposed regions remain amorphous, and the degree of crystallinity increases with substrate temperature and with proximity to the beam center. A separate room-temperature STEM experiment shows that a 200 keV electron beam can convert an initially amorphous TiO2 film to a crystalline one, with crystallinity increasing with dose. The authors attribute both observations to electron-beam radiolysis and argue that radiolysis can provide the energy needed for crystallization at low substrate temperatures.

Significance. If the radiolysis mechanism is correct, this is a genuinely novel low-temperature crystallization route for MBE and potentially other growth techniques. The internal controls are a clear strength: the unexposed regions of the same 150 °C film remain amorphous while the beam-exposed stripe crystallizes, and the room-temperature STEM dose series directly demonstrates electron-beam-induced amorphous-to-crystalline transformation without substrate heating. The paper also provides a quantitative degree-of-crystallinity metric and shows dose/temperature trends. However, the mechanistic attribution to radiolysis is not fully secured: the RHEED beam current is assumed rather than measured, and no experiment directly discriminates radiolysis from local beam heating. These gaps are load-bearing because the title and conclusion rest on radiolysis as the driving force.

major comments (2)

[Methods (MBE growth) and Results (Fig. 3, Conclusion)] The central attribution to radiolysis is not uniquely established. The Methods states that 'we rely on an assumed beam current of 100 nA for all calculations'; the actual RHEED beam current is not measured. Since 14 keV is below the knock-on threshold, the two plausible beam effects are radiolysis and local heating. The observed crystallization could in principle result from beam-induced temperature rise, especially if the real current is tens to hundreds of microamps. The room-temperature STEM experiment (Fig. 4) shows crystallization at 200 keV where heating is negligible, but the geometry, beam energy, and dose rate differ markedly from the 14 keV grazing-incidence RHEED beam on a bulk substrate during growth. The sentence in the Results that the data are 'directly confirming that indeed the radiolysis is the driving force' is therefore too strong. Please provide a measured beam curre
[Eq. (1), Fig. 1(c), Conclusion] The paper uses Eq. (1) to argue that radiolysis efficiency increases with decreasing beam energy, and concludes that 'the efficiency of this radiolysis-assisted approach can be further improved by lowering electron beam energy.' This is a falsifiable prediction of the radiolysis mechanism, but no experiment varying E0 is reported. A test comparing crystallization under, e.g., 10 keV vs 20 keV RHEED beams would separate radiolysis from heating: radiolysis predicts a larger effect at lower energy, while heating scales approximately with beam power (roughly proportional to E0 at fixed current). Without such a test, the energy-scaling discussion remains speculative and does not strengthen the mechanistic claim.

minor comments (4)

[Eq. (2)] The text describes the degree of crystallinity as the ratio of average Bragg-spot intensities, but Eq. (2) is written as a ratio of sums. Please define n explicitly and clarify that the normalization by the zero-spot intensity is applied to the sums as written.
[Fig. 2(d) caption] The caption says 'High resolution HAADF-STEM images of film cross-sections' but does not state that these images are from the beam-exposed region. Since Fig. 2(a) shows both exposed and unexposed regions, please specify which region is shown to avoid ambiguity.
[References and text] Reference (8) contains a typo: 'Depostion' should be 'Deposition'. There are also several spacing artifacts in the text (e.g., 's how s', 'b y irradiating') that should be cleaned up in the final version.
[Methods (MBE growth)] The phrase 'for all calculations' suggests that quantitative dose calculations are performed, but no numerical RHEED dose values appear in the paper. Either provide the relevant dose estimates or remove the phrase to avoid implying calculations that are not shown.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular reduction; radiolysis attribution is underdetermined but not self-referential.

full rationale

The paper's central observations are direct measurements: RHEED-exposed regions crystallize at 130–150 °C while unexposed regions remain amorphous (Figs. 2 and 3), and room-temperature STEM scanning converts amorphous TiO2 to crystalline with increasing dose (Fig. 4). The crystallinity metric S (Eq. 2) is an image-analysis definition, not a fitted parameter later called a prediction. Eq. (1) is cited from prior literature (ref. 16) and used only to argue that lower beam energy increases radiolysis cross-section; no parameters are fitted to the observed crystallinity and then reused. The only self-citation bearing on the mechanism is ref. 25, the same group's prior TiO2 radiolysis study; it is used as supporting evidence that TiO2 can undergo radiolytic atomic motion, but the present stripe-crystallization and STEM-dose experiments stand independently of that citation, so it is not load-bearing. The Methods limitation—'Due to technical limitations in measuring the beam current, we rely on an assumed beam current of 100 nA for all calculations'—and the lack of a direct heating or knock-on control weaken the mechanistic attribution to radiolysis, but these are external-validity/identification concerns, not circularity. The phrase 'directly confirming that indeed the radiolysis is the driving force behind crystallization' overstates what the controls establish, but it is an interpretive conclusion, not a definitional reduction. No equation in the paper is equivalent to its inputs by construction.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central claim does not introduce new physical entities or fitted parameters that force the result. However, it depends on several domain assumptions: applicability of Eq. (1), the FFT crystallinity proxy, the unexposed-region control, and neglect of knock-on/heating. The assumed RHEED beam current is an unmeasured experimental input to all quantitative dose statements.

free parameters (2)

Assumed RHEED beam current = 100 nA
Used in all dose and cross-section calculations; the authors state the beam current could not be measured and was assumed (Methods). Quantitative dose claims scale with this value.
FFT sampling disk radius = 1 nm real-space equivalent
Chosen by hand to define the integration disk for Bragg and zero spots in the degree-of-crystallinity metric S; different disk sizes would change absolute crystallinity values (Results, Fig. 3).

assumptions (4)

domain assumption The radiolysis cross-section formula (Eq. 1, from Hobbs) with efficiency ζ applies to TiO2 under 14 keV grazing and 200 keV STEM irradiation.
Used to argue that 10–20 keV RHEED beams are roughly 10× more efficient than 200–300 keV STEM beams; not re-derived or measured here.
domain assumption Intensity of Bragg spots in the FFT of HAADF-STEM images is a monotonic, comparable measure of local crystallinity after zero-spot normalization.
Underlies the degree-of-crystallinity metric S (Eq. 2); no calibration against known standards is provided.
domain assumption Regions far from the beam footprint are valid amorphous controls with identical growth conditions.
Used to attribute SEM stripe contrast and STEM crystallinity differences to electron-beam exposure rather than thickness or growth variations.
domain assumption Knock-on damage is negligible compared with radiolysis at the energies and doses used.
The paper states knock-on effects are 'almost always destructive' and assumes radiolysis dominates, but does not measure displacement damage.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Electron Beam Radiolysis-Assisted Growth of Rutile TiO2 Thin Films." pith.science (2026). https://pith.science/paper/ISQ5G3TN

@misc{pith2026260713851,
  author       = {Pith},
  title        = {Pith review of: Electron Beam Radiolysis-Assisted Growth of Rutile TiO2 Thin Films},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ISQ5G3TN}},
  note         = {Machine review of arXiv:2607.13851}
}

read the original abstract

A new approach for growing crystalline thin films is developed that takes advantage of electron beam radiolysis being a constructive force to rearrange atoms into a crystalline structure. It is demonstrated that by irradiating the surface of a TiO2 film by an electron beam supplied by a reflection high energy electron diffraction (RHEED) gun inside the MBE chamber during growth, a crystalline film can be grown at much lower substrate temperatures, where deposited films typically appear amorphous. Here, rutile TiO2 films were grown using hybrid molecular beam epitaxy (MBE) allowing atomic level control of growth as well as an observation of radiolysis-driven crystallization. Analysis was carried out using a combination of SEM and atomic-resolution STEM imaging. It is also shown that by tuning the temperature of the substrate and the dose of the electron beam, the degree of crystallinity of the film can be controlled.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

6 extracted references · 6 canonical work pages

[68]

(15) Du, Y.; Zhang, K

https://doi.org/10.3390/IOCN2023-14481. (15) Du, Y.; Zhang, K. H. L.; Varga, T.; Chambers, S. A. Reflection High -Energy Electron Diffraction Beam-Induced Structural and Property Changes on WO3 Thin Films. Appl. Phys. Lett. 2014, 105 (5), 051606. https://doi.org/10.1063/1.4892810. (16) Hobbs, L. W. Introduction to Analytical Electron Microscopy; Hren J. J...

work page doi:10.3390/iocn2023-14481 2014
[437]

A.; Thomas, M

(17) Ugurlu, O.; Haus, J.; Gunawan, A. A.; Thomas, M. G.; Maheshwari, S.; Tsapatsis, M.; Mkhoyan, K. A. Radiolysis to Knock-on Damage Transition in Zeolites under Electron Beam Irradiation. Phys. Rev. B 2011, 83 (11), 113408. https://doi.org/10.1103/PhysRevB.83.113408. (18) Pascucci, M. R.; Hutchison, J. L.; Hobbs, L. W. The Metamict Transformation in Alp...

work page doi:10.1103/physrevb.83.113408 2011
[1995]

D.; Liu, R.; Chen, Y.; Li, Q.; Leake, S

(9) Marks, S. D.; Liu, R.; Chen, Y.; Li, Q.; Leake, S. J.; Savage, D. E.; Babcock, S. E.; Schülli, T. U.; Evans, P. G. Crystallographic Rotation during Solid -Phase Epitaxy of SrTiO 3 from Nanoscale Seed Crystals. Cryst. Growth Des. 2022, 22 (7), 4043 –4048. https://doi.org/10.1021/acs.cgd.1c01389. (10) Tashman, J. W.; Lee, J. H.; Paik, H.; Moyer, J. A.; ...

work page doi:10.1021/acs.cgd.1c01389 2022
[1998]

J.; Stemmer, S

(29) Jalan, B.; Engel -Herbert, R.; Wright, N. J.; Stemmer, S. Growth of High -Quality SrTiO 3 Films Using a Hybrid Molecular Beam Epitaxy Approach. J. Vac. Sci. Technol. A 2009, 27 (3), 461–464. https://doi.org/10.1116/1.3106610. (30) Jalan, B.; Engel -Herbert, R.; Cagnon, J.; Stemmer, S. Growth Modes in Metal -Organic Molecular Beam Epitaxy of TiO 2 on ...

work page doi:10.1116/1.3106610 2009
[2000]

T.; Thompson, A

(3) Ferguson, I. T.; Thompson, A. G.; Barnett, S. A.; Long, F. H.; Feng, Z. C. Epitaxial Film Growth and Characterization. In Frontiers of Thin Film Technology; Francombe, M. H., Ed.; Thin Films; Elsevier, 2001; Vol. 28, pp 1–69. https://doi.org/10.1016/S1079-4050(01)80017- X. (4) Oke, J. A.; Jen, T.-C. Atomic Layer Deposition and Other Thin Film Depositi...

work page doi:10.1016/s1079-4050(01)80017- 2001
[6005]

(26) Shibata, N.; Goto, A.; Choi, S.-Y.; Mizoguchi, T.; Findlay, S

https://doi.org/10.1038/s41467-023-41781-x. (26) Shibata, N.; Goto, A.; Choi, S.-Y.; Mizoguchi, T.; Findlay, S. D.; Yamamoto, T.; Ikuhara, Y. Direct Imaging of Reconstructed Atoms on TiO 2 (110) Surfaces. Science 2008, 322 (5901), 570–573. https://doi.org/10.1126/science.1165044. (27) Berger, S. D.; Macaulay, J. M.; Brown, L. M. Radiation Damage in TiOx a...

work page doi:10.1038/s41467-023-41781-x 2008

Pith tools

Reviewed August 2, 2026 · model on record in the stance chip above.

[1] [68]

(15) Du, Y.; Zhang, K

https://doi.org/10.3390/IOCN2023-14481. (15) Du, Y.; Zhang, K. H. L.; Varga, T.; Chambers, S. A. Reflection High -Energy Electron Diffraction Beam-Induced Structural and Property Changes on WO3 Thin Films. Appl. Phys. Lett. 2014, 105 (5), 051606. https://doi.org/10.1063/1.4892810. (16) Hobbs, L. W. Introduction to Analytical Electron Microscopy; Hren J. J...

work page doi:10.3390/iocn2023-14481 2014

[2] [437]

A.; Thomas, M

(17) Ugurlu, O.; Haus, J.; Gunawan, A. A.; Thomas, M. G.; Maheshwari, S.; Tsapatsis, M.; Mkhoyan, K. A. Radiolysis to Knock-on Damage Transition in Zeolites under Electron Beam Irradiation. Phys. Rev. B 2011, 83 (11), 113408. https://doi.org/10.1103/PhysRevB.83.113408. (18) Pascucci, M. R.; Hutchison, J. L.; Hobbs, L. W. The Metamict Transformation in Alp...

work page doi:10.1103/physrevb.83.113408 2011

[3] [1995]

D.; Liu, R.; Chen, Y.; Li, Q.; Leake, S

(9) Marks, S. D.; Liu, R.; Chen, Y.; Li, Q.; Leake, S. J.; Savage, D. E.; Babcock, S. E.; Schülli, T. U.; Evans, P. G. Crystallographic Rotation during Solid -Phase Epitaxy of SrTiO 3 from Nanoscale Seed Crystals. Cryst. Growth Des. 2022, 22 (7), 4043 –4048. https://doi.org/10.1021/acs.cgd.1c01389. (10) Tashman, J. W.; Lee, J. H.; Paik, H.; Moyer, J. A.; ...

work page doi:10.1021/acs.cgd.1c01389 2022

[4] [1998]

J.; Stemmer, S

(29) Jalan, B.; Engel -Herbert, R.; Wright, N. J.; Stemmer, S. Growth of High -Quality SrTiO 3 Films Using a Hybrid Molecular Beam Epitaxy Approach. J. Vac. Sci. Technol. A 2009, 27 (3), 461–464. https://doi.org/10.1116/1.3106610. (30) Jalan, B.; Engel -Herbert, R.; Cagnon, J.; Stemmer, S. Growth Modes in Metal -Organic Molecular Beam Epitaxy of TiO 2 on ...

work page doi:10.1116/1.3106610 2009

[5] [2000]

T.; Thompson, A

(3) Ferguson, I. T.; Thompson, A. G.; Barnett, S. A.; Long, F. H.; Feng, Z. C. Epitaxial Film Growth and Characterization. In Frontiers of Thin Film Technology; Francombe, M. H., Ed.; Thin Films; Elsevier, 2001; Vol. 28, pp 1–69. https://doi.org/10.1016/S1079-4050(01)80017- X. (4) Oke, J. A.; Jen, T.-C. Atomic Layer Deposition and Other Thin Film Depositi...

work page doi:10.1016/s1079-4050(01)80017- 2001

[6] [6005]

(26) Shibata, N.; Goto, A.; Choi, S.-Y.; Mizoguchi, T.; Findlay, S

https://doi.org/10.1038/s41467-023-41781-x. (26) Shibata, N.; Goto, A.; Choi, S.-Y.; Mizoguchi, T.; Findlay, S. D.; Yamamoto, T.; Ikuhara, Y. Direct Imaging of Reconstructed Atoms on TiO 2 (110) Surfaces. Science 2008, 322 (5901), 570–573. https://doi.org/10.1126/science.1165044. (27) Berger, S. D.; Macaulay, J. M.; Brown, L. M. Radiation Damage in TiOx a...

work page doi:10.1038/s41467-023-41781-x 2008