The Bimodal Neutron and X-ray Imaging Driven by a Single Electron Linear Accelerator
Round 1
Reviewer 1 Report
The paper describes a facility permitting simultaneous neutron and X-ray imaging.
The application both radiations simultaneously provides additional information about the chemical composition of the object. However, some questions are keeping open:
- The thermal neutron flux is very low compared to the fluxes offered by research reactors or spallation sources. Long illumination times are needed to obtain an acceptable counting statistics and the effect to nois ratio should be lower. Can the authors give information about typical illumination times? What was the total illumination time applied for the two examples?
- The examples was detected with a pixel size of 0.2 mm. Please discuss the lateral resolution limit.
Author Response
We are grateful to the editor and all the reviewers for their effort reviewing our manuscript and making their constructive remarks and useful suggestions, which has significantly improved the quality of the manuscript.
Please see the attachment for the point-by-point responses.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This manuscript describes a very creative approach to use electron-induced radiation for imaging. The authors show that such a source of both neutrons and X-rays can be used as a "mobile" imaging station, possibly for industrial applications (or for small university labs?). It would help the reader if the authors would mention the amount of time required to collect 1 transmission image (pair).
This work is definitely worth publishing, but it also needs to be put in the proper scope of future possibilities (authors suggest industry) and other electron-induced neutron sources in the world.
I noted below several separate questions and remarks while reading the manuscript. Towards the end, I noted more elaborately some things to address in the over-all structure and story line.
________________
Please state an intensity comparison of n and X-rays.
what is the delta lambda photons?
measuring time of fig 6?
attenuation instead of XS should be used, as this work concerns transmission.
XS 30.4@25.4meV --> isn't this 80 barns for H?
noise in grey value over the pixels in one "material" --> influence on material determination --> how much materials could you discern?
grey values are of line integrals over "object voxels", rather than individual voxels
the text says: "Figure 8b and d are the photon image and neutron image for a blade with residual gadolinium tracer (gadolinium oxide powder in this study), respectively, while Figure 8c and e are that for a blade without residual gadolinium tracer, respectively."
while the caption says (e) is neutrons+Gd
What is the use of Gd in turbine blades?
200: for blade without gadolinium
201: Because of the existence of gadolinium tracer,
203: the two curves in which conform to each other and does not show
as stated, beam hardening causes deviations: please quantify these as well. What is the spectral width of the photon/neutron beam?
explain choice of 6 specific regions in fig 9a
figure sub-sub-numbering is confusing.
Discussion:
discussion of why electrons should be moved to Introduction; protons are not the research of this manuscript.
Explain the argumentation of adding non-zero rest mass, taking into account the production of the neutrons and the fact that electrons have a rest mass.
The use of bi-modal capability mostly really suffers from the brightness of the neutron source being so much lower than the photon beam (which the authors also demonstrate implicitly in the quality of the images). Using a moderate-intensity X-ray tube at the brightest neutron sources (spallation/reactors) somewhat compensate for this. The comparable quality of n- and X-images makes data fusion more meaningful, but the authors claim that the orientation of the beam with respect to the object under inspection is the major advantage of their setup.
What is not clear to me in this argumentation, is why a setup of an n-beam at 90deg to a X-beam would not work as well. It would need a simple rotation of the object by 90deg and will allow an optimized camera for either n- or X-ray beam. Surely the spatial resolution of the data shown does not require any advanced positioning or orienting of any of the components.
Can the authors more explicitly elaborate on the advantages, and dis-advantages, of their design.
Conclusions:
where does the environmental footprint come in for industrial applications? Can the authors such industrial applications that would otherwise be prohibit the use of neutrons/x-rays due to environmental footprint?
The fusion of the data seems to be limited to overplotting of the two (n,x) data sets for particular combinations of attenuation coefficients. Although the priciple is shown, the authors do not relate measured attenuation (pair) coefficients to the known values of (for instance) the keys, or the clamp. It would greatly improve the impact of the manuscript if the analysis can be taken one step further by stating the "resolution" in grey-values pairs that can be discerned through this approach. Would the authors be capable of discerning steel from iron, or iron from copper or copper from brass? Such an outlook would largely determine the feasible industrial applications.
Annex:
The ratio of mu_n and mu_p is called Slope, and it would help to clarify which "slope" it is.
The Slope is stated with one mu_n, but the spectrum of neutrons is much broader than the typical 25meV (different from the mean of this facility too).
The Annex discusses the number of materials that could be discerned and the possible use of ToF of the neutron pulse. Both discussions should move the appropriate Section inside the manuscript.
The absorption cross section scales with 1/v for neutrons: how does this enable to discern more materials, as this rule hold for all isotopes (when not in resonance scattering)?
Only in the Annex, the authors mention experiments with 'isotope identification': this should be included in this work in a much more elaborate way (and again; in the Discussion Section).
Author Response
We are grateful to the editor and all the reviewers for their effort reviewing our manuscript and making their constructive remarks and useful suggestions, which has significantly improved the quality of the manuscript.
Please see the attachment for the point-by-point responses.
Author Response File: Author Response.pdf
Reviewer 3 Report
The manuscript presents an interesting approach to combined neutron and Xray imaging.
The manuscript seems to be original and can be in the reviewer's opinion accepted for publication after the following points are addressed:
Title: The reveiwer strongly suggests to replace word 'photon' with 'X-ray'
lines 31-34: About the historical development of the technique, please include the reference to a recent review paper by Lehmann et al (2021).
https://www.mdpi.com/2076-3417/11/9/3825
line 54: '...acquired simultaneously...' well technically, it is not simultaneously. Please rephase ...
line 54: '...free from different beam geometries...'. This is in contradiction to the statements in the chapter 2.3.
line 82: square brackets missing
Figure 1: Is it really so that the sample is placed inside the flight tube (as suggested from the sketch 1a.
Line 106: '...a 2500 n/cm2/s thermal neutron flux at 10 metres away can be anticipated....'
This is a total neutron flux. Has this been measured? Or is this based on a simulation only? How exactly is thermal range defined here?
The reviewer suggests -for the readers comfort- to specify this used thermal region in Figure 3 (spectra) and add the number of 2500 n/cm2/s into the Figure as well.
Chapter 2.2. Is there any overlap of cold neutron signal with the X-ray signal? The reviewer supposes that this is negligable, but please address this issue here.
Line 133-136: This is not quite clear. Is there a beam defining aperture in the system? Can collimation (L/D) ratio be specified?
Line 139-142: This sentence needs better (more detailed) explanation.
Figure 5: Missing units for the axes.
Line 158: Provide the reference for the 1H cross-section.
In general: All the images in the Figures are missing scalebars and colorbars. Also apart from the Figure 10, the total acquisition times have to be specified clearly.
Linked with this issue of the total acquisitions times the authors should discuss the potential of this technique for the investigations of fast processes (both repetitive DOI: 10.4271/2010-32-0013 and non-repetitive doi: 10.1016/j.mex.2016.10.001). What spatial resolution is expected to be achieved from their detection system for such processes?
Figure 7 : The reviewer trusts that Figure 7c, does not bring much and can be removed (including the corresponding text) from the manuscript.
Conclusions:
Line 254: 'one-source-one-detector' system
Author Response
We are grateful to the editor and all the reviewers for their effort reviewing our manuscript and making their constructive remarks and useful suggestions, which has significantly improved the quality of the manuscript.
Please see the attachment for the point-by-point responses.
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
The reveiwer thanks the authors for the quick address of the raised point
The reviewer strongly suggest the inclusion of the author's discussion to the Point 14 in answer to the reviewers (with improvement of the English language editing) into the manuscript. In other words, the readers should receive the clear message that for the fast repetitive processes this development might be useful (even though a rather long total acquisition times will be needed, while for non-repetitive fast processes this methodology is not applicable.
The reviewer agrees with keeping the Figure 7c in the manuscript.
Author Response
Dear Editors and Reviewers,
Thank you for your effort reviewing our manuscript.
The manuscript has been revised based on the advice of reviewer 3.
Please see the attachment for more details.
Yigang Yang.
Author Response File: Author Response.pdf
