arfc/msbr-reproc

holds a paper describing an MSBR online reprocessing demonstration.

2018-msbr-reproc

holds a paper describing an MSBR online reprocessing demonstration.

msbr-reproc's Issues

In section 3.3 ....

I'm not sure whether this sentence is necessary.

Thus, a significant amount fissile material will be needed to make the reactor critical.

Review Changes

Review all of the changes by looking over this pdf:
msbr-reproc-diff.pdf

• Andrei review complete
• Katy review complete
• Teddy review complete

( @jbae11 , if you don't have time to look this over, that's fine, I just want to make sure you're ok with the final submission since your name is on it. We're submitting tomorrow.)

Comments - Mark

Comments in document attached.
andre-msbr-reproc - kamuda reviewed.pdf

Comments - Gwen

Cool Paper Andrei! Here are my comments (my comments only start from page 18 onwards):
msbr-reproc-gwenedits.pdf

Figure captions say "both" but only one graph is shown

I just noticed: Figures 13 and 14 say :

• Figure 13: Normalized power density for both initial and equilibrium fuel salt composition.
• Figure 14: 232Th neutron capture reaction rate normalized by total flux for both initial and equilibrium fuel salt composition.

But, only one plot is shown for each. Maybe two plots were intended??

Abstract should be restructured

The abstract is somewhat confusing, I think because it doesn't follow the structure I usually expect. Here are two possible approaches to structuring an abstract:

http://s3-service-broker-live-19ea8b98-4d41-4cb4-be4c-d68f4963b7dd.s3.amazonaws.com/uploads/ckeditor/attachments/7808/2c_Summary_para.pdf

http://mitcommlab.mit.edu/be/commkit/journal-article-abstract/

From the MIT one, we should fit the following into 5-8 sentences:

General Background. Something that everyone in your audience cares about. (e.g. climate change, safety, licensing molten salt reactors)
Specific Background. Zoom in from the thing everyone cares about to the thing you did. (i.e. people need tools for analyzing liquid fueled molten salt reactor depletion and fuel processing)
Statement of Problem or Knowledge Gap. What specific problem or phenomenon do we not understand in this field of study? (i.e. needed a high fidelity wrapper for serpent)
Here we show. One sentence about what you learned or did, and how that fulfills the demonstrated gap. (i.e. describe saltproc)
Approach & Results. Only the very highest-level methodology results. So what? What do your results mean for the thing everyone cares about? (i.e. SaltProc was demonstrated for the MSBR and compared well to previous results from lower fidelity analyses.)

First Methods paragraph

The first methods paragraph is a little redundant (first sentence has already been stated in the paper). Ideally, this paragraph should just prepare the reader for the following subsections.

This might have some useful insights.

http://mitcommlab.mit.edu/be/commkit/journal-article-methods/

Read review response, editing for kindness

Make sure the review response reads nicely (and not disparaging toward SERPENT2 extension).
The current pdf is here:
review-responses.pdf

This issue can be closed when all authors approve the state of the review response.

• @andrewryh approves
• @katyhuff approves
• @jbae11 approves

Final comments after acceptance

Reviewer #3: In this paper, a SaltProc-Serpent2 code system was developed to simulate the online reprocessing of the Molten Salt Breeder Reactor (MSBR). However, a lot of similar works have been published. This work needs to be improved further to show the difference and innovation.

• 1. In Page 7, the differences between this work and Park et al.'s work were minor, and the uniqueness of this work is doubtable.
• 2. In Page 6, the comments on Aufiero's work were not reasonable. The following sentence "only two publications using these capabilities are available, reflecting the reproducibility challenge inherent in this feature" was not appropriate in this kind of scientific research article.
• 3. Page 6, Figure 11, please explain how the normalization of neutron spectrum was performed. The neutron spectrums were different from those in Park et al.'s work, especially in the fast neutron range. Could the authors explain the differences? The caption of figures should not be too long.
• 4. Page 45, Figure 13, only flux distribution at initial state was given.

Reviewer 2: minor comments

Abstract

• Considering the scope of the journal, mentioning the names of companies and start-up is not appropriate. I would suggest removing them.

Introduction

• The sentence "Immediate advantages over traditional, solid-fueled, reactors include near-atmospheric pressure in the 15 primary loop, relatively high coolant temperature, outstanding neutron economy, improved safety parameters," may suggest improved neutron economy vs solid-fuel fast reactors. This is rarely the case, especially vs Pu-based SFRs. I would suggest reformulating.

• The sentence "With regard to the nuclear fuel cycle, the thorium cycle produces a reduced quantity of plutonium and minor actinides (MAs) compared to the traditional uranium fuel cycle" is correct, but the fact that this is an advantage is questionable. The pros&cons of thorium cycle have been long debated and there is no consensus on its advantage in terms of exposure of workers, exposure of public, geological repository, etc. I would suggest removing the sentence.

• "Methods listed in references [14, 17, 24, 25, 28, 29, 30] as well as the current work also employ a batch-wise approach". As a matter of fact, the work in [14] allows for continuous reprocessing via introduction of "reprocessing" time constants. The work from Aufiero (mentioned in the following paragraph) actually used the methodology previously developed in [14] for verification purposes.

• Table 3. It is not clear what "effective cycle times" are. Please clarify.

• The removal of fission products is made batch wise in the described algorithms. However it is not clear how the fission products with the longest "effective cycle times" are removed. Part of them at every batch? Or all of them at the end of the "effective cycle times". Please clarify. And please clarify the relation between "effective cycle times", batches and the average time spent by a fission product in the reactor.

• The oscillatory behavior shown in Fig. 7 on the time scale of months/years is hard to explain. Is this because the fission product with longer residence time are batch-wise removed at their "effective cycle time"? In case, why not to remove part of them at every depletion step?

• "This thorium consumption rate is in good agreement with a recent online reprocessing study by ORNL [29]." Please notice that in a reactor with only Th as feed, and near equilibrium, the Th consumption rate is exclusively determined by the reactor power and by the energy released per fission.

• Are you considering the effect of the gradual poisoning of the graphite with fission products? If not, it would be worthwhile briefly discussing its effect.

• "The current work results show that a true equilibrium composition cannot exist but balance
  between strong absorber accumulation and new fissile material production can be achieved to keep the reactor critical." Not clear. Do you mean that the equilibrium composition cannot be achieved in a lifetime of the reactor? Please clarify

Review Andrei SaltProc Paper

Hi Folks,

This week, @andrewryh has kindly agreed to allow us to provide feedback on his paper in this repository. It will be submitted as a journal article (probably to Annals of Nuclear Energy?) pending this final round of revisions.

Here is the current pdf that you can review:
msbr-reproc.pdf

If possible, please review by September 5.

Once you're done reading this paper:

1. open a new issue in this repository summarizing your review and suggestions (perhaps as a marked up pdf, a list of questions, a review report, or similar)
2. check your name off of the list below.

• Mark
• Teddy
• Sun Myung
• Ansh
• Greg
• Gwen

Reviewer #1: minor comments

• Abstract: Given the recent announcement about TransAtomic Power, you may want to remove them from your list?
• Table 1, page 3: SCALE/TRITON is fast as well as thermal reactor capable.
• Page 4, Line 48: Should read "...and refill using a single or multiple unit cell ...."
• Page 4, Line 55. Also worth noting that the latest SCALE release will have the same functionality using continuous removal (B. R. Betzler, J. J. Powers, N. R. Brown, and B. T. Rearden, "Molten Salt Reactor Neutronics Tools in SCALE," Proc. M&C 2017 - International Conference on Mathematics & Computational Methods Applied to Nuclear Science and Engineering, Jeju, Korea, Apr. 16-20 (2017).)
• Page 8, Fig 2: Appears the image is cut off at the top?
• Page 13, line 209: The description of the Pa removal, although correct, isn't quite fully correct. The reason the Pa is removed from the core and hence flux is to then enable the Pa to decay to U233. If it was left in the core, it would transmute further and hence not be able to produce the U233 that is necessary for this breeding cycle to work.
• Table 3: "cycle time" is not defined in the paper. Please add.
• Page 14, line 224. The 3 day time step as the "optimum" for Th fuel cycles in an MSR was first described and concluded by Powers et al. Please add a reference to their initial work.
• Page 18, line 327: Are those elements removed every 3435 days, or is it that the entire salt core is discharged?
• Page 19, Figure 8 (and same for Fig 9): y-axis in grams/kgs or mass units would be better for the reader.
• Page 20, Fig 9: What are the wiggles and dips , especially seen for Np235?
• Page 20, line 351: It is more than just the Pu isotopes that makes the spectrum harder? What about the other MAs etc?
• Fig 12: units on y-axis?
• Page 24, line 389: Should that be "233U" and not "233Th"?
• Table 5: Please provide some comments on the uncertainties - where do they come from? Also, the "reference" results need to state whether "initial" or "equilibrium"
• Page 26, line 425: "Relatively large" compared with what? Perhaps results for an LWR case would be a good basis for comparison?
• Page 27, section 3.8: It needs to be made more clear that these results were calculated, and that they are taken from the code output.

Reviewer #3 minor comments

• 1. What are the main differences of this work with the previous works, especially with the work published by Park at al. in "Whole core analysis of molten salt breeder reactor with online fuel reprocessing"?
• 2. How did the authors verify the coupled SaltProc and Serpent code system?
• 3. In Page 3, the title of "Table 1" should not only contain the "fast spectrum system". The work published by Zhou and Yang et al. with the title of "Fuel cycle analysis of molten salt reactors based on coupled neutronics and thermal-hydraulics calculations" needs to be included in Table 1 for completeness.
• 5. In Page 10, "SERPENT generates the problem-dependent nuclear data library", how did SERPENT generate the problem-dependent nuclear data library? What kind of nuclear data library did SERPENT generate?
• 6. In Page 14, "depletioncalculations" should be "depletion calculations".
• 7. In Page 21, it looks like the neutron spectrum is not normalized in Figure 10. It is recommended to normalize the neutron spectrum for comparison.
• 8. In Page 22, "Figure 13 reflects the normalized power distribution of the MSBR quarter core, which is the same at both the initial and equilibrium states" contradicts the following statement of "The spectral shift during reactor operation results in different power fractions at startup and equilibrium".
• 9. In Page 24, it is hard to agree with the statement that "the majority of 233Th is produced in zone II.". How did the authors draw this conclusion?
• 10. In Page 24 and 25, why did the normalized power density distribution and the 232Th neutron capture reaction rate distribution share the same figure for both initial and equilibrium fuel salt compositions?
• 11. Too much background information was contained in the abstract, which deteriorates the readability of the abstract.

Reviewer 2 (comments related with Reviewer 3):

• Can the proposed tool adjust reactivity?

Review - Park

Here are my comments. Good luck!

msbr-reproc-review-park.pdf

Confusing

I think the opposite of this statement is true.

When the graphite temperature increases, the density of graphite decreases creating additional space for fuel salt.

I believe this is true:

When the graphite temperature decreases, the density of graphite decreases creating additional space for fuel salt.

Table 1

I'm not sure about the use of Table 1. The distinction between "neutronic code" and "depletion code" isn't particularly clear. Instead of column 1, perhaps we could just refer to the works by their refs.

Confusing lit review words. Please clarify.

In lines 40-60, I'm confused about which methods employ batchwise and which employ a continous approach.

I think it's appropriate to add the word also to the paragraph below, to be more clear:

ORNL researchers have developed ChemTriton, a Python-based script for SCALE/TRITON which uses the batch-wise approach to simulate a continuous reprocessing. ChemTriton models salt treatment, separations, discharge, and refill using a unit-cell MSR SCALE/TRITON depletion simulation over small time steps to simulate continuous reprocessing and deplete the fuel salt [27]. Methods listed in references [14, 17, 24, 25, 28, 29, 30] as well as the current work also employ a batch-wise approach.

Similarly, I suspect the word "Meanwhile" in the following paragraph is misleading. Meanwhile is usually used to indicate a contrast between two similar things. For example "Jenny was an exceptional student. Meanwhile, her twin sister Jessy was always in trouble. "

Accounting for continuous removal or addition presents a greater challenge 55 since it requires adding a term to the Bateman equations. In SCALE [26], ORIGEN [16] solves a set of Bateman equations using spectrum-averaged fluxes and cross sections generated from a deterministic transport calculation. Meanwhile, approaches listed in references [11, 12, 19, 31] model true continuous feeds and removals.

In the above paragraph, are you trying to indicate that the methods in [11, 12, 19, 31] use this SCALE/ORIGEN capability, or that they use some other strategy and thereby acheive continuous feeds and removals much like this SCALE/ORIGEN capability?

K-eff and Th refill graphs are hard to read due to huge swings (Reviewer #1)

Discussion needed to resolve these comments:

• Page 18, Figure 7: The figure is hard to interpret or see clearly what is going on. Could an additional figure or a zoomed in portion be added to show what the swing in k is over a much shorter time interval? k seems to be swinging dramatically but over what time period and how would this be controlled in reality? The graph almost suggests that the core is unstable??
• Page 29, Figure 15: Similar comment to above regarding Fig 7 - the results are difficult to see and interpret with such notable swings.

Explain why we don't use SERPENT2 built-in reprocessing feature

Reviewer 1:

• Page 14, line 234 onwards: Doesn't SERPENT already have an MSR removal capability? If so, what is different about using SaltProc with SERPENT?

Reviewer 2:

• The most critical point of the work is that the built-in capabilities for online reprocessing of Serpent 2 have not been used. Their use is mentioned in the future work, but it is not clear why these capabilities have not been used in the current work. To the author's knowledge, they have been available in Serpent 2 since quite a while. The authors should clarify this point at the beginning of the paper, and not only in the "Future work" section. Even though the technical work was done without using these capabilities, they should highlight what SaltProc adds to the built-in Serpent capabilities, and they should at least try to extrapolate on the potential advantages of combining SaltProc and Serpent capabilities. Based on this, they should slightly restructure the paper in order to prove the claimed advantages over Serpent 2.

Reviewer 3:

• 4. In Page 5, the following sentence needs to be explained. "We employed this extended SERPENT 2 for a simplified unit-cell geometry of thermal spectrum thorium-fueled MSBR and obtained results which contradict existing MSBR depletion simulations."

Comments - Ansh

msbr-reproc-edited.pdf

Due to @andrewryh 's request for speeding up the review process, @gtw2 and I are reviewing the first half. Refer to @smpark7 and @gwenchee 's issues for comments on the second half.

Paragraph is confusing

The paragraph beginning...

Liquid-fueled MSR designs have online separations and/or feeds

...is pretty confusing. In particular, the order is a bit odd and it's not obvious which methods are which (without repeatedly referring back to the table.) Probably some of this could be rewritten in a more traditional literature review style.

Temperature feedback: add row with density and Doppler in Table 5

Reviewer 2:

• "The main physical principle underlying the reactor temperature feedback is an expansion of material that is heated". This sentence would deserve some support data. Can you please calculate separately the effect of temperature (Doppler in fuel and spectral shift in graphite) and density?
• How uncertainties have been calculated in Table 5? Is this just statistical uncertainty from Serpent calculations? Please clarify
• For calculating the coefficients in table 5, are you only changing densities, or also dimensions? Please clarify and provide a justification.
• "The fuel temperature coefficient (FTC) is negative for both initial and equilibrium fuel compositions due to thermal Doppler broadening of the resonance capture cross sections in the thorium." What is the effect of density?

Reviewer 2: numerical approximations vs physics assumptions

In the manuscript it is not always clear when the authors refer to numerical approximations of physical situations. For instance, the authors write "Figure 16 demonstrates that batch-wise removal of strong absorbers every 3 days did not necessarily leads to fluctuation in results but rare earth elements 480 removal every 50 days causes an approximately 600 pcm jump in reactivity." These 600 pcm are an effect of a numerical approximation, but the way things are presented can be confusing to the reader. Please try to explicitly separate physical and numerical effects. And try to related numerical effects to physical effects. For instance, how does these numerical "jumps" affect the results? In this sense, why the batch-wise removal of strong absorbers every 3 days was not done? And why a fraction of rare earths is not removed every three days?

I totally can't understand what Reviewer 2 wanna from me. Completely

Liquid-fueled MSBR design paragraph

The paragraph begining "Liquid-fueled MSBR design in this work" seems quite out of place. Is it necessary? Maybe it can be deleted?

Section Methodology ---> SaltProc description to write

Tentative structure of Methods section:

1. MSBR design description and challenges
2. SERPENT2 model of MSBR
3. Reprocessing method

• Fuel material flows
• Online separations and feeds
  - The SaltProc tool description

Use my thesis, Section 4.3 but make it up-to-date.

Final sentences

The final sentences of the conclusions section seem like too much detail and not totally supported by the work.

The existing full-core SERPENT 2 model and equilibrium fuel material composition would be
employed to generate problem-oriented nuclear data libraries for further usage in accident
transient analysis. The final goal of this effort is to develop a fast-running computational model
which could study the dynamic behavior of generic MSRs, performing detailed safety analysis
and design optimization for a variety of reactor concepts.

Comments Teddy

Great Job Andrei!
msbr-reproc.pdf

Figure 16 caption

The caption of fig 16 says " The confidence interval ±σ is shaded. " Is that really true? I don't see shading.

Confusing statement

...zone I produced much more thermal neutrons than zone II, indicating that the majority of fissions occured in the central part of the core...

Thermal neutrons cause fission, not the other way around. Fissions cause fast neutrons. Moderation causes thermal neutrons. Yes, fissions and thermal neutrons must co-exist, but perhaps it's clearer to say :

...zone I produced more thermal neutrons than zone II, corresponding to a majority of fissions occuring in the central part of the core...

normalization explanation

I'm sufficiently confused by the following sentence that I can't reword it: "Figure 8 shows major nuclides which have strong influence on the reactor core physics normalized separately for each isotope by average atomic density, at the beginning of each depletion timestep."

Graphical Abstract can only contain one image

The graphical abstract guidelines are thus:

Although a graphical abstract is optional, its use is encouraged as it draws more attention to the online article. The graphical abstract should summarize the contents of the article in a concise, pictorial form designed to capture the attention of a wide readership. Graphical abstracts should be submitted as a separate file in the online submission system. Image size: Please provide an image with a minimum of 531 × 1328 pixels (h × w) or proportionally more. The image should be readable at a size of 5 × 13 cm using a regular screen resolution of 96 dpi. Preferred file types: TIFF, EPS, PDF or MS Office files. You can view Example Graphical Abstracts on our information site. Authors can make use of Elsevier's Illustration Services to ensure the best presentation of their images and in accordance with all technical requirements.

The graphical abstract currently includes all the figures. Please pick just one that is representative of the article.