the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 14 Feb 2024
Technical note: a new analytical protocol for apatite (U-Th)/He and trace element analysis (incorporating a continuous ramped heating measurement system for the He)
Abstract. This study details the procedures developed between 2021 and 2023 at the Géosciences Rennes lab (GeOHeLiS plateform, Rennes University, France) for obtaining precise (U–Th)/He thermochronological data on apatite. The methodology involves crystal selection, measuring helium (4He) content by heating crystals, and analyzing them using noble gas mass spectrometry. Additionally, determining U, Th, and Sm contents is achieved through crystal dissolution and subsequent solution analysis with triple-quadrupole mass spectrometry (QQQ-MS). Emphasis is placed on a new approach to quadrupole 4He sensitivity calibration based on Durango apatite, ensuring a precision of ~3.9 % in 4He content determination. The study also highlights the development of a "ramped heating – direct analysis" protocol used to screen apatite diffusion behaviour. In addition, a new quantification protocol for major and trace elements measured in apatite is presented, employing a standard concentration range approach rather than the isotopic spiking technique commonly used for (U-Th)/He studies.
This preprint has been withdrawn.
-
Withdrawal notice
This preprint has been withdrawn.
-
Preprint
(2087 KB)
-
Supplement
(1916 KB)
-
This preprint has been withdrawn.
- Preprint
(2087 KB) - Metadata XML
-
Supplement
(1916 KB) - BibTeX
- EndNote
Interactive discussion
Status: closed
-
AC1: 'Comment on gchron-2024-6', Alexis Derycke, 20 Feb 2024
Authors apologise for the initial omission of the tables and point out that the tables are now available as a supplement.
Citation: https://doi.org/10.5194/gchron-2024-6-AC1 -
AC2: 'Comment on gchron-2024-6', Alexis Derycke, 21 Feb 2024
The editing process revealed some missing parts in the supplement, which has therefore been updated as follow :
- Sup. Mat. 1: chemical data (used to produce Fig. 5)
- Sup. Mat. 2: Durango age and ThU-SmTh ratio (used to produce Fig. 8)
- Sup. Mat. 3: raw data from standards (e.g. He, U, Th, Sm and Ca content)
- Sup. Mat. 4: ramped heating raw data (e.g. He concentrations, temperature, degassed fraction and Ln(D/a²))
- Sup. Mat. 5: automated data reduction file for (U-Th)/He analysisAuthors apologize for the initial oversight.
Citation: https://doi.org/10.5194/gchron-2024-6-AC2 -
RC1: 'Comment on gchron-2024-6', Anonymous Referee #1, 29 Feb 2024
Please refert to the attached PDF file for my review
-
AC3: 'Final response to the reviewers', Alexis Derycke, 24 May 2024
The manuscript Gchron 2024-6 was reviewed by four different reviewers (R2: James Metcalf, R4: Bruce Idleman and R1-3 anonymous), who are greatly thanked for the time they took to do so and for the extensive comments and recommendations they provided. These reviews contain similar or related comments, and in the following we will attempt to address and clarify those general comments.
Please find attached to this comment a .pdf files including the general responses.
To supplement this overarching response, if the manuscript is accepted for review, individual and precise replies will be produced as responses to each review on the Gchron 2024-6.
-
AC3: 'Final response to the reviewers', Alexis Derycke, 24 May 2024
-
RC2: 'Comment on gchron-2024-6', James Metcalf, 12 Mar 2024
This technical note presents specific details and results of how the GeOHeLiS (U-Th)/He lab at Rennes University operates. As a (U-Th)/He lab manager I found this paper very interesting, and am very supportive of it, and other technical descriptions of analytical set ups, appearing as technical notes in EGU Geochronology. This lab specifically seems to be targeting three larger challenges: 1) how to calibrate a noble gas QMS, 2) how to determine total U, Th, and Sm without expensive and difficult to obtain spike materials, and 3) how to execute ramped heating of geologic materials.
In general I think the paper is well written and informative. I believe though that there are, however, areas where the contribution would be greatly improved with some additional clarifications, discussions, and modifications.
The paper would benefit from a re-organization, I often found myself flipping back and forth to look at figures from earlier in the paper. One strategy might be to handle all of the He measurement things first, then move on to dissolution. Some parts, like grain selection and measurement, are relatively standard and can probably be shortened.
Importantly, I think that because the authors are presenting, in some cases, alternative ways to calibrate He, U, Th, and Sm measurements, that the paper would benefit greatly from data collected more traditional ways. On the same solutions, for example, how does this method compare to isotope dilution? Same with using Durango as the 4He calibration source, it seems like a direct comparison would be much more informative for people considering building a lab.
I think most of the figures are in good shape, although I have noted some places where items are not explained or noted.
I believe my concerns would fall somewhere between major and minor revisions, and I strongly encourage the authors to make the modifications (I would really like to see lab descriptions as detailed as this appearing regularly).
General Comments
It seems like one motivating factor for this machine design is to not have a 4He tank or pipette on the line and instead use Durango shards as the 4He calibration material. While calibrating pipettes and tanks is certainly difficult and time consuming, I think the paper would benefit from a discussion of the rationale behind excluding this more common method of QMS calibration. I understand the logic of trying to avoid using spikes for U, Th, and Sm since they can be hard to obtain, however I don’t understand the motivation for not having a 4He tank and pipette on the line.
This is especially important considering the very high 4He measurement uncertainties in the data supplied in supplementary materials. Looking at the Durango data you provided in Supp. Mat 3, it seems like the uncertainties and Durango ages for grains < 10 ug are much more scattered. The data seems to imply that the uncertainties at least are the result of high 4He uncertainties. The total 4He values don’t seem too low, can you comment on how well this method handles small amounts of gas? Are those uncertainties (some over 100%) because of the calibration method or were these specific measurements due to blank or other problems? Also, is it common for the small Durango shards used in the calibrations to have these kinds of uncertainties?
Specific comments and questions
• There are a number of places where items are described as “computer controlled thanks to LabView assets provided by manufacturers.” Do all of these vi’s and sub-vi’s operate within an in-house piece of code?
• Where in the laser process is the electric lens? I don’t see it in Figure 1B.
• For a technical note there are a few things that would be interesting (as a fellow lab manager) to know
o What type of filament and settings (emission, SEM voltage) do you run on the QMG and did you do any testing to arrive at the best settings?
o How did you determine the volumes of your pipette and 3He tank?
o What is the internal volume of your line?
• When filling the 3He tank you say the target pressure is 1E-6 mbar, this is at the low end of capacitance manometer detection so I assume you are backfilling an empty tank with a higher pressure trapped and measured in the pipette?
• Line 107 - The Ie (ionization efficiency term) is adjusted for each run, is it monitored throughout a run? With a pan holding as many samples as yours does it seems like some changes might happen during the time of a run.
• Line 115 – do you consider uncertainties in 4He blanks in the calculations?
• Are S and Ie the same for really really large samples and small samples? Are these effects linear?
• Line 151 – how long does the heating and pumping typically take? Is this on the turbo pump only or are you gettering the volume?
• Figure 2 – Is the top dark red line the 4He or is it the cold head temp over time? I find this a little difficult to read.
• I think the description of how the ramped heating measurements are accomplished it really excellent.
• Line 226 – is the process that after the hot plate the samples are diluted so the total volume is 2 or 10 mL? Is this done assuming none of the original acid has evaporated (meaning you just pipette in the right amount), or are you filling to a mark on the individual vials? I guess I am curious about the problems of evaporation on hot plates mostly.
• For measuring the Ca – my experience is that keeping Ca and Th in solution together can require some HF, and my attempts to do stoichiometric apatite volume determinations has not been overly reliable. Do you have external checks (micro CT for example) for the volume? For samples obviously this is important but also for the Durangoes used for the initial He calibration, are all of the volumes stoichiometric?
• Line 261 – Are your Nb tubes also contaminated to this degree? We’ve switched over to using Nb for everything because they are more consistently and reliably uncontaminated.
• Figure 7B – since it is a log scale and only U and Th are really important for age reliability, can that part be blown up? It looks like the QQQ line is well below the others for Th?
• Section 4.1.3 – I don’t quite follow the argument for why the linear relationship between Th/U and Sm/Th isn’t a chemical issue. You state that it hasn’t been reported before but then conclude that it is a real thing, do you mean just for this crystal? Could this be checked by isotope dilution?
• Section 4.2.1 – are the Durangoes used for the calibration dissolved and analyzed for U, Th, and Sm? And then the volumes are all stoichiometric?
• Figure 9 – what is encompassed in the y-axis uncertainties in figure 9A?
• Figure 9B – what are the green sections labeled “R.H.”?
• Line 345 – how small of a shard is used? Is it similar in volume to a typical apatite?
• Line 440 – I think this point warrants more discussion, especially since this relationship isn’t apparent in all Durango datasets.
Minor Line Edits• Line 83 – should be SAES getters
• Line 111 – should be influences not influence
• Line 135 – I might use “pure apatite separate” instead of “powder” here, to me powder implies the grains are completely pulverized.
• Line 157 – no need to capitalize apatite
• Line 172 – it’s minor but apatites should be degassed after one heating step, and the second should be just blank level, right?
• Line 232 – should read “checked”
• Line 310 – should be “Crystals” and not “crystal”
• Line 390 – could the end trend and the global trend be labeled on figure 4?James R. Metcalf
Citation: https://doi.org/10.5194/gchron-2024-6-RC2 -
AC3: 'Final response to the reviewers', Alexis Derycke, 24 May 2024
The manuscript Gchron 2024-6 was reviewed by four different reviewers (R2: James Metcalf, R4: Bruce Idleman and R1-3 anonymous), who are greatly thanked for the time they took to do so and for the extensive comments and recommendations they provided. These reviews contain similar or related comments, and in the following we will attempt to address and clarify those general comments.
Please find attached to this comment a .pdf files including the general responses.
To supplement this overarching response, if the manuscript is accepted for review, individual and precise replies will be produced as responses to each review on the Gchron 2024-6.
-
AC3: 'Final response to the reviewers', Alexis Derycke, 24 May 2024
-
RC3: 'Comment on gchron-2024-6', Anonymous Referee #3, 13 Mar 2024
The technical note by Derycke et al. describes their newly built He-line setup and analytical protocols for apatite (U-Th-Sm)/He (AHe) analyses, including a ramped heating protocol (developed during 2021-2023 at the Géosciences Rennes lab, France). The main changes compared to previous approaches are their calibration of quadrupole sensitivity (simplified from Gautheron et al. (2021)) and the “standard concentration range approach” for trace element measurements.
I appreciate the authors' efforts to set up the presented protocols and share this technical report. However, I cannot support the publication in its current form. My main issues relate to the organisation of the manuscript, missing details of some descriptions, and confusing or possibly inconsistent presentation of data. See below for general issues and line- and section-specific comments and suggestions. I hope these help to improve the manuscript and that it will be published after a thorough revision.
Inconsistency of results and data presentation:
The Abstract talks of ~3.9% “precision in 4He content determination”, Section 4.2.2 mentions ~3.9% dispersion of Durango ages (but not for all of the ages? I'm not sure; this part is confusing), and the Conclusions talk of <7% age error. This makes it difficult to follow where the numbers come from and whether the terminology is correctly used in different places. Are the 3.9% from the Abstract the same as in Section 4.2.2? Could you not use the same type of uncertainty when describing it?
The presentation of data in the supplementary materials and the descriptions in the text of how many samples were used and for what were somewhat confusing. One reason may be that the supplementary materials are not referred to in the main text, and the supplementary tables have no proper titles and captions. From the tables, text, and your short note in the GChron discussion forum on what the supplementary tables are, I cannot reconstruct which data are shown and why, and which data are omitted from tables and figures and why. So, what is actually shown in the tables? What are the numbers 34, 20, and 45 Durango fragments in the text, what are they used for (calibration vs test measurement? ramped heating?), and which data are shown in the figures? And which are not shown but are shown in some tables (but not in others)? And how many MK-1 fragments are there? 10 or 15? Can’t you show all data together in fewer tables and indicate which data might be omitted from plots, central age calculations, … and why? Either it’s inconsistent or confusing; I'm not sure.
Structure:
The structure of the manuscript is somewhat confusing. Should Sections 2-3 be combined under a Methods section? In addition to my comment on l.25-34, I suggest adding a paragraph at the end of the Introduction to help guide the reader. Furthermore, it would be helpful to have the same (similar) subsection headings in Methods and Results so that one knows where to find the results (like you have it for Sections 3.2.2, 3.2.3 and 4.3 concerning ramped heating).
It might also be good to separate Results and Discussion. The Discussion should include a more condensed text on what has been improved with your protocols, which aspects need more work, etc. I know some places in the text address these points, but having them in one place with corresponding subheadings would be helpful.
Language and style:
Please check the text for English grammar, the use of articles, and word choice. Sometimes, words or parts of words are missing. There are issues in many places that I will not point out individually at this point, as I think the text may change during revision. Also, check punctuation. Carefully check the figures and tables and consistently use the English language and style.
Title: Try to rephrase the title without using parentheses. Delete “for the He.”
Standard style guides suggest writing out numbers up to ten. You keep switching between doing and not doing this. Please be consistent in your writing style.
Add spaces between values and their units throughout the text.
Do not use bullet points in your text (Conclusions). You may number your main results but integrate them into the continuous text.
Make sure to add proper titles and captions to tables in the supplementary materials (as you would for the main text). Please also explain, in captions, how uncertainties were determined (in your and previous studies). Furthermore, every supplementary material should be referred to in the main text. You currently don’t refer to any.
Provide a separate reference list for the supplementary materials.
I suggest changing “ages” to “AHe ages” in tables and figures (including captions) and throughout the text.
Be consistent with how you write MK1 or MK-1.
Line-specific comments:
l.25-34. The introduction could be more helpful if you’d describe current analytical protocols, challenges, and aspects needing improvement. I don’t think an introduction to the method’s history is necessary. You only allude to “recent technical development” (l.34-35). I suggest focusing on this part so it becomes clearer from the beginning what your new contribution is.
l.43. “QQQ-MS” is introduced as an abbreviation in the abstract, but I suggest doing it again in the first section.
l.44. Write out VBA.
l.83. SAES
l.93. With “first publications”, do you refer to House et al. and Farley? Be more specific (or use “former” instead of “first”).
l.97-98. Here, you should elaborate on the Gautheron et al. method and how and why you modify it. There are some instances throughout the text where you allude to something and return to it later or not at all. Please check carefully again and rewrite so that when you bring up a point, you finish the description/argument. It isn’t easy to follow otherwise.
l.102. How is the internal pressure adjusted?
l.112-113. Unclear what this sentence means. This seems to be the reasoning behind modifying the Gautheron method, so it’s important. But I don’t understand it.
l.122. Can you specify the “range of sizes”?
l.128. Given that the calibration is based on Durango, how useful is it to check it with Durango? Ah, I see it’s the combination with MK-1 (l.130). Could you elaborate on this?
l.140. What is meant by “automatic grain detection and size measurement”? And what does the other option do (ToupView, l.140-141)?
l.141. Either state the “conventional criteria” (which size, which morphology?) and/or give a reference.
l.142-143. It reads like you are introducing the standards, but you already discussed them in Section 2. You might want to give these references and ages when discussing Durango and MK-1 for the first time in the text.
l.145. How do you measure the fragments’ size? Is this the widest diameter?
l.146. Are the MK-1 standards crystals and the Durango fragments?
l.151-152. In which way “optimised”? The form or depth of the holes?
l.172. What do you mean by “systematically reheated”? Do you decide this for each aliquot? Do you routinely reheat apatite once or more often?
l.172-173. “should occur during the second heating” Shouldn’t it occur during the first heating, and the re-extract should be at blank level?
l.213. Be more specific about the filter.
l.222. “(INFO)”?
l.236. What is meant by “sample homogeneity”?
l.243-244. I suggest you do describe this approach in more detail, given that this is one of the main changes in your protocol. Also, provide references to this “classic” approach.
l.248. Maybe also add another sentence describing briefly the isotope dilution approach.
Section 4.1. Should this be in the methods and get a different subtitle (something more general to trace element concentrations)?
l.260-261. Have you tried Nb capsules?
l.274. What do the 35-40 µm refer to? Rs?
l.281-282. I don’t understand this sentence. Please rephrase.
l.288. Would it be such a hassle to always weigh all?
l.292. What makes you interpret the Durango as “internally homogeneous”?
l.294-297. Could you give more details on these two measurement protocols? How many fragments were measured by LA-ICP-MS and at SARM?
l.305-313. I can’t follow the argument. Why is this trend not seen in previously published Durangos? It’s not present in the Gautheron et al. data set (Fig. 8, right panel). Are there other datasets? Do you have other evidence than “internal lab experience”? Is this “experience” based on the same analytical protocol presented here? Please clarify this paragraph.
l.309. I might just be a bit confused, but what do you mean by “crystal heterogeneity” and the homogeneity you talked about before (e.g., line 295)?
l.318. Why are quadrupole calibration and standard ages combined in one subsection (4.2)?
l.320. Didn’t you mention 20 before (instead of 34) for the calibration?
Section 4.2.1. Shouldn’t this be in the Methods?
l.329. Delete space between 6 and 1 in 6142.
l.350, 357, 408. Be consistent with the type of age you present. Why present central ages and weighted mean ages?
l.351-352. I don’t understand this sentence. When you have different populations, they are not “central ages”.
l.357. Before (l.145), you give 18.0 +- 0.7 Ma for MK-1 instead of +- 0.5.
l.364. Rephrase to better describe what this subsection is about. This will also help with the manuscript structure. “Feedback” is not the right word, I think.
l.380-381. Areas A and B are labelled in Fig. 3, but I cannot see a grey area in the figure.
l.386. I assume ‘cooking’ means that you heated the sample holder before sample degassing. Was the “pollution” reduced when you did this or the same? To what temperatures did you heat the sample holder, and for how long?
l.442. Delete “briefly”.
Sup. Mat. 1:
The uncertainty in apatite masses seems constant for most grains, except for Durango (6) grains. Could you please clarify how uncertainties are determined?
- Some of the MK-1 sample names are used twice. Are you missing the ‘P2’ part of the naming scheme from row 107?
- Column AO: Add in the (new) caption what (Ca) means and how you determined mass - I know what it is, but you should describe your headers well.
- Rethink how you name the supplementary files and tables. For example, it’s unclear what “standard concentration” is (name of Sup. Mat. 1 file).
Sup. Mat. 2:
- Individual Durango grains are referred to as “Sample” in Sup. Mat. 1, but here as “Name”. “Name” (or rather “Nom”) in Sup. Mat. 1, however, is just numbered Durango and MK-1. Please make this consistent throughout.
- What is the difference between Durangos given in Sup. Mats. 1 and 2 (your study only)? Not all in Table 1 are shown in Table 2.
- Column I: Change “Ages” to “AHe ages”.
Sup. Mat. 3:
- This table gives 62 Durango analyses; the table in Sup. Mat. 2 gives 61. However, it doesn’t seem like just one is missing. I think DG6D, -E, -F, and -G are missing from Sup. Mat. 2. D23P6D, -E, -F, -G are missing from Sup. Mat. 3. Somewhere, another one must be missing. Or maybe these are not supposed to be the same samples? It’s unclear.
- There are 10 MK-1 samples, while there are 15 in Sup. Mat. 1.
Sup. Mat. 4:
Please present the data so that one can understand what is given. This can, of course, be in a read-me file or something like that for a raw data file.
Sup. Mat. 5:
It is unclear what this is supposed to show the reader. Is this for the data reduction you refer to in l.43-44 and make available “upon request”? It may be a problem on my side, but I can’t read all the sheets, and some links don’t seem to work. If you want to make this file available with the article after all, please ensure it works for everyone and include your data as example input and comments to make it idiot-proof.
Fig. 1:
- Check French/English use of words. Check other figures as well, there are some more instances (e.g., in axes labels).
- Check font size (especially in legend).
- What are the grey, red, blue, and black colours in the sample holder in B?
Figs. 3, 4:
- This sample D22P3F is missing from Sup. Mat. 2 and 3.
Fig. 8:
- If it can be quickly done, make the colour scale the same for the left and right panels.
- Label left and right panels A and B.
Fig. 9:
- Remind readers what Ti and CH are.
- What is R.H.?
Fig. 10:
- l.360. I can’t see a grey area in the figure.
- Add units to the axis in B.
- What is the central age (B)?
- Add an explanation of what’s going on at pipette numbers <350 (A) to the caption. What “change”? Wasn’t there a significant change after pipette number ~420 (according to Fig. 9B)? Sorry, I can’t find it right now; did you describe this somewhere in the text?
Fig. 11:
- Label all panels of the figure (C for the left panel).
Fig. 12:
- Are these the data presented in Sup. Mat. 4? You need to refer to all supplementary materials in your text.
Citation: https://doi.org/10.5194/gchron-2024-6-RC3 -
AC3: 'Final response to the reviewers', Alexis Derycke, 24 May 2024
The manuscript Gchron 2024-6 was reviewed by four different reviewers (R2: James Metcalf, R4: Bruce Idleman and R1-3 anonymous), who are greatly thanked for the time they took to do so and for the extensive comments and recommendations they provided. These reviews contain similar or related comments, and in the following we will attempt to address and clarify those general comments.
Please find attached to this comment a .pdf files including the general responses.
To supplement this overarching response, if the manuscript is accepted for review, individual and precise replies will be produced as responses to each review on the Gchron 2024-6.
-
AC3: 'Final response to the reviewers', Alexis Derycke, 24 May 2024
-
RC4: 'Comment on gchron-2024-6', Bruce Idleman, 23 Mar 2024
Review of Derycke et al. (submitted to Geochronology)
This paper describes the development of analytical facilities for U-Th/He thermochronology at the Géosciences Rennes lab at Rennes University. The U-Th/He technique is currently a topic of considerable interest and discussion, and there have been many advances made in both measurement protocols and interpretation over the past few years. Despite this, relatively few labs have published detailed descriptions of their analytical hardware, software, and experimental protocols. Detailed technical accounts of laboratory practices represent useful benchmarks for guiding current and future researchers in the field, and in general papers of this type are quite appropriate for publication in Geochronology.
Derycke et al. describe an analytical system based around a small quadrupole mass spectrometer that is similar in many respects to those found in other U-Th/He labs. Sample heating is accomplished using a fiber-coupled laser with a coaxial pyrometer for temperature measurement, a system somewhat simpler than that described by Idleman et al. (2018) but suitable for helium degassing experiments using the continuous ramped heating (CRH) method. Their experimental protocols are a mixture of tried-and-true methods used and described previously for both conventional and CRH, and a few new protocols of their own design. The new methods are interesting and perhaps deserve further scrutiny, but I have significant reservations about several of them (see below).
General Comments:
The content of the paper can be divided into several themes: a description of the analytical hardware in the Rennes lab, a summary of the analytical procedures used, issues surrounding calibration of both He and U/Th/Sm measurements (for which new protocols are presented), and treatment of various artifacts of measurement (particularly with regard to CRH experiments). Unfortunately, I believe a lack of organization renders these discussions difficult to follow, and the significance of some passages is lost when the reader has to jump back and forth through the manuscript to follow the thought train. I suggest that the authors consider a major reorganization of the text, at a minimum separating the sections dealing with the He mass spectrometry and those describing the LA-ICP-MS measurements for U, Th, and Sm. Additionally, text describing instrumentation and basic analytical methods should not be intermixed with that addressing data treatment and analytical complications.
In terms of the various themes themselves, I found the portions of the paper detailing the instrumentation in the lab to be fairly straightforward, with some specific issues and omissions as outlined below. Sections dealing with issues of calibration and artifacts for the CRH and LA-ICP-MS experiments, some of which invoke new analytical methodology, are somewhat more problematic.
Derycke et al. present a complicated and somewhat unorthodox protocol for calibrating their 4He measurements that utilizes both a 3He spike to monitor run-to-run variations, and reference to measured Durango apatite ages for the primary calibration of instrument sensitivity. For conventional U-Th/He analyses, the use of 3He spike to monitor system sensitivity is standard procedure, but calibration of this spike is generally performed by comparison with other well-analyzed spikes or by careful preparation using calibrated volumes and pressure measurements (i.e., first principles). I question the validity of using a material with a somewhat poorly constrained age like Durango apatite as a primary standard for helium measurement calibration – it can be done, but it introduces both additional complexity and uncertainty. It’s not a step forward in my view.
The use of the 3He spike for CRH experiments introduces further complications, particularly with regard to corrections for gas consumption, 3He/4He mass discrimination issues that can only be resolved by reaching back once again to Durango apatite measurements for resolution, etc. Some of these issues I outline more specifically below. In summary, most of these problems could be addressed much more effectively by the addition of a 4He spike on the extraction line, as described initially by Idleman et al. (2018).
For U/Th/Sm measurements Deryck et al. describe a procedure that also relies on using Durango apatite as a primary compositional standard. In this case the conventional method is somewhat onerous in that it requires a radioactive spike for isotope dilution measurements. The methodology outlined in the paper could prove useful as appropriate spikes become more useful to procure, and would allow labs not set up to handle radioactive materials to conduct the required measurements. However, I question the selection of Durango apatite as the standard for developing this technique given its potential heterogeneity and anomalous Th/U.
Aside from its organizational issues, the manuscript itself is somewhat unpolished and probably would have benefited from an additional round of editing. Unfortunately, some sections suffer from somewhat stilted phrasing and spelling errors that could reflect imprecise translation from French to English. These issues need to be cleaned up, and a careful read of the final manuscript by a native English speaker might be a good idea.
Does this paper represent a useful contribution to the advancement of the U-Th/He technique? By themselves the basic descriptions of instrumentation and lab protocols would be useful to other researchers in the field, although they would need to be reorganized and cleaned up before publication. The major new “contributions” in the paper are centered around the somewhat novel approaches taken to calibration, both for helium and U/Th/Sm measurements. It’s not clear to me that these approaches offer significant advantages over those in current use elsewhere, particularly the protocol for helium calibration, and they have significant disadvantages (complexity, reduced precision). Personally, I would be hesitant to recommend that a researcher new to the field consider adopting these approaches without careful evaluation of their potential benefits. I would like to see the authors provide a critical evaluation of their new methods that addresses their strengths and weakness relative to others in current use.
In summary, I believe this paper would benefit from moderate to major revision and potentially another editorial review prior to consideration for publication.
Specific Comments:
Line 61 – Fused silica is somewhat pervious to He. Do you have any evidence that it is contributing to the blank during long CRH runs?
Line 63 – Suggest “LABVIEW software provided by the manufacturer.” See also lines 77 and 86.
Line 72 – How large is the pyrometer spot relative to the sample packet?
Line 73 – “Electric lens” – what is this, and why is it not shown in fig. 1b?
Line 91 – More details about the mass spectrometer operation are needed in this section. Was measurement performed with a faraday detector or electron multiplier? What was the multiplier voltage/gain? What electron emission setting was used?
Line 99 – This is a large pipette for the tank size (1:1000). How often do you have to refill the tank? Pipettes of <0.5 ml are available, and would allow you to run for years on the same calibration.
Line 101 – Given the proportionally large size of your 3He spike (see fig. 3), 0.1% 4He contamination in the spike could make a significant contribution to the 4He signal. Are you correcting for any 4He in the spike?
Lines 107-110 – This methodology is confusing, and I believe your explanation is somewhat incorrect. Some of the factors you describe can directly influence mass spec sensitivity, such as power failures. The sensitivity of electron-impact source mass spec is inherently somewhat variable over time, for a number of reasons. In short, there probably is no such thing as constant instrument sensitivity (your S). I can see the usefulness of separating true mass spec sensitivity (your S) from other physical factors that influence signal size, such as adding/removing volume (what I would include in Ie), but don’t mix these up. Additionally, calling your Ie term “ionization efficiency” is simply misleading.
Line 126-127 – Why not add a 4He spike? Is there any advantage to your method? If so, you should point it out. I don’t see cost as a significant advantage, since the cost of an additional pipette and tank is fairly small with respect to the total cost of the system.
Line 168 – Suggest you replace “pollution” with “contamination” here and elsewhere.
Line 172 – What does “systematically reheated” mean? Some particularly badly behaving apatites can still contain some 4He at 1050°C. Are you increasing the temperature, or just repeating the same experiment?
Line 194-195 – This is confusing. Is the temperature ramp paused for “re-centre”? What does “reaction to the laser beam” mean? The shape of the gas release curve obtained during a CRH run is dependent upon ramp rate, and curves can’t be compared easily if their heating schedules aren’t exactly the same.
Line 205 – Ah, but gas consumption rate depends upon partial pressure, as pointed out by Idleman et al. (2018).
Lines 205-209 – There are actually two things happening here, which you have not explained very clearly: measurement and correction of the static blank using the 4He signal, and measurement and correction of the 4He consumption rate using the 3He signal.
Line 213 – Exactly what kind of filter did you use?
Lines 231-233 – why not do the dilutions by weight (much more accurate)?
Lines 247-248 – what does “below the sensitivity of the ICP-MS” mean? Below the detection limit? Below some level of measurement precision?
Line 288 – Again, how difficult would it be to do the dilutions by weight, and how much improvement in precision would you expect?
Lines 308-309 – Why was this correlation not seen by Gautheron et al. (2021)?
Lines 320-329 – I admit that this “works”, but it is analytically much more complicated and of lower precision than using a properly calibrated 4He spike with low pipette/tank volume ratio. Other than cost, is there any reason to recommend this method? Would you recommend it to someone considering building a helium analysis system?
Lines 336-344 and fig. 9b – See my comments for lines 107-110. Again, I don’t see the justification for separating out two components of “sensitivity” as you have done, and the idea that fundamental mass spec sensitivity (signal/partial pressure) stays constant over time is simply incorrect. Another way of saying this is that some of the changes you put into your Ie parameter are actually changes in S. In the end what really matters is the “total system sensitivity”, e.g., how much gas do I need to put into the system to get a particular signal. Separating out an “Ie” parameter does have some usefulness with regard to volume changes, e.g., opening a separate volume to the line for some analyses, and closing it for others. When such changes are made over a short time span (minutes to perhaps a day), this kind of correction factor makes sense. However, other factors that directly influence the mass spectrometer, such as power outages, electron multiplier and filament aging, etc. most certainly directly change the intrinsic mass spec sensitivity.
Line 355 – Don’t count helium out as a source of Durango dispersion. Some shards of Durango have been shown to produce anomalous gas spikes during CRH runs (see McDannell et al. (2018) and more recent publications by the Lehigh group.
Lines 365-369 and fig. 11b – Despite the apparent correlation, I’m concerning about the degree of coupling (or lack of coupling) between your sample packet and thermocouple. For fig. 11b it might be more instructive to plot temperature difference vs. temperature.
Lines 371-373 – Heat transfer at these lower temperatures is slow, and good coupling is required (see comment above). Additionally, restricting temperature measurement to >350°C effectively eliminates using the system for kinetic parameter determination, since most of the useful information comes from lower temperatures.
Lines 385-388 – Have you demonstrated that the active gas contaminants don’t suppress the 4He signal in your experiments (see Idleman et al., 2018)? Additionally, Durango apatite generally produces little of this contamination – have you analyzed any unknow samples that show more contamination?
Lines 390-404 – As pointed out by Idleman et al., (2018), gas consumption rates are highly dependent upon partial pressure, sometimes in a complex and non-linear way. Applying consumption rate corrections derived from a large 3He signal to 4He signals several orders of magnitude smaller is certain to lead to incorrect results. I suspect a significant number of your misbehaving samples would prove to be fine with appropriate corrections applied. Reducing the size of your 3He spike would help improve this situation.
Lines 440-441 – This is an interesting relationship and deserves more treatment in the text.
Fig. 1 – Change French terms to English. Show “electric lens”.
Fig. 3 – For zone 2 label change “rampe” to “ramped”. In caption change “blows” to “release”.
Fig. 4 - For zone 2 label change “rampe” to “ramped”. It might also be useful to show the 3He beam evolution and how the gas consumption was corrected. For fig. 4c, the convention is to show f vs. temperature. You might need a separate plot for that.
Fig. 7b – Change “caraterisation” to “characterization” or possibly “characterization”.
Fig. 9b – What are the “R.H.” segments?
Fig. 11b – Show as temp. difference vs. temperature.
Bruce Idleman
Citation: https://doi.org/10.5194/gchron-2024-6-RC4 -
AC3: 'Final response to the reviewers', Alexis Derycke, 24 May 2024
The manuscript Gchron 2024-6 was reviewed by four different reviewers (R2: James Metcalf, R4: Bruce Idleman and R1-3 anonymous), who are greatly thanked for the time they took to do so and for the extensive comments and recommendations they provided. These reviews contain similar or related comments, and in the following we will attempt to address and clarify those general comments.
Please find attached to this comment a .pdf files including the general responses.
To supplement this overarching response, if the manuscript is accepted for review, individual and precise replies will be produced as responses to each review on the Gchron 2024-6.
-
AC3: 'Final response to the reviewers', Alexis Derycke, 24 May 2024
Interactive discussion
Status: closed
-
AC1: 'Comment on gchron-2024-6', Alexis Derycke, 20 Feb 2024
Authors apologise for the initial omission of the tables and point out that the tables are now available as a supplement.
Citation: https://doi.org/10.5194/gchron-2024-6-AC1 -
AC2: 'Comment on gchron-2024-6', Alexis Derycke, 21 Feb 2024
The editing process revealed some missing parts in the supplement, which has therefore been updated as follow :
- Sup. Mat. 1: chemical data (used to produce Fig. 5)
- Sup. Mat. 2: Durango age and ThU-SmTh ratio (used to produce Fig. 8)
- Sup. Mat. 3: raw data from standards (e.g. He, U, Th, Sm and Ca content)
- Sup. Mat. 4: ramped heating raw data (e.g. He concentrations, temperature, degassed fraction and Ln(D/a²))
- Sup. Mat. 5: automated data reduction file for (U-Th)/He analysisAuthors apologize for the initial oversight.
Citation: https://doi.org/10.5194/gchron-2024-6-AC2 -
RC1: 'Comment on gchron-2024-6', Anonymous Referee #1, 29 Feb 2024
Please refert to the attached PDF file for my review
-
AC3: 'Final response to the reviewers', Alexis Derycke, 24 May 2024
The manuscript Gchron 2024-6 was reviewed by four different reviewers (R2: James Metcalf, R4: Bruce Idleman and R1-3 anonymous), who are greatly thanked for the time they took to do so and for the extensive comments and recommendations they provided. These reviews contain similar or related comments, and in the following we will attempt to address and clarify those general comments.
Please find attached to this comment a .pdf files including the general responses.
To supplement this overarching response, if the manuscript is accepted for review, individual and precise replies will be produced as responses to each review on the Gchron 2024-6.
-
AC3: 'Final response to the reviewers', Alexis Derycke, 24 May 2024
-
RC2: 'Comment on gchron-2024-6', James Metcalf, 12 Mar 2024
This technical note presents specific details and results of how the GeOHeLiS (U-Th)/He lab at Rennes University operates. As a (U-Th)/He lab manager I found this paper very interesting, and am very supportive of it, and other technical descriptions of analytical set ups, appearing as technical notes in EGU Geochronology. This lab specifically seems to be targeting three larger challenges: 1) how to calibrate a noble gas QMS, 2) how to determine total U, Th, and Sm without expensive and difficult to obtain spike materials, and 3) how to execute ramped heating of geologic materials.
In general I think the paper is well written and informative. I believe though that there are, however, areas where the contribution would be greatly improved with some additional clarifications, discussions, and modifications.
The paper would benefit from a re-organization, I often found myself flipping back and forth to look at figures from earlier in the paper. One strategy might be to handle all of the He measurement things first, then move on to dissolution. Some parts, like grain selection and measurement, are relatively standard and can probably be shortened.
Importantly, I think that because the authors are presenting, in some cases, alternative ways to calibrate He, U, Th, and Sm measurements, that the paper would benefit greatly from data collected more traditional ways. On the same solutions, for example, how does this method compare to isotope dilution? Same with using Durango as the 4He calibration source, it seems like a direct comparison would be much more informative for people considering building a lab.
I think most of the figures are in good shape, although I have noted some places where items are not explained or noted.
I believe my concerns would fall somewhere between major and minor revisions, and I strongly encourage the authors to make the modifications (I would really like to see lab descriptions as detailed as this appearing regularly).
General Comments
It seems like one motivating factor for this machine design is to not have a 4He tank or pipette on the line and instead use Durango shards as the 4He calibration material. While calibrating pipettes and tanks is certainly difficult and time consuming, I think the paper would benefit from a discussion of the rationale behind excluding this more common method of QMS calibration. I understand the logic of trying to avoid using spikes for U, Th, and Sm since they can be hard to obtain, however I don’t understand the motivation for not having a 4He tank and pipette on the line.
This is especially important considering the very high 4He measurement uncertainties in the data supplied in supplementary materials. Looking at the Durango data you provided in Supp. Mat 3, it seems like the uncertainties and Durango ages for grains < 10 ug are much more scattered. The data seems to imply that the uncertainties at least are the result of high 4He uncertainties. The total 4He values don’t seem too low, can you comment on how well this method handles small amounts of gas? Are those uncertainties (some over 100%) because of the calibration method or were these specific measurements due to blank or other problems? Also, is it common for the small Durango shards used in the calibrations to have these kinds of uncertainties?
Specific comments and questions
• There are a number of places where items are described as “computer controlled thanks to LabView assets provided by manufacturers.” Do all of these vi’s and sub-vi’s operate within an in-house piece of code?
• Where in the laser process is the electric lens? I don’t see it in Figure 1B.
• For a technical note there are a few things that would be interesting (as a fellow lab manager) to know
o What type of filament and settings (emission, SEM voltage) do you run on the QMG and did you do any testing to arrive at the best settings?
o How did you determine the volumes of your pipette and 3He tank?
o What is the internal volume of your line?
• When filling the 3He tank you say the target pressure is 1E-6 mbar, this is at the low end of capacitance manometer detection so I assume you are backfilling an empty tank with a higher pressure trapped and measured in the pipette?
• Line 107 - The Ie (ionization efficiency term) is adjusted for each run, is it monitored throughout a run? With a pan holding as many samples as yours does it seems like some changes might happen during the time of a run.
• Line 115 – do you consider uncertainties in 4He blanks in the calculations?
• Are S and Ie the same for really really large samples and small samples? Are these effects linear?
• Line 151 – how long does the heating and pumping typically take? Is this on the turbo pump only or are you gettering the volume?
• Figure 2 – Is the top dark red line the 4He or is it the cold head temp over time? I find this a little difficult to read.
• I think the description of how the ramped heating measurements are accomplished it really excellent.
• Line 226 – is the process that after the hot plate the samples are diluted so the total volume is 2 or 10 mL? Is this done assuming none of the original acid has evaporated (meaning you just pipette in the right amount), or are you filling to a mark on the individual vials? I guess I am curious about the problems of evaporation on hot plates mostly.
• For measuring the Ca – my experience is that keeping Ca and Th in solution together can require some HF, and my attempts to do stoichiometric apatite volume determinations has not been overly reliable. Do you have external checks (micro CT for example) for the volume? For samples obviously this is important but also for the Durangoes used for the initial He calibration, are all of the volumes stoichiometric?
• Line 261 – Are your Nb tubes also contaminated to this degree? We’ve switched over to using Nb for everything because they are more consistently and reliably uncontaminated.
• Figure 7B – since it is a log scale and only U and Th are really important for age reliability, can that part be blown up? It looks like the QQQ line is well below the others for Th?
• Section 4.1.3 – I don’t quite follow the argument for why the linear relationship between Th/U and Sm/Th isn’t a chemical issue. You state that it hasn’t been reported before but then conclude that it is a real thing, do you mean just for this crystal? Could this be checked by isotope dilution?
• Section 4.2.1 – are the Durangoes used for the calibration dissolved and analyzed for U, Th, and Sm? And then the volumes are all stoichiometric?
• Figure 9 – what is encompassed in the y-axis uncertainties in figure 9A?
• Figure 9B – what are the green sections labeled “R.H.”?
• Line 345 – how small of a shard is used? Is it similar in volume to a typical apatite?
• Line 440 – I think this point warrants more discussion, especially since this relationship isn’t apparent in all Durango datasets.
Minor Line Edits• Line 83 – should be SAES getters
• Line 111 – should be influences not influence
• Line 135 – I might use “pure apatite separate” instead of “powder” here, to me powder implies the grains are completely pulverized.
• Line 157 – no need to capitalize apatite
• Line 172 – it’s minor but apatites should be degassed after one heating step, and the second should be just blank level, right?
• Line 232 – should read “checked”
• Line 310 – should be “Crystals” and not “crystal”
• Line 390 – could the end trend and the global trend be labeled on figure 4?James R. Metcalf
Citation: https://doi.org/10.5194/gchron-2024-6-RC2 -
AC3: 'Final response to the reviewers', Alexis Derycke, 24 May 2024
The manuscript Gchron 2024-6 was reviewed by four different reviewers (R2: James Metcalf, R4: Bruce Idleman and R1-3 anonymous), who are greatly thanked for the time they took to do so and for the extensive comments and recommendations they provided. These reviews contain similar or related comments, and in the following we will attempt to address and clarify those general comments.
Please find attached to this comment a .pdf files including the general responses.
To supplement this overarching response, if the manuscript is accepted for review, individual and precise replies will be produced as responses to each review on the Gchron 2024-6.
-
AC3: 'Final response to the reviewers', Alexis Derycke, 24 May 2024
-
RC3: 'Comment on gchron-2024-6', Anonymous Referee #3, 13 Mar 2024
The technical note by Derycke et al. describes their newly built He-line setup and analytical protocols for apatite (U-Th-Sm)/He (AHe) analyses, including a ramped heating protocol (developed during 2021-2023 at the Géosciences Rennes lab, France). The main changes compared to previous approaches are their calibration of quadrupole sensitivity (simplified from Gautheron et al. (2021)) and the “standard concentration range approach” for trace element measurements.
I appreciate the authors' efforts to set up the presented protocols and share this technical report. However, I cannot support the publication in its current form. My main issues relate to the organisation of the manuscript, missing details of some descriptions, and confusing or possibly inconsistent presentation of data. See below for general issues and line- and section-specific comments and suggestions. I hope these help to improve the manuscript and that it will be published after a thorough revision.
Inconsistency of results and data presentation:
The Abstract talks of ~3.9% “precision in 4He content determination”, Section 4.2.2 mentions ~3.9% dispersion of Durango ages (but not for all of the ages? I'm not sure; this part is confusing), and the Conclusions talk of <7% age error. This makes it difficult to follow where the numbers come from and whether the terminology is correctly used in different places. Are the 3.9% from the Abstract the same as in Section 4.2.2? Could you not use the same type of uncertainty when describing it?
The presentation of data in the supplementary materials and the descriptions in the text of how many samples were used and for what were somewhat confusing. One reason may be that the supplementary materials are not referred to in the main text, and the supplementary tables have no proper titles and captions. From the tables, text, and your short note in the GChron discussion forum on what the supplementary tables are, I cannot reconstruct which data are shown and why, and which data are omitted from tables and figures and why. So, what is actually shown in the tables? What are the numbers 34, 20, and 45 Durango fragments in the text, what are they used for (calibration vs test measurement? ramped heating?), and which data are shown in the figures? And which are not shown but are shown in some tables (but not in others)? And how many MK-1 fragments are there? 10 or 15? Can’t you show all data together in fewer tables and indicate which data might be omitted from plots, central age calculations, … and why? Either it’s inconsistent or confusing; I'm not sure.
Structure:
The structure of the manuscript is somewhat confusing. Should Sections 2-3 be combined under a Methods section? In addition to my comment on l.25-34, I suggest adding a paragraph at the end of the Introduction to help guide the reader. Furthermore, it would be helpful to have the same (similar) subsection headings in Methods and Results so that one knows where to find the results (like you have it for Sections 3.2.2, 3.2.3 and 4.3 concerning ramped heating).
It might also be good to separate Results and Discussion. The Discussion should include a more condensed text on what has been improved with your protocols, which aspects need more work, etc. I know some places in the text address these points, but having them in one place with corresponding subheadings would be helpful.
Language and style:
Please check the text for English grammar, the use of articles, and word choice. Sometimes, words or parts of words are missing. There are issues in many places that I will not point out individually at this point, as I think the text may change during revision. Also, check punctuation. Carefully check the figures and tables and consistently use the English language and style.
Title: Try to rephrase the title without using parentheses. Delete “for the He.”
Standard style guides suggest writing out numbers up to ten. You keep switching between doing and not doing this. Please be consistent in your writing style.
Add spaces between values and their units throughout the text.
Do not use bullet points in your text (Conclusions). You may number your main results but integrate them into the continuous text.
Make sure to add proper titles and captions to tables in the supplementary materials (as you would for the main text). Please also explain, in captions, how uncertainties were determined (in your and previous studies). Furthermore, every supplementary material should be referred to in the main text. You currently don’t refer to any.
Provide a separate reference list for the supplementary materials.
I suggest changing “ages” to “AHe ages” in tables and figures (including captions) and throughout the text.
Be consistent with how you write MK1 or MK-1.
Line-specific comments:
l.25-34. The introduction could be more helpful if you’d describe current analytical protocols, challenges, and aspects needing improvement. I don’t think an introduction to the method’s history is necessary. You only allude to “recent technical development” (l.34-35). I suggest focusing on this part so it becomes clearer from the beginning what your new contribution is.
l.43. “QQQ-MS” is introduced as an abbreviation in the abstract, but I suggest doing it again in the first section.
l.44. Write out VBA.
l.83. SAES
l.93. With “first publications”, do you refer to House et al. and Farley? Be more specific (or use “former” instead of “first”).
l.97-98. Here, you should elaborate on the Gautheron et al. method and how and why you modify it. There are some instances throughout the text where you allude to something and return to it later or not at all. Please check carefully again and rewrite so that when you bring up a point, you finish the description/argument. It isn’t easy to follow otherwise.
l.102. How is the internal pressure adjusted?
l.112-113. Unclear what this sentence means. This seems to be the reasoning behind modifying the Gautheron method, so it’s important. But I don’t understand it.
l.122. Can you specify the “range of sizes”?
l.128. Given that the calibration is based on Durango, how useful is it to check it with Durango? Ah, I see it’s the combination with MK-1 (l.130). Could you elaborate on this?
l.140. What is meant by “automatic grain detection and size measurement”? And what does the other option do (ToupView, l.140-141)?
l.141. Either state the “conventional criteria” (which size, which morphology?) and/or give a reference.
l.142-143. It reads like you are introducing the standards, but you already discussed them in Section 2. You might want to give these references and ages when discussing Durango and MK-1 for the first time in the text.
l.145. How do you measure the fragments’ size? Is this the widest diameter?
l.146. Are the MK-1 standards crystals and the Durango fragments?
l.151-152. In which way “optimised”? The form or depth of the holes?
l.172. What do you mean by “systematically reheated”? Do you decide this for each aliquot? Do you routinely reheat apatite once or more often?
l.172-173. “should occur during the second heating” Shouldn’t it occur during the first heating, and the re-extract should be at blank level?
l.213. Be more specific about the filter.
l.222. “(INFO)”?
l.236. What is meant by “sample homogeneity”?
l.243-244. I suggest you do describe this approach in more detail, given that this is one of the main changes in your protocol. Also, provide references to this “classic” approach.
l.248. Maybe also add another sentence describing briefly the isotope dilution approach.
Section 4.1. Should this be in the methods and get a different subtitle (something more general to trace element concentrations)?
l.260-261. Have you tried Nb capsules?
l.274. What do the 35-40 µm refer to? Rs?
l.281-282. I don’t understand this sentence. Please rephrase.
l.288. Would it be such a hassle to always weigh all?
l.292. What makes you interpret the Durango as “internally homogeneous”?
l.294-297. Could you give more details on these two measurement protocols? How many fragments were measured by LA-ICP-MS and at SARM?
l.305-313. I can’t follow the argument. Why is this trend not seen in previously published Durangos? It’s not present in the Gautheron et al. data set (Fig. 8, right panel). Are there other datasets? Do you have other evidence than “internal lab experience”? Is this “experience” based on the same analytical protocol presented here? Please clarify this paragraph.
l.309. I might just be a bit confused, but what do you mean by “crystal heterogeneity” and the homogeneity you talked about before (e.g., line 295)?
l.318. Why are quadrupole calibration and standard ages combined in one subsection (4.2)?
l.320. Didn’t you mention 20 before (instead of 34) for the calibration?
Section 4.2.1. Shouldn’t this be in the Methods?
l.329. Delete space between 6 and 1 in 6142.
l.350, 357, 408. Be consistent with the type of age you present. Why present central ages and weighted mean ages?
l.351-352. I don’t understand this sentence. When you have different populations, they are not “central ages”.
l.357. Before (l.145), you give 18.0 +- 0.7 Ma for MK-1 instead of +- 0.5.
l.364. Rephrase to better describe what this subsection is about. This will also help with the manuscript structure. “Feedback” is not the right word, I think.
l.380-381. Areas A and B are labelled in Fig. 3, but I cannot see a grey area in the figure.
l.386. I assume ‘cooking’ means that you heated the sample holder before sample degassing. Was the “pollution” reduced when you did this or the same? To what temperatures did you heat the sample holder, and for how long?
l.442. Delete “briefly”.
Sup. Mat. 1:
The uncertainty in apatite masses seems constant for most grains, except for Durango (6) grains. Could you please clarify how uncertainties are determined?
- Some of the MK-1 sample names are used twice. Are you missing the ‘P2’ part of the naming scheme from row 107?
- Column AO: Add in the (new) caption what (Ca) means and how you determined mass - I know what it is, but you should describe your headers well.
- Rethink how you name the supplementary files and tables. For example, it’s unclear what “standard concentration” is (name of Sup. Mat. 1 file).
Sup. Mat. 2:
- Individual Durango grains are referred to as “Sample” in Sup. Mat. 1, but here as “Name”. “Name” (or rather “Nom”) in Sup. Mat. 1, however, is just numbered Durango and MK-1. Please make this consistent throughout.
- What is the difference between Durangos given in Sup. Mats. 1 and 2 (your study only)? Not all in Table 1 are shown in Table 2.
- Column I: Change “Ages” to “AHe ages”.
Sup. Mat. 3:
- This table gives 62 Durango analyses; the table in Sup. Mat. 2 gives 61. However, it doesn’t seem like just one is missing. I think DG6D, -E, -F, and -G are missing from Sup. Mat. 2. D23P6D, -E, -F, -G are missing from Sup. Mat. 3. Somewhere, another one must be missing. Or maybe these are not supposed to be the same samples? It’s unclear.
- There are 10 MK-1 samples, while there are 15 in Sup. Mat. 1.
Sup. Mat. 4:
Please present the data so that one can understand what is given. This can, of course, be in a read-me file or something like that for a raw data file.
Sup. Mat. 5:
It is unclear what this is supposed to show the reader. Is this for the data reduction you refer to in l.43-44 and make available “upon request”? It may be a problem on my side, but I can’t read all the sheets, and some links don’t seem to work. If you want to make this file available with the article after all, please ensure it works for everyone and include your data as example input and comments to make it idiot-proof.
Fig. 1:
- Check French/English use of words. Check other figures as well, there are some more instances (e.g., in axes labels).
- Check font size (especially in legend).
- What are the grey, red, blue, and black colours in the sample holder in B?
Figs. 3, 4:
- This sample D22P3F is missing from Sup. Mat. 2 and 3.
Fig. 8:
- If it can be quickly done, make the colour scale the same for the left and right panels.
- Label left and right panels A and B.
Fig. 9:
- Remind readers what Ti and CH are.
- What is R.H.?
Fig. 10:
- l.360. I can’t see a grey area in the figure.
- Add units to the axis in B.
- What is the central age (B)?
- Add an explanation of what’s going on at pipette numbers <350 (A) to the caption. What “change”? Wasn’t there a significant change after pipette number ~420 (according to Fig. 9B)? Sorry, I can’t find it right now; did you describe this somewhere in the text?
Fig. 11:
- Label all panels of the figure (C for the left panel).
Fig. 12:
- Are these the data presented in Sup. Mat. 4? You need to refer to all supplementary materials in your text.
Citation: https://doi.org/10.5194/gchron-2024-6-RC3 -
AC3: 'Final response to the reviewers', Alexis Derycke, 24 May 2024
The manuscript Gchron 2024-6 was reviewed by four different reviewers (R2: James Metcalf, R4: Bruce Idleman and R1-3 anonymous), who are greatly thanked for the time they took to do so and for the extensive comments and recommendations they provided. These reviews contain similar or related comments, and in the following we will attempt to address and clarify those general comments.
Please find attached to this comment a .pdf files including the general responses.
To supplement this overarching response, if the manuscript is accepted for review, individual and precise replies will be produced as responses to each review on the Gchron 2024-6.
-
AC3: 'Final response to the reviewers', Alexis Derycke, 24 May 2024
-
RC4: 'Comment on gchron-2024-6', Bruce Idleman, 23 Mar 2024
Review of Derycke et al. (submitted to Geochronology)
This paper describes the development of analytical facilities for U-Th/He thermochronology at the Géosciences Rennes lab at Rennes University. The U-Th/He technique is currently a topic of considerable interest and discussion, and there have been many advances made in both measurement protocols and interpretation over the past few years. Despite this, relatively few labs have published detailed descriptions of their analytical hardware, software, and experimental protocols. Detailed technical accounts of laboratory practices represent useful benchmarks for guiding current and future researchers in the field, and in general papers of this type are quite appropriate for publication in Geochronology.
Derycke et al. describe an analytical system based around a small quadrupole mass spectrometer that is similar in many respects to those found in other U-Th/He labs. Sample heating is accomplished using a fiber-coupled laser with a coaxial pyrometer for temperature measurement, a system somewhat simpler than that described by Idleman et al. (2018) but suitable for helium degassing experiments using the continuous ramped heating (CRH) method. Their experimental protocols are a mixture of tried-and-true methods used and described previously for both conventional and CRH, and a few new protocols of their own design. The new methods are interesting and perhaps deserve further scrutiny, but I have significant reservations about several of them (see below).
General Comments:
The content of the paper can be divided into several themes: a description of the analytical hardware in the Rennes lab, a summary of the analytical procedures used, issues surrounding calibration of both He and U/Th/Sm measurements (for which new protocols are presented), and treatment of various artifacts of measurement (particularly with regard to CRH experiments). Unfortunately, I believe a lack of organization renders these discussions difficult to follow, and the significance of some passages is lost when the reader has to jump back and forth through the manuscript to follow the thought train. I suggest that the authors consider a major reorganization of the text, at a minimum separating the sections dealing with the He mass spectrometry and those describing the LA-ICP-MS measurements for U, Th, and Sm. Additionally, text describing instrumentation and basic analytical methods should not be intermixed with that addressing data treatment and analytical complications.
In terms of the various themes themselves, I found the portions of the paper detailing the instrumentation in the lab to be fairly straightforward, with some specific issues and omissions as outlined below. Sections dealing with issues of calibration and artifacts for the CRH and LA-ICP-MS experiments, some of which invoke new analytical methodology, are somewhat more problematic.
Derycke et al. present a complicated and somewhat unorthodox protocol for calibrating their 4He measurements that utilizes both a 3He spike to monitor run-to-run variations, and reference to measured Durango apatite ages for the primary calibration of instrument sensitivity. For conventional U-Th/He analyses, the use of 3He spike to monitor system sensitivity is standard procedure, but calibration of this spike is generally performed by comparison with other well-analyzed spikes or by careful preparation using calibrated volumes and pressure measurements (i.e., first principles). I question the validity of using a material with a somewhat poorly constrained age like Durango apatite as a primary standard for helium measurement calibration – it can be done, but it introduces both additional complexity and uncertainty. It’s not a step forward in my view.
The use of the 3He spike for CRH experiments introduces further complications, particularly with regard to corrections for gas consumption, 3He/4He mass discrimination issues that can only be resolved by reaching back once again to Durango apatite measurements for resolution, etc. Some of these issues I outline more specifically below. In summary, most of these problems could be addressed much more effectively by the addition of a 4He spike on the extraction line, as described initially by Idleman et al. (2018).
For U/Th/Sm measurements Deryck et al. describe a procedure that also relies on using Durango apatite as a primary compositional standard. In this case the conventional method is somewhat onerous in that it requires a radioactive spike for isotope dilution measurements. The methodology outlined in the paper could prove useful as appropriate spikes become more useful to procure, and would allow labs not set up to handle radioactive materials to conduct the required measurements. However, I question the selection of Durango apatite as the standard for developing this technique given its potential heterogeneity and anomalous Th/U.
Aside from its organizational issues, the manuscript itself is somewhat unpolished and probably would have benefited from an additional round of editing. Unfortunately, some sections suffer from somewhat stilted phrasing and spelling errors that could reflect imprecise translation from French to English. These issues need to be cleaned up, and a careful read of the final manuscript by a native English speaker might be a good idea.
Does this paper represent a useful contribution to the advancement of the U-Th/He technique? By themselves the basic descriptions of instrumentation and lab protocols would be useful to other researchers in the field, although they would need to be reorganized and cleaned up before publication. The major new “contributions” in the paper are centered around the somewhat novel approaches taken to calibration, both for helium and U/Th/Sm measurements. It’s not clear to me that these approaches offer significant advantages over those in current use elsewhere, particularly the protocol for helium calibration, and they have significant disadvantages (complexity, reduced precision). Personally, I would be hesitant to recommend that a researcher new to the field consider adopting these approaches without careful evaluation of their potential benefits. I would like to see the authors provide a critical evaluation of their new methods that addresses their strengths and weakness relative to others in current use.
In summary, I believe this paper would benefit from moderate to major revision and potentially another editorial review prior to consideration for publication.
Specific Comments:
Line 61 – Fused silica is somewhat pervious to He. Do you have any evidence that it is contributing to the blank during long CRH runs?
Line 63 – Suggest “LABVIEW software provided by the manufacturer.” See also lines 77 and 86.
Line 72 – How large is the pyrometer spot relative to the sample packet?
Line 73 – “Electric lens” – what is this, and why is it not shown in fig. 1b?
Line 91 – More details about the mass spectrometer operation are needed in this section. Was measurement performed with a faraday detector or electron multiplier? What was the multiplier voltage/gain? What electron emission setting was used?
Line 99 – This is a large pipette for the tank size (1:1000). How often do you have to refill the tank? Pipettes of <0.5 ml are available, and would allow you to run for years on the same calibration.
Line 101 – Given the proportionally large size of your 3He spike (see fig. 3), 0.1% 4He contamination in the spike could make a significant contribution to the 4He signal. Are you correcting for any 4He in the spike?
Lines 107-110 – This methodology is confusing, and I believe your explanation is somewhat incorrect. Some of the factors you describe can directly influence mass spec sensitivity, such as power failures. The sensitivity of electron-impact source mass spec is inherently somewhat variable over time, for a number of reasons. In short, there probably is no such thing as constant instrument sensitivity (your S). I can see the usefulness of separating true mass spec sensitivity (your S) from other physical factors that influence signal size, such as adding/removing volume (what I would include in Ie), but don’t mix these up. Additionally, calling your Ie term “ionization efficiency” is simply misleading.
Line 126-127 – Why not add a 4He spike? Is there any advantage to your method? If so, you should point it out. I don’t see cost as a significant advantage, since the cost of an additional pipette and tank is fairly small with respect to the total cost of the system.
Line 168 – Suggest you replace “pollution” with “contamination” here and elsewhere.
Line 172 – What does “systematically reheated” mean? Some particularly badly behaving apatites can still contain some 4He at 1050°C. Are you increasing the temperature, or just repeating the same experiment?
Line 194-195 – This is confusing. Is the temperature ramp paused for “re-centre”? What does “reaction to the laser beam” mean? The shape of the gas release curve obtained during a CRH run is dependent upon ramp rate, and curves can’t be compared easily if their heating schedules aren’t exactly the same.
Line 205 – Ah, but gas consumption rate depends upon partial pressure, as pointed out by Idleman et al. (2018).
Lines 205-209 – There are actually two things happening here, which you have not explained very clearly: measurement and correction of the static blank using the 4He signal, and measurement and correction of the 4He consumption rate using the 3He signal.
Line 213 – Exactly what kind of filter did you use?
Lines 231-233 – why not do the dilutions by weight (much more accurate)?
Lines 247-248 – what does “below the sensitivity of the ICP-MS” mean? Below the detection limit? Below some level of measurement precision?
Line 288 – Again, how difficult would it be to do the dilutions by weight, and how much improvement in precision would you expect?
Lines 308-309 – Why was this correlation not seen by Gautheron et al. (2021)?
Lines 320-329 – I admit that this “works”, but it is analytically much more complicated and of lower precision than using a properly calibrated 4He spike with low pipette/tank volume ratio. Other than cost, is there any reason to recommend this method? Would you recommend it to someone considering building a helium analysis system?
Lines 336-344 and fig. 9b – See my comments for lines 107-110. Again, I don’t see the justification for separating out two components of “sensitivity” as you have done, and the idea that fundamental mass spec sensitivity (signal/partial pressure) stays constant over time is simply incorrect. Another way of saying this is that some of the changes you put into your Ie parameter are actually changes in S. In the end what really matters is the “total system sensitivity”, e.g., how much gas do I need to put into the system to get a particular signal. Separating out an “Ie” parameter does have some usefulness with regard to volume changes, e.g., opening a separate volume to the line for some analyses, and closing it for others. When such changes are made over a short time span (minutes to perhaps a day), this kind of correction factor makes sense. However, other factors that directly influence the mass spectrometer, such as power outages, electron multiplier and filament aging, etc. most certainly directly change the intrinsic mass spec sensitivity.
Line 355 – Don’t count helium out as a source of Durango dispersion. Some shards of Durango have been shown to produce anomalous gas spikes during CRH runs (see McDannell et al. (2018) and more recent publications by the Lehigh group.
Lines 365-369 and fig. 11b – Despite the apparent correlation, I’m concerning about the degree of coupling (or lack of coupling) between your sample packet and thermocouple. For fig. 11b it might be more instructive to plot temperature difference vs. temperature.
Lines 371-373 – Heat transfer at these lower temperatures is slow, and good coupling is required (see comment above). Additionally, restricting temperature measurement to >350°C effectively eliminates using the system for kinetic parameter determination, since most of the useful information comes from lower temperatures.
Lines 385-388 – Have you demonstrated that the active gas contaminants don’t suppress the 4He signal in your experiments (see Idleman et al., 2018)? Additionally, Durango apatite generally produces little of this contamination – have you analyzed any unknow samples that show more contamination?
Lines 390-404 – As pointed out by Idleman et al., (2018), gas consumption rates are highly dependent upon partial pressure, sometimes in a complex and non-linear way. Applying consumption rate corrections derived from a large 3He signal to 4He signals several orders of magnitude smaller is certain to lead to incorrect results. I suspect a significant number of your misbehaving samples would prove to be fine with appropriate corrections applied. Reducing the size of your 3He spike would help improve this situation.
Lines 440-441 – This is an interesting relationship and deserves more treatment in the text.
Fig. 1 – Change French terms to English. Show “electric lens”.
Fig. 3 – For zone 2 label change “rampe” to “ramped”. In caption change “blows” to “release”.
Fig. 4 - For zone 2 label change “rampe” to “ramped”. It might also be useful to show the 3He beam evolution and how the gas consumption was corrected. For fig. 4c, the convention is to show f vs. temperature. You might need a separate plot for that.
Fig. 7b – Change “caraterisation” to “characterization” or possibly “characterization”.
Fig. 9b – What are the “R.H.” segments?
Fig. 11b – Show as temp. difference vs. temperature.
Bruce Idleman
Citation: https://doi.org/10.5194/gchron-2024-6-RC4 -
AC3: 'Final response to the reviewers', Alexis Derycke, 24 May 2024
The manuscript Gchron 2024-6 was reviewed by four different reviewers (R2: James Metcalf, R4: Bruce Idleman and R1-3 anonymous), who are greatly thanked for the time they took to do so and for the extensive comments and recommendations they provided. These reviews contain similar or related comments, and in the following we will attempt to address and clarify those general comments.
Please find attached to this comment a .pdf files including the general responses.
To supplement this overarching response, if the manuscript is accepted for review, individual and precise replies will be produced as responses to each review on the Gchron 2024-6.
-
AC3: 'Final response to the reviewers', Alexis Derycke, 24 May 2024
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 574 | 166 | 31 | 771 | 31 | 13 | 12 |
- HTML: 574
- PDF: 166
- XML: 31
- Total: 771
- Supplement: 31
- BibTeX: 13
- EndNote: 12
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Alexis Derycke
Nathan Cogné
Dominique Bavay
David Vilbert
Lionel Dutruch
Sebastien Ternois
Gilles Ruffet
Marc Jolivet
Kerry Gallagher
This preprint has been withdrawn.
- Preprint
(2087 KB) - Metadata XML
-
Supplement
(1916 KB) - BibTeX
- EndNote