Investigation of the 176Yb Interference Correction during Determination of the 176Hf/177Hf Ratio by Laser Ablation and Solution Analysis on the Neoma MC-ICP-MS
Round 1
Reviewer 1 Report
The manuscript conducted three kinds of Yb-correction models to both laser ablation analysis on zircons and solution analysis on the newly released NeomaTM MC-ICP-MS platform. Their result showed that the NeomaTM platform can be eligible for Hf isotope analysis by the empirical approach like the NeptuneTM platform.
To my knowledge, the manuscript is suitable for publication in minerals after some modifications.
Detailed suggestions are listed in below:
Line 32:
Here there is a duplicated “TM”.
Line 116-117:
The N2 flow at ~0.4L/min (400mL/min)? In Hu et al. (2012), the N2 flow is much smaller (<10mL/min). I wonder whether it is a typo here.
If possible, please point out the effect of the N2 addition and compare the effects with N2 and without N2 addition.
Line 126:
It might be a waste of the faraday cups. Neoma has 11 cups, thus there is no difficulty to collect the 171Yb~174Yb beams (from 171-180).
Line 135:
I think it would better change the “178Hf/177Hf=1.46720082” to “178Hf/177Hf=1.46720” thinking of the precision of instruments.
Line 173-174:
The authors using the 173Yb/171Yb to calculate the beta Yb here. Why don’t use the 173Yb/172Yb or 174Yb/171Yb?
Again, since Neoma has 11 faraday cups by standard, there is no difficulty to collect the 171Yb~174Yb beams. After that, the effects using different ratios to calculate beta Yb can be compared.
Line 181:
Miss a “)” after "176Lu(calc)".
Line 277, Figure 5b
There is quite a different of beta Hf when the Hf voltage is the same. Why? I didn’t catch any interpretation to this in the text.
All the figures:
The figures are all needed to be redrawn and aligned. At lease put a, b or c in the corner for readers as the index.
Author Response
Reviewer #1 (R1) Comments & Author Responses (AR):
R1: The manuscript conducted three kinds of Yb-correction models to both laser ablation analysis on zircons and solution analysis on the newly released NeomaTM MC-ICP-MS platform. Their result showed that the NeomaTM platform can be eligible for Hf isotope analysis by the empirical approach like the NeptuneTM platform.
To my knowledge, the manuscript is suitable for publication in minerals after some modifications.
AR: We thank the reviewer for taking the time to provide a detailed review and appreciate his/her opinion that the manuscript is suitable for publication after revision.
R1: Detailed suggestions are listed in below:
AR: The reviewer is thanked for providing these detailed comments; we have endeavored to incorporate all of them.
R1: Line 32: Here there is a duplicated “TM”.
AR: Corrected
R1: Line 116-117: The N2 flow at ~0.4L/min (400mL/min)? In Hu et al. (2012), the N2 flow is much smaller (<10mL/min). I wonder whether it is a typo here. If possible, please point out the effect of the N2 addition and compare the effects with N2 and without N2 addition.
AR: This was indeed a typo; we used 4 mL/min. This has been corrected in the revised version. Thank you for pointing this out.
R1: Line 126: It might be a waste of the faraday cups. Neoma has 11 cups, thus there is no difficulty to collect the 171Yb~174Yb beams (from 171-180).
AR: While it is true that there are 11 faraday cups, our instrument is equipped with the ‘nuclear package’ which means that additional SEM and CCD detectors are interspersed with the faraday detectors on the low side of the axial cup. This introduces some cup spacing issues that make it impossible to sequentially collect 11 different isotopes with single mass unit spacing.
R1: Line 135: I think it would better change the “178Hf/177Hf=1.46720082” to “178Hf/177Hf=1.46720” thinking of the precision of instruments.
AR: This has been changed.
R1: Line 173-174: The authors using the 173Yb/171Yb to calculate the beta Yb here. Why don’t use the 173Yb/172Yb or 174Yb/171Yb? Again, since Neoma has 11 faraday cups by standard, there is no difficulty to collect the 171Yb~174Yb beams. After that, the effects using different ratios to calculate beta Yb can be compared.
AR: As part of our data processing, we did experiment with utilizing different Yb isotope ratios to apply the interference correction. However, we found that the issue with trying to use the observed Yb beta factors persisted regardless of which Yb isotope pairs were used. Furthermore, we did not notice any trends that would lead to meaningful comparisons between different beta Yb calculations. For example, for the analyses that resulted in an accurate 176Hf/177Hf when utilizing the observed beta Yb, the choice of which Yb isotope pairs were utilized to determine the beta factor had very little effect. In contrast, for the analyses which required the empirical beta Yb determination to produce and accurate 176Hf/177Hf, utilizing different Yb isotope pairs did not ameliorate the resultant 176Hf/177Hf. Additionally, we felt that the 173Yb/171Yb natural ratio was the best constrained (based on the literature) which is why we utilized it. An important point, however, is that all of the raw data is found in the electronic appendix so a reader could theoretically apply their own correction regimes to perform their own investigation of the Yb beta factor calculation.
R1: Line 181: Miss a “)” after "176Lu(calc)".
AR: Thank you for pointing this out; the parenthesis has been added.
R1: Line 277, Figure 5b There is quite a different of beta Hf when the Hf voltage is the same. Why? I didn’t catch any interpretation to this in the text.
AR: This is a good observation by the reviewer, but we do discuss this in the text (lines 317 – 326 as well as lines 345 – 357). We believe that subtle differences in the instrument setup and configuration between the different sessions may be responsible for the subtle differences in the mass bias behavior. However, as described in the text, further work specifically targeting these questions will need to be done.
R1: All the figures: The figures are all needed to be redrawn and aligned. At lease put a, b or c in the corner for readers as the index.
AR: We have re-drafted all of the figures. We have included legends (where appropriate), and rescaled many of the figures (also in response to the comments from the other reviewers).
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This manuscript presents new Hf isotopic data, on both standard solutions as well as well-known zircon standards via laser ablation, using the new NeomaTM MC-ICP-MS instrument provided by ThermoFisher Scientific. The main goal of the manuscript is to investigate different Yb isotopic corrections needed to produce accurate Hf isotopic data via laser ablation, while comparing this new instrument platform to its predecessor, the NeptuneTM MC-ICP-MS. This initial dataset is a welcome contribution for future users that will be utilizing the NeomaTM platform, however there are some important modifications and/or clarifications that need to be made before publication.
1) Variability in laser ablation data needs further interpretation.
The laser ablation data requires more investigation and discussion. The authors use a nanosecond laser ablation system. These laser systems are known to impart a larger thermal imprint on the sample, well outside of the ablation pit itself. This thermal imprint can cause elemental fractionation and is exasperated at higher energy levels (e.g., the 8.14 J/cm2). While unlikely to make a huge impact on the conclusions here, this may explain some of the variability between laser sessions (rather than solely a change in signal intensity) and should be commented on.
Similarly, the two different power settings used in the laser ablation analyses are going to produce two very different ablation pits, with the higher energy analyses producing a much deeper ablation pit and more importantly a different pit aspect ratio (spot size vs depth) than the lower energy analyses. This alone can produce some of the variability also mentioned in the above paragraph, however it also brings up the issue of down-hole fractionation.
Based on the data reduction section, I assume that during a single laser ablation spot: each integration cycle is processed as its own data point (background subtraction, Hf beta calculation, Yb beta calculation, final ratio). I also assume that all these cycles are then averaged to produce a single value and uncertainty for each ablation spot? I think the final calculated Hf isotopic ratios are OK when calculated this way, but not the “final” calculated Hf and Yb beta values (e.g., Fig. 5).
How do the Beta values change over each cycle during the ablation, as the pit depth increases and the aspect ratio (spot size vs. depth) gets smaller? I suspect that the beta values change, but maybe not significantly, during the low energy ablations. However, the 8.14 J/cm2 energy is almost guaranteed to produce some significant downhole fractionation. The pit depth is going to be much deeper, and the aspect ratio will likewise suffer.
I think this deserves some discussion and/or an example of the calculated ratios and beta values over the duration of the ablation (for both the low and high energy ablations). Figure 5 may need to be updated to account for this or at a minimum a few sentences in the main text have to be added if there is any observed downhole fractionation in the ratios and/or beta values.
2) The figures throughout the manuscript need adjustments.
The detailed comments below go through each figure in the manuscript with comments and edits that need to be made. In general, each figure needs a legend and for that legend to be described in detail within the figure caption. Additionally, many of the figures have a Y-axis that is inappropriate for the data. For example, the data in figure 8C have Y values that only range from ~1.000 to ~1.002, however the Y-axis ranges from 0.992 to 1.002. This bunches up the data and doesn’t allow the reader to see at the necessary resolution.
3) There are a number of other instrument parameters on the MC-ICP-MS that need to be detailed.
With the major point of the manuscript being an investigation into the use of a new MC-ICP-MS system, there needs to be more details provided about the instrument parameters used during the analyses. Currently there is only a nebulizer gas flow rate. At a minimum, the RF power, cooling Ar flow, Auxiliary Ar flow, and mass resolution (likely Low Res, but should still be explicitly mentioned) should be listed. A statement about roughly how long the instrument was on and warming up before the analyses would be useful to future users as well. With other MC-ICP-MS instruments, instrument (and measurement) stability increases with the longer the warmup.
Minor comments throughout:
1) There should be a data table reporting the mean of measured values for each zircon standard. This table should also have the epsilon values, in addition to the ratios, as the precision of the epsilon values is mentioned a few times in the main text.
2) All of the equations in the main text are written in paragraph form and should be re-written with an equation editor and numbered accordingly.
Below are my detailed comments by line number:
In the main manuscript:
Line 48: The epsilon description should be written out in an equation editor format instead of or at least in addition to the written text.
Line 52: Change “comply” to “commonly”.
Line 57: Delete “time”.
Line 97-101: It’s mentioned in LINES 80-86 that the relationship between the Yb and Hf mass bias factors, as well as the best correction regime, changes not only between labs with the same instrumentation, but also between sessions. The laser sessions listed in Table 2 do not have the same parameters over multiple sessions, so the inter-session variability can’t be investigated. Session 2 & 3 of the solution work, however, can be compared. A large portion of the variability might be simply explained by lab environmental changes, so it would be helpful if the authors comment on the variability between these two solution sessions and how it can fit into the larger variability story.
Line 98: Change “several session and a variety” to “several sessions and from a variety”.
Line 126: The MC-ICPMS and Laser Ablation information (e.g., model, gas flows, power settings, etc.) should be presented in a table format as well as being included in the main text of section 2.1.
Line 128: Related to the above comment, what was the duration of the ablation for each spot analysis? I can guess based on the integration time, that these were 25-30s ablations… but it’s not clear, as these cycles may also include the gas blank measurements.
Table 2: The laser power needs units added.
Line 145: Delete “as well as”.
Lines 159-191: It would help the reader if the equations listed in this section are written out using an equation editor with each equation numbered. Although understandable, it’s difficult to work out all the variables and 3 data reduction techniques when the equations are written in a paragraph format.
Line 189: Change “176Yb/173Yb” to “176Yb/173Yb”.
Line 197: Change “These averages where” to “These averages were”.
Line 204: Is there a reason that standard error is used here rather than standard deviation? Since the laser standards are natural minerals with some variability, I would think that standard deviation would be a better choice.
Figure 1: The same symbol scheme should be used in this figure as all the others. There may be some additional variability between the laser session, but it’s not clear as the symbols are all the same. It would also be useful to add a weighted mean value (with statistics) for each standard and how that compares to the “true value”. This could be added as a small text box next to the data for each standard. Also, I’m assuming that the red lines are the “true value” of the standards? This should be mentioned in the figure caption.
Figure 2: Although mentioned in Table 2, a legend is needed for the different symbols used.
Figure 3: The current Y-axis does not allow for interpretation of the dataset. The data seem to cluster around 1.000 to 1.001, but these are rounded values (shown by each Y-axis value being repeated e.g., 0.999, 0.999). There needs to be an added decimal place. Additionally, the Y-axis could be zoomed in, as all the data seem to fall between 1.001 and 0.999. Although mentioned in Table 2, a legend is needed for the different symbols used.
Line 246-248: This statement is not evident, as mentioned above, as the Y-axis is does not allow this level of interpretation.
Line 256: Are there quantitative ways to support the statement that the empirical beta factor produces the most accurate ratios regardless of Yb signal? Fixing the Y-axis will help, but to the two methods seem to produce very similar results. Perhaps, calculating an average obs/true value for each correction method? There may also be some differences in the variability, which would be useful to see quantitatively.
Figure 4 a-c: Same comment as above for Figure 3.
Figure 5 a-b: As mentioned above, a legend is needed. Also, the figures and caption do not specify what correction method was used to produce the data that are graphically shown.
Figure 6: As mentioned above, a legend is needed.
Figure 7: The variability that occurs at ~1.5V 177Hf during session 3 is alarming. There is not much mention in the main text as to what may have been a cause of this. As mentioned above, a legend is needed.
Figure 8 a-c: All figures need a legend. Figure 8b and 8c need the Y-axis rescaled (zoomed in). Figure 8a is the same as Figure 6 and is not needed. Until the Y-axis are rescaled, it is difficult to assess which correction method is better. This would again be aided by a quantitative comparison.
Line 300-302: There is no Figure 9b (Yb beta factors) included in the manuscript file.
Supplement
- If the final version of the published supplement is going to be a single Excel workbook, then it would be beneficial to have the first sheet of the workbook be a description of the contents. A few sentences describing what each sheet is going to be reporting. This can also be where the conditions for each laser session are re-reported (i.e., which session is the higher fluence?).
- The uncertainties need to be labelled. Are these standard error or standard deviation… 1-sigma or 2-sigma?
Author Response
Reviewer #2 (R2) Comments & Author Responses (AR):
R2: This manuscript presents new Hf isotopic data, on both standard solutions as well as well-known zircon standards via laser ablation, using the new NeomaTM MC-ICP-MS instrument provided by ThermoFisher Scientific. The main goal of the manuscript is to investigate different Yb isotopic corrections needed to produce accurate Hf isotopic data via laser ablation, while comparing this new instrument platform to its predecessor, the NeptuneTM MC-ICP-MS. This initial dataset is a welcome contribution for future users that will be utilizing the NeomaTM platform, however there are some important modifications and/or clarifications that need to be made before publication.
AR: We thank the reviewer for taking the time to make such detailed comments and suggestions. We also appreciate that the reviewer feels this will be an important contribution after revision.
R2: 1) Variability in laser ablation data needs further interpretation.
The laser ablation data requires more investigation and discussion. The authors use a nanosecond laser ablation system. These laser systems are known to impart a larger thermal imprint on the sample, well outside of the ablation pit itself. This thermal imprint can cause elemental fractionation and is exasperated at higher energy levels (e.g., the 8.14 J/cm2). While unlikely to make a huge impact on the conclusions here, this may explain some of the variability between laser sessions (rather than solely a change in signal intensity) and should be commented on.
Similarly, the two different power settings used in the laser ablation analyses are going to produce two very different ablation pits, with the higher energy analyses producing a much deeper ablation pit and more importantly a different pit aspect ratio (spot size vs depth) than the lower energy analyses. This alone can produce some of the variability also mentioned in the above paragraph, however it also brings up the issue of down-hole fractionation.
Based on the data reduction section, I assume that during a single laser ablation spot: each integration cycle is processed as its own data point (background subtraction, Hf beta calculation, Yb beta calculation, final ratio). I also assume that all these cycles are then averaged to produce a single value and uncertainty for each ablation spot? I think the final calculated Hf isotopic ratios are OK when calculated this way, but not the “final” calculated Hf and Yb beta values (e.g., Fig. 5).
How do the Beta values change over each cycle during the ablation, as the pit depth increases and the aspect ratio (spot size vs. depth) gets smaller? I suspect that the beta values change, but maybe not significantly, during the low energy ablations. However, the 8.14 J/cm2 energy is almost guaranteed to produce some significant downhole fractionation. The pit depth is going to be much deeper, and the aspect ratio will likewise suffer.
I think this deserves some discussion and/or an example of the calculated ratios and beta values over the duration of the ablation (for both the low and high energy ablations). Figure 5 may need to be updated to account for this or at a minimum a few sentences in the main text have to be added if there is any observed downhole fractionation in the ratios and/or beta values.
AR: This is an interesting point by the reviewer, and the questions raised regarding the effect of different laser power settings on the down-hole mass fractionation behavior are certainly valid. In response to this comment, we generated plots examining the cycle-by-cycle beta factors for a subset of the analyses conducted throughout this study. However, we did not see any systematic shifts in the calculated beta factor as a function of depth. We did observe that the cycle-by-cycle Yb beta factors are significantly more scattered than the Hf beta factors within a given analysis. However, this is the expected result given that the Yb beta factor is generally poorly constrained (hence the need to use the empirical relationship to produce accurate 176Hf/177Hf ratios) as already explained in the paper. We have decided not to include a plot of the down-hole beta factors since there were not discernable trends other than the one already discussed in the text (namely that beta Yb is poorly constrained). If the reviewer feels strongly that inclusion of some representative plots of the down-hole beta factors are necessary, then we can certainly do this to get the paper published. However, we also want to point out that the raw data (e.g. cycle-by-cycle for each analysis) is found in the electronic appendix. Therefore, a reader specifically interested in the effect of different laser conditions on down-hole mass bias relationships can easily generate these types of plots. Because we did not see any trends indicating strong shifts in the mass bias regimes as a function of depth in the pit (regardless of the laser power), we did not want to add another figure. However, as requested by the reviewer, we did add a statement explaining that we did not observe any systematic shifts in the calculated beta factors with increasing pit depth regardless of the laser condition utilized (lines 326 – 328).
R2: 2) The figures throughout the manuscript need adjustments.
The detailed comments below go through each figure in the manuscript with comments and edits that need to be made. In general, each figure needs a legend and for that legend to be described in detail within the figure caption. Additionally, many of the figures have a Y-axis that is inappropriate for the data. For example, the data in figure 8C have Y values that only range from ~1.000 to ~1.002, however the Y-axis ranges from 0.992 to 1.002. This bunches up the data and doesn’t allow the reader to see at the necessary resolution.
AR: This is a valid point. We have completely redrafted the figures. We have added a legend for both the laser and solution data in figures 2 (for laser) and 6 (for solution), and then referenced this legend in subsequent figures as appropriate. We have also re-scaled the Y axis of many of the figures to ensure that the individual data points can be resolved. We hope that the reviewer is pleased with the changes and will certainly be happy to make additional changes if requested.
R2: 3) There are a number of other instrument parameters on the MC-ICP-MS that need to be detailed.
With the major point of the manuscript being an investigation into the use of a new MC-ICP-MS system, there needs to be more details provided about the instrument parameters used during the analyses. Currently there is only a nebulizer gas flow rate. At a minimum, the RF power, cooling Ar flow, Auxiliary Ar flow, and mass resolution (likely Low Res, but should still be explicitly mentioned) should be listed. A statement about roughly how long the instrument was on and warming up before the analyses would be useful to future users as well. With other MC-ICP-MS instruments, instrument (and measurement) stability increases with the longer the warmup.
AR: This information has been added to section 2.1; lines 129 – 132. We did not address the warmup time as this is somewhat subjective and it is implied that an instrument operator would not start making analyses until they were satisfied that the instrument was behaving in a stable manner.
R2: Minor comments throughout:
R2: 1) There should be a data table reporting the mean of measured values for each zircon standard. This table should also have the epsilon values, in addition to the ratios, as the precision of the epsilon values is mentioned a few times in the main text.
AR: We have now added this information to summary figure 1, where the average 176Hf/177Hf and corresponding population standard deviations for each standard are reported with the plot of the individual analyses and their uncertainties. If the reviewer feels strongly that this warrants its own table, we can certainly accommodate. However, now that this information is presented in figure 1 we feel that adding a table would be redundant.
R2: 2) All of the equations in the main text are written in paragraph form and should be re-written with an equation editor and numbered accordingly.
AR: Because none of the equations are new (i.e. we did not derive them ourselves), we feel that it is more appropriate to simply list them in paragraph form. We feel that the formatting of the equations in the templated manuscript does not detract from the reader’s ability to read and understand the equations. However, if acceptance of the manuscript hinges on putting the equations into a different format, then of course we will accommodate this. However, listing the equations in the text is consistent with the way that other papers covering this topic have handled the equations except in the cases where new data processing mathematical expressions are being presented.
R2: Below are my detailed comments by line number:
In the main manuscript:
AR: The reviewer is thanked for providing such detailed comments. We have endeavored to incorporate all of them.
R2: Line 48: The epsilon description should be written out in an equation editor format instead of or at least in addition to the written text.
AR: See our comment above; we do not feel that the equations used in this paper warrant this treatment because they are not new equations.
R2: Line 52: Change “comply” to “commonly”.
AR: Corrected; thanks.
R2: Line 57: Delete “time”.
AR: Deleted.
R2: Line 97-101: It’s mentioned in LINES 80-86 that the relationship between the Yb and Hf mass bias factors, as well as the best correction regime, changes not only between labs with the same instrumentation, but also between sessions. The laser sessions listed in Table 2 do not have the same parameters over multiple sessions, so the inter-session variability can’t be investigated. Session 2 & 3 of the solution work, however, can be compared. A large portion of the variability might be simply explained by lab environmental changes, so it would be helpful if the authors comment on the variability between these two solution sessions and how it can fit into the larger variability story.
AR: This seems to be a case where the reviewer may have commented a bit pre-emptively since we do actually have several figures dealing with the inter-session mass bias factor relationships (figure 5a-b for the laser ablation analyses and figure 9a-b for the solution mode analyses). We also cover the mass bias behavior in sections 4.2 and 4.3.
R2: Line 98: Change “several session and a variety” to “several sessions and from a variety”.
AR: This has been changed.
R2: Line 126: The MC-ICPMS and Laser Ablation information (e.g., model, gas flows, power settings, etc.) should be presented in a table format as well as being included in the main text of section 2.1.
AR: As stated by the reviewer, this information is already present in the text (Lines 112 – 117). Our opinion is that adding another table would be redundant. However, if adding another table is required for acceptance, then we will of course accommodate this request.
R2: Line 128: Related to the above comment, what was the duration of the ablation for each spot analysis? I can guess based on the integration time, that these were 25-30s ablations… but it’s not clear, as these cycles may also include the gas blank measurements.
AR: We now specify that each ablation lasted 30 seconds (e.g. the duration of time the laser was actually firing onto the surface of the sample). This can be found in lines 125 – 127.
R2: Table 2: The laser power needs units added.
AR: This has been added; the reviewer is thanked for pointing out this omission.
R2: Line 145: Delete “as well as”.
AR: We have replaced ‘as well as’ with ‘and’.
R2: Lines 159-191: It would help the reader if the equations listed in this section are written out using an equation editor with each equation numbered. Although understandable, it’s difficult to work out all the variables and 3 data reduction techniques when the equations are written in a paragraph format.
AR: We feel that reporting the equations in paragraph form is appropriate considering that we have not derived any new equations. If the reader is interested in seeing the original derivation of the equations, we have provided the citations where they can go to find the original treatment. We feel that presenting the equations in their own figure (or using an equation editor) might give the appearance that we somehow derived these equations. As stated by the reviewer, the mathematical expressions are understandable and can be easily followed in their current format. If this is going to be a sticking point for getting the paper accepted, then we will certaintly explore other options for depicting the equations.
R2: Line 189: Change “176Yb/173Yb” to “176Yb/173Yb”.
AR: This formatting issues has been corrected; the reviewer’s attention to detail is greatly appreciated.
R2: Line 197: Change “These averages where” to “These averages were”.
AR: This has been corrected. The reviewer’s attention to detail is greatly appreciated.
R2: Line 204: Is there a reason that standard error is used here rather than standard deviation? Since the laser standards are natural minerals with some variability, I would think that standard deviation would be a better choice.
AR: Typically the within-run error (e.g. the scatter of various cycles making up an individual analysis) is represented by the standard error of the mean whereas the standard deviation is utilized for depicting the scatter of the population (e.g. the scatter of the individual analyses of a particular standard within a given session). We do utilize the standard deviation to depict the degree of scatter observed for each of the zircon standards that were measured over the course of the study (see figure 1).
R2: Figure 1: The same symbol scheme should be used in this figure as all the others. There may be some additional variability between the laser session, but it’s not clear as the symbols are all the same. It would also be useful to add a weighted mean value (with statistics) for each standard and how that compares to the “true value”. This could be added as a small text box next to the data for each standard. Also, I’m assuming that the red lines are the “true value” of the standards? This should be mentioned in the figure caption.
AR: We have now modified figure 1 to include the standard deviation associated with the various standards. We have also made some additional cosmetic modifications to the figure. If the reviewer would like to see additional changes prior to publication, we can certainly accommodate if necessary.
R2: Figure 2: Although mentioned in Table 2, a legend is needed for the different symbols used.
AR: A legend has now been added to figure 2.
R2: Figure 3: The current Y-axis does not allow for interpretation of the dataset. The data seem to cluster around 1.000 to 1.001, but these are rounded values (shown by each Y-axis value being repeated e.g., 0.999, 0.999). There needs to be an added decimal place. Additionally, the Y-axis could be zoomed in, as all the data seem to fall between 1.001 and 0.999. Although mentioned in Table 2, a legend is needed for the different symbols used.
AR: We have now re-scaled the figure. We also refer to the new legend that has been added to figure 2 (same symbols).
R2: Line 246-248: This statement is not evident, as mentioned above, as the Y-axis is does not allow this level of interpretation.
AR: Now that the figure has been re-scaled, we feel that this statement is better supported.
R2: Line 256: Are there quantitative ways to support the statement that the empirical beta factor produces the most accurate ratios regardless of Yb signal? Fixing the Y-axis will help, but to the two methods seem to produce very similar results. Perhaps, calculating an average obs/true value for each correction method? There may also be some differences in the variability, which would be useful to see quantitatively.
AR: Now that the figure Y-axes have been re-scaled, it is very evident that the use of the internally determined Yb mass bias factor produces highly erroneous data compared to the empirical relationship.
R2: Figure 4 a-c: Same comment as above for Figure 3.
AR: We have re-scaled the Y-axes of the figures and feel that this makes interpretation of the data much clearer.
R2: Figure 5 a-b: As mentioned above, a legend is needed. Also, the figures and caption do not specify what correction method was used to produce the data that are graphically shown.
AR: We have now added a legend to figure 2, which is also now referenced in the caption of figure 5 (since it’s the same symbols). An important note is that figure 5 depicts the independently calculated Yb and Hf beta factors and not the final 176Hf/177Hf, so the comment about a correction method does not apply. We specify in the figure caption that these are the Yb and Hf beta factors.
R2: Figure 6: As mentioned above, a legend is needed.
AR: A legend has now been added to this figure (since this is now the solution data, not the laser data). This legend is also now referenced in the subsequent figures presenting the solution data.
R2: Figure 7: The variability that occurs at ~1.5V 177Hf during session 3 is alarming. There is not much mention in the main text as to what may have been a cause of this. As mentioned above, a legend is needed.
AR: We honestly do not know what the cause of this variability is, but we do speculate in the text about the various parameters that warrant further examination as the method becomes more widely implemented on the new platform. The legend in figure 6 is now referenced in the caption of figure 7.
R2: Figure 8 a-c: All figures need a legend. Figure 8b and 8c need the Y-axis rescaled (zoomed in). Figure 8a is the same as Figure 6 and is not needed. Until the Y-axis are rescaled, it is difficult to assess which correction method is better. This would again be aided by a quantitative comparison.
AR: We have re-scaled the Y-axes of the figures to make it easier for the reader to pull out the relationships between the different datasets described in the text. We feel that the graphical representation of the data is now much stronger and clearer, so the addition of quantitative comparisons is not necessary (e.g. it is abundantly clear from the relationships on the figures that certain correction regimes produce erroneous final calculated values). Also, the legend added to figure 6 has been referenced in the caption of this figure.
R2: Line 300-302: There is no Figure 9b (Yb beta factors) included in the manuscript file.
AR: This appears to have been a mistake introduced during templating of the manuscript. We have now ensured that the figure is present in the revised version and apologize for this.
R2: Supplement
If the final version of the published supplement is going to be a single Excel workbook, then it would be beneficial to have the first sheet of the workbook be a description of the contents. A few sentences describing what each sheet is going to be reporting. This can also be where the conditions for each laser session are re-reported (i.e., which session is the higher fluence?).
AR: A summary sheet has been added to the workbook. The names of the individual worksheets have also been slightly modified for clarity.
R2: The uncertainties need to be labelled. Are these standard error or standard deviation… 1-sigma or 2-sigma?
AR: We now specify in the summary sheet that the uncertainties on the isotope ratios are the 1 sigma standard error. This was explained in the paper but we agree that adding this to the summary sheet provides additional clarity.
Author Response File: Author Response.pdf
Reviewer 3 Report
see file attached
Comments for author File: Comments.pdf
Author Response
Reviewer #3 (R3) Comments & Author Responses (AR):
R3: I have read the manuscript and have few commentaries to enhance the manuscript quality. Certainly, the results deserve to be published.
AR: The reviewer is thanked for providing these comments, as well as for his/her opinion that the results are deserving of publication.
R3:Corrections (in yellow the phrase to be corrected).
Abstract
Page 1, Line 13: improve the spelling.
Neptune™™
AR: We have corrected the double ‘TM’ symbol. We cannot find any other issue in this line.
R3:Page 1, Line 15: improve the language
with higher signal analyses being able produce...
AR: We have changed this to ‘producing’.
R3:Introduction
Page 1. Line 32. improve the spelling.
Neptune™™
AR: We have removed the double ‘TM’ symbol.
R3:Methods
Page 3, Lines 104-107. Why to include 180Hf isotope in the cup configuration?
... For the two different cup configurations, the only difference is that one configuration
included 180Hf on the ‘highest’ faraday cup whereas the other configuration did not....
AR: This is a valid question; As this is an entirely new system and our was the first one to be installed in the USA, there are many issues that need to be worked out. We were having a minor issue with the cup movement that prevented us from getting 180Hf aligned properly during one of the sessions. The issue ended up being a slight bug in the software that controls the cup motors and was eventually resolved by Thermo. Because we do not use this isotope as part of the data processing, we decided that not having this isotope during one of the sessions did not in any way detract from the observations and conclusions presented in the paper.
R3:Page 2, Line 116. The sample gas flow of about 1L/min. How about the influence of He flow
on the signal intensity?
AR: We now specify that we tuned for maximum signal intensity while aiming for the lowest oxide production rate.
R3: During laser ablation analyses, the sample gas flow was ~1 L/min He…
AR: This is correct.
R3: Page 2, Line119. Sample gas?
AR: We now specify that this is the sample gas.
R3: A nebulizer gas flow of ~1.1 L/min
AR: This is correct.
R3: Table 1
In L1 Faraday detector there are two isotopes (174Hf and 174Yb)
AR: This is a valid point and why we don’t actually use the 174Yb to do any calculations. While the 174Hf interference should be minor, it still may skew the results. Many labs do not even collect the 174Yb.
R3: Table 2.
Int.(s) is integration time?
AR: This is correct; we now specify this in the caption of table 2.
R3: Page 4, Line 142. Please, better explain the three standards utilized.
For the solution analyses, three standards were utilized…
AR: We have added some additional description and clarification of the three standards utilized during the solution measurements.
R3: Results
Page 4, Lines 168, 171, 172 and 174
Please, explain the M in the equation (M179Hf = 178,945815 and M177Hf = 176,94322), etc.
beta Hf’ = LN(179Hf/177Hf(true)/179Hf/177Hf(obs))/LN(M179Hf/M177Hf).
AR: We now specify that the ‘M’ denotes the isotopic mass of each isotope.
R3: Page 4, Lines 178, 179, and 182
Again. Please, explain the M in the equation (M173 = 172,938207 and /M176= 175,942568).
176Yb(calc) = 173Yb(obs) x 176Yb/173Yb(true) x (M173Yb/M176Yb)^(beta Hf)
AR: We now specify that ‘M’ denotes the isotopic mass in the preceding paragraph. If the reviewer feels that mentioning this again would provide additional clarity, we can certainly accommodate if necessary. We also mention this again in line 195.
R3: Page 4, Line 172. Misspelling
. . The observed…
AR: We have checked this line and cannot find the misspelling (observed seems to be spelled correctly).
R3: Page 5, Lines 188-191. Discuss why to use the Patchett and Tastumoto (1981) values
instead Thirlwall (2004) and Segal et al., (2003).
The true values utilized in this study are 179Hf/177Hf = 0.7325 (Patchett and Tastumoto 1981), 176Yb/173Yb = 0.79618, 173Yb/171Yb = 1.132685, and 176Lu/175Lu = 0.02655 (all from sources reported in Fisher et al 2011). The atomic masses are those reported in Baum et al
2002.
M.F. Thirlwall, R. Anczkiewicz, Multidynamic isotope ratio analysis using MC-ICP-MS
and the causes of secular drift in Hf, Nd and Pb isotope ratios, Int. J. Mass Spectrom. 235
(2004) 59 – 81
Segal, I., Halicz, L. & Platzner, I.T. 2003. Accurate isotope ratio measurements of
ytterbium by multiple collection inductively coupled plasma mass spectrometry applying
erbium and hafnium in an improved double external normalization procedure. Journal of
Analytical Atomic Spectrometry, 18: 1217–1223.
AR: This is a valid question, but one that we feel is not necessarily within the scope of this paper. For example, in the Segal et al paper it is stated that ‘(179Hf/177Hf)corr = 0.732 49(2) was obtained, in very good agreement with the IUPAC value of 0.732 50’; basically meaning that their value is well within uncertainty of the IUPAC value (which is the same as the Patchett and Tastumoto value utilized here). The issue with the Yb isotope ratios is a bit more complex as there has been some debate in the literature over the actual natural isotope ratios of Yb. However, we specify in the paper that the sources of the values utilized are based on the work of Fisher et al 2011, which contains a discussion of the rational for using the various available Yb isotopic compositions in the context of the zircon laser ablation Hf isotope analytical method. There have actually been a few studies that examine the effect of utilizing the different Yb isotopic compositions on the final calculated 176Hf/177Hf ratios- in practice, the effect is quite small. We do not want to delve into this topic in this paper since it is not the focus of this study.
R3:Discussion
AR: Page 5: Lines 204. Please, specify the line and include in the Figure caption.
the 176Hf/177Hf ratios (± 1 se) for each of the laser ablation analyses are shown relative to the
accepted values of the various zircon standards.
Page 6 Figure 1: Please, cite the line (accepted values) in the caption.
AR: We have now added this information to the caption of figure 1 (was already present in the text in section 2.2).
R3: Page 7 Figure 3: This figure is not cited in the text.
AR: The reviewer is thanked for pointing this omission out. We have added the reference to figure 3 to line 225 where it is appropriate (was supposed to be there but somehow got left out).
R3: Page 7 Line 236: Figure 6 is cited before Figures 4 and 5. Page 7 Line 244: Figure 8 is cited before Figure 7.
AR: We thank the reviewer for pointing this out. Our preference was to present all of the laser ablation data in a series of figures, followed by all of the solution data. Unfortunately, this means that a few of the figures do not actually get mentioned in the text in the correct order in which they appear. While we recognize that this is not ideal, the alternative is that the figures presenting laser ablation data would get intercalated with the figures presenting the solution data which would potentially result in confusion for the reader. We believe that the presentation of the figures is logical and therefore our preference is not to change this. However, if this becomes a sticking point we can certainly add a few sentences at various points in the manuscript to make sure that the figures are called out in the same order they appear.
R3: Page 8. Figure 4a, b and c. Please, label a, b and c.
AR: This has been added
R3: Page 9. Figure 5a and b. Please, label a and b.
4
AR: This has been added
R3: Page 10. Figure 7: This figure is not cited in the text.
AR: We apologize for this oversight; we have added the reference to figure 7 where it was originally intended (section 4.1).
R3: Pages 11-12. Figures 8a, b and c. Please, label a, b and c.
AR: This has been added
R3: Figure 9a: This figure is missing. Please, label a and b.
AR: This has been corrected; a & b have also been added.
R3: Avenues for Future Study
Page 13, Line 313: References? The manuscript doesn’t present Neptune results.
AR: Because of the fact that there are thousands of papers presenting Neptune results, our preference in this paper was to generalize some of the existing observations of laser ablation zircon Hf data collected on the Neptune. While we could certainly add some Neptune data to the paper, we feel that it may complicate presentation of the Neoma data due to the simple fact that the Neptune data is highly variable in terms of how the interference correction is treated, instrument conditions, etc (as discussed in the text).
R3: The data collected in this study indicate that, broadly speaking, the Neoma™ performs
similarly to the Neptune™™ series MC-ICP-MS with regard to the Yb interference
correction.
AR: This is correct. We have removed the extra ‘TM’.
R3: General Comments
1) The manuscript is an important contribution since the Neoma will replace Neptune.
But the authors should discuss or present in the introduction what They think about
Nu instrument ICP-MC-MS Lu-Hf results.
AR: We now specify in the introduction that the other commercially available MC-ICP-MS system is offered by Nu™. The new system is certainly good and may even have some advantages over the thermos systems. However, we do not want to turn this manuscript into a sales pitch for either platform and therefore will reserve discussion regarding the Nu system.
R3: 2) LA-ICP-MS technique is very well explored to investigate Lu and Yb
interferences. What about the importance of tuning the instrument for low oxides?
AR: Oxide production is certainly an issue (for nearly all ICP-MS analytical regimes). We now specify in the methods section that we tuned to achieve the lowest observed oxide signals. The issue of oxides relevant to the zircon Hf laser ablation method has also been explored in several other papers, so we do not want to shift the focus of this paper onto the issue of dealing with oxide interferences.
R3: 3) The new machine allows to investigate new techniques on new minerals moving
forward with low-volume Lu-Hf analyses?
AR: Ideally that will be possible but for now we opted to use a conventional approach.
R3: 4) Missing references
Nan-Chin Chu, Rex N. Taylor, Valerie Chavagnac, Robert W. Nesbitt, Rose M. Boella, J.
Andrew Milton, Christopher R. German, Germain Bayon and Kevin Burton. 2002 Hf isotope
ratio analysis using multi-collector inductively coupled plasma mass spectrometry: an
evaluation of isobaric interference corrections. J. Anal. At. Spectrom. 17, 1567–1574
Ann M. Bauer, Matthew S.A. Horstwood (2018) small-volume Lu-Hf and U-Pb isotope
determination of complex zircons by solution and laser ablation MC-ICP-MS. Chemical
Geology, 476 85–99
5
Arnaud Goolaerts, Nadine Mattielli, Jeroen de Jong, Dominique Weisa, James S. Scoates Hf
and Lu isotopic reference values for the zircon standard 91500 by MC-ICP-MS. Chemical
Geology 206 (2004) 1 – 9.
Janne Blichert-Toft 2001 On the Lu-Hf Isotope Geochemistry of Silicate Rocks.
GEOSTANDARDS NEWSLETTER. The Journal of Geostandards and Geoanalysis. Vol. 25
— N°1 p.41-56
Lapen, Thomas & Mahlen, Nancy & Johnson, CM & Beard, BL. (2004). High precision Lu
and Hf isotope analyses of both spiked and unspiked samples: A new approach.
Geochemistry Geophysics Geosystems. 5. 10.1029/2003GC000582.
AR: We thank the reviewer for providing these suggested references. We have incorporated the Goolaerts et al reference into the paper because is provides additional information on the 91500 standard. We have also added the Bauer and Horstwood reference since it provides context for where the frontiers of this analytical method currently lie (see lines 389 – 392). We have chosen not to include the other references for the common reason that many of them present information that has since been incorporated into more recent studies that are already cited in the paper. For example, the very important early discussion of Yb mass bias correction regimes by Chu et al 2012 has evolved considerably and is now captured by the more recent references already presented in the paper. Some of the other papers are not necessarily relevant to this study, for example, the Lapen et al paper focuses on how to achieve chemical separation of rock and mineral samples for Lu-Hf isotope analysis whereas our paper is focused solely on the actual isotope ratio measurement.
Round 2
Reviewer 1 Report
The authors answered all my comments in the last round.
Though the figures have been updated, they are not friendly enough to publish. I suggest the authors to put the needed figures in a Powerpoint page to align them and then out put the figure (e.g. Figure 4, 5, 8 and 9). Thus, a minor revision is needed.
