A Comparison of Different Natural Groundwaters from Repository Sites—Corrosivity, Chemistry and Microbial Community

Round 1

Reviewer 1 Report

The current manuscript is devoted to investigation whether groundwater environment effects steel containers for the radioactive waste. Importance in choosing a right location is of immediate concern for the public and scientific community. The authors examine chemistry and microbial composition of groundwater samples, taken from sites of interest.  Different types of steel are exposed to groundwater taken from different sites. Subsequently, corrosion during this exposure is compared to the one in simulated groundwater. As a result, expected effect of high chloride on the development of corrosion has been confirmed. However, the mechanisms driving the corrosion are not clear. Additionally, quantification and sequencing of microbial 16S rRNA has been performed in order to show that different sites do differ in microbial composition. In general, electrochemical experiments and sequencing results create a strong background for a sound scientific contribution, however manuscript requires substantial improvements First, experimental design should be presented more clearly. Some graphs should be changed: It is not appropriate in a scientific manuscript to show 1,0E+7 instead of 10⁷.  The corrosion experiments and description of microbial community seem to be like two separate stories. More attention should be devoted on discussion how microbial communities from the sites of interest might influence different types of steel used in the study. Which scenarios are expected? Are microorganisms expected to produce sulfate, to scavange, to changes the sulfate/chloride ratio? If no bacterial activity has been measured during the electrochemical experiments and microorganisms have been neglected there, what was the reason for sequencing 16 S rRNA? What is typical microbial composition of the repositories sites worldwide (some examples would be appreciated)? In general, is it a typical practice to deposit radioactive waste in this kind of environments with this kind of chemical composition (this could be a good piece of information for the introduction). In the end, which site would be better for storage of radioactive waste?

Below are more detailed comments per lines.

Line 34. Spelling mistake. Decomposition?

Line 47. Here and further below, it is assumed that 16S rRNA gene copies directly correspond to microbial cell numbers, which is not the case. Here is also given a reference to a paper of Haveman et al., 2002. There qPCR was not performed! Cell numbers were determined via epifluorescence microscopy and concentration was expressed not in copies in ml per in cells per ml. Please, add here a reference to a paper, if known, that gene copies are equal to cell numbers or please, refer to an appropriate estimation of microbial cell numbers based on qPCR.

Lines 54-55. The SO42-/Cl- ratio causing inhibition increases with increasing Cl-concentration and temperature of the groundwater but decreases with alloying of the steel. This sentence is confusing. Please, rephrase.

Line 60. Pdease add reference here.

Lines 63-65. It would be nice to have graphical presentation of locations. Where those sites are located? Is it allowed to give this information? If not, please, indicate so and try at least give more details on the sites.  Simulated groundwater – please, give details how it was prepared? What is also prepared under anoxic conditions? How?

Line 97. “from triplicate 1-3.3 l”. What does it mean? When was 1 L used, when was it 3.3? Does it mean that not equal amount of water has been pumped for each triplicate?

Line 189. It is not clear from description where groundwater sample was introduced? And how? How was it done? Under anaerobic conditions? Please, give hier in results section also details on replicate measurements.

Line 232. In general, it would be better to use instead of “waters” – groundwater samples, instead of “Water A” – Site A.  Why some values for simulated water are missing? Are there values for nitrate?

Line 240. Again. Please, not “waters” but groundwater samples from Sites…

Figure 1. Please, update the figure from the excel-like one to the more manuscript appropriate. Annotation of y-axis should scientific. Please, remove the header of the graph. You do not present the number of microorganisms here. Plus, there are no colors in this manuscript version. The graph is in the gray scale.

Table 3. Could this table be changed into graph for more obvious representation?

Figure 3. From annotation: what are different “water samples” and “water types”? Please, use terms groundwater samples and sites!

Line 302. Did you mean here Figure 4a instead of Fig 2b?

Figures 5,8,9 – please, use scientific annotations for x-axis

Figure 10. How many replicates? Please, indicate.

Line 448. What are possible mechanisms for driving the corrosion?

Line 465. The experiment could be done in the anaerobic chamber. Would that be possible?

Lines 466-473. Again here: in this study not microbes per ml were detected , rather gene copies per ml!Could you please, also refer to have many cells normally are there in the groundwaters and what at typical TOC? In the presetine groundwater TOC is usually around 0.2 – 0.1 mg per ml (which is lower than at your site C) and bacterial cell numbers are around 10⁴. Which is higher than at Site C. What could be further explanations there?

Lines 479-482. Could you, please, provide here more details on the Sulphur cycle. Do Sulfuricurum produce sulfate and sulfide from sulfur? To what proportion? What is about possible nitrate content in the sites presented in this study?

Line 528. Could you, please, provide more detail how bacteria/microbes can contribute to the sulfate content and subsequently to the corrosion? How it could be possible to decrease sulfate content knowing bacterial physiology?

Line 549. Could you, please, give suggestions on what it could be “something else”?

Author Response

We thank the reviewer for constructive comments towards our manuscript and feel that those comments helped to improve the quality of the manuscript.

The major changes made to the manuscript are highlighted in the manuscript.

To summarize the major changes in the manuscript:

• The data in Table 3 is converted to Figure 3
• More discussion is added e.g. hypothesis of corrosion mechanisms
• Technical improvements have been made in figures
• The experimental setup is now described in more detailed

Author Response ：

The current manuscript is devoted to investigation whether groundwater environment effects steel containers for the radioactive waste. Importance in choosing a right location is of immediate concern for the public and scientific community. The authors examine chemistry and microbial composition of groundwater samples, taken from sites of interest.  Different types of steel are exposed to groundwater taken from different sites. Subsequently, corrosion during this exposure is compared to the one in simulated groundwater. As a result, expected effect of high chloride on the development of corrosion has been confirmed. However, the mechanisms driving the corrosion are not clear. Additionally, quantification and sequencing of microbial 16S rRNA has been performed in order to show that different sites do differ in microbial composition. In general, electrochemical experiments and sequencing results create a strong background for a sound scientific contribution, however manuscript requires substantial improvements First, experimental design should be presented more clearly. Some graphs should be changed: It is not appropriate in a scientific manuscript to show 1,0E+7 instead of 10⁷.  The corrosion experiments and description of microbial community seem to be like two separate stories. More attention should be devoted on discussion how microbial communities from the sites of interest might influence different types of steel used in the study. Which scenarios are expected? Are microorganisms expected to produce sulfate, to scavange, to changes the sulfate/chloride ratio? If no bacterial activity has been measured during the electrochemical experiments and microorganisms have been neglected there, what was the reason for sequencing 16 S rRNA? What is typical microbial composition of the repositories sites worldwide (some examples would be appreciated)? In general, is it a typical practice to deposit radioactive waste in this kind of environments with this kind of chemical composition (this could be a good piece of information for the introduction). In the end, which site would be better for storage of radioactive waste?

Response: We would like to thank the reviewer for the considerate evaluation of our manuscript and the comments that helped us to improve the quality of our manuscript. The steel material is actually present in waste material itself as well as in container (steel drums) in some of the concepts for low and intermediate level waste repositories. This is now phrased more clearly in the manuscript. Also the experiment design is now described more clearly and potential corrosion mechanisms are discussed.

The manuscript is fundamental site characterization study highlighting the chemical and biological differences of the ground water in three sites and susceptibility of steel materials to corrosion in these chemical environments. This does not include long-term studies for example of microbe-metal interaction, that of course is an important aspect and needs to be studied in the future. All of these three sites will be used for repository of low and intermediate level waste but research is needed to ensure that the concept design is suitable to support long-term safety.

Below are more detailed comments per lines.

Line 34. Spelling mistake. Decomposition?

Response: Decommissioning waste is a commonly used term and refers to the waste material generated at the decommissioning phase of power plants, containing radioactively contaminated structural materials.

Line 47. Here and further below, it is assumed that 16S rRNA gene copies directly correspond to microbial cell numbers, which is not the case. Here is also given a reference to a paper of Haveman et al., 2002. There qPCR was not performed! Cell numbers were determined via epifluorescence microscopy and concentration was expressed not in copies in ml per in cells per ml. Please, add here a reference to a paper, if known, that gene copies are equal to cell numbers or please, refer to an appropriate estimation of microbial cell numbers based on qPCR.

Response: We agree with the reviewer that 16S rRNA gene numbers measured by qPCR is not the same as cell numbers quantified by microscopy. In this case qPCR was used as a proxy for the amount of bacteria, archaea, sulfate reducers and fungi, which is easy to compare between different sampling sites. In addition, it is not possible to distinguish between bacteria, archaea, sulfate reducers etc using microscopy and thus the differences in the abundance between these groups would be lost. We have previously compared total cell numbers to bacterial 16S rRNA gene numbers and have found that the measurements correlate quite well (e.g. Bomberg et al., 2016), although differences may be detected. These differences may be due to different number of 16S rRNA gene copies/cell, poor matching of primers to specific bacterial taxa, or e.g. high abundance of archaea, which would not be detected by the bacterial qPCR and could not be separated from the bacteria using microscopy. This is why we have chosen to use the number of bacterial 16S rRNA genes here and not the number of cells. The reference to the paper of Haveman et al. has been substituted with another one in the manuscript.

Lines 54-55. The SO42-/Cl- ratio causing inhibition increases with increasing Cl-concentration and temperature of the groundwater but decreases with alloying of the steel. This sentence is confusing. Please, rephrase.

Response: Sentence revised accordingly.

Line 60. Pdease add reference here.

Response: Reference added.

Lines 63-65. It would be nice to have graphical presentation of locations. Where those sites are located? Is it allowed to give this information? If not, please, indicate so and try at least give more details on the sites. Simulated groundwater – please, give details how it was prepared? What is also prepared under anoxic conditions? How?

Response: Unfortunately, not all power companies agreed to be identified and therefore the locations where groundwater samples were collected are identified only as sites A, B and C.

Regarding the simulated groundwater recipe and preparation, more information is added to the manuscript. Shortly, it is also anoxic, achieved by purging with argon gas.

Line 97. “from triplicate 1-3.3 l”. What does it mean? When was 1 L used, when was it 3.3? Does it mean that not equal amount of water has been pumped for each triplicate?

Response: Different amounts of groundwater were retrieved from different sites, this is now indicated in the manuscript more clearly. qPCR results are normalized by ml.

Line 189. It is not clear from description where groundwater sample was introduced? And how? How was it done? Under anaerobic conditions? Please, give hier in results section also details on replicate measurements.

Response: The description of experimental design of electrochemical measurements is now extended. Shortly, electrochemical measurements were run for two or three replicate individual specimen and replacing the groundwater or simulated water every time so that fresh water was used for each measurement. The electrochemical measurements were performed in laboratory under argon flow to maintain the anoxic conditions using the natural groundwaters collected earlier from 3 sites and the simulated groundwater. The groundwater samples were collected and stored in air-tight bottles and protective gas was used in each step of handling the waters.

Line 232. In general, it would be better to use instead of “waters” – groundwater samples, instead of “Water A” – Site A.  Why some values for simulated water are missing? Are there values for nitrate?

Response: Naming is revised accordingly. Regarding the simulated groundwater, the recipe was kept really simple containing only sulphate and chloride The basis for simulated water was distilled water, where sulphate was introduced in form of Na₂SO₄ and chloride content in form of NaCl. The pH 7 was adjusted using NaOH or HCl.

Line 240. Again. Please, not “waters” but groundwater samples from Sites…

Response: Revised accordingly throughout the manuscript.

Figure 1. Please, update the figure from the excel-like one to the more manuscript appropriate. Annotation of y-axis should scientific. Please, remove the header of the graph. You do not present the number of microorganisms here. Plus, there are no colors in this manuscript version. The graph is in the gray scale.

Response: Revised accordingly.

Table 3. Could this table be changed into graph for more obvious representation?

Response: Graph for alpha diversity measures provided is added in revised manuscript (Figure 3). The current table is also remained in the manuscript.

Figure 3. From annotation: what are different “water samples” and “water types”? Please, use terms groundwater samples and sites!

Response: Terms corrected for figure and figure text.

Line 302. Did you mean here Figure 4a instead of Fig 2b?

Response: Yes. Referred figure numbering corrected.

Figures 5,8,9 – please, use scientific annotations for x-axis

Response: Revised.

Figure 10. How many replicates? Please, indicate.

Response: Revised.

Line 448. What are possible mechanisms for driving the corrosion?

Response: The discussion of protective and corrosion driving mechanism in each groundwater is now added to the manuscript.

Line 465. The experiment could be done in the anaerobic chamber. Would that be possible?

Response: The reviewer is correct, the experiment setup could be performed in anaerobic chamber at least to some extent.

Lines 466-473. Again here: in this study not microbes per ml were detected , rather gene copies per ml! Could you please, also refer to have many cells normally are there in the groundwaters and what at typical TOC? In the presetine groundwater TOC is usually around 0.2 – 0.1 mg per ml (which is lower than at your site C) and bacterial cell numbers are around 10⁴. Which is higher than at Site C. What could be further explanations there?

Response: Revised accordingly. These all detected TOC concentrations are on the “normal” range of crystalline bedrock groundwaters at this depth in Fennoscandian area.

Lines 479-482. Could you, please, provide here more details on the Sulphur cycle. Do Sulfuricurum produce sulfate and sulfide from sulfur? To what proportion? What is about possible nitrate content in the sites presented in this study?

Response. Sulfuricurvum and Sulfurimonas are both sulfur oxidizing bacteria. Some species are known to oxidice S°, sulfide or sulfite and CO₂ utilize via carbon fixation coupled to sulfur oxidation.

Nitrate was below the detection limit of the analysis assay.

Line 528. Could you, please, provide more detail how bacteria/microbes can contribute to the sulfate content and subsequently to the corrosion? How it could be possible to decrease sulfate content knowing bacterial physiology?

Response: Decrease in sulphate concentration would increase the aggressivity of chloride. If SRBs use sulphate as electron acceptor, they would in most cases generate sulphide. Sulphide is aggressively corrosion inducing chemical for steels. However, the microbial effect on sulpur cycle is a long-term effect and is not considered to play a role in current short term corrosion experiment.

Line 549. Could you, please, give suggestions on what it could be “something else”?

Response: More detailed discussion of corrosion inducing and inhibiting effects have been added to manuscript. 

***********************************************************************************

Reviewer 2 Report

The paper investigates and report the chemical, biological and corrosion characteristic of different natural groundwater. The focus of the paper is not clear. The paper kind of conclude that the chemical characteristic of the groundwater found to have the most important role in corrosivity of the metals, and after giving so much emphasis on Bacteria count and microbial community size, the role of bacteria on corrosion was not discussed at all. It is recommended to organize the data toward the purpose of the paper.  Below you can find the detail comment regarding the text.

Line 42. It is important to also add the information on Microbiologically influence corrosion (MIC)  by SRB and the related electromechanical process.

Line 54. More supporting literature review may be needed in this section , as the manuscript will mostly focus on the role of chemical composition on corrosion.

Line 65. Were the steel coupons installed in the field or the testing was done in the laboratory in the sample of groundwater solution? Is the corrosion cell was only used for simulated ground water solution or also other cases of experiment too?

Line 200. which direction was the Tafel scan made?

Line 200. The polarization amount for the Tafel curves were up to 250 mV , which may cause some level of surface oxidation . Did the polarization affect the steel surface condition? Did this affect the analyses for Ecorr and icorr?

Line 217. Were the visual examination was done on same coupons with cyclic polarization ? Polarization up to +1000 mV will affect the surface condition.

 Line 325.  The difference in the Open Circuit Potential (OCP) and the corrosion potential (Ecorr) must be due the electrochemical testing , as the test coupans were polarized up to 250 mV for Tafel measurement and this level of polarization can change the corrosion potential.

Figure 6.  Based on the complicated polarization as shown in figure 5, How the I (corr) was determined from the graphs?

Discussion. Clarify the role of bacteria on electrochemical corrosion behavior

 Author Response

We thank the reviewer for constructive comments towards our manuscript and feel that those comments helped to improve the quality of the manuscript.

The major changes made to the manuscript are highlighted in the manuscript.

To summarize the major changes in the manuscript:

• The data in Table 3 is converted to Figure 3
• More discussion is added e.g. hypothesis of corrosion mechanisms
• Technical improvements have been made in figures
• The experimental setup is now described in more detailed

Author Response ：

The paper investigates and report the chemical, biological and corrosion characteristic of different natural groundwater. The focus of the paper is not clear. The paper kind of conclude that the chemical characteristic of the groundwater found to have the most important role in corrosivity of the metals, and after giving so much emphasis on Bacteria count and microbial community size, the role of bacteria on corrosion was not discussed at all. It is recommended to organize the data toward the purpose of the paper.  

Response: We would like to thank the reviewer for the considerate evaluation of our manuscript and the comments that helped us to improve the quality of our manuscript.

We have made the suggested revisions to improve the focus of the manuscript.

The aim of this study was to survey the differences in the corrosion behavior and in the microbiological community between different sites through short-term measurements and analyses of the groundwater. However, these measurements can be considered as a starting point and cannot straightforward predict the long-term behavior. Microbial corrosion inducing mechanisms are expected to be long-term effects that need to be studied further.

Detailed comments regarding the text:

Line 42. It is important to also add the information on Microbiologically influence corrosion (MIC) by SRB and the related electromechanical process.

Response: Indeed, SRB are known to have huge impact on corrosion of steel in anoxic conditions. Extended descriptions and references regarding of this area is now added.

Line 54. More supporting literature review may be needed in this section, as the manuscript will mostly focus on the role of chemical composition on corrosion.

Response: More literature references has been added.

Line 65. Were the steel coupons installed in the field or the testing was done in the laboratory in the sample of groundwater solution? Is the corrosion cell was only used for simulated ground water solution or also other cases of experiment too?

Response: All corrosion studies were conducted in laboratory. More detailed description of experiment setup is now added to the manuscript.

Line 200. which direction was the Tafel scan made?

Response: Tafel scan (for carbon steel) was performed from -250 mV to + 250 mV (vs OCP), this is now indicated in manuscript.

Line 200. The polarization amount for the Tafel curves were up to 250 mV, which may cause some level of surface oxidation. Did the polarization affect the steel surface condition? Did this affect the analyses for Ecorr and icorr?

Response: Tafel polarization was performed for carbon steel samples. Replicate measurements were done for replicate specimens.

Line 217. Were the visual examination was done on same coupons with cyclic polarization? Polarization up to +1000 mV will affect the surface condition.

Response: Indeed, polarization affects the surface conditions. The visual examination was here decided to conduct for the polarized specimens, and we agree that they do not represent stabile immersion test specimens that would be the traditional method. However, we believe that it is reasonable to study these surfaces. By driving the corrosion by electrochemical polarization differences between corrosion behaviour of different materials and groundwater were achieved. The information that specimens were polarized is now emphasized in the manuscript.

Line 325.  The difference in the Open Circuit Potential (OCP) and the corrosion potential (Ecorr) must be due the electrochemical testing, as the test coupons were polarized up to 250 mV for Tafel measurement and this level of polarization can change the corrosion potential.

Response: OCP was recorded in the beginning for pristine specimens. This is now indicated in manuscript and the chapter is rephrased.

Figure 6.  Based on the complicated polarization as shown in figure 5, How the I (corr) was determined from the graphs?

Response: The i(corr) was determined by extrapolating the linear part of the anodic curve, i.e., the Tafel slope, to the E(corr). That is now added to the manuscript.

Discussion. Clarify the role of bacteria on electrochemical corrosion behaviour.

Response: In short electrochemical measurements, the microbial community is not expected to have time to interact with steel surface and thus would not directly play a role. However, microbes are expected to have major role in long-term corrosion when metallic waste becomes in contact with native groundwater and its microbial community.

This is now clarified in the discussion section.

Round 2

Reviewer 1 Report

Dear authors,

thank you for considering all comments. Few last comments from my side:

 - could you also add to the manuscript text your nice explanation on why you have chosen to do qPCR instead of microscopy?

• could you, please, also ,add to the manuscript that TOC detected are normal for Fennoscandian area and add a corresponding reference?
• I also did not find you response on "Sulfuricurvum and Sulfurimonas are both sulfur oxidizing bacteria. Some species are known to oxidice S°, sulfide or sulfite and CO₂ utilize via carbon fixation coupled to sulfur oxidation" in the revised manuscript text. " Especially, what considers carbon fixation. Or may I just overlooked it?

Author Response

Thank you very much for your valuable comments for improving our manuscript.

The responses to the last comments:

1. could you also add to the manuscript text your nice explanation on why you have chosen to do qPCR instead of microscopy?

Response: Revised accordingly.

2. could you, please, also ,add to the manuscript that TOC detected are normal for Fennoscandian area and add a corresponding reference?

Response: Revised accordingly and reference added.

3. I also did not find you response on "Sulfuricurvum and Sulfurimonasare both sulfur oxidizing bacteria. Some species are known to oxidice S°, sulfide or sulfite and CO₂utilize via carbon fixation coupled to sulfur oxidation" in the revised manuscript text. " Especially, what considers carbon fixation. Or may I just overlooked it?

Response: Revised accordingly. Information added to the manuscript.

Reviewer 2 Report

The Authors have applied the requested changes.

Author Response

Thank you very much for your valuable comments for improving our manuscript.