Combining benzalkonium chloride addition with filtration to inhibit dissolved inorganic carbon alteration during the preservation of seawater in radiocarbon analysis

Takahashi, Hiroshi A.; Minami, Masayo

doi:https://doi.org/10.5194/egusphere-2024-3349

Preprints

https://doi.org/10.5194/egusphere-2024-3349

Preprints

06 Nov 2024

| 06 Nov 2024

Combining benzalkonium chloride addition with filtration to inhibit dissolved inorganic carbon alteration during the preservation of seawater in radiocarbon analysis

Hiroshi A. Takahashi and Masayo Minami

Abstract. Benzalkonium chloride (BAC) addition has shown great promise as a disinfectant for measuring δ¹³C and ¹⁴C of dissolved inorganic carbon (DIC) in freshwater samples. However, it was reported that the effectiveness of BAC to prevent DIC change was reduced for the use of seawater samples. The present study aimed to evaluate the effectiveness of adding BAC as a disinfectant in carbon isotopic analyses of DIC in seawater samples. We compared the efficacy of BAC addition, filtration (0.22 μm PTFE or 0.2–0.45 μm PES filters), and a combination of BAC addition and filtration in preventing DIC alterations caused by biological activity. The combined procedure was effective in preserving seawater, although this assessment was based on results from a single seawater sample. The ¹⁴C concentration of samples treated with both BAC addition and filtration exhibited minimal changes, ranging from 0.2–0.4 pMC over 41 weeks, despite the addition of sugar included to increase biological activity several-fold. Although the complete elimination of biological effects may be challenging with the combined method, the observed changes remained within practical limits. Concerns about CO₂ contamination during sample filtration were also addressed and found to be negligible. These results suggest that combining filtration and BAC addition is an effective method for suppressing biological DIC alterations in ¹⁴C analysis, even in seawater samples.

Received: 28 Oct 2024 – Discussion started: 06 Nov 2024

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.

Download & links

Preprint (PDF, 694 KB)

Supplement (108 KB)

Download & links

Hiroshi A. Takahashi and Masayo Minami

Status: final response (author comments only)

RC1:
'Comment on egusphere-2024-3349', Anonymous Referee #1, 17 Nov 2024
General comments:
Overall this is an interesting study that may provide more data to inform our understanding of water sample collection and storage prior to analysis, and the ways in which different preservation methods may affect accuracy, pending the authors responses to the comments below.
Specific comments:
The experimental design needs to be made explicitly clear. The manuscript must state how many independent replicates were used for each treatment in order for the reader to better understand what the results were. At present, it appears n = 1 or 2 for some or all experiments which is too low to warrant publication.

Where appropriate, add 'groundwater' to the title and Abstract. The study tested both groundwater and seawater.

Line 16: Edit for clarity. Microbial activity was not directly measured.

Line 83. Please add information on the salinity of the SW and GW to Table S1 and in the Abstract. The reader should be able to view this information to better understand what this text is talking about and/or more easily make an assessment whether it is appropriate for the SW sample to be called seawater, estuarine water, brackish water or something else.

Figure 1, Figure 2, Table 1, and all supplementary Table captions should state what the data is (means?) and the number of replicates.

Figure 1. There appear to be duplicate data points shown for some treatments (e.g. unfiltered, GF) but there is no explanation of what each represents. Are these replicates or different treatments?

Line 165. It is incorrect to conclude that filtration does not cause a change in 13C. There is not enough evidence for or against. Indeed, as there appears to be very low or no replication (n = 1 or 2) any differences in 13C among treatments may simply be an artefact.

Line 186: “Beet sugar is more easily degraded”. Compared to what?

Line 204: “ …reduced changes in DIC….”. Compared to the control treatment?

Line 213: This is not true. Filtration + BAC essentially stopped changes in DIC for both GW and SW.

Technical corrections:

Line 36. Remove “have been” and “. They “

Line 63. Replace “is” with ‘are’

Line 210-211: Edit for clarity.

Line 273: Remove full stop after “However”
Citation: https://doi.org/10.5194/egusphere-2024-3349-RC1
- AC1: 'Reply on RC1', Hiroshi Takahashi, 13 Dec 2024
  
  We would like to thank the anonymous referee for their time and the thoughtful feedback on our manuscript. We will revise the manuscript according to the reviewer’s comments. The main points of revision are summarized as follows:
  (1) The Materials and procedures section will be revised to clearly explain the experimental design, especially the number of samples.
  (2) While the number of analyses conducted in this study was limited, we believe that the conclusions would remain unchanged even with an increase in the number of analyses. We will include an explanation of this in the Results and discussion section.
  All responses to the reviewer’s comments are listed below. We believe that we have addressed all of the comments from the reviewer.
  
  <Specific comment from RC1>
  The experimental design needs to be made explicitly clear. The manuscript must state how many independent replicates were used for each treatment in order for the reader to better understand what the results were. At present, it appears n = 1 or 2 for some or all experiments which is too low to warrant publication.
  <Answer>
  The description of the experimental design will be revised to include the number of samples that were processed or measured. The measured values excluding GW-Control2, GW-PES2 (0.2 µm), and GW-PES2 (0.45 µm) were derived from a single measurement of an individual treatment or a preservation bottle. The three excluded samples are presented as the mean and standard deviation of the analytical values from the three preservation bottles of each. The description will be revised to include the number of samples, and errors.
  In response to the comment that the number of analyses is insufficient, we offer the following counterargument:
  Background assessment: While not previously delineated in the initial manuscript, the filtration background assessment was conducted using a single NaHCO₃ solution, with the order of unfiltered, PES, GF, GF, and unfiltered samples. The ¹⁴C concentrations and δ¹³C values of the NaHCO₃ solution before and after the filtration experiment were identical, indicating that the DIC of the NaHCO₃ solution itself did not change during the evaluation experiment. A comparison of the filtered and unfiltered samples showed a high degree of consistency in the analytical values, indicating that filtration did not change the ¹⁴C concentration. While the individual data points were obtained from a single experiment, the results of the assessment should be evaluated as being derived from the results of a series of experiments. Therefore, we believe it is reasonable to conclude that the results are not merely due to chance but reflect the absence of an influence of filtration. The revised manuscript will include the processing order for both unfiltered and filtered samples and a discussion of these results.
  Comparison of treatments: As same above, the results were obtained as a time-series, although the individual data points were derived from a single experiment. While the detailed values pertaining to isotopic changes may not be entirely accurate, it is possible to discern the presence or absence of alterations and trends. If the objective is to ascertain an exact amount of change, repeated measurements are necessary. However, if the objective is to determine whether change has occurred or to identify its direction of change, the results of this study are sufficient for this purpose.
  
  <Specific comment from RC1>
  Where appropriate, add 'groundwater' to the title and Abstract. The study tested both groundwater and seawater.
  <Answer>
  The title will be revised from "seawater" to "water sample" to encompass both seawater and groundwater. The result of groundwater will be incorporated into the abstract as follows: “The freshwater sample that had undergone a BAC addition treatment showed the no alteration of DIC”.
  
  <Specific comment from RC1>
  Line 16: Edit for clarity. Microbial activity was not directly measured.
  <Answer>
  As the reviewer commented, the degree of biological activity was not directly measured in this study. Instead, changes in dissolved inorganic carbon (DIC) were attributed to biological activity, and thus the expression "biological activity" was used in the text. However, this description is not entirely accurate and will be revised to "DIC changes" for clarity.
  
  <Specific comment from RC1>
  Line 83. Please add information on the salinity of the SW and GW to Table S1 and in the Abstract. The reader should be able to view this information to better understand what this text is talking about and/or more easily make an assessment whether it is appropriate for the SW sample to be called seawater, estuarine water, brackish water or something else.
  <Answer>
  The salinities of the SW and GW will be added to the abstract, the Materials and procedures section, and Table S1. The salinities of SW and GW were 20.4% and less than 0.05%, respectively. Therefore, SW is categorized as brackish water, and GW is categorized as freshwater.
  
  <Specific comment from RC1>
  Figure 1, Figure 2, Table 1, and all supplementary Table captions should state what the data is (means?) and the number of replicates.
  <Answer>
  All values indicated in Figures 1 and 2 and Tables S2 and S3 are derived from a single measurement taken for each respective treatment. In contrast, the values shown in Table 1 represent the mean values of the respective measurements taken of the three preservation bottles for each period.
  The figure and table captions will be revised to clearly indicate whether the analytical values represent mean values or individual treatment sample values.
  
  <Specific comment from RC1>
  Figure 1. There appear to be duplicate data points shown for some treatments (e.g. unfiltered, GF) but there is no explanation of what each represents. Are these replicates or different treatments?
  <Answer>
  The respective data points were obtained from a single treatment, therefore, the duplicate points of unfiltered and GF indicate two measurements from two individual respective treatments.
  Figure 1 will be rearranged to show the order of filtration treatments using the NaHCO₃ solution. Please refer to the supplement file for additional details.
  
  <Specific comment from RC1>
  Line 165. It is incorrect to conclude that filtration does not cause a change in 13C. There is not enough evidence for or against. Indeed, as there appears to be very low or no replication (n = 1 or 2) any differences in 13C among treatments may simply be an artefact.
  <Answer>
  As you correctly noted, the description that filtration does not cause a δ¹³C change was not accurate and therefore will be deleted. Moreover, we concur with your comment that the observed differences in δ¹³C among treatments may be attributed to an artefact. However, we maintain that the changes in δ¹³C were not caused by atmospheric CO₂ contamination or degassing, and we propose that the observed δ¹³C change may be attributed to an unidentified artifact factor other than filtration. This conclusion and a clarification stating that the effect of filtration on ¹⁴C analysis is negligible will be added.
  
  <Specific comment from RC1>
  Line 186: “Beet sugar is more easily degraded”. Compared to what?
  <Answer>
  The materials under comparison were BAC or other organic materials suspended in water. This description will be added.
  
  <Specific comment from RC1t>
  Line 204: “ …reduced changes in DIC….”. Compared to the control treatment?
  <Answer>
  Yes. It is compared to the Control sample. The description will be revised.
  
  <Specific comment from RC1t>
  Line 213: This is not true. Filtration + BAC essentially stopped changes in DIC for both GW and SW.
  <Answer>
  The description was inaccurate and will be deleted.
  
  < Technical corrections from RC1t>
  <Answer>
  The manuscript will be revised in accordance with all technical recommendations.
  
  Citation: https://doi.org/10.5194/egusphere-2024-3349-AC1
RC2:
'Comment on egusphere-2024-3349', Anonymous Referee #2, 06 Feb 2025

The authors presented a study to examine whether a bacteria inhibiting compound can be used in conjunction with filtration (at different pore sizes) in storing dissolved inorganic carbon samples for both isotope analysis of both seawater and groundwater (note groundwater needs to be included in the title).
The results of these different treatment are as somewhat expected, that filtration with small pore filter and bacteria inhibition help to preserve DIC samples, but with limited success (i.e., relatively short storage time). However, the experiment section needs much articulation based on what is presented in the manuscript. A flow chart that includes many of the treatment details may be helpful for readers to comprehend the procedure in the experiment. For example, please include the volume of the initial sample vessel, how samples were distributed into these different treatments, and number of replicate samples/analysis (N) for each treatment.
While there is nothing wrong with using metric units (mg/L) for solutes, as an analytical chemistry-oriented study, I would recommend using molar units (Table S1 needs to include salinity) to facilitate the comparison with seawater.
For the information provided in Tables S2 and S3, it is unclear what the somewhat consistent uncertainty values are. Having a more detailed experimental setup will help to clear this up. Fig. 2 has no error bars, was each sample analyzed only once, what is the analytical precision and accuracy? Even though DIC concentration isn’t a focus of this study but isotopes, it would be helpful to present the DIC concentration changes through time in these different treatments along with analytical uncertainty, if such information is available.
Minor comments:
Line 101, ReCEIT needs to be briefly explained in addition to the citation.
Line 191, remove “salt”.
Line 192-195, explain what the “boost effect” is.
Line 213-217, this seems quite wordy as it’s already in the table.
Table 1, show number of replicate analysis.

Citation: https://doi.org/10.5194/egusphere-2024-3349-RC2
- AC2: 'Reply on RC2', Hiroshi Takahashi, 26 Feb 2025
  
  We would like to thank the anonymous reviewer for his/her time and the thoughtful feedback on our manuscript. We will revise the manuscript according to the reviewer's comments. Please find below the responses to the reviewer's comments. We believe that we have addressed all of the comments.
  
  <Comment from RC2>
  The authors presented a study to examine whether a bacteria inhibiting compound can be used in conjunction with filtration (at different pore sizes) in storing dissolved inorganic carbon samples for both isotope analysis of both seawater and groundwater (note groundwater needs to be included in the title).
  <Answer>
  Another reviewer made a similar suggestion regarding the paper's title. Considering these suggestions, we would like to modify the title to "water samples," which encompasses samples that are not limited to "seawater."
  
  <Comment from RC2>
  The results of these different treatment are as somewhat expected, that filtration with small pore filter and bacteria inhibition help to preserve DIC samples, but with limited success (i.e., relatively short storage time).
  <Answer>
  We anticipated that filtration would not have enough effect to inhibit DIC changes during the water preservation. However, a recent publication asserted the efficacy of filtration alone, prompting us to investigate its effect. Our findings aligned with our initial expectations, but we also discovered that both filtration and BAC should be utilized in seawater samples.
  
  <Comment from RC2>
  However, the experiment section needs much articulation based on what is presented in the manuscript. A flow chart that includes many of the treatment details may be helpful for readers to comprehend the procedure in the experiment. For example, please include the volume of the initial sample vessel, how samples were distributed into these different treatments, and number of replicate samples/analysis (N) for each treatment.
  <Answer>
  The Materials and procedures section will be revised to clearly explain the experiments, especially the number of samples. The flowchart will be added (please see the supplement with this reply).
  
  <Comment from RC2>
  While there is nothing wrong with using metric units (mg/L) for solutes, as an analytical chemistry-oriented study, I would recommend using molar units (Table S1 needs to include salinity) to facilitate the comparison with seawater.
  <Answer>
  The description of chemical concentration will change to molar units in the revised manuscript and tables, and the salinity will be added to Table S1. They are 20.4 % for SW, and <0.05% for GW.
  
  <Comment from RC2>
  For the information provided in Tables S2 and S3, it is unclear what the somewhat consistent uncertainty values are. Having a more detailed experimental setup will help to clear this up.
  <Answer>
  The description of uncertainty values, which are indicated to be 1σ of analytical error, will be added to the explanations in Tables S2 and S3. The standard deviations for ¹⁴C measurements have been described into the “¹⁴C concentration and δ¹³C measurements” section of the manuscript, and the method to obtain the error of ¹⁴C measurement will be described as a standard calculation method in ¹⁴C analysis with a citation. Since the description of δ¹³C error had not been described, it will be incorporated into the “¹⁴C concentration and δ¹³C measurements” section of the revised manuscript.
  For ¹⁴C: The standard deviations for ¹⁴C measurements were 0.02–0.04 pMC for waters with concentrations below 1 pMC, and less than 0.8% of ¹⁴C concentration for waters above 10 pMC when measured at the Nagoya University. The precision of the quantitative analysis of carbon was better than 3%, and the background was below 2 × 10⁻¹⁵ at KIGAM. The error of ¹⁴C measurement was represented in 1σ based on a standard calculation method in ¹⁴C analysis (Scott et al., 2007).
  For δ¹³C: The standard deviation of multiple δ¹³C measurements by IRMS is less than 0.01‰, and that in individual measurements was represented in 1σ calculated from the variations of the dual inlet measurement. …. The standard deviation of multiple measurements of water samples by GC-IRMS is 0.04‰ (1σ). …. The δ¹³C value of each sample was represented the averaged value over the three vials with the standard deviation (1σ).
  
  <Comment from RC2>
  Fig. 2 has no error bars, was each sample analyzed only once, what is the analytical precision and accuracy?
  <Answer>
  The error bars in Fig. 2 were not displayed, since the errors of ¹⁴C and δ¹³C are equivalent to or less than the size of the plotting symbols. This will be mentioned in the figure caption. The one plot was obtained from the single sample treatment. The statement regarding analytical error will be incorporated into the “¹⁴C concentration and δ¹³C measurements” section of the revised manuscript as mentioned above.
  
  <Comment from RC2>
  Even though DIC concentration isn’t a focus of this study but isotopes, it would be helpful to present the DIC concentration changes through time in these different treatments along with analytical uncertainty, if such information is available.
  <Answer>
  Unfortunately, due to limitations in the analytical equipment, we were unable to measure the DIC concentration. It is possible to calculate it from the weight of the water sample used for CO₂ extraction and the amount of CO₂ collected. However, this approach did not provide sufficient precision to distinguish between subtle DIC changes.
  
  <Minor comment from RC2>
  Line 101, ReCEIT needs to be briefly explained in addition to the citation.
  <Answer>
  The explanation of the ReCEIT will be added in the revised manuscript.
  The ReCEIT (repeated cycles of extraction, introduction, and trapping) method is a simple and carrier gas-free method that can handle a variety of water samples with a wide range of DIC concentrations (0.4–100 mmol/L, in the case of 1 mmol of carbon) and produces high CO₂ yields. The procedure is composed of repeating the cycles of CO₂ extraction from water into the headspace of the reaction container, expansion of the extracted gas into the vacuum line, and cryogenic trapping of CO₂. This method, which extracts CO₂ without bubbling, is particularly well-suited for BAC-added samples, which tend to foam.
  
  <Minor comment from RC2>
  Line 191, remove “salt”.
  <Answer>
  The description of “nutrient salts” will be changed to “nutrients” in the revised manuscript.
  
  <Minor comment from RC2>
  Line 192-195, explain what the “boost effect” is.
  <Answer>
  The explanation of the boost effect will be added in the revised manuscript.
  The boost effect was defined as an index of how many times the DIC change in the sugar-added sample is greater than the DIC concentration change in the no-sugar sample during the preservation by Takahashi and Minami (2022).
  
  <Minor comment from RC2>
  Line 213-217, this seems quite wordy as it’s already in the table.
  <Answer>
  To streamline the text, the manuscript will be changed to delete isotopic changes, but to remain the sample name.
  
  <Minor comment from RC2>
  Table 1, show number of replicate analysis.
  <Answer>
  The explanation of Table 1 will be changed to show the number of replicate analysis: N=3.
  
  Citation: https://doi.org/10.5194/egusphere-2024-3349-AC2

Hiroshi A. Takahashi and Masayo Minami

Supplement

https://doi.org/10.5194/egusphere-2024-3349-supplement

Hiroshi A. Takahashi and Masayo Minami

Viewed

Total article views: 277 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
180	65	32	277	36	10	9

HTML: 180
PDF: 65
XML: 32
Total: 277
Supplement: 36
BibTeX: 10
EndNote: 9

Views and downloads (calculated since 06 Nov 2024)

Month	HTML	PDF	XML	Total
Nov 2024	80	21	6	107
Dec 2024	40	14	1	55
Jan 2025	9	6	1	16
Feb 2025	28	12	3	43
Mar 2025	12	9	19	40
Apr 2025	11	3	2	16

Cumulative views and downloads (calculated since 06 Nov 2024)

Month	HTML	PDF	XML	Total
Nov 2024	80	21	6	107
Dec 2024	40	14	1	55
Jan 2025	9	6	1	16
Feb 2025	28	12	3	43
Mar 2025	12	9	19	40
Apr 2025	11	3	2	16

Viewed (geographical distribution)

Total article views: 271 (including HTML, PDF, and XML) Thereof 271 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 23 Apr 2025

Short summary

A combined procedure of benzalkonium chloride (BAC) addition and filtration was investigated to preserve water samples for radiocarbon analysis of dissolved inorganic carbon (DIC). The results indicated that DIC changes can be effectively suppressed during sample storage for up to 41 weeks. This procedure offers a practical, environmentally friendly alternative to conventional mercury-disinfected methods, effectively water sample preservation samples in aquatic environments.


Total:	0
HTML:	0
PDF:	0
XML:	0