the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Stability of alkalinity in Ocean Alkalinity Enhancement (OAE) approaches – consequences for durability of CO2 storage
- 1Institute for Geology, Universität Hamburg, Bundesstrasse 55, D-20146 Hamburg, Germany
- 2Faculty of Physics/Electrical Engineering, Universität Bremen, Otto-Hahn-Allee 1, 28359 Bremen, Germany
- 3Geomar Helmholtz Centre for Ocean Research Kiel, Kiel, Germany
- 4Instituto de Oceanografía y Cambio Global, Universidad de Las Palmas de Gran Canaria, Las Palmas, Spain
- 5School of Engineering and Physical Sciences, Heriot-Watt University, Edinburgh, EH14 4AS, UK
- These authors contributed equally to this work.
- 1Institute for Geology, Universität Hamburg, Bundesstrasse 55, D-20146 Hamburg, Germany
- 2Faculty of Physics/Electrical Engineering, Universität Bremen, Otto-Hahn-Allee 1, 28359 Bremen, Germany
- 3Geomar Helmholtz Centre for Ocean Research Kiel, Kiel, Germany
- 4Instituto de Oceanografía y Cambio Global, Universidad de Las Palmas de Gran Canaria, Las Palmas, Spain
- 5School of Engineering and Physical Sciences, Heriot-Watt University, Edinburgh, EH14 4AS, UK
- These authors contributed equally to this work.
Abstract. According to modelling studies, ocean alkalinity enhancement (OAE) is one of the proposed carbon dioxide removal (CDR) approaches with large potential, and the beneficial side effect of counteracting ocean acidification. The real-world application of OAE, however, remains unclear as most basic assumptions are untested. Before large-scale deployment can be considered, safe and sustainable procedures for the addition of alkalinity to seawater must be identified and governance established. One of the concerns is the stability of alkalinity when added to seawater. Seawater is already supersaturated with respect to calcium carbonate minerals, and an increase in total alkalinity together with a corresponding shift in carbonate chemistry towards higher carbonate ion concentrations would result in further increase in supersaturation, and potentially to solid carbonate precipitation. Precipitation of carbonate minerals consumes alkalinity and increases dissolved CO2 in seawater, thereby reducing the efficiency of OAE for CO2 removal. In order to address the application of alkaline solution as well as fine particulate alkaline solids, a set of six experiments was performed using natural seawater with alkalinity of around 2,400 μmol/kgw. The application of CO2-equilibrated alkaline solution bears the lowest risk of losing alkalinity due to carbonate formation if added total alkalinity (ΔTA) is less than 2,400 μmol/kgw. The addition of reactive alkaline solids can cause a net loss of alkalinity if ΔTA > 600 μmol/kgw (e.g., for Mg(OH)2). Commercially available Ca(OH)2 causes in general a net loss in TA for the tested amounts of TA addition, which has consequences for suggested use of slurries supplied from ships. The application of excessive amounts of Ca(OH)2, exceeding a threshold for alkalinity loss, resulted in a massive increase in TA (> 20,000 μmol/kgw) at the cost of lower efficiency and resultant high pH values > 9.5.
Our results indicate that using an alkaline solution instead of reactive alkaline particles can avoid carbonate formation, unless alkalinity addition shifts the system beyond critical supersaturation levels. To avoid the loss of alkalinity and dissolved inorganic carbon (DIC) from seawater, the application of reactor techniques can be considered. These techniques produce an equilibrated solution from alkaline solids and CO2 prior to application. Differing behaviours of tested materials suggest that standardized engineered materials for OAE need to be developed to achieve safe and sustainable OAE with solids, if reactors technologies should be avoided.
Jens Hartmann et al.
Status: final response (author comments only)
-
RC1: 'Comment on bg-2022-126', olivier sulpis, 26 Jun 2022
The manuscript presents results from a set of experiments assessing the fate of alkalinity added under various forms to natural seawater. Predicting alkalinity stability is key to quantify the potential of OAE to neutralize atmospheric CO2. Hence, this study is particularly timely.
The particularity of this study is that it is composed of many different experiments, providing a vast quantity of results, all quite convincing, and numerous take-home messages. The authors seem to have designed their series of experiments incrementally, guided by the preliminary results obtained along the way. As a result, they went very far into the search for answers to their initial research questions, and the diversity and quality of results they produced is impressive. The discussion section nicely summarizes and explains the key findings, analyzes them in the context of preexisting literature, and offers guidance for implementing OAE techniques. This study is overall a pleasure to read, and it will certainly have broad implications. That said, the fact that numerous experiments were carried on leads to some lack of clarity in the methods. A proper statistical treatment on the collected data is also missing. Finally, the absence of analyses on the recovered precipitates is frustrating, as knowing what these precipitates are and how they form would be capital to better understand how to deal with them.
General comments:
- It is clear from the water chemistry data that carbonate minerals are precipitating. In abiotic experiment II, the authors observed immediate precipitation of centimetric needle shaped precipitates floating at the center of the bottles, suggesting spontaneous nucleation of aragonite. After 1 day, they observed precipitates on the walls of the reactors. They also observed these precipitates on the walls in experiment III. In all other cases, it seems that precipitation happened on preexisting particles, such as natural particles smaller than 55 microns, or minerals added as a TA source. In all those cases, precipitation occurred presumably without the need to nucleate in the first place, as growth on preexisting nuclei. What is precipitating exactly is unknown, but the results shown on Fig. 7 suggest that it may be a phase more soluble than aragonite. The majority of CaCO3 precipitates observed by Moras et al. (under review) were aragonite, and a little of vaterite. It would be enlightening to provide visuals of what the precipitates observed in the present study look like, and analyses of their mineralogy, as other similar studies have done using a combination of scanning electron microscope or energy dispersive X-Rays (Moras et al., under review; Fuhr et al., 2022). At minimum, it is needed to discuss the hypothesis of ACC precipitates in comparison with observations from the Moras et al. study.
- There is a lack of proper statistical treatment. Results should be given with an associated uncertainty that should reflect uncertainties arising from (i) the experiments, from (ii) the instruments/analyses and from (iii) the computations (errors on thermodynamic constants, uncertainty propagation, etc.). (i) it seems that only experiment V was conducted in replicates, but any resulting uncertainty quantification is absent in Fig. 7. (ii) in the measurements section, no precision is provided for the instruments used, and even though it is stated that some samples were taken and measured in duplicates, the resulting uncertainties are not provided in the text or shown in figures (apart from Fig.4). (iii) the authors should use the new CO2SYS versions recently published, available in excel (Orr et al., 2018), that allow to include and propagate all uncertainties including those from the thermodynamic constants.
- Reading through the methods section was challenging, due to the large number of experiments and to the lack of details provided, see list of specific comments. In particular, the motivations for the experimental design are not well explained. It would be helpful that the paper reflects the same succession of thoughts and actions that the authors went through when designing their study.
Specific comments:
L19: precise which minerals and that it does not apply for the deep ocean
L26: precise that the carbonates mentioned are carbonate minerals
L28: precise what kind of slurry
L31-32: “unless alkalinity addition shifts the system beyond critical supersaturation levels”: this portion of the sentence is unclear, please rephrase
L45: field experiments are lacking but not laboratory studies, although more are needed, please precise that and add relevant references
L48: modeling studies have also focused on the OAE applied specifically in coastal environments and in oceanic regions of high CO2 outgassing, e.g., Fakhraee et al., under review, please precise that
L52: here and after, please precise carbonate minerals
L54: adding one sentence here on nucleation itself being the greatest energy barrier to encompass before crystal growth can continue would be insightful, possibly referring to Sun et al. 2016.
L60: please precise that the optimal conditions refer to natural, typical sea surface environments
Methods: there are a lot of experiments and variables, and the methods section describing them is sometimes hard to follow. Table 1 really helps. In addition, in the methods, it would be helpful to explain which research questions enumerated in the introduction relates to which experiment, to guide the reader and explain why all these different experiments were needed.
Table 1: precising the type of seawater used (North Sea or Gran Canaria) in this table would be useful. In addition, please provide elsewhere eventual details on dissolved silica and phosphate concentrations for both seawaters, as well as the TA and S for the North Sea seawater.
L101: please provide further details, e.g., how much of NaHCO3 and Na2CO32- were added, what was the pCO2 of the solution and of the air?
L110: is there a reason for not using the same Dickson bottles for all experiments?
L114: please precise why should airspace be avoided in the case of Ib, since this experiment is in principle equilibrated
L115: why not keeping the same alkalinity addition steps than for the abiotic experiments?
L119: replace “gained” by “retrieved” or “collected”
L119: at this stage we have no information on what those precipitated carbonates are. What are the mineralogy, grain size, composition?
L120: that the experiment took place in the harbor introduces temperature as a new and potentially important variable: what was the temperature and its difference between day and night, and how were the effects of temperature kept track of when interpreting the results and comparing them to experiments carried out in temperature-controlled environments?
L121: how much liquid was sampled?
L121: replace “the 10 days of processing” by “10 days of experiment,”
L138: was only the liquid sampled or some solid with it?
L142: how many Erlenmeyer-flasks were there and why?
L142: it is a different type of seawater than for the previous experiments. What was the composition of that seawater? What was the temperature for those experiments?
L143: do “timesteps” here correspond to sampling times? Please clarify, and precise how much liquid was sampled each time.
L145: please precise why the need for a shaking table and why 125 rpm.
L145: delete “value”
L149: how many different solid masses were investigated? How much solid was added in each? Were the experiments conducted in triplicates?
L152: replace “as” by “than”
L155: please provide more details on how airtight filtrations were done
L157: were dissolved silica and phosphate not measured in the two types of natural seawater used? If not, how were their contributions to total alkalinity removed in CO2SYS?
L165: what scale was the pH expressed on and how was this conversion made?
L183: the term “carbonate chemistry remained stable” is unclear. In fact, it seems that this last sentence does not bring anything new relative to what was said earlier in that paragraph and could be removed.
Fig.3 I suggest adding dashed lines in the background showing the 2:1 ratio, as a reference for the reader to assess the decline in TA:DIC ratio observed in the experiments
Fig.3: is “added TA” similar to “ΔTA”? If so, precise or update to ΔTA for consistency with the text and the other figures
L243: were floating precipitates, as described in experiment II, also observed here?
L246: “due to other surfaces abundant”: this portion of the sentence is unclear, please rephrase
Fig.5: are the black/grey, and light blue/dark blue lines replicates? This is not indicated. If so, it would be clearer and make more sense to combine them and use the spread between them as a measure of uncertainty (which should also include and reflect the other uncertainty sources).
L255: can more information on the dissolution rate of brucite be provided, e.g., approximatively how much was left after 4 days?
L260: the use of “aspired” seems odd, perhaps change to “targeted”
L273: what did the aggregates look like, and couldn’t they be disaggregated with gently shaking the bottles?
L299: the title of section 3.7 mentions CO2 equilibration, but Table 1 says the opposite. If I understand correctly, there is an ambiguity here because Ca(OH)2 dissolving induces a disequilibrium with the atmosphere which is then compensated by CO2 uptake from the air due to the open flasks. Here and in general, it would be good to distinguish more clearly between “CO2-equilibrated” as in preparing a solution with amounts of TA and DIC targeted to produce a certain pCO2 that would match that of the air and “CO2-equilibrated” as in actual exchanges between air and seawater until both are at equilibrium.
L327: please rephrase “wave line”
L342-343: “The reason is that the carbonate system (based on W of the bulk solution) stayed within boundaries to avoid mineral precipitation” and L345-346: “The reason is that, for the same DTA, perturbations of the carbonate system that affect mineral precipitation are much stronger in this non-equilibrated scenario”: because it is a very important issue for anyone implementing OAE to understand, I believe those two sentences should be explained in more details
L360: the threshold observed by Morse and He is for nucleation, which is more specific than spontaneous carbonate formation. It should be made clear here and in the intro that nucleation and growth are two different things with different energy barriers
L367: “between adding carbonate precipitates and not” this portion of the sentence is understandable but could be better phrased
Data availability: the data should be made available at the time of submission for reviewers to be able to see it
References:
Fakhraee et al., under review, https://doi.org/10.21203/rs.3.rs-1475007/v1
Fuhr et al., 2022, https://doi.org/10.3389/fclim.2022.831587
Moras et al., under review, https://doi.org/10.5194/bg-2021-330
Orr et al., 2018, https://doi.org/10.1016/j.marchem.2018.10.006
Sun et al., 2016, https://doi.org/10.1073/pnas.1423898112
-
AC1: 'Reply on RC1 and RC2', Jens Hartmann, 25 Sep 2022
The comment was uploaded in the form of a supplement: https://bg.copernicus.org/preprints/bg-2022-126/bg-2022-126-AC1-supplement.pdf
-
RC2: 'Comment on bg-2022-126', Scott C. Doney, 09 Jul 2022
Overall this is an important study presenting a series of small-scale experiments regarding the potential for mineral precipitation during ocean alkalinity addition, with the possibility of mineral formation significantly counteracting the intended net change in total alkalinity. The authors do a good job framing the experimental results of the study in terms of the efficiency or efficacy of ocean alkalinity enhancement approaches. The study highlights a number of key nuances in terms of the amount and character of the added alkaline solutions or alkaline solid phases. I did not identify any significant conceptual issues with the study or with the main data analysis approaches and results presentations. The 6 sets of experiments involve a wide range of conditions, source waters, materials added, time durations, and underlying scientific questions. Interpreting these different experiments, therefore, requires some effort, and most of my comments below involve possible clarifications to the text that may help improve the readability of the study.
Abstract
Line 25-27:
"The application of CO2-equilibrated alkaline solution bears the lowest risk of losing alkalinity due to carbonate formation if added total alkalinity (Delta TA) is less than 2,400 micromol/kgw. The addition of reactive alkaline solids can cause a net loss of alkalinity if Delta TA > 600 μmol/kgw (e.g., for Mg(OH)2)."If I am interpreting the text correctly, the notation "Delta TA" is used in three different, and somewhat confusing, ways:
1) initial amount of total alkalinity added to the seawater solution prior to mineral precipitation (e.g., Delta TA_1200, etc.
2) net change in total alkalinity reflecting both the amount of alkalinity added as well as the loss of alkalinity due to mineral precipitation (e.g., Delta TA in Figure 8).
3) loss of alkalinity due to mineral precipitation following alkalinity addition (e.g., Figure 4)
The addition of an equation along the following lines may be useful:
net TA change = TA_final - (TA_initial + TA_added + TA_loss)
It would be very helpful to avoid multiple usages for Delta TA and be more specific about the quantity being presented.
Also, as a general comment it would be helpful to clearly state at each use of the term what is meant by "loss of alkalinity". In some extreme cases in the experiments "loss" refers to a "net loss" relative to the natural background concentration. In other situations, "loss" appears to refer loss of alkalinity relative to the expected increase in alkalinity following the addition of alkaline solutions or solids.
Line 43
"(NAS, 2021)" (also Line 655 in references)
Please update the citation and reference to "NASEM, 2022" for the final report:National Academies of Sciences, Engineering, and Medicine, 2022: A Research Strategy for Ocean-based Carbon Dioxide Removal and Sequestration, Washington, DC, The National Academies Press, https://doi.org/10.17226/26278
Line 49:
"negative TA could be a consequence"I think this should read: "negative net change in TA could be a consequence"
Line 70:
The authors present a well-formulated set of research questions. Very helpful.Lines 74 to 85 and Table 1:
The study used quite a number of different experiments with different conditions for filtering, material added, storage, etc. The list of the six main experiments in the text and in Table 1 provide a reasonable description, but the details perhaps could be amplified. In particular, a better link early in the Methods section to how the experiments address the science questions at the end of Section 1 (Line ~70) would be helpful. This is done well for filtering in Line 88-93 but could be done for other conditions within the experiment introduction section.Line 96
For CO2SYS, a brief description should be added of any choices made in equilibrium coefficients, carbonate saturation coefficients, etc.Line 99 - 109
It may be beneficial for readers who are not seawater carbonate chemists to add a sentence or two on the difference chemically between "equilibrated" (Ia) and "non-equilibrated" (IIa) experiments using NaHCO3/Na2CO3 versus NaOH. While perhaps obvious to chemists, the distinction may not be immediately obvious to the full range of potential readers or why the added alkaline solution is clearly equilibrated with the atmosphere CO2.Line 134
From the text, it appears that the particle experiments incorporated mixing only for 5 minutes at the beginning of the experiment. Some comment is needed on the affect of no mixing over several days in the tubes for particle dissolution kinetics and particle boundary layer dynamics relative to what would happen to particles added to the real ocean. I am not sure the kinetics are actually that comparable to ship disposal as mentioned in line 139 where most schemes assume a turbulent dispersion component during discharge plus natural background turbulence in the water column.Line 141
"This experiment focused on redissolution of alkalinity"Perhaps consider an alternate phrasing:
"This experiment focused on addition of alkalinity due to the redissolution of alkaline precipitates"Line 148-149:
"The experiment was set up to check if potential threshold values for TA loss/gain after 24 hours exist in case larger amounts of solids are dissolved than in the previous experiments, resulting in the loss of TA."Perhaps add clarification "for net TA loss/gain with the combined effects of alkaline material addition plus mineral precipitation ..."
Line 154:
Are the particles captured on the filters for each experiment analyzed for the amount and character of the material? In particular, was anything done to analyze precipitates?Line 169:
which coefficients are used for carbonate saturation state?Line 172-175:
Perhaps worth noting that the notation Delta TA_2400 (or Delta TA_1200) is only approximate as the the actual increases are not exactly 2400 or 1200 but are close, for example 2400 is actually 4750 - 2411 = 2339.Line 178-179:
"Neither in the abiotic (0.2 micrometer filtered) nor in the biotic (55 micrometer filtered) treatment, including living planktonic organisms, a detectable loss in TA could be observed after 1 and 4 days."Perhaps reorganize this sentence as:
"No decidable loss in TA was observed after 1 and 4 days following the initial TA addition at time 0 for either the abiotic (0.2 micrometer filtered) or in the biotic (55 micrometer filtered) treatment, including living planktonic organisms."Also, this sentence helps better understand some of the earlier text in the Abstract and the Methods regarding the meaning of "alkalinity loss". Perhaps a sentence or two could be added along the lines of:
"The seawater TA was monitored after the initial alkalinity addition to determine if any alkalinity was lost subsequently do to mineral precipitation processes."Line 180
"while carbonate chemistry remained stable in all TA treatments [after the initial alkalinity addition]."From Figure 1 there does appear to be a noticeable, if relatively small, decline in pH and aragonite saturation state in the abiotic-equilibrated experiments. This should be mentioned briefly in the text.
Line 185, figure 1
The change in seawater carbonate system in the experimental treatments is over-determined with the measurements of 3 variables of the seawater carbonate system (pH, alkalinity, and DIC). Would be good to clarify for plots what is measured versus what is calculated from CO2SYS (only saturation state?) and if the measurement and CO2SYS calculated values are consistent.Minor point but in all of the captions it would be good to add the number and name of the experimental treatment to connect back to the notation on Table 1. So add Ia and Ib in the bolded text in the caption.
Line 221-226:
"Consequently, while precipitation was absent in the abiotic and biotic set-up in the equilibrated treatment, strong alkalinity loss through precipitation occurred in the non-equilibrated experiment ... The loss ratio in TA:DIC was about 2:1 (Fig. 3b, d) in all treatments, indicating the loss of alkalinity due to the precipitation of carbonates."This is a good line and perhaps something like this would be helpful in the Abstract.
Line 250, Figure 4:
"development of alkalinity loss (Delta TA) over time"
Here is a third different usage of "Delta TA", here the "loss" due to mineral precipitation.Line 330, Figure 8
A point about notation. Here Delta TA label on y-axis of both sub-panels indicates the net change in alkalinity rather than the use of the notation Delta TA in some previous figures for TA initially added via solution or solids. Perhaps change label to "net Delta TA".Line 335
Section 4.1 General discussion
This is a well written summary of the many of the key results.Line 377-378:
"Not only particle surface processes on added solid alkaline particles are to be considered."Perhaps rephrase as, "In addition, particle surface processes ..."
Line 385:
"... partly be released again. Here probably facilitated ..."I think these should be a single sentence (second sentence may be a sentence fragment), so
"... partly be released again, here probably facilitated ..."
partly be released again. Here probably facilitated
Line 408:
"... lead not to a positive TA ..."
perhaps change to
"... did not lead to a positive change in TA ..."-
RC3: 'Reply on RC2', Scott C. Doney, 10 Jul 2022
I noticed an error in my post near the beginning. The base equation should be:
Delta_net TA = TA_final - TA_initial = Delta_add TA + Delta_loss TA
- AC4: 'Reply on RC3', Jens Hartmann, 27 Sep 2022
-
AC2: 'Reply on RC1 and RC2', Jens Hartmann, 25 Sep 2022
The comment was uploaded in the form of a supplement: https://bg.copernicus.org/preprints/bg-2022-126/bg-2022-126-AC2-supplement.pdf
- AC3: 'Reply on RC2', Jens Hartmann, 27 Sep 2022
-
RC3: 'Reply on RC2', Scott C. Doney, 10 Jul 2022
Jens Hartmann et al.
Jens Hartmann et al.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 1,128 | 510 | 23 | 1,661 | 11 | 23 |
- HTML: 1,128
- PDF: 510
- XML: 23
- Total: 1,661
- BibTeX: 11
- EndNote: 23
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1